FRONTIERS OF FUNDAMENTAL PHYSICS
Frontiers of Fundamental Physics Proceedings of the Sixth International Symposium “Frontiers of Fundamental and Computational Physics ”, Udine, Italy, 2629 September 2004
Edited by
B.G. SIDHARTH B.M. Birla Science Centre, Adarsh Nagar, Hyderabad, India
F. HONSELL University of Udine, Italy and
A. DE ANGELIS University of Udine, Italy
A C.I.P. Catalogue record for this book is available from the Library of Congress.
ISBN10 ISBN13 ISBN10 ISBN13
1402041519 (HB) 9781402041518 (HB) 1402043392 (ebook) 9781402043390 (ebook)
Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com
Printed on acidfree paper
All Rights Reserved © 2006 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.
TABLE OF CONTENTS
Introduction
xi
I
1
Field Theory, Relativity and Cosmology
Cosmological Theories of Special and General Relativity  I
3
M. Carmeli Cosmological Theories of Special and General Relativity  II
13
M. Carmeli Carmeli’s Cosmology: the Universe is Spatially Flat Without Dark Matter
21
J.G. Hartnett Black Holes and the Information Paradox
29
G. t’ Hooft A Quantum Approach to Cosmology
31
A. AlfonsoFaus My Focus on the Quantum Source of Gravity
37
S.G. Goradia Vacuum Decay by pbranes Production
45
L. Sindoni, S. Ansoldi Superstrong Interacting Gravitons as a Main Engine of the Universe Without Expansion or Dark Energy
51
M.A. Ivanov Generalized Cosmological Constraints on Neutrino Oscillations Relaxed or Strengthened D.P. Kirilova
55
Attractions of Aﬃne Quantum Gravity
61
J.R. Klauder
vi
Table of Contents
Gravitational Tunnelling of Relativistic Shells S. Ansoldi, L. Sindoni Threedimensional Relativistic Simulations of Rotating Neutronstar Collapse to a Kerr Black Hole.
69
75
L. Baiotti, I. Hawke, P.J. Montero, F. L¨ oﬄer, L. Rezzolla, Ni. Stergioulas, J.A. Font, E. Seidel Gravitational Wave Sources V. Ferrari
83
Model Analysis of Gravitational Shell Collapses M. Seriu
93
The Deuteron and the Big Bang M. Mizushima
103
Astrophysical Applications of the Theory of Scale Relativity
107
L. Nottale Making Maps of the ReesSciama Eﬀect M.J. Fullana, D.P. S´ aez
115
Stationary Points of Scalar Fields Coupled to Gravity
123
H. Kr¨ oger, G. Melkonyan, F. Paradis, S.G. Rubin The Density Matrix Deformation in Quantum and Statistical Mechanics of the Early Universe
131
A.E. ShalytMargolin, V.I. Strazhev
II
Foundations of Physics
135
Dark Energy, Chaotic Fields, and Fundamental Constants C. Beck
137
How Fundamental is Gravitation?
147
B.G. Sidharth Scaledependent Stochastic Quantization M. Altaisky
155
Principia Geometrica Physicae J. Keller
163
Table of Contents
vii
Do Real Numbers Obscure Real Physics? G.N. Ord
175
Cantorian Space in Nature and Dynamical Systems
183
G. Iovane, P. Giordano, S. Salerno Mathematical Structure of Individual Quantum States S.N. Mayburov
191
Space and Time Physics with the Lorentz Ether: the Clock Paradox
195
F. Selleri
III Nuclear and HighEnergy Particle Physics and As trophysics Clusters of Matter and Antimatter  a Mechanism for Cold Compression
209
211
W. Greiner Understanding the Nucleon Spin F. Bradamante
221
The Mass and Spin of the Mesons, Baryons, and Leptons E.L. Koschmieder
231
Equality and Identity and (In)distinguishability in Classical and QuantumMechanics from the Point of View of Newton’s Notion of State
239
P. Enders Numerical Modelling of Quantum Statistics in Highenergy Physics
247
O.V. Utyuzh, G. Wilk, Z. Wlodarczyk Frontiers of High Energy Cosmic Rays M. Pimenta
255
Outlooks on Gamma Ray Astrophysics R. Giannitrapani
263
Will Antihydrogen Light Shine? A. Variola
265
Physics Potential of the ATLAS Experiment M. Cobal, for the ATLAS Collaboration
267
viii
Table of Contents
A Global Optimization Algorithm for Finite Density Quark Matter R. Buﬀa Multiphoton Approach on Pair Production under the Light of Recent Experimental and Theoretical Investigations
273
279
C. Kaberidis, I. Tsohantjis, S. Moustaizis Nielsen Identity, Wilson Line and Constrained Eﬀective Action: the High Temperature Standard Model
285
R. Buﬀa The MAGIC Experiment and its First Results D. Bastieri, R. Bavikadi, C. Bigongiari, E. Bisesi, P. Boinee, A. De Angelis, B. De Lotto, A. Forti, T. Lenisa, F. Longo, O. Mansutti, M. Mariotti, A. Moralejo, D. Pascoli, L. Peruzzo, A. Saggion, P. Sartori, V. Scalzotto and The MAGIC collaboration
291
Neural Networks for GammaHadron Separation in MAGIC P. Boinee, F. Barbarino, A. De Angelis, A. Saggion, M. Zacchello
297
GammaRay Astrophysics with AGILE 303 F. Longo, A. Argan, G. Barbiellini, M. Basset, F. Boﬀelli, A. Bulgarelli, P. Caraveo, P.W. Cattaneo, E. Celesti, A. Chen, V. Cocco, E. Costa, E. Del Monte, G. Di Cocco, G. Di Persio, I. Donnarumma, M. Feroci, M. Fiorini, T. Froysland, M. Frutti, M. Galli,F. Gianotti, A. Giuliani,C. Labanti, I. Lapshov, F. Lazzarotto, F. Liello, P. Lipari, M. Marisaldi, M. Mastropietro, A. Mauri, F. Mauri, S. Mereghetti, E. Morelli, A. Morselli, L. Pacciani, A. Pellizzoni, F. Perotti, P. Picozza, C. Pittori, C. Pontoni, G. Porrovecchio, M. Prest, M. Rapisarda, E. Rossi, A. Rubini, P. Soﬃtta,M. Tavani, A. Traci, M. Trifoglio, E. Vallazza, S. Vercellone, D. Zanello Simulating the High Energy GammaRay Sky Seen by the GLAST Large Area Telescope
309
F. Longo, P. Azzi, D. Bastieri, G. Busetto, Y. Lei, R. Rando, O. Tibolla, L. Baldini, M. Kuss, L. Latronico, N. Omodei, M. Razzano, G. Spandre, P. Boinee, A. De Angelis, M. Frailis, M. Brigida, F. Gargano, N. Giglietto, F. Loparco, M.N. Mazziotta, C. Cecchi, P. Lubrano, F. Marcucci, M. Pepe, G. Tosti, A. Lionetto, A. Morselli, C. Pittori Dark Matter Detection in Gamma Astroparticle Experiments E. Bisesi, M. Mariotti, V. Scalzotto
315
Table of Contents Contribution of Pulsars to the GammaRay Background and their Observationwith the Space Telescopes GLAST and AGILE E. Bisesi
ix
321
Data Mining in Gamma Astrophysics Experiments M. Frailis, A. De Angelis, V. Roberto
327
Grid Services for the MAGIC Experiment A. Forti, S.R. Bavikadi, C. Bigongiari, G. Cabras, A. De Angelis, B. De Lotto, M. Frailis, M. Hardt, H. Kornmayer, M. Kunze, M. Piraccini and the MAGIC Collaboration
333
IV
339
Complex Systems
Towards a Physical Model for Memory L. Cooper Patterns and Dissipative Waves, including Solitons, in Lattices and at Interfaces
341
343
M. G. Velarde Lattice Protein Models: A Computational Approach to Folding and Aggregation Phenomena
355
M. Citossi, G. Giugliarelli Interface Depinning from Wedges with a Central Ridge G. Giugliarelli
359
V
365
New Approaches to Physics Teaching
Learning Problems Related to the Concept of Field F. Bradamante, M. Michelini, A. Stefanel
367
Elastic Waves: Mental Models and Teaching/Learning Sequences
381
G. Tarantino Index of Authors
385
INTRODUCTION
One of the ﬁrst Computer Science sites in Italy, in recent years, the Friuli region has become a very active hub in Computational Physics and other applications of Informatics to Human and Natural Sciences. In particular the University of Udine has developed a tradition in innovative crossdisciplinary research areas involving Computer Science and Physics, providing digital tools for laboratories such as NASA and CERN. The sixth International Symposium “Frontiers of Fundamental and Computational Physics” (FFP6) aimed at providing a platform for a wide range of physicists to meet and share thoughts on the latest trends in various research areas including High Energy Physics, Theoretical Physics, Gravitation and Cosmology, Astrophysics, Condensed Matter Physics, Fluid Mechanics. Such frontier lines were uniﬁed by the use of computers as an, often primary, research instrument, or dealing with issues related to information theory. The present Sixth International Symposium in the series was organized at the University of Udine, Italy from 26th to 29th of September 2004. The Univer sity of Udine and the B.M. Birla Science Centre in Hyderabad have collaborated in the organization of this Symposium and the edition of these Proceedings, under the auspices of their joint initiative the International Institute of Applicable Mathematics and Information Sciences. The contributions in the Proceedings are grouped as follows: • Field Theory, Relativity and Cosmology • Foundations of Physics and of Information Sciences • Nuclear and HighEnergy Particle Physics and Astrophysics; Astroparticle Physics • Complex Systems; Fluid Mechanics • New Approaches to Physics Teaching This Symposium had an attendance of over 100 participants. There were 63 papers/presentations, including 4 introductory invited lectures delivered by the Nobel Laureates L. Cooper and G. ‘t Hooft, and by the eminent physicists Y. Ne’eman and W. Greiner. As was the case for the previous editions, the conference was a great success. Full papers in these Proceeding underwent a refereeing process. When full papers have not been submitted, abstracts have been included. The Editors Udine, March 2005
xii
Introduction
Previous Symposia • 5th International Symposium on Frontiers of Fundamental Physics 06  08 January 2003 B M Birla Science Center  Adarsh Nagar  HYDERABAD 500463 India • 4th International Symposium on Frontiers of Fundamental Physics 09  13 December 2000 B M Birla Science Center  Adarsh Nagar  HYDERABAD 500463 India • 3rd International Symposium on Frontiers of Fundamental Physics B M Birla Science Center  Adarsh Nagar  HYDERABAD 500463 India • 2nd International Symposium on Frontiers of Fundamental Physics 30 December 1998  01 January 1999 B M Birla Science Center  Adarsh Nagar  HYDERABAD 500463 India • 1st International Symposium on Frontiers of Fundamental Physics December 1997 B M Birla Science Center  Adarsh Nagar  HYDERABAD 500463 India
Part I
Field Theory, Relativity and Cosmology
COSMOLOGICAL THEORIES OF SPECIAL AND GENERAL RELATIVITY  I
MOSHE CARMELI a
a
Department of Physics, Ben Gurion University of the Negev, Beer Sheva 84105, Israel
Abstract In the standard cosmological theory one uses the Einstein concepts of space and time as were originally introduced for the special theory of relativity and the general relativity theory. According to this approach all physical quantities are described in terms of the continuum spatial coordinates and time. Using general relativity theory a great progress has been made in understanding the evolution of the Universe. Cosmologists usually measure spatial distances and redshitfs of faraway galaxies as expressed by the Hubble expansion. In recent years this fact was undertaken to develop new theories in terms of distances and velocities (redshift). While in Einstein’s relativity the propagation of light plays the major role, in the new theory it is the expansion of the Universe that takes that role and appears at the outset. The cosmic time becomes crucial in these recent theories, which in the standard theory is considered to be absolute but here it is relative. In this lecture this new approach to cosmology is presented.
1
Introduction
It is well known that both Einstein’s theories are based on the fact that light propagates at a constant velocity. However, the Universe also expands at a constant rate when gravity is negligible. Moreover, cosmologists usually measure spatial distances and redshifts of faraway galaxies as expressed by the Hubble expansion. In recent years this fact was undertaken to develop new theories in terms of distances and velocities (redshift). While in Einstein’s special relativity the propagation of
3 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 3–11. © 2006 Springer. Printed in the Netherlands.
4
Theories of Special and General Relativity  I
light plays the major role, in the new theory it is the expansion of the Universe that takes that role. It is the concept of cosmic time that becomes crucial in these recent theories. In the standard theory the cosmic time is considered to be absolute. Thus we talk about the Big Bang time with respect to us here on Earth as an absolute quantity. Consider, for example, another galaxy that has, let us say, a relative cosmic time with respect to us of 1 billion year. Now one may ask what will be the Big Bang time with respect to this galaxy. Will it be the Big Bang time with respect to us minus 1 billion year? A person who lives in that galaxy will look at our galaxy and say that ours is far away from him by also 1 billion year. Will that mean, with respect to him, our galaxy is closer to the Big Bang time by 1 billion year? Or will we seem to him to be farther by 1 billion year? All this leads to the conclusion that there is no absolute cosmic time. Rather, it is a relative concept. Based on this assumption, we present in this lecture a theory that relates distances between galaxies and their relative velocities. These are actually the dynamical variables that astronomers measure. In the ﬁrst part the theory will be relating distances to velocities with very weak gravitational ﬁeld (special relativity). We then, in the second part, extend the theory to include the gravitational ﬁeld of the Universe. Before doing that, we present a brief review of Einstein’s special relativity theory.
2
Einstein’s special relativity: A review
We will not give a full detail of this very important theory. Rather, we outline some of its fundamentals [1,2] 1. The MichelsonMorley experiment for the constancy of the speed of light. 2. Einstein’s two postulates: (a) Constancy of the speed of light; (b) Validity of the laws of physics in internal coordinate systems. (See Figure 1) 3. The Lorentz transformation. This is given by ct = (ct − βx) / 1 − β 2 , x = (x − βct) / 1 − β 2 , (2.1)
where β = v/c, and y = y, z = z. 4. Minkowski’s uniﬁcation of space and time. 5. Invariance of the laws of physics under the Lorentz transformation. 6. Minkowski’s line element ds2 = c2 dt2 − dx2 + dy 2 + dz 2
(2.2)
7. The light cone (see Figure 2).
3
Cosmological special relativity
We outline this theory very brieﬂy with the following points [3, 4, 5]. 1. Hubble’s law R = τ v, where R is the distance to a galaxy, v is the receding
M. Carmeli
5
Figure 1. Two coordinate systems K and K , one moving with respect to the other with a velocity v in the xdirection.
velocity of the galaxy, τ is universal constant equal to 12.486 Gyr (Big Bang time). 2. Cosmic time is not absolute but a relative concept. Example: Another galaxy is one billion year with respect to us. One may ask, what is the Big Bang time with respect to this galaxy. Will it be BB time with respect to us minus 1 billion year?From the point of view of that galaxy ours is far away from it by also 1 billion year. Does that mean, with respect to that galaxy, ours is closer to the BB time by 1 billion year?Or our galaxy will seem to be farther by 1 billion year? All this leads to the conclusion that there is no absolute cosmic time. Rather, it is a relative concept. 3. Universe with negligible gravity. 4. Line element of an expanding Universe with negligible gravity: ds2 = τ 2 dv 2 − (dx2 + dy 2 + dz 2 ).
(3.1)
It is equal to zero for the Hubble expansion but is not vanishing at cosmic times smaller than τ . Notice the similarity to Minkowskian line element ds2 = c2 dt2 − (dx2 + dy 2 + dz 2 )
(3.2)
6
Theories of Special and General Relativity  I
Figure 2. The light cone in two dimensions, x0 (= ct) and x1 (= x). The propagation of two light signals in opposite directions passing through x = 0 at time t = 0, is represented by the two diagonal straight lines. (Compare the galaxy cone given in the sequel.)
which vanishes for light propagation but is diﬀerent from zero for particles of ﬁnite mass. 5. Postulates of cosmological special relativity: (1) The laws of physics are valid at all cosmic times; (2) τ has the same value at all cosmic times. (Similarly to Einstein’s special relativity postulates.) 6. The cosmological transformation. This is the analog to the Lorentz transformation. It relates physical quantities at diﬀerent cosmic times (similarly to the Lorentz transformation that relates quantities at diﬀerent velocities): x − (τ − t) v x = , (t/τ ) (2 − t/τ )
y = y,
τ v − x (τ − t) /τ τ v = , (t/τ ) (2 − t/τ )
(3.3)
z = z,
0 ≤ t ≤ τ , t = 0 at the Big Bang, t = τ , now. 7. Example: Denote the temperature of the Universe at a cosmic time t by T and
M. Carmeli
7
Figure 3. The galaxy cone in cosmological relativity, describing the cone in the x − v space satisfying τ 2 v 2 − x2 = 0, where x represents the threedimensional space. The heavy dots describe galaxies. The galaxy cone represents the locations of the galaxies at a given time rather than their path of motion in the real space.
that at present by T0 (=2.73K), we then have T = T0 / (t/τ ) (2 − t/τ ).
(3.4)
At t/τ = 1/2 we get T = 3.15K. (This result assumes negligible gravity and needs a correction by a factor of 13, and thus the temperature at t/τ = 1/2 is 41K.) 8. The galaxy cone. This is the analog to the light cone in Einstein’s special relativity. It represents the expansion of the Universe with negligible gravity. It is a fourdimensional cone. The coordinates are the velocity and the other three coordinates are the space coordinates. At the cone surface the Hubble expansion is represented, whereas the inner part of the cone represents events at cosmic times less than τ . The similarity to the light cone is remarkable. 9. Inﬂation at the early Universe. From the cosmological transformation we obtain the relationship between the mass density ρ0 now to its value ρ at a backward time t: ρ = ρ0 / 1 − t2 /τ 2 . (3.5)
8
Theories of Special and General Relativity  I
The volume of the Universe is inversely proportional to its density, hence the ratio of the volumes at two backward cosmic times t1 and t2 is V2 /V1 = (1 − t22 /τ 2 ) / (1 − t21 /τ 2 ). (3.6)
For t1 , t2 very close to τ we have V2 /V1 =
t2 /t1 ,
(3.7)
where primes indicate times with respect to the Big Bang. For t2 − t1 ≈ 10−32 s, and t2 much less than 1s, we have V2 /V1 = 10−16 / t1 . (3.8)
For example, at t1 ≈ 10−132 s, we obtain V2 /V1 ≈ 1050 describing inﬂation. The above introduction gives a brief review of a new special relativity (cosmological special relativity, for more details see [5]). Obviously the Universe is ﬁlled up with gravity and therefore one has to go to a Riemannian space with the Einstein gravitational ﬁeld equations in terms of space and redshift (velocity). This is done in the second part of these lectures. Before that we outline Einstein’s general relativity theory.
4
General relativity theory: A brief outline
1. Postulates: (a) Principle of general covariance; (b) Principle of equivalence. 2. Riemannian curved space for the gravitational ﬁeld. The metric tensor gµν as the gravitational potential [6]. 3. The Einstein ﬁeld equations: 1 Gµν = Rµν − gµν R = κTµν 2
(4.1)
4. The geodesic equation as the equation of motion: α β d2 xρ ρ dx dx + Γ = 0. αβ ds2 ds ds
5
(4.2)
Extension to curved space
The theory presented here, cosmological general relativity, uses a Riemannian fourdimensional presentation of gravitation in which the coordinates are those of Hubble, i.e. distances and velocity rather than the traditional space and time. We solve the ﬁeld equations and show that there are three possibilities for the
M. Carmeli
9
Universe to expand. The theory describes the Universe as having a threephase evolution with a decelerating expansion, followed by a constant and an accelerating expansion, and it predicts that the Universe is now in the latter phase. It is shown, assuming Ωm = 0.245, that the time at which the Universe goes over from a decelerating to an accelerating expansion, i.e., the constantexpansion phase, occurs at 8.5 Gyr ago. Also, at that time the cosmic radiation temperature was 146K. Recent observations of distant supernovae imply that the Universe’s growth is accelerating, contrary to what has always been assumed, that the expansion is slowing down due to gravity. Our theory conﬁrms these recent experimental results by showing that the Universe now is deﬁnitely in a stage of accelerating expansion. The theory predicts also that now there is a positive pressure, p = 0.034g/cm2 , in the Universe. It is worthwhile mentioning that the theory has no cosmological constant. It is also shown that the threedimensional space of the Universe is Euclidean, as the Boomerang, Maxima, DASI and CBI microwave telescopes have shown. Comparison with general relativity theory is ﬁnally made and it is pointed out that the classical experiments as well as the gravitational radiation prediction follow from the present theory, too.
6
Cosmology in spacevelocity
In the framework of cosmological general relativity (CGR) gravitation is described by a curved fourdimensional Riemannian spacevelocity. CGR incorporates the BB constant τ at the outset. The Hubble law is assumed in CGR as a fundamental law. CGR, in essence, extends Hubble’s law so as to incorporate gravitation in it; it is actually a distribution theory that relates distances and velocities between galaxies. The theory involves measured quantities and it takes a picture of the Universe as it is at any moment. The following is a brief review of CGR as was originally given by the author in 1996 [7]. The foundations of any gravitational theory are based on the principles of equivalence and general covariance. These two principles lead immediately to the realization that gravitation should be described by a fourdimensional curved spacetime, in our theory spacevelocity, and that the ﬁeld equations and the equations of motion should be generally covariant. Hence these principles were adopted in CGR also. Use is made in a fourdimensional Riemannian manifold with a metric gµν and a line element ds2 = gµν dxµ dxν . The diﬀerence from Einstein’s general relativity is that our coordinates are: x0 is a velocitylike coordinate (rather than a timelike coordinate), thus x0 = τ v where τ is the Big Bang time and v the velocity. The coordinate x0 = τ v is the comparable to x0 = ct in ordinary general relativity. The other three coordinates xk , k = 1, 2, 3, are spacelike, just as in general relativity theory. An immediate consequence of the above choice of coordinates is that the null condition ds = 0 describes the expansion of the Universe in the curved spacevelocity (generalized Hubble’s law with gravitation) as compared to the propagation of light in the curved spacetime in general relativity. This means one solves the
10
Theories of Special and General Relativity  I
ﬁeld equations (to be given in the sequel) for the metric tensor, then from the null condition ds = 0 one obtains immedialety the dependence of the relative distances between the galaxies on their relative velocities. As usual in gravitational theories, one equates geometry to physics. The ﬁrst is expressed by means of the Einstein tensor. The physical part is expressed by the energymomentum tensor which now has a diﬀerent physical meaning from that in Einstein’s theory. More important, the coupling constant that relates geometry to physics is now also diﬀerent. Accordingly the ﬁeld equations are 1 Rµν − gµν R = κTµν , 2
(6.1)
exactly as in Einstein’s theory, with κ given by κ = 8πk/τ 4 , (in general relativity it is given by 8πG/c4 ), where k is given by k = Gτ 2 /c2 , with G being Newton’s gravitational constant, and τ the Big Bang constant time. When the equations of motion are written in terms of velocity instead of time, the constant k will replace G. Using the above equations one then has κ = 8πG/c2 τ 2 . The energymomentum tensor T µν is constructed, along the lines of general relativity theory, with the speed of light being replaced by τ . If ρ is the average mass density of the Universe, then it will be assumed that T µν = ρuµ uν , where uµ = dxµ /ds is the fourvelocity. In general relativity theory one takes T00 = ρ. In Newtonian gravity one has the Poisson equation ∇2 φ = 4πGρ. At points where ρ = 0 one solves the vacuum Einstein ﬁeld equations in general relativity and the Laplace equation ∇2 φ = 0 in Newtonian gravity. In both theories a null (zero) solution is allowed as a trivial case. In cosmology, however, there exists no situation at which ρ can be zero because the Universe is ﬁlled with matter. In order to be able to have zero on the righthand side of (6.1) one takes T00 not as equal to ρ, but to ρef f = ρ − ρc , where ρc is the critical mass density, a constant in CGR given by ρc = 3/8πGτ 2 , whose value is ρc ≈ 10−29 g/cm3 , a few hydrogen atoms per cubic meter. Accordingly one takes T µν = ρef f uµ uν ; ρef f = ρ − ρc for the energymomentum tensor. Moreover, the above choice of the energymomentum tensor is the only possibility that yields a constant expansion when ρ = ρc as it should be. In Part II we apply CGR to obtain the accelerating expanding Universe and related subjects.
References [1] A. Einstein, Autobiographical Notes, P.A. Schlipp, Ed., Open Court (1979) [2] A. Einstein, The Meaning of Relativity, 5th Edition, Princeton Univ. Press, Princeton (1955) [3] M. Carmeli, Found. Phys. 25, 1029 (1995) [4] M. Carmeli, Found. Phys. 26, 413 (1996) [5] M. Carmeli, Cosmological Special Relativity, 2nd Edition, World Scientiﬁc (2002)
M. Carmeli
11
[6] M. Carmeli, Classical Fields: General Relativity and Gauge Theory, Wiley, New York (1982) [reprinted by World Scientiﬁc (2001)] [7] M. Carmeli, Commun. Theor. Phys. 5, 159 (1996)
COSMOLOGICAL THEORIES OF SPECIAL AND GENERAL RELATIVITY  II
MOSHE CARMELI a
a
Department of Physics, Ben Gurion University of the Negev, Beer Sheva 84105, Israel
Abstract Astronomers measure distances to faraway galaxies and their velocities. They do that in order to determine the expansion rate of the Universe. In Part I of these lectures the foundations of the theory of the expansion of the Universe was given. In this part we present the theory. A formula for the distance of the galaxy in terms of its velocity is given. It is √ very simple: r(v) = cτ /β sinh βv/c, where τ is the Big Bang time, β = 1 − Ωm , and Ωm is the mass density of the Universe. For Ωm < 1 this formula clearly indicates that the Universe is expanding with acceleration, as experiments clearly show.
1
Gravitational ﬁeld equations
In the fourdimensional spacevelocity the spherically symmetric metric is given by (1.1) ds2 = τ 2 dv 2 − eµ dr2 − R2 dθ2 + sin2 θdφ2 , where µ and R are functions of v and r alone, and comoving coordinates xµ = (x0 , x1 , x2 , x3 ) = (τ v, r, θ, φ) have been used. With the above choice of coordinates, the zerocomponent of the geodesic equation becomes an identity, and since r, θ and φ are constants along the geodesics, one has dx0 = ds and therefore uα = uα = (1, 0, 0, 0) . The metric (1.1) shows that the area of the sphere r = constant is given by 4πR2 and that R should satisfy R = ∂R/∂r > 0. The possibility that R = 0 at a point r0 is excluded since it would allow the lines r = constants at the
13 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 13–19. © 2006 Springer. Printed in the Netherlands.
14
Theories of Special and General Relativity  II
neighboring points r0 and r0 + dr to coincide at r0 , thus creating a caustic surface at which the comoving coordinates break down. As has been shown in Part I the Universe expands by the null condition ds = 0, and if the expansion is spherically symmetric one has dθ = dφ = 0. The metric (1.1) then yields τ 2 dv 2 − eµ dr2 = 0, thus dr = τ e−µ/2 . dv
(1.2)
This is the diﬀerential equation that determines the Universe expansion. In the following we solve the gravitational ﬁeld equations in order to ﬁnd out the function µ (r.v). The gravitational ﬁeld equations, written in the form Rµν = κ (Tµν − gµν T /2) ,
(1.3)
Tµν = ρef f uµ uν + p (uµ uν − gµν ) ,
(1.4)
where with ρef f = ρ − ρc and T = Tµν g µν , are now solved. One ﬁnds that the only nonvanishing components of Tµν are T00 = τ 2 ρef f , T11 = c−1 τ peµ , T22 = c−1 τ pR2 and T33 = c−1 τ pR2 sin2 θ, and that T = τ 2 ρef f − 3c−1 τ p. One obtains three independent ﬁeld equations (dot and prime denote derivatives with v and r) ¨ + R˙ 2 + 1 − R2 = −κτ c−1 eµ R2 p, eµ 2RR (1.5) 2R˙ − R µ˙ = 0,
2 2 1 R 2 1 ˙ R˙ 1 −µ e Rµ − − R + Rµ˙ + + 2 = κτ 2 ρef f . R R R R R R
2
(1.6)
(1.7)
Solution of the ﬁeld equations
The solution of (1.6) satisfying the condition R > 0 is given by eµ = R2 / (1 + f (r)) ,
(2.1)
where f (r) is an arbitrary function of the coordinate r and satisﬁes the condition f (r) + 1 > 0. Substituting (2.1) in the other two ﬁeld equations (1.5) and (1.7) then gives ¨ + R˙ 2 − f = −κc−1 τ R2 p, 2RR (2.2) 1 1 2R˙ R˙ − f + 2 R˙ 2 − f = κτ 2 ρef f , (2.3) RR R
15
M. Carmeli
respectively. The simplest solution of the above two equations, which satisﬁes the condition R = 1 > 0, is given by R = r. Using this in Eqs. (2.2) and (2.3) gives f (r) = κc−1 τ pr2 , and f + f /r = −κτ 2 ρef f r, respectively. Using the values of κ = 8πG/c2 τ 2 and ρc = 3/8πGτ 2 , we obtain f (r) = (1 − Ωm ) r2 /c2 τ 2 ,
(2.4)
where Ωm = ρ/ρc . We also obtain 1 − Ωm c 1 − Ωm = = 4.544 (1 − Ωm ) × 10−2 g/cm2 , 3 κcτ τ 8πG
(2.5)
e−µ = 1 + f (r) = 1 + τ c−1 κpr2 = 1 + (1 − Ωm ) r2 /c2 τ 2 .
(2.6)
p=
Accordingly, the line element of the Universe is given by ds2 = τ 2 dv 2 −
dr2 − r2 dθ2 + sin2 θdφ2 , 1 + (1 − Ω) r2 /c2 τ 2
(2.7)
dr2 − r2 dθ2 + sin2 θdφ2 . 2 1 + (κτ /c) pr
(2.8)
or, ds2 = τ 2 dv 2 −
This line element is the comparable to the FRW line element in the standard theory. It will be recalled that the Universe expansion is determined by Eq. (1.2), dr/dv = τ e−µ/2 . The only thing that is left to be determined is the sign of (1 − Ωm ) or the pressure p. Thus we have
dr 1 − Ωm −1 2 = τ 1 + κτ c pr = τ 1 + 2 2 r2 . (2.9) dv c τ
3
Physical meaning
For Ωm > 1 one obtains r (v) =
cτ v sin α , α c
α=
Ωm − 1.
(3.1)
This is obviously a closed Universe, and presents a decelerating expansion. For Ωm < 1 one obtains cτ v r (v) = sinh β , β = 1 − Ωm . (3.2) β c
This is now an open accelerating Universe. For Ωm = 1 we have, of course, r = τ v.
16
Theories of Special and General Relativity  II
Figure 1. Hubble’s diagram describing the threephase evolution of the Universe according to cosmological general relativity theory.
4
The accelerating Universe
From the above one can write the expansion of the Universe in the standard Hubble form v = H0 r with (4.1) H0 = h 1 − (1 − Ωm ) v 2 /6c2 , where h = τ −1 . Thus H0 depends on the distance it is being measured [12]. It is wellknown that the farther the distance, the lower the value for H0 is measured. This is possible only for Ωm < 1, i.e. when the Universe is accelerating. In that case the pressure is positive. Figure 1 describes the Hubble diagram of the above solutions for the three types of expansion for values of Ωm from 100 to 0.245. The ﬁgure describes the threephase evolution of the Universe. Curves (1)(5) represent the stages of decelerating expansion according to Eq. (3.1). As the density of matter ρ decreases, the Universe goes over from the lower curves to the upper ones, but it does not have enough time to close up to a Big Crunch. The Universe subsequently goes over to curve (6) with Ωm = 1, at which time it has a constant expansion for a fraction of a second. This then followed by going to the upper curves (7) and (8) with Ωm < 1, where the Universe expands with acceleration according to Eq. (3.2). Curve no. 8 ﬁts the present situation of the Universe. For curves (1)(4) in the diagram we use the cutoﬀ when the curves were at their maximum.
M. Carmeli 5
17
Theory versus experiment
To ﬁnd out the numerical value of τ we use the relationship between h = τ −1 and H0 given by Eq. (4.1)(CR denote values according to Cosmological Relativity): 2 H0 = h 1 − 1 − ΩCR (5.1) z /6 , m 2 where z = v/c is the redshift and ΩCR m = ρm /ρc with ρc = 3h /8πG. (Notice that our ρc = 1.194 × 10−29 g/cm3 is diﬀerent from the standard ρc deﬁned with H0 .) The redshift parameter z determines the distance at which H0 is measured. We choose z = 1 and take for ΩCR m = 0.245, its value at the present time (corresponds to 0.32 in the standard theory), Eq. (5.1) then gives H0 = 0.874h. At z = 1 the corresponding Hubble parameter H0 according to the latest results from HST can be taken [20] as H0 = 70km/sMpc, thus h = (70/0.874)km/sMpc, or h = 80.092km/sMpc, and τ = 12.486Gyr = 3.938 × 1017 s. CR ST ST What is left is to ﬁnd the value of ΩCR = Λ . We have ΩΛ = ρc /ρc , where ρc 2 2 CR 3H02 /8πG and ρc = 3h2 /8πG. Thus ΩCR = (H /h) = 0.874 , or Ω = 0.764. 0 Λ Λ As is seen from the above equations one has CR ΩT = ΩCR m + ΩΛ = 0.245 + 0.764 = 1.009 ≈ 1,
(5.2)
which means the Universe is Euclidean. Our results conﬁrm those of the supernovae experiments and indicate on the existance of the dark energy as has recently received conﬁrmation from the Boomerang cosmic microwave background experiment [21, 22], which showed that the Universe is Euclidean.
6
Comparison with general relativity
One has to add the time coordinate and the result is a ﬁvedimensional theory of spacetimevelocity. One can show that all the classical experiments predicted by general relativity are also predicted by CGR. Also predicted a wave equation for gravitational radiation. In the linear approximation one obtains 1 ∂2 1 ∂2 2 − ∇ + γµν = −2κTµν , (6.1) c2 ∂t2 τ 2 ∂v 2
where γµν is a ﬁrst approximation term, gµν ≈ ηµν + hµν = ηµν + γµν − ηµν γ/2,
(6.2)
γ = η αβ γαβ .
(6.3)
Hence CGR predicts that gravitational waves depend not only on space and time but also on the redshift of the emitting source.
18
Theories of Special and General Relativity  II
Table 1. Cosmological parameters in cosmological general relativity and in standard theory.
Theory type Expansion type Present expansion Pressure Cosmological constant
ΩT = Ω m + Ω Λ Constantexpansion occurs at Constantexpansion duration Temperature at constant expansion
7
COSMOLOGICAL RELATIVITY
STANDARD THEORY
Spacevelocity Triphase: decelerating, constant, accelerating Accelerating (predicted) 0.034g/cm2 1.934 × 10−35 s−2 (predicted) 1.009 8.5Gyr ago (Gravity is included) Fraction of second 146K (Gravity is included)
Spacetime One phase
One of three possibilities Negative Depends
Depends No prediction Not known No prediction
New developments on dark matter
Using the theory presented here, John Hartnett has recently shown that there is no need for the existence of dark matter in spiral galaxies. We only give the references to this work by Hartnett [25, 26, 27]. References [1] M. Carmeli, Cosmological Special Relativity, 2nd Edition, World Scientiﬁc (2002) [2] M. Carmeli and T. Kuzmenko, ”Value of the cosmological constant”, astroph/0110590 [3] M. Carmeli, Proc. 20th Texas Symp., AIP Conf. Proc., 586 (2001), astroph/012033 [4] M. Carmeli, ”Accelerating Universe: Theory vs experiment”, astroph/0205396 [5] M. Carmeli, ”Fivedimensional cosmological theory of uniﬁed space, time and velocity”, Nuclear Phys B124, 258 (2003); asrtoph/0205113 [6] M. Carmeli, ”The line element in the Hubble expansion”, In: Gravitation and cosmology, A. Lobo et al. Eds., Universitat de Barcelona, 2003, pp. 113130 (Proceedings of the Spanish Relativity Meeting, Menorca, Spain, 2224 September 2002); astroph/0211043 [7] M. Carmeli, Commun. Theor. Phys. 5, 159 (1996) [8] M. Carmeli, Classical Fields: General Relativity and Gauge Theory, Wiley, New York (1982) [reprinted by World Scientiﬁc (2001)]
M. Carmeli
19
[9] S. Behar and M. Carmeli, Intern. J. Theor. Phys. 39, 1375 (2000), astroph/0008352 [10] M. Carmeli and S. Behar, ”Cosmological general relativity”, pp. 526, in: Quest for Mathematical Physics, T.M. Karade et al. Eds., New Delhi (2000) [11] M. Carmeli and S. Behar, ”Cosmological relativity: a general relativistic theory for the accelerating Universe”, Talk given at Dark Matter 2000, Los Angeles, February 2000, pp.182– 191, in: Sources and Detection of Dark Matter/Energy in the Universe, D. Cline, Ed., Springer (2001) [12] P.J.E. Peebles, Status of the big bang cosmology, in: Texas/Pascos 92: Relativistic Astrophysics and Particle Cosmology, C.W. Akerlof and M.A. Srednicki, Editors, p. 84, New York Academy of Sciences, New York (1993) [13] P.M. Garnavich et al., Astrophys. J. 493, L53 (1998) [HiZ Supernova Team Collaboration (astroph/9710123)] [14] B.P. Schmidt et al., Astrophys. J. 507, 46 (1998) [HiZ Supernova Team Collaboration (astroph/9805200)] [15] A.G. Riess et al., Astronom. J. 116, 1009 (1998) [HiZ Supernova Team Collaboration (astroph/9805201)] [16] P.M. Garnavich et al., Astrophys. J. 509, 74 (1998) [HiZ Supernova Team Collaboration (astroph/9806396)] [17] S. Perlmutter et al., Astrophys. J. 483, 565 (1997) [Supernova Cosmology Project Collaboration (astroph/9608192)]. [18] S. Perlmutter et al., Nature 391, 51 (1998) [Supernova Cosmology Project Collaboration (astroph/9712212)] [19] S. Perlmutter et al., Astrophys. J. 517, 565 (1999) [Supernova Cosmology Project Collaboration (astroph/9812133)] [20] W.L. Freedman et al., ”Final results from the Hubble Space Telescope”, Talk given at the 20th Texas Symposium on Relativistic Astrophysics, Austin, Texas 1015 December 2000, (astroph/0012376) [21] P. de Bernardis et al., Nature 404, 955 (2000), (astroph/0004404) [22] J.R. Bond et al., in Proc. IAU Symposium 201 (2000), (astroph/0011378) [23] A.V. Filippenko and A.G. Riess, p.227 in: Particle Physics and Cosmology: Second Tropical Workshop, J.F. Nieves, Editor, AIP, New York (2000) [24] A.G. Riess et al., Astrophys. J., in press, astroph/0104455 [25] J. Hartnett, ”Carmeli’s accelerating Universe is spatially ﬂat without dark matter”, Intern. J. Theor. Phys., in press (accepted for publication) (2004), grqc/0407082 [26] J. Hartnett, ”Can the Carmeli metric correctly describe spiral galaxy rotation curves?” 2004, grqc/0407083 [27] J. Hartnett, ”Carmeli’s cosmology: The Universe is spatially ﬂat without dark matter”, invited talk given at the international conference ”Frontiers of Fundamental Physics 6”, held in Udine, Italy, September 2629, 2004
CARMELI’S COSMOLOGY: THE UNIVERSE IS SPATIALLY FLAT WITHOUT DARK MATTER
JOHN G. HARTNETT a
a
School of Physics, University of Western Australia, 35 Strirling Hwy, Crawley, WA 6009 Australia
Abstract Carmeli’s 5D brane cosmology has been applied to the expanding accelerating universe and it has been found that the distance redshift relation will ﬁt the data of the highz supernova teams without the need for dark matter. Also the vacuum energy contribution to gravity, ΩΛ indicates that the universe is asymptotically expanding towards a spatially ﬂat state, where the total mass energy density Ω + ΩΛ → 1.
1
Introduction
The Carmeli cosmology [3] [1] is as revolutionary in its implementation as it is in its interpretation. The metric used by Carmeli is unique in that it extends the number of dimensions of the universe by either one dimension if we consider only the radial velocity of the galaxies in the Hubble ﬂow or by three if we consider all three velocity components. We will conﬁne the discussion in this paper to only one extra dimension as does Carmeli. In that case the line element in ﬁve dimensions becomes ds2 = (1 + Φ)c2 dt2 − dr2 + (1 + Ψ)τ 2 dv 2 (1) where dr2 = (dx1 )2 + (dx2 )2 + (dx3 )2 and Φ are Ψ potential functions to be determined. The time t is measured in the observer’s frame. The new dimension (v) is the radial velocity of the galaxies in the expanding universe, in accordance with Hubble ﬂow. The parameter τ is a constant at any epoch and its reciprocal (designated h) is approximately the Hubble constant H0 .
21 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 21–28. © 2006 Springer. Printed in the Netherlands.
22
Carmeli’s Cosmology: universe is flat
The line element represents a spherically symmetric isotropic universe, and the expansion is the result of spacetime expansion. The expansion is observed at a deﬁnite time and thus dt = 0. Taking into account dθ = dφ = 0 (isotropy condition) and equation (1) becomes −dr2 + (1 + Ψ)τ 2 dv 2 = 0
2
(2)
Phase space equation
The solution of equation (2)(given by equation B.38 and solved in section B.10 in [3] is reproduced here. dr =τ dv
1 + (1 − Ω)
r2
(3)
c2 τ 2
The parameter Ω is the mass/energy density of the universe expressed as a fraction of the critical or “closure” density, i.e. Ω = ρm /ρc where ρm is the averaged matter/energy density of the universe. In this model, ρc =
3 = 10−29 gcm−3 8πGτ 2
Then (3) may be integrated exactly to get r(v) = √
v√ cτ sinh 1−Ω c 1−Ω
∀Ω
(4)
Carmeli has expanded (4) in the limit of small z = v/c and small Ω to get v2 r = τ v 1 + (1 − Ω) 2 6c
⇒
r z2 = z 1 + (1 − Ω) cτ 6
∀Ω < 1, z < 1
(5)
(6)
Thus we can write the expansion in terms of normalized or natural units r/cτ . Equation (6) is plotted in ﬁg. 1 for various values of Ω = 1, 0.24 and 0.03. Let us now rewrite (4) in terms of natural units and for z small but arbitrary Ω. √ r sinh(z 1 − Ω) √ = cτ 1−Ω
(7)
Equation (7) produces curves almost indistinguishable from (6) so this veriﬁes that the approximations work well for z < 1.
J.G. Hartnett
23
Figure 1. Plot of (6), r/cτ vs redshift (z) for Ω = 1 (broken line), Ω = 0.245 (solid black line) and Ω = 0.03 (solid grey line).
3
Density verses redshift
Now let us consider what happens to the density of matter as we look back in the cosmos with redshift, z. It was assumed in ﬁg. 1 that the value of Ω is ﬁxed for each curve. Carmeli does this also in ﬁgure A4, page 134 in ref [3]. However, more correctly Ω varies as a function of z. For ﬂat space we assume the following relation to hold, ρm Ω = (1 + z)3 = (8) ρ0 Ω0
where ρm is a function of the redshift z, and ρ0 is the averaged matter density of the universe locally. The parameter Ω0 is then the local averaged matter density expressed as a fraction of “closure” density. Equation (8) results from the fact that as the redshift increases the volume decreases as (1 + z)3 . Notice at z = 1 that the universe is 8 times smaller in volume and therefore 8 times more dense, that is, at z = 1, Ω = 8Ω0 . Substituting (8) into (7) we get sinh z 1 − Ω0 (1 + z)3 r = (9) cτ 1 − Ω0 (1 + z)3
Carmeli was able to simulate the form of the 0.1 < z < 1 redshift data of [6] published in 1998 which announced an accelerating universe following the observations of [4] and [5]. See ﬁgure A4, page 134 in [3]. But in fact he had predicted this
24
Carmeli’s Cosmology: universe is flat
Figure 2. Plot of (7) with Ω = 1 (broken line) and Ω = 0.245 (solid black line) and (9) with Ω0 = 0.03 (solid grey line). Note: the top two curves lay on top of each other .
in 1996 [2]. So this means that Carmeli assumed a value of total matter (normal + dark matter) density Ω = 0.245, which was the accepted value in 1998. Now let’s plot (7) with Ω = 0.245 and (9) with Ω0 = 0.03. See ﬁg. 2. This means that my modiﬁed equation (9) with Ω0 = 0.03 gives the same result as Carmeli’s unapproximated equation (7) with his assumed value of Ω = 0.245, but this includes dark matter. In fact, comparing (7) and (9), a local matter density of only Ω0 = 0.03–0.04 is necessary to have agreement. This eﬀectively eliminates the need for the existence of dark matter on the cosmic scale. Table I shows the critical data from the comparison at redshifts between z = 0.25 and z = 1. It can be seen that the diﬀerence between the two equations over the domain of the measurements is much less signiﬁcant than the ﬁt to the data. If we assume Ω0 = 0.04 instead of Ω0 = 0.03, since both are within measured parameters, we get closer agreement at smaller redshifts but worse near z = 1. Table 1. Comparison of equations (7) and (9)
Redshift z r/cτ from (7) with Ω = 0.245 r/cτ from (9) with Ω0 = 0.03 % diﬀerence with Ω0 = 0.03 % diﬀerence with Ω0 = 0.04
0.25 0.251984 0.252459 0.19 0.17
0.5 0.515984 0.518935 0.57 0.43
0.75 0.804591 0.810416 0.72 0.23
1.0 1.13157 1.13157 0.00 1.28
In any case (7) and (9) really must be modiﬁed as z → 1 to allow for relativistic
25
J.G. Hartnett
Figure 3. Plot of (10) with Ω = 0.245 (solid curve) and (11) with Ω0 = 0.03 (broken curve). Note: the two curves separate for z > 1.2.
eﬀects, by replacing v/c with the relativistic form v/c = ((1+z)2 −1)/((1+z)2 +1). Therefore we can rewrite (7) and (9) respectively as r 1 (1 + z)2 − 1 √ √ = sinh 1−Ω (10) cτ (1 + z)2 + 1 1−Ω
and r 1 = sinh cτ 1 − Ω0 (1 + z)3
(1 + z)2 − 1 1 − Ω0 (1 + z)3 (1 + z)2 + 1
(11)
where the varying matter density has been taken into account. In ﬁg. 3, (10) and (11) are compared. The density approximation may be no longer valid past z = 1, because it is shown below that the vacuum energy term dominates and the universe is far from ﬂat.
4
Hubble parameter
Based on the above analysis we can rewrite equation A.54 from [3] for H0 as z2 H0 = h 1 − (1 − Ω0 (1 + z)3 ) ∀Ω < 1, z < 1 (12) 6
which according to equation A.51 from [3] may be further generalized without approximation, and using the relativistic form of the redshift. Still that equation
26
Carmeli’s Cosmology: universe is flat
Figure 4. Plot of (12) (solid curve) and (13) (broken curve). Note: the two curves intersect at z = 1 where H0 = 70kms−1 M pc−1 .
may only be approximate for z > 1 because of the density assumptions. However it becomes ξ H0 = h (13) sinh ξ 2 −1 where ξ = (1+z) 1 − Ω0 (1 + z)3 . (1+z)2 +1 Both (12) and (13) have been plotted in ﬁg. 4, and for Carmeli’s chosen value of H0 ≈ 70kms−1 M pc−1 at z = 1 in (12) yields h ≈ 80.2kms−1 M pc−1 (very close to Carmeli’s value) but (13) yields h ≈ 73.27kms−1 M pc−1 . This means without the small z approximation the value of h is reduced when compared to that in [3].
5
Dark energy
The vacuum or dark energy parameter ΩΛ does not appear explicitly in Carmeli’s model. It is only by a comparison with FL models can an assignment be made. On page 138 of [3] by comparing with the standard model it is shown that ΩΛ = (H0 /h)2 , therefore we can write 2 ξ ΩΛ = . (14) sinh ξ From (14) it is expected that using the unapproximated equation (13) for H0
J.G. Hartnett
27
Figure 5. Plot of ΩΛ (broken curve) and total density Ω + ΩΛ (solid curve) as a function of redshift z. Notice that ΩΛ tends to unity as z tends to zero and the total density tends to the local matter density Ω0 plus the vacuum energy density ΩΛ .
the value of ΩΛ will be larger than Carmeli’s value using the form of (6). Fig. 5 shows the values for the vacuum energy density ΩΛ (broken curve) and for the total energy density Ω + ΩΛ (solid curve) as a function of redshift, z. From (14) it follows that as the universe expands the total density tends to the vacuum energy density ΩΛ → 1 (since Ω0 → 0). This means a totally 3D spatially ﬂat universe in a totally relaxed state. For small z the total density becomes Ω + ΩΛ ≈ (1 + Ω0 ) + 3zΩ0 .
(15)
It follows from (15) that for Ω0 = 0.03 at z = 0 the total density Ω+ΩΛ ≈ 1.03. This value is consistent with Carmeli’s result of 1.009. However, it follows from (8) and (15) that the universe will always be open, Ω < 1 as it expands. From ﬁg. 5 the total density Ω + ΩΛ is always greater than unity and as the universe expands, it asymptotically approaches unity—therefore a spatially ﬂat universe devoid of dark matter.
6
Conclusion
The 5D brane world of Moshe Carmeli has been has been applied to the expanding accelerating universe and the redshift distance relation has been generalised for redshifts up to at least z = 1.2. It has been found that if a certain form of the
28
Carmeli’s Cosmology: universe is flat
dependence of baryonic matter density on redshift is assumed then the resulting distanceredshift relation will ﬁt the data of the highz supernova teams without the need for dark matter. Even though it does not explicitly appear in the Carmeli spacevelocity metric, the vacuum energy contribution to gravity, ΩΛ tends to unity as a function of decreasing redshift. Also since the baryonic matter density Ω0 → 0 as the universe expands, the total mass/energy density Ω+ΩΛ → 1. This indicates that the universe, though always open because Ω < 1, is asymptotically expanding towards a spatially ﬂat state.
7
Acknowledgment
I would like to thank Prof. Moshe Carmeli for many valuable discussions. References [1] S. Behar, M. Carmeli, Cosmological Relativity: A New Theory of Cosmology, Int. J. Theor. Phys. 39(5) (2000) [2] M. Carmeli, Commun. Theor. Phys. 5:159 (1996) [3] M. Carmeli, Cosmological Special Relativity, Singapore, World Scientiﬁc. (2002). [4] P.M. Garnavich et al, Constraints on Cosmological Models from Hubble Space Telescope Observations of Highz Supernovae, Bulletin of the American Astronomical Society 29(7) (1997). [5] S. Perlmutter et al, emphCosmology From Type Ia Supernovae: Measurements, Calibration Techniques, and Implications., Bulletin of the American Astronomical Society 29(5): 1351 (1997). [6] A.G. Riess et al, Observational evidence from supernovae for an accelerating universe and a cosmological constant, Astron. J. 116(Sept): 10091038 (1998)
A QUANTUM APPROACH TO COSMOLOGY
ANTONIO ALFONSOFAUS a
a
E.U.I.T. Aeron´ autica, Plaza Cardenal Cisneros s/n, 28040 Madrid, Spain
Abstract We present a theory based upon the treatment of the gravitational ﬁeld as a sea of gravity quanta, as deﬁned elsewhere. The resultant model for the Universe is a static one, like Einstein ﬁrst saw, with a new feature: a local shrinking quantum world that completely explains the Hubble red shift under a new point of view. The presently accepted expansion of the Universe is interpreted here as an apparent eﬀect, as seen from the Lab system of reference. The static Universe has immersed in it a local shrinking atomic world: a fundamental change in the interpretation of the Hubble’s observations. The conservation principles (momentum, angular momentum and energy) can be dealt with under 2 diﬀerent points of view: local (apparent) and COSMOLOGICAL (”real”). The 2 are in complete agreement with observation. They are also free of well known contradictions or paradoxes/incoherencies (i.e. in the Big Bang model). By dealing now with very well known ﬁrst principles (Heisenberg, Mach, de Broglie, Weinberg’s relation) under the same 2 points of view, we arrive at the conclusion that our new approach is in accordance with the Einstein’s ﬁeld equations of General Relativity, and Quantum Mechanics. We consider this to be a promising ﬁrst step towards the way of dealing with the gravitational ﬁeld coherently both from the General Relativity and from the Quantum Mechanical theories. The agreement with the present values of the cosmological parameters is very satisfactory.
31 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 31–35. © 2006 Springer. Printed in the Netherlands.
32 1
A Quantum Approach to Cosmology Introduction
The point of view of considering the gravitational ﬁeld as a sea of gravity quanta [1] has been dealt with elsewhere. The published result there for the mass mg of this quantum is given by the relation mg = /(c2t), where t is the age of the Universe (then today mg is of the order of 2 × 10−66 grams). The ﬁrst important consequence of the above approach is the need to introduce a new concept, that we call the Mass Boom, [2], [3], [4] and [5]. In essence it expresses the property of any gravitational mass that, due to the emission of these gravity quanta, having a negative energy, its mass increases linearly with cosmological time. This linear dependence between mass and time makes it possible to identify the mass of the Universe M with the cosmological time t. A philosophical statement like we are made of time obviously follows and merits a deep reﬂexion. Clearly this approach is of the Machean type. The second important consequence of this approach is that the speed of light must decrease with time. In fact it can be equated to the inverse of t, c = 1/t [4] and [5] Then the resultant model for the Universe is a static one, as Einstein ﬁrst proposed, and mathematically stated as a(t) = ct = constant, i.e. a constant cosmological scale factor. The expansion of the Universe, a generally accepted interpretation of the red shift, is interpreted here as an apparent eﬀect seen from the laboratory system of reference. The reinterpretation of the Hubble red shift is that the quantum world is shrinking, an eﬀect coming directly from the time variation of Planck’s constant [3], proportional to 1/t2 or equivalently to c2. The work we present here is based upon the above results. We analyze, from this new point of view, the conservation principles, and solve both: the Schr¨ odinger equation together with the Einstein cosmological equations, which represents a ﬁrst step in the harmonization of Quantum Mechanics and Relativity. The conclusion is that the whole approach is very promising and liberates present theories (like the Big Bang) from contradictions ands paradoxes. The agreement with the known numerical values for the cosmological parameters, as accepted today, is very satisfactory.
2
New concepts in the conservation principles
A summary of the new concepts is as follows: • The mass of the gravity quanta, mg = /(c2t). • The Mass Boom eﬀect: any gravitational mass m has a time dependence as m = constant t (t the age of the Universe). • The decrease of the speed of light with time, c = 1/t. • The G = c3 relation, following the Action Principle.
Antonio AlfonsoFaus
33
• Heisenberg and the De Broglie wavelength (perhaps the Compton wavelength as an alternative), /mc = constant in the Lab. • The v/c constancy as seen from the LAB system in order to conserve the constancy of the relativistic relations at any time. • The Mass Boom is always present (as long as gravity is present). • The decrease of the speed of light with time as c = 1/t, always present as a consequence of the constancy of momentum (in the absence of mechanical perturbations). • The apparent interpretation of the cosmological expansion, following the Hubble’s observations. • The h = c2 relation (which explains the contraction of the quantum world). • Weinberg’s relation under a new point of view: not only explains the quantum of mass at the local Lab. It explains the Universe as a quantum black hole whose mass increases linearly with time. • The introduction of H = 1, the cosmological Planck’s constant, given by the relation H = t2 = 1, which is the essence of the quantum approach to cosmology. • The Cosmological Planck’s units using H, instead of , deﬁning the cosmological quantum given by the whole Universe (mass, size and time M = t, and size ct = 1 = 1028 cm with the constant homogeneous tic given by Planck’s time). • The ﬂuctuation of the whole Universe, seen as a quantum black hole (corroborated by the Weinberg’s relation using H). • The determination of the age of the Universe as t = 1061 units of time (the age of the Universe today) and given by the ratio of Planck’s length at t = 1 (the constant length 1028 cm) and the present value of 10−33 cm). • The solution to the cosmological Schr¨ odinger equation coupled with the Einstein’s cosmological equations (harmonization of General Relativity and Quantum Mechanics). • The new entropy concept, that includes the gravitational entropy: S = kM/mg = M = t (for the Universe, [6]).
34 3
A Quantum Approach to Cosmology The solution to the Schr¨ odinger cosmological equation coupled to the Einstein cosmological equations
The Schr¨ odinger equation can be formulated from a cosmological point of view by using the ”cosmological” Planck’s constant H = t2 = 1 (a real constant). The resultant equation is then: H2 ∂2Ψ ∂Ψ + V Ψ = iH 2m ∂x2 ∂t We see that all the terms in this equation vary as 1/t. Then multiplying by t we have both members of the equation constant, as in the normal quantum mechanical treatment. The solution is then, assuming the wave function to be represented by a product of two functions: one depending on space and the other depending on time only, in the usual way one has:
Ψ(x, t) = const (sinx) t2 On the other hand the Einstein cosmological equations have the solution a(t) = t2 [5] which coincides with the above time dependence, as seen from the Lab reference system. Hence we have the same time dependent solution for both: General Relativity and Quantum Mechanics.
4
Conclusions
The main implications of this work are: a change in basic paradigms, perhaps the most important one is the new explanation for gravitation, in quantum mechanical terms, and coherent with general relativity. Also a new approach to the entropy concepts, in particular to the HawkingBekenstein treatments, where the entropy of a black hole is here deﬁned as linear with mass [7]. A return to the Einstein initial cosmological model is the most signiﬁcant change in the formulation of cosmological models. References [1] A. AlfonsoFaus, ”Gravity Quanta, Entropy and Black Holes”, Physics Essays 12, n. 4 (1999). Also in arXiv: grqc/0002065 v1, February 18, 2000. [2] A. AlfonsoFaus, ”Mass Boom versus Big Bang: Einstein was right”. International Symposium no V ”Frontiers in Fundamental Physics”. Hyderabad, India 2003. arXiv:physics 0302058 [3] A. AlfonsoFaus, ”Mass Boom versus Big Bang: the role of Planck’s constant”, Journal of Theoretics June 12 (2003) (arXiv: physics/0309108) [4] A. AlfonsoFaus, ”Quantum Gravity and General Relativity Consistent with a Decreasing Speed of Light and Mach’s principle”. Spacetime and Substance 3 (13) (2002) 130 [5] A. AlfonsoFaus, ”Laboratory Physics and Cosmology”, Physics Essays, July 2004 and arXiv: physics/0407150 v1
Antonio AlfonsoFaus
35
[6] A. AlfonsoFaus, ”Entropy in the Universe: A new Approach”, online Entropy, 2 (2000) 168 (at www.mdpi.orgg/entropy/) [7] A. AlfonsoFaus, ”Black Hole Entropy: Linear or Quadratic?” arXiv:astroph/0211053 v1, November 4, 2002
BLACK HOLES AND THE INFORMATION PARADOX
GERARD ’T HOOFT a
a
Institute for Theoretical Physics, University of Utrecht, The Nederlands
Abstract In electromagnetism, like charges repel, opposite charges attract. A remarkable feature of the gravitational force is that like masses attract. This gives rise to an instability: the more mass you have, the stronger the attractive force, until an inevitable implosion follows, leading to a “black hole”. It is in the black hole where an apparent conﬂict between Einstein’s General Relativity and the laws of Quantum Mechanics becomes manifest. Most physicists now agree that a black hole should be described by a Schr¨ odinger equation, with a Hermitean Hamiltonian, but this requires a modiﬁcation of general relativity. Both General Relativity and Quantum mechanics are shaking on their foundations.
29 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 29–29. © 2006 Springer. Printed in the Netherlands.
MY FOCUS ON THE QUANTUM SOURCE OF GRAVITY
SHANTILAL G. GORADIA Gravity Research Institute, Inc., 3003 Royal Huntsman Court, South Bend, Indiana 46637, USA, [emailprotected]
Abstract If Newtonian gravitation is modiﬁed to use surfacetosurface separation between particles, it can have the strength of nuclear force between nucleons. This may be justiﬁed by possible existence of quantum wormholes in particles. All gravitational interactions would be between coupled wormholes, emitting graviton ﬂux in proportional to particle size, allowing for the pointlike treatment above. When the wormholes are 1 Planck length apart, the resultant force is 1040 times the normal gravitational strength for nucleons. I explain the stability of the nucleus by predicting that the force mediated by pions is repulsive, not attractive.
1
Introduction
Newtonian gravity encounters issues for microscopic dimensions and cannot explain the nuclear binding force. Experimentalists and string theorists face a yet incomplete task of detecting and incorporating the spin 2 graviton into a fully quantized and renormalized theory. If we use the surfacetosurface separation between these particles to quantify the gravitational attraction instead of the centertocenter separation, at small separations relative to the particle radii the force between these particles grows much larger than classical gravity, and may resolve the above issues. The ﬁrst step in the road map suggested by Richard Feynman for consistency in our physical theories is to see if Newton’s law can be modiﬁed to be consistent with Einstein’s law and can be further modiﬁed to be consistent with the uncertainty principle [4].
37 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 37–44. © 2006 Springer. Printed in the Netherlands.
38 2
Quantum Source of Gravity Modiﬁcation of the Inverse Square Law
As an example, for two coupled nucleons (Fig. 1a), I chose the Planck length L = (Gh/c3 )0.5 as the surface separation, as it is the minimum possible spatial distance that makes any sense in physics. Assuming zero separation distance would imply that the two particles are joined to form one particle, losing their distinctions as separate particles. The diameter of the nucleon is about 1 fm (10−15 meters). The Newtonian gravitational force is then FN = Gm2 /D2 ,
(1)
where D is the centertocenter distance, ∼ 1 fm. If we select the surfacetosurface separation instead, the force would become FP = Gm2 /d2 ,
(2)
with d = L = 10−20 fm. The ratio of these two forces is FP D2 = 2 = 1040 , FN d
(3)
which is also the strength of the proposed gravity relative to Newtonian gravity. As the nucleons are separated, D/d shrinks, and FP rapidly approaches FN . Mathematically, D lim = 1. (4) D→∞ d A similar analysis can be made of the quarklepton interaction (Fig. 1b). Nucleons are responsible for over 99 percent of gravity, therefore they are the primary focus of this paper. For nucleons, I recover Newtonian gravity at practically 1000 fm. This modiﬁcation yields a force with high intensity at short range, rapidly falling oﬀ to a very low intensity at long range. The values of a ﬁeld and its rate of change with time are like the position and velocity of a particle. This modiﬁcation meets the uncertainty principle requirement that the ﬁeld can never be measured to be precisely zero. “Einstein, in a paper written in 1919, attempted to demonstrate that his gravitational ﬁelds play an important role in the structure and stability of elementary particles. His hypothesis was not accepted because of gravity’s extreme weakness” [10]. While Einstein’s attempt is worth mentioning, it is not the foundation of my theory. Einstein could be wrong, but it seems he may not be. “It has been proposed that the gravitational constant inside a hadron is very large, ∼ 1038 times the Newtonian G” [10]. This “strong gravity” inside the hadron is similar to my proposed modiﬁcation, but in my modiﬁcation, instead of needing to change G itself, I change the distance measurement and get the same result. My theory does not create a conﬂict with the color force theory either. Strong gravity is consistent with string theory. The short range forces are weakened at long range by a high order of magnitude. This makes other attributes of the short range forces, inﬁnitesimal at long range.
39
S. G. Goradia
a
1 fm
Nucleon
Nucleon L
b
.001 fm
Quark
L
Lepton
Figure 1. Pictorial view of gravitational interaction showing surface and center separations (not to scale). L is the Planck length, 10−20 fm. a, Two nucleons at minimum separation; b, A quark and a lepton, also at minimum separation. The standard inversesquare law would use the centertocenter distances to calculate the force between the particles; using the surfacetosurface distance yields a much stronger force for these separations, equal to the relative strengths of the strong and weak nuclear forces, respectively.
One may question the mathematically simple application of the Planck scale to a problem where the relevant distances seem to be fm. After I ﬁrst published my ﬁndings in early 1999, Frank Wilczek wrote a series of articles [13], explaining how these scales can be reconciled and provided responses. While this may seem simplistic, it seems to be mathematically valid, and frequently signiﬁcant problems can be solved simply in the end, as also illustrated by Morris and Thorne [8]. Complexity in physics lies in the abstraction of simplicity. Classical centers of shapes and therefore surfaces, though used here only for intuitive reasoning are invoked in nuclear coupling constants by implicit comparison to Newtonian gravity and in other descriptions in modern physics. My model is very consistent and therefore suggestive, however it does not reconcile the fact that nucleons overlap. Thanks are due to Gerald ’t Hooft for this comment. Quantum wormholes, as currently theorized, may resolve this issue and give a mathematical foundation to
40
Quantum Source of Gravity
my model. If quantum wormholes do not resolve the issue, we face a challenge to investigate some other veriﬁable quantum entity to explain this phenomenon.
3
Quantum Wormhole Connection
I postulate that each nucleon has a quantum mouth, potentially matching the mouth of a quantum wormhole [3]. The existence of quantum wormholes was examined by Visser [12]. The wormhole’s mouth then represents the entire mass of the particle and propagates its 1/r potential to the rest of the universe. All gravitational interactions become interactions between these wormholes. Radiation by nucleons would consist of energy being emitted by the mouth of the wormhole. The high concentration of bosons, whether gravitons or photons, is consistent with the exclusion principle. This would justify a quantum source of gravity. The mouth emitting the gravitational radiations does not have to be at the surface, allowing the nucleons to overlap. This may sound like a radical approach, but it is not. The direction of my proposal coincides with that in the particle related article by Einstein and Rosen, introducing what is now known as Einstein–Rosen bridges [1]. The abundance of Plancklength size wormholes required could have evolved from perturbations in the initial bigbang density. Stable wormholes require “exotic”, negative energy matter. “... it is not possible to rule out the existence of such material; and quantum ﬁeld theory gives tantalizing hints that such material might, if fact, be possible [8].” The stability of wormholes is on ﬁrmer grounds now. “...the theoretical analysis of Lorentzian wormholes is “merely” an extension of known physicsno new physical principle or fundamentally new physical theories are involved [11].” Literature search reveals no detection of any central force within nucleons, raising a question about the existence of gravitons within nucleons. Fig. 2 shows the mental picture of the graviton ﬂux from nucleons with some background data. Richard Feynman seems to have investigated transfusion of two particles into gravitons [2], but not in this context. The structure of the quantum spacetime is foamy [6]. The potential conversion of two gluons into one graviton and vice versa would be debatable. However, such foamy structure may give a green light for some other form of a particle mechanism. Since the spindependent nuclear force could be positive or negative, my theory maybe suggestive of photonlike particles as mediators of gravity. Some long range forces are potentially simple, cumulative long range manifestations of their short range counter parts and vice versa with their intermediate range immeasurable by microscopic or macroscopic means. My model showing the strong gravity as a function of D2 instead of particle mass (logical function of D3 ) is consistent with the holographic principle. This is analogous to entropy’s proportionality to the horizon area or D2 . Mach principle may imply that the universe spinning in the reference frame of nucleons may subject the nucleons to some form of gravity, not residual color force. As long as the observable characteristics of the proposed wormholes are stable, their stability and types are of secondary importance because the coupling constants are averages of observations.
S. G. Goradia
41
Figure 2. Mental image of nuclear interactions via quantum wormholes. The graviton ﬂux would be proportional to the mass of the interacting particle, yielding couplings of 1040 for nucleons, 1034 for lighter quarklepton pairs and ∼ 1 for pointlike leptons.
The understanding of the coupling constants lies at the heart of our understanding other important issues. “Using the concept of strong gravity, one can show the stability and structure of elementary particles, which could not be achieved by weak gravity” [10]. The sudden decrease in nuclear potential near the surfaces of nucleons may be a result of pion intervention pushing the nucleons apart as needed to stabilize the nucleus against the potential collapse by strong gravity. Pions might not be pulling the nucleons together as originally theorized. Since pions are observed to be spinzero and their range matches the size of nuclei, this possibility cannot be ruled out. A toy bird thrown in the air follows deterministic path, but a real bird has a soul. Therefore, it does not follow a deterministic path. A particle does likewise. This raises a question: does a particle have a soul? Per ancient Hindu Vedas (Katha Upanishad 1.2.20), a particle has a particlesoul (anuatma), and it is also connected to the Supreme, which is omnipresent (paramatma). I see some qualitative consistency between this Vedic doctrine and my proposal based on a potentially justiﬁable speculation that anuatma and paramatma are analogous to particle and normal spacetime respectively, connected by some quantum entity.
4
Double Slit Experiment
Per my theory, in a twoslit experiment (Fig. 3), the network of geodesics downstream of the slits would depend upon whether both slits are open or only one of them is open, not upon the number of slits used for shooting the photons at the same time. Strong gravity at the edges of the slits would strongly impact the network of geodesics downstream geodesics. It does not matter whether the experiment shoots the photons through one slit or both. The screen pattern is a function of the network of geodesics.
42
Quantum Source of Gravity
Figure 3. Depiction of double slit experiment. As shown, the screen pattern is independent of whether the left, right, or both slits are used, as long as the slits are open.
5
The Uncertainty Principle
If one assumes the uncertainty created for the information crossing the quantum wormholes as the Planck time multiplied by the energy of the wormhole (Planck energy in this speciﬁc case), the product of the two closely equals the Planck constant used in the uncertainty principle as shown below, potentially supporting my postulation. ∆E = (1019 GeV × 109 eV /GeV )/(1.6 × 1019 eV /J) = 0.6 × 109 J
(5)
∆t = 10−43 s
(6)
Multiplying the above equations ∆E × ∆t = 0.6 × 1019 J × 10−43 s = 0.6 × 10−34 · s.
(7)
This yields Heisenburg’s Uncertainty, which is ∆E × ∆t ≥ 0.5 × 10−34 J · s /2
(8)
In my theory, I do not have to express the range of nuclear force as “short” with an unanswered question as to precisely how short. The diﬀerence between the two large dynamic numbers of proposed strong gravity and the repulsive nuclear force manifests into observed short range, short enough to ﬁx the size of the nucleus.
S. G. Goradia 6
43
The Early Universe
If God created the universe from nothing, my theory shows that mass energy on one side of the throat of the quantum wormhole is equal to the gravitational ﬁeld energy on the other side of its throat, both canceling each other. This would imply that the gravitational ﬁeld is negative energy considering that mass is positive energy. The appearance of binding energy as mass defect in nuclear engineering calculations may lead to justify the binding energy (result of strong gravity) as negative energy. This is consistent with inﬂationary universe, no matter how big is the universe. “There is nothing known that places any limit on the amount of inﬂation that can occur while the total energy remains exactly zero [5].” 7
Prediction
My model provides a consistent, intuitive and simplistic, but mathematical explanation of the observed relative values of coupling constants, something no other theory has done. If a theory explains observations, it need not predict. Experimentally, my theory may be explored by a careful examination of the nuclear force at distances above 10 fm. Recently published test results veriﬁed the gravitational inverse square law down to 218µm [7]. The test results do not verify the higher dimensional theories that motivated the test, but they are not in conﬂict with my theory, as at these separations my modiﬁed force should be indistinguishable from Newtonian gravity. The generalized equation in the conclusion predicts a string coupling constant of (10−35 )2 = 10−70 . My theory does away with the need to renormalize gravity, since the value of “r” is never zero. 8
Conclusion
In summary, in the early part of last century, when the nuclear force was declared to be a separate force, the Planck length and its implications were not well understood. Planck’s system of fundamental units was considered heretical until came the proposal by Peres and Rosen [9]. The weakness of gravity was unquestioned. Therefore, it was impossible to explain strong gravity force in terms of Newtonian gravity and Einstein’s view was undermined. In light of my article this issue needs to be revisited. My consistent results show that strong gravity creates an illusion of a diﬀerent force between nucleons. Mathematically, the strong force coupling constant Cs = D2 , where D = nucleon diameter in Planck lengths. 9
Acknowledgments
I thank the staﬀ at the Physics Department, University of Notre Dame for comments.
44
Quantum Source of Gravity
References [1] Einstein, A. and Rosen, N., Phys. Rev. 48, 73 (1935). [2] Feynman, R., Acta Phys. Pol. 24, 697 (1963). [3] Goradia, S. G., “Potential Link: Gravity and Nuclear Force,” physics/0210040 (2002). [4] Feynman, R., Six Easy Pieces, Perseus Books Group, Cambridge, Massachusetts (1995). [5] Guth, A. H., Phys. Rev. D 23, 347 (1981). [6] Hawking, S. W., Phys. Rev. D 46, 603 (1992). [7] Hoyle, C. D., et al., Phys. Rev. Lett. 86, 1418 (2001). [8] Morris, M. S. and Thorne, K. S., Am. J. Phys. 56, 395 (1988). [9] Peres, A. and Rosen, N., Phys. Rev. 118, 335 (1960). [10] Shrivastava, S. K. Aspects of Gravitational Interactions, Nova Science Publishers, Commack, New York (1998). [11] Visser, M., Lorentzian Wormholes, From Einstein to Hawking, SpringerVerlag, New York, New York (1996). [12] Visser, M., “Quantum Wormholes,”Physical Review D 43, 402 (1991). [13] Wilczek, F., Physics Today 54, 12 (2001).
VACUUM DECAY BY P BRANES PRODUCTION
LORENZO SINDONI a
b
c
a
AND STEFANO ANSOLDI
b,c
Dipartimento di Fisica, Universit` a degli Studi di Trieste, via Valerio 2, I34127 Trieste (TS), Italy Dipartimento di Matematica ed Informatica, Universit` a degli Studi di Udine, via delle Scienze 206, I33100 Udine (UD), Italy INFN, Sezione di Trieste, Italy
Abstract We present a generalization to the N dimensional case for the nucleation coeﬃcient of a spherical pbrane, separating two (anti)de Sitter spacetimes. We use a semiclassical approximation based on the analytical continuation to the Euclidean sector of a suitable eﬀective action describing a pbrane in General Relativity.
1
Introduction
Vacuum decay can be seen as a phase transition in spacetime and a long time ago the relevance of gravity for the process was studied [1]. The standard treatment of this process makes use of a scalar ﬁeld, known as the inﬂaton, that drives the transition between the false and true vacuum states. This situation can be described by instanton calculations as, for instance, the Colemande Luccia and the HawkingMoss instantons. Here we present a diﬀerent approach, generalizing past works of one of the authors [2, 3]. In particular we are going to use (anti)de Sitter solutions in N spacetime dimensions. In this background we put a spherically symmetric (N −1)brane that splits spacetime into two domains. The system can be described by Israel junction conditions [4], which provide the equations of motion for the
45 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 45–49. © 2006 Springer. Printed in the Netherlands.
46
Vacuum decay by pbranes production
timelike brane. The associated solutions are of two kinds: the ﬁrst one consists of a degenerate brane of zero radius, while the second one consists of a bounce brane collapsing from inﬁnity towards a ﬁnite nonzero turning point, and then reexpanding. To model vacuum decay we consider the tunnelling from the zero radius solution to the bounce solution. The corresponding physical picture is the following: a very small brane1 inside a de Sitter geometry with cosmological constant Λ+ , due to quantum eﬀects, has a nonvanishing probability to tunnel into a brane, containing a de Sitter spacetime with a diﬀerent cosmological constant Λ− . This represents the formation of a bubble of a diﬀerent vacuum phase that then expands to inﬁnity, realizing a transition of the whole spacetime geometry. We can obtain an expression for the probability of such a process using an eﬀective action for this system.
2
Classical Dynamics
The stressenergy tensor for a distribution of matter localized on an hypersurface Σ (the pbrane we mentioned above) can be written in the form Sµν δ(η), where δ is a Dirac delta, and η can be thought as a transverse coordinate to Σ. In N dimensional General Relativity it is possible to write down the equations of motion for this inﬁnitesimally thin distribution of matter by splitting Einstein equations in the tangential and transverse part (see [4] for the 4dimensional case; it can be extended to higher dimensions). Israel junction conditions, then, are [Kij − hij K] ∝ Sij , where Kij and K are, respectively, the extrinsic curvature tensor and its trace and hij is the induced metric on Σ. Here we introduced the standard notation [A] = limη→0+ {A(η)−A(−η)}. Israel junction conditions describe how the (N −1)brane is embedded in the (in principle diﬀerent) geometries of the two spacetime domains that it separates. For our purposes, we are going to write down these equations for a spherical brane with surface stress energy tensor Sij = khij separating two de Sitter spacetimes. This can be done explicitly in terms of the radius R of the brane2 . Then Israel junction conditions reduce to the single diﬀerential equation 2 ˙ ˙ H(R, R) = R + f (R) R(N −3) − kRN −2 = 0; (1)
k is the constant tension of the brane, are signs to be determined by the equation itself [5], and f (R) = 1 − ΛR2 is the metric function appearing in the static line
1 In the mathematical treatment of the classical situation the brane has, in fact, zero radius; from the physical point of view, with quantum gravity in mind, we can imagine this brane as a result of zero point quantum ﬂuctuations. 2 We are going to consider R as a function R(τ ) of the proper time τ of an observer sitting on the brane and denote by an overdot the derivative with respect to τ .
47
L. Sindoni, S. Ansoldi
element adapted to the spherical symmetry for the (anti)de Sitter spacetime3 . For suitable values of the cosmological constants equation (1) has two types of solutions: the ﬁrst is R ≡ 0, while the second represents a brane collapsing and reexpanding from and to inﬁnity. For our purpose it is also important to note that equation (1) can also be obtained by an eﬀective action, which in the N dimensional case can be written as ˙ R −1 ˙ − H(R, R) Seﬀ. = RN −3 R˙ χth dτ, (2) R˙ 2 + f (R)
with the additional constraint H = 0 that has to be imposed on the solutions of the corresponding EulerLagrange equation [5].
3
Tunnelling
The action (2) is crucial in our semiclassical quantization program, since it can be used to quantize the system via a path integral approach. Here we are going to consider the tunnelling process from the R ≡ 0 solution to the bounce solution, within the saddlepoint approximation. This gives the possibility to estimate the following approximated amplitude (e) As.p. ∝ exp −Seﬀ. , (3) (e)
where Seﬀ. is the Euclidean eﬀective action obtained by analytically continuing the action (2) to the Euclidean sector. In order to simplify some expressions, we introduce the following adimensional quantities: x = kR
,
t = kτ
,
α=
Λ − + Λ+ k2
,
β=
Λ− − Λ+ . k2
Moreover it is a well known result that the adimensional version of the equation of motion (1) can be cast in the following form (x )2 + V (x) = 0, where the prime now denotes the derivative with respect the adimensional time t. The potential V (x) is given by V (x) = 1 −
x2 , x20
(4)
3 Since the compact notation could be misleading, we remember that we have two spacetimes with diﬀerent cosmological constants Λ± and, thus, two metric functions f± (R); please, also remember the meaning of the square brackets deﬁned above.
48
Vacuum decay by pbranes production
where x0 = 2/ (1 + 2α + β 2 ) is the adimensional turning radius, provided that the argument of the square root is positive. If this condition holds, there is a bounce trajectory, otherwise we have only the x = 0 solution. Starting from (2), (3) and (4) it is possible to evaluate the Euclidean action on the tunnelling trajectory, i.e on the segment [0, x0 ]. The ﬁnal result can be expressed as (e)
Seﬀ =
(N −3)/2
x0 [(β − )J(N, Cσ )] , 2(N − 2)k (N −2)
(5)
where J(N, p) is a function of the dimension of spacetime, N , and of Cσ = 1 −
β−σ 2
2
x20 ;
with σ± = ±1. A detailed description of the functional form of J(N, p) is beyond the scope of this contribution and can be found elsewhere [7]. We just remark one important feature of it, namely that J(N, p) is deﬁned only when p < 1, a condition that, according to the form of Cσ , is always satisﬁed in our case.
4
Conclusions
In this contribution we have summarized how it is possible to compute the tunnelling amplitude for a spherical pbrane: the process, in view of our short discussion in the introduction, can be used to model the transition between two diﬀerent vacuum phases, one being the de Sitter spacetime with cosmological constant Λ+ and the other being the inﬂating inside of the pbrane, which is characterized by the cosmological constant Λ− . This computation was already performed in four dimensions in the seminal work by Coleman and de Luccia [1] and that result was reproduced in [3]. Here we have carried out its generalization to arbitrary dimensions (referring to [7] for a more detailed analysis). ACKNOWLEDGMENTS One of us (L.S.) wants to thank the Physics Department of University of Udine for ﬁnancial support.
References [1] S. Coleman and F. de Luccia, Phys. Rev. D 21 (1980) 3305 [2] V. A. Berezin, V. A. Kuzmin and I. I. Tkachev, Phys. Rev. D 36 (1987) 2919; E. Fahri, A. H. Guth and J. Guven, Nucl. Phys. B 339 (1990) 417; S. Ansoldi, Class. Quantum Grav. 19 (2002) 6321. [3] S. Ansoldi, A. Aurilia, R. Balbinot and E. Spallucci, Class. Quantum Grav. 14 (1997) 2727;
L. Sindoni, S. Ansoldi
49
[4] W. Israel, Nuovo Cimento 1 (1966) 1; W. Israel, Nuovo Cimento 2 (1967) 463 (errata); C. Barrabes and W. Israel, Phys. Rev. D 43 (1991) 1129. [5] L. Sindoni, MSc Thesis, Universit` a di Trieste (in italian). [6] S. Ansoldi and L. Sindoni, see this conference proceedings. [7] S. Ansoldi and L. Sindoni, in preparation.
SUPERSTRONG INTERACTING GRAVITONS AS A MAIN ENGINE OF THE UNIVERSE WITHOUT EXPANSION OR DARK ENERGY
MICHAEL A. IVANOV a
a
Physics Dept., Belarus State University of Informatics and Radioelectronics, 6 P. Brovka Street, BY 220027, Minsk, Republic of Belarus. Email: [emailprotected]
Abstract The basic cosmological conjecture about the Dopplerian nature of redshifts may be false if gravitons are superstrong interacting particles. A quantum mechanism of classical gravity and the main features of a new cosmological paradigm based on it are described here.
If we assume that the background of superstrong interacting gravitons exists, then the classical gravitational attraction between any pair of bodies arises due to a Le Sage’s kind mechanism. A net force of attraction and repulsion will be nonzero if one suggests that graviton pairs exist and these pairs are destructed by collisions. This pairing is like to the one having place in a case of superconductivity. The portion of pairing gravitons, 2¯ n2 /¯ n, a spectrum of single gravitons, f (x), and a spectrum of subsystem of pairing gravitons, f2 (2x), are shown on Fig. 1 as functions of the dimensionless parameter x ≡ ω/kT (for more details, see [1]). By the Planckian spectra of gravitons we ﬁnd for the Newtonian constant [1]: 2 ·D2 c(kT )6 · I2 (1) 3 π 3 3 where I2 = 2.3184 · 10−6 , T is an eﬀective temperature of the background, and D is some new dimensional constant. It is necessary to accept for a value of this constant: D = 1.124 · 10−27 m2 /eV 2 . G=
51 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 51–54. © 2006 Springer. Printed in the Netherlands.
52 Superstrong interacting gravitons as a main engine of the universe
Figure 1. The portion of pairing gravitons, 2¯ n2 /¯ n, (solid line), a spectrum of single gravitons, f (x), (dashed line), and a spectrum of graviton pairs, f2 (2x), (dotted line) versus the dimensionless parameter x.
In a presence of the graviton background, which is considered in a ﬂat spacetime, an energy of any photon should decrease with a distance r, so we have for a redshift z [2]: z = exp(ar) − 1, where a = H/c with the Hubble constant: H=
1 D · ¯ · (σT 4 ), 2π
(2)
where ¯ is an average graviton energy, σ is the StephanBoltzmann constant. It means that in this approach the two fundamental constants, G and H, are connected between themselves: H = (G
45 σT 4 I42 1/2 ) , 32π 5 c3 I2
(3)
with I4 = 24.866. Using the known value of G, one can compute:H = 3.026 · 10−18 s−1 = 94.576 km · s−1 · M pc−1 by T = 2.7K. From another side, an additional relaxation of any photonic ﬂux due to nonforehead collisions of gravitons with photons leads to a luminosity distance DL : DL = a−1 ln(1 + z) · (1 + z)(1+b)/2 ≡ a−1 f1 (z), where b = 3/2 + 2/π = 2.137.
(4)
M.A. Ivanov
53
Figure 2. The ratio of observed to theoretical functions f1obs (z)/f1 (z) (dots); observational data are taken from Table 5 of [3]. If this model is true, the ratio should be equal to 1 for any z (solid line).
This function ﬁts supernovae data well for z < 0.5 [4]. It excludes a need of any dark energy to explain supernovae dimming. If one introduces distance moduli µ0 = 5 log DL + 25 = 5 log f1obs + c1 , where c1 is a constant, f1obs (z) is an observed analog of f1 (z), we can compute the ratio f1obs (z)/f1 (z) using recent supernovae observational data from [3] (see Fig. 2). In this approach, every massive body would be decelerated due to collisions with gravitons [2] that may be connected with the Pioneer 10 anomaly [5].
References [1] M.A. Ivanov. Screening the graviton background, graviton pairing, and Newtonian gravity [grqc/0207006]. [2] M.A. Ivanov, General Relativity and Gravitation, 33, 479 (2001); Erratum: 35, 939 (2003); [astroph/0005084 v2]. [3] A.G. Riess et al. Type Ia Supernova Discoveries at z > 1 From the Hubble Space Telescope ..., [astroph/0402512] (to appear in ApJ, 2004).
54 Superstrong interacting gravitons as a main engine of the universe [4] M.A. Ivanov. Another possible interpretation of SN 1a data  without kinematics: Comments on the paper astroph/0402512 by A. Riess et al. [astroph/0403130]. [5] J.D. Anderson et al. Phys. Rev. Lett. 81, 2858 (1998); Phys. Rev. D65 082004 (2002); [grqc/0104064 v4]
GENERALIZED COSMOLOGICAL CONSTRAINTS ON NEUTRINO OSCILLATIONS  RELAXED OR STRENGTHENED
DANIELA P. KIRILOVA a
a
Institute of Astronomy, Bulgarian Academy of Sciences, Soﬁa, Bulgaria
Abstract I discuss cosmological constraints, based on BBN production of He4, on electronsterile neutrino oscillations proceeding after electron neutrino freezeout. The general case of sterile neutrino state partially ﬁlled at the onset of oscillations is discussed 0 < δNs < 1. It is shown that δNs = 0 has twofold eﬀect on He4: (a) it enhances the energy density, increasing the overproduction of He4 and strengthening BBN constraints on oscillations and (b) it suppresses the kinetic eﬀects of oscillations on BBN, especially neutrino spectrum distortion, thus decreasing He4 overproduction and relaxing BBN constraints.
1
Introduction
Evidences for neutrino oscillations were obtained by atmospheric, solar and terrestrial experiments. Solar and atmospheric neutrino anomalies were resolved in terms of neutrino oscillations, ﬂavour oscillations being the dominant channel. Experimental constraints on the impact of the sterile neutrino νs in the neutrino anomalies were obtained. Stringent constraints on activesterile oscillations were obtained also from Big Bang Nucleosynthesis (BBN) considerations [1, 2, 3, 4, 5, 6]. For more details concerning BBN constraints see refs. [7, 8]. LMA and LOW activesterile solar oscillation solutions and atmospheric activesterile solutions were excluded by BBN many years before global analysis of experimental neutrino data pointed to the preference of ﬂavour oscillations for solving the neutrino anomalies.
55 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 55–59. © 2006 Springer. Printed in the Netherlands.
56
Oscillations Constraints
Update of BBN constraints on oscillations with δm2 > 10−6 eV2 , assuming equilibrium neutrino energy spectrum, is available in refs. [9, 10]. In the nonresonant case they can be approximated by: (δm2νe νs /eV2 ) sin4 2θνe νs = 3.16 × 10−5 (δNν )2 (δm2νµ νs /eV2 ) sin4 2θνµ νs = 1.74 × 10−5 (δNν )2 In case of oscillations eﬀective after νe decoupling, i.e. (δm2 /eV2 ) sin4 2θ < 10 , the repopulation of active neutrino becomes slow and the spectrum distortion of νe is considerable. For that nonequilibrium oscillations case BBN constraints were discussed in [5, 6]. The analytical ﬁts to the exact constraints, corresponding to 3% He4 overproduction, are: −7
δm2 (sin2 2ϑ)4 ≤ 1.5 × 10−9 eV2 δm2 > 0 δm2  < 8.2 × 10−10 eV2 δm2 < 0, large ϑ, All these BBN constraints assumed initially zero sterile state δNs = 0 before oscillations. δNs = nνs /nνe and nνe is the equilibrium number density of nνe . However, νs may be present at BBN epoch as is predicted by diﬀerent models. Here we discuss BBN with neutrino oscillations in the case δNs = 0. We present the results of the analysis of δNs = 0 eﬀects on oscillations and on BBN. We provide generalized BBN constraints on νe ↔ νs oscillations for 0 < δNs < 1.
2
BBN with νe ↔ νs and nonempty initially νs
4
He is abundantly produced (25% by mass), precisely measured (∼ 3% uncertainty) and calculated (∼ 0.1% uncertainty) and has simple postBBN chemical evolution. Besides it is very sensitive to the kinetics of the nucleon transitions before the nucleosynthesis epoch, and therefore, as far as nucleons freezing depends strongly on the expansion rate, it is considered the best speedometer. Hence, 4 He is the preferred primordial element for probing nonstandard physics. 4 He yield essentially depends on the freezing of the reactions: νe + n ↔ p + e− , + e + n ↔ p + ν˜e , deﬁned by the equilibration of expansion rate H ∼ gef f T 4 and the weak rate Γw ∼ GF E 2 Nν3 : Γw ∼ H. The mass fraction of He4 is roughly estimated as: Yp ∼ 2(n/p)f /(1 + n/p)f exp(−t/τn ) where the frozen neutrontoproton ratio is to a good approximation given by (n/p)f ∼ exp{−(mn − mp )/Tf }. It depends on relativistic degrees of freedom at BBN, gef f = 10.75 + 7/4 δNs , which enter through H, on the electron neutrino spectrum and on the neutrinoantineutrino asymmetry through Γw .
D.P. Kirilova
57
Activesterile oscillations exert two types of eﬀects on BBN: • dynamical eﬀect Through bringing an additional neutrino type into equilibrium they increase 1/2 the expansion rate [1] H(t) ∼ gef f , leading to earlier n/pfreezing, Tf ∼ (gef f )1/6 , and 4 He overproduction. This eﬀect gives up to 5% 4 He overproduction. Hence, initially present δNs = 0 increases the dynamical eﬀect of oscillations, as far as it further increases gef f and exaggerates He overproduction. • kinetic eﬀect due to distortion of the electron neutrino spectrum Oscillations between nonequilibrium νs , δNs = 1, and active neutrinos lead to deviations from the equilibrium ν spectrum because oscillation rate is energy dependent Γ ∼ δm2 /E [2]. The distortion leads both to a depletion of the active neutrino number densities Nν and a decrease of the Γw , ﬁnally overproducing 4 He. This kinetic eﬀect in case of oscillations eﬀective after ν decoupling can be as strong as δNs = 6! [11] Larger δNs decreases the spectrum distortion, and hence, it decreases the oscillations kinetic eﬀect. In conclusion, δNs = 0 has twofold eﬀect on He4: (a) it enhances the energy density, increasing the overproduction of He4 and strengthening BBN constraints on oscillations and (b) it suppresses the kinetic eﬀects of oscillations on BBN, especially neutrino spectrum distortion, thus decreasing He4 overproduction and relaxing BBN constraints. The ﬁgure shows the inﬂuence of diﬀerent eﬀects on helium4 production Yp . The dotted curve present the kinetic eﬀect, the dashed curve – the dynamical eﬀect of initially present νs . There is an interplay between the two eﬀects on He due to the dependence of the kinetic eﬀect on the initial population of νs .
3
BBN constraints on oscillations
Numerical analysis of He4 production in the presence of νe ↔ νs , eﬀective after neutrino decoupling, in the general case of partially ﬁlled initially sterile neutrino state Yp (δNs , δm2 , sin2 2ϑ) was provided in ref. [12]. Cosmological constraints on oscillation parameters corresponding to δYp /Yp = 3% overproduction and diﬀerent initial δNs were calculated in ref. [13]. They strengthen with the increase of δNs . BBN constraints corresponding to δYp /Yp = 5% and diﬀerent νs populations were calculated in ref. [14]. These constraints relax with the increase of δNs . The constraints for δNs = 0.5 and δNs = 0 and diﬀerent He overproduction are illustrated in the ﬁgure 2. The two uppermost contours present the constraints for δYp /Yp = 5% overproduction, the two lower curves – for δYp /Yp = 3%. In conclusion BBN constraints may be either strengthened or relaxed with the increase of the initial population of the sterile state, depending on the level of the
58
Oscillations Constraints
Figure 1. The solid curve presents frozen neutron number density relative to nucleons Xnf = Nnf /Nnuc as a function of the sterile neutrino initial population, at δm = ±10−8 eV2 , sin2 2θ = 1. The dotted curve presents the kinetic eﬀect, while the lower dashed curve presents energy density increase eﬀect. The uppermost long dashed curve corresponds to the total eﬀect when the decrease of the kinetic eﬀect is not accounted for, i.e. in case the eﬀects were additive.
Figure 2. The lower dashed curve presents BBN constraints corresponding to 3% He overproduction and δNs = 0, while the lowest curve presents the strengthened constraints due to higher δNs , namely δNs = 0.5. The upper curve gives the relaxed 5% He overproduction contour corresponding to δNs = 0.5, while the upper dashed curve corresponds to 5% He and δNs = 0.
D.P. Kirilova
59
He overproduction. The results are important for studying neutrino properties, for deﬁning the role of the sterile neutrino in resolving the solar and atmospheric neutrino anomalies, for constraining models the presence of sterile neutrinos during BBN epoch. References [1] A. Dolgov : Sov. J. Nucl. Phys. 33, 700 (1981). [2] D. Kirilova : JINR E288301 (1988). [3] R. Barbieri, and A. Dolgov : Phys. Lett. B 237, 440 (1990); Nucl. Phys. B 349, 743 (1991). [4] K. Enqvist, K. Kainulainen, J. Maalampi : Phys. Lett. B 244, 186 (1990); K. Enqvist, K. Kainulainen, M. Thompson : Nucl. Phys. B 373, 498 (1992). [5] M. Chizhov, D. Kirilova : in Neutrino96, 478 (1996); Phys. Lett. B 393, 375 (1997); D. Kirilova, and M. Chizhov : Phys. Rev. D 58, 073004 (1998); Nucl. Phys. B 591, 457 (2000). [6] D. Kirilova, M. Chizhov : Nucl. Phys. Suppl. B 100, 360 (2001). [7] A. Dolgov : Phys. Rept. 370, 333 (2002). [8] D. Kirilova : Central Eur.J.Phys 2, 467 (2004). [9] A. Dolgov : hepph/0306154 (2003). [10] A. Dolgov, F. Villante : hepph/0308083 (2003). [11] D. Kirilova : Astropart. Phys 19, 409 (2003). [12] D. Kirilova : Int. J. Mod. Phys. D 13, 831 (2004). [13] D. Kirilova : ULBTH/0401, ULB (2004). [14] D. Kirilova, M. Panayotova : IC/IR2004/13 (2004).
ATTRACTIONS OF AFFINE QUANTUM GRAVITY
JOHN R. KLAUDER a
a
Departments of Physics and Mathematics University of Florida Gainesville, FL 32611. Email: [emailprotected]ﬂ.edu
Abstract All attempts to quantize gravity face several diﬃcult problems. Among these problems are: (i) metric positivity (positivity of the spatial distance between distinct points), (ii) the presence of anomalies (partial secondclass nature of the quantum constraints), and (iii) perturbative nonrenormalizability (the need for inﬁnitely many distinct counterterms). In this report, a relatively nontechnical discussion is presented about how the program of aﬃne quantum gravity proposes to deal with these problems.
1
Introduction and Survey
The program of aﬃne quantum gravity diﬀers from that of string theory or loop quantum gravity: speciﬁcally, it diﬀers in the insistence on a spatial metric tensor that is strictly positive deﬁnite; in the simultaneous and uniform treatment of both ﬁrst and secondclass operator constraints; in dealing with nonperturbative renormalizability; and in maintaining a close connection with the motivating classical (Einstein) gravity theory. A suitable realization of these principles is most readily presented within a formalism that is, for the most part, generally unfamiliar to many readers. The purpose of this paper is to provide a relatively simple introduction to several concepts used to study quantum gravity from this new perspective.
61 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 61–67. © 2006 Springer. Printed in the Netherlands.
62 2
Attractions of Affine Quantum Gravity Reproducing Kernel Hilbert Spaces
This important idea can be readily explained. Let l ∈ H, for all l ∈ L, denote a set of states (chosen to be normalized, but that is not a requirement) that span the separable Hilbert space H of interest. In addition we assume that (when it is ﬁnite dimensional) the space L is locally equivalent to a Euclidean space, and that the states l are continuously labeled by the (multidimensional) labels l. We will refer to the set of states {l} as coherent states. Since the coherent states span H, it follows that two elements of a dense set of states may be given by ψ = ΣJj=1 αj lj , φ =
ΣK k=1 βk l(k)
J 0 in the sense of operators for all nonvanishing dxa . Moreover, we insist that gˆab (x) becomes self adjoint when smeared with a suitable real test function. In canonical quantization one chooses the canonical (ADM) momentum π ab (x) as the ﬁeld to promote to an operator, π ˆ ab (x). However, since the momentum acts to translate the metric, such a choice is inconsistent with the preservation of metric positivity. Instead, it is appropriate to choose the mixed valence momentum ﬁeld πca (x) ≡ π ab (x)gbc (x) to promote to an operator π ˆca (x). This choice is dictated by the relation ! ! a b 3 a b 3 ei γb (y) πˆa (y) d y gˆcd (x) e−i γb (y) πˆa (y) d y = (eγ(x)/2 )ec gˆef (x) (eγ(x)/2 )fd , a relation that manifestly preserves metric positivity. The full set of kinematical commutation relations is given by [1] ˆdc (y)] = 21 i[δbc π [ˆ πba (x), π ˆda (x) − δda π ˆbc (x)] δ(x, y) , ˆdc (y)] = 21 i[δac gˆbd (x) + δbc gˆad (x)] δ(x, y) , [ˆ gab (x), π
[ˆ gab (x), gˆcd (y)] = 0 . These are the socalled aﬃne commutation relations appropriate to the aﬃne ﬁelds π ˆca (x) and gˆab (x), both of which may be taken as self adjoint when smeared with real test functions. These commutation relations provide a realization of the group IGL(3, R), and as such they are more in the spirit of a current algebra than traditional canonical commutation relations. It is important to add that by choosing π ˆca (x) as the partner ﬁeld to go with ab gˆab (x), it follows that the momentum ﬁeld π ˆ (x) does not make an operator when smeared but only a form.
4
Quantization of Constraints
There are several schemes in common usage to quantize canonical systems with constraints. Traditionally, these schemes treat ﬁrst and secondclass constraints diﬀerently. Gravity is not a traditional gauge theory since the set of classical constraints form an open ﬁrst class system, which means that the Poisson brackets among the constraints have the form of a Lie algebra except that the structure constants are actually structure functions depending on the canonical variables. On quantization, these structure functions become operators that do not commute with the constraints, and as a consequence, the quantum constraints are partially second class in character. As noted above this usually entails a separate procedure for their analysis. However, the recently introduced projection operator method [2] to incorporate quantum constraints treats ﬁrst and secondclass constraints in an identical fashion and thereby it seems ideal to apply to gravity. Here, we content ourselves with a sketch of how this procedure is applied to simple systems.
64
Attractions of Affine Quantum Gravity
We start by assuming that φα (p, q) = 0, α = 1, ..., A, represent a set of real classical constraints for some multidimensional system. We choose some quantization procedure and identify {Φα (P, Q)} as a set of selfadjoint operators representing the constraints. Ideally, following Dirac, we would identify the physical Hilbert space Hphys ⊂ H as composed of vectors ψphys that have the property that Φα (P, Q)ψphys = 0 for all α. Consistency of this procedure requires that (i) ψphys ψphys < ∞, and (ii) [Φα (P, Q), Φβ (P, Q)]ψphys = 0. Unfortunately, for certain constrained systems, either one or both of these consistency conditions is violated. In that case it is useful to propose another scheme. One alternative procedure, known as the projection operator method, involves a projection operator IE = IE(Σα Φ2α (P, Q) ≤ δ()2 ) , an expression which means that 0 ≤ IEΣα Φ2α (P, Q)IE ≤ δ()2 I . In these expressions, δ() denotes a small, positive cutoﬀ, generally dependent on , that can be reduced to a suitable level. In this approach the (regularized) physical Hilbert space is taken as Hphys = IEH. It is pedagogically useful to illustrate this procedure with three simple examples: (1) Let Φα = J1 , J2 , J3 denote the generators of SU(2), and the desired physical Hilbert space satisﬁes Jk ψphys = 0 for k = 1, 2, 3. We can secure the physical Hilbert space of interest by choosing IE(Σk Jk2 ≤ 21 2 ) .
This example represents a ﬁrstclass constrained system. (2) Let Φα = P, Q, a pair of canonical operators. In this case we choose IE(P 2 + Q2 ≤ ) , which projects onto states ψphys that satisfy (Q + iP )ψphys = 0. This example represents a secondclass constrained system. (3) Let Φα = Q, a single operator with zero in the continuous spectrum and for which Qψphys = 0 has no normalizable solution. here we choose IE(Q2 ≤ δ 2 ) , with no dependence necessary. As δ → 0, this projection operator passes strongly (hence weakly) to the zero operator. To overcome this fact, we rescale the projection operator and take a suitable limit as δ goes to zero. As one example, we introduce coherent states p, q ≡ e−iqP eipQ 0 ,
J.R. Klauder
65
and we consider p , q p , q ≡ lim
δ→0
p , q IE(Q2 ≤ δ 2 )p , q . 0IE(Q2 ≤ δ 2 )0
The resultant expression forms a suitable reduced reproducing kernel which can be used to characterize the physical Hilbert space as a reproducing kernel Hilbert space. If P and Q form an irreducible pair, and for the sake of illustration we choose 0 as a normalized solution of (Q + iP )0 = 0, i.e., 0 is the oscillator ground state, then 1
p , q p , q = e− 2 (q
2
+ q 2 ) ,
a reproducing kernel which characterizes a onedimensional Hilbert space. Diﬀerent choices of the ﬁducial vector 0 may lead to diﬀerent functional representatives, but they nevertheless still describe onedimensional Hilbert spaces. It is noteworthy that path integral expressions exist that directly generate matrix elements of any desired projection operator. For example, staying with elementary examples, coherent state path integrals that generate expressions such as p , q IEp , q may formally be written as conventional phasespace path integrals save for one change, namely, the choice of the integration measure for the Lagrange multipliers [2]. How these general ideas may be applied to quantum gravity can be found in [1].
5
Perturbative Nonrenormalizability
One of the most challenging aspects of conventional approaches to quantum gravity is its perturbative nonrenormalizability. Divergences can be regularized by the introduction of cutoﬀs, as usual, and then counterterms developed on the basis of perturbation theory can be identiﬁed and included in the formalism. For renormalizable theories there are only a ﬁnite number of distinct types of counterterms, while for nonrenormalizable theories – such as gravity – an inﬁnite set of qualitatively distinct counterterms is mandated by perturbation theory. It is no wonder that the morass created by renormalized perturbation theory has driven many workers to alternative approaches such as string theory. On the other hand, perhaps we are deceiving ourselves; could it be that perturbatively suggested counterterms to nonrenormalizable models are in fact irrelevant? This heretical viewpoint is indeed suggested by the hardcore picture of nonrenormalizable interactions which we now outline [3]. To present the essential ideas as simply as possible let us initially examine certain singular potentials in quantum mechanics. In particular, consider the Euclideanspace path integral for a free particle in the presence of an additional
66
Attractions of Affine Quantum Gravity
potential V (x) ≥ 0. In symbols, let us study Wλ ≡ N
x(T )=x
x(0)=x
! ! 1 ˙ 2 dt − λ V (x(t))dt Dx . e− 2 x(t)
As λ → 0+ , it appears self evident that Wλ passes to the expression W0 ≡ N
x(T )=x
x(0)=x
! 2 1 2 1 e− 2 x˙ (t)dt Dx = √ e−(x −x ) /2T 2πT
appropriate to a free particle. Whatever the analytic dependence of Wλ − W0 for small λ (e.g., O(λ), O(λ1/3 ), O(e−1/λ ), etc.), it is tacitly assumed that as λ → 0+ , Wλ → W0 , i.e., that Wλ is continuously connected to W0 . However, this limiting behavior is not always true. Consider the example V (x) = x−4 . In this case the singularity at x = 0 is so strong that the contribution from all paths that reach or cross the origin is ! completely suppressed since x(t)−4 dt = ∞ for such paths, no matter how small λ > 0 is chosen. As a consequence, as λ → 0+ for V (x) = x−4 , it follows that θ(x x ) −(x −x )2 /2T −(x +x )2 /2T √ lim Wλ = W0 ≡ . −e e λ→0+ 2πT
Stated otherwise, when V (x) = x−4 , Wλ is decidedly not continuously connected to the free theory W0 , but is instead continuously connected to an alternative theory – called a pseudofree theory – that accounts for the hardcore eﬀects of the interaction. The interacting theory may well possess a perturbation expansion about the pseudofree theory (to which it is continuously connected), but the interacting theory will not possess any perturbation expansion about the free theory (to which it is not even continuously connected). Let us next pass to scalar ﬁeld theory and the Euclideanspace functional integral ! ! ! Sλ (h) ≡ N exp{ hφdnx − 21 [(∇φ)2 + m2 φ2 ]dnx − λ φ4 dnx} Dφ
appropriate to the φ4n model in n spacetime dimensions. We recall for such expressions that there is a Sobolevtype inequality to the eﬀect that ! ! { φ(x)4 dnx}1/2 ≤ K [(∇φ(x))2 + m2 φ(x)2 ]dnx holds for ﬁnite K (e.g., K = 4/3) whenever n ≤ 4, but which fails to hold (i.e., K = ∞) whenever n ≥ 5. Thus for nonrenormalizable interactions φ4n , for which n ≥ 5, it follows that there are ﬁelds φ for which the free action is ﬁnite while the interaction action is inﬁnite. Just as in the elementary example, there is no reason to believe that counterterms suggested by a regularized perturbation analysis (the underlying premise of which is to maintain a continuous connection with the free theory!) should have any relevance in deﬁning the pseudofree theory S0 (h).
J.R. Klauder
67
It is noteworthy that proposals have been advanced to deﬁne S0 (h) and thereby to develop a meaningful and nontrivial theory of nonrenormalizable scalar ﬁelds [4]. Monte Carlo studies of such proposals are currently under way. Lastly we observe that gravity is also a theory for which the free action (limited to quadratic terms) does not dominate the interaction action (remaining terms), and consequently gravity would seem to be a candidate theory to be understood on the basis of a hardcore interaction, which, when regularized, leads to its perturbatively nonrenormalizable behavior. As plausible as this scenario seems, it will involve a considerable eﬀort to establish it convincingly. References [1] J.R. Klauder, “The Aﬃne Quantum Gravity Programme”, Class. Quant. Grav. 19, 817826 (2002). See also J.R. Klauder, “Noncanonical Quantization of Gravity. I. Foundations of Aﬃne Quantum Gravity”, J. Math. Phys. 40, 58605882 (1999); “Ultralocal Fields and their Relevance for Reparametrization Invariant Quantum Field Theory”, J. Phys.A: Math. Gen. 34 32773288 (2001); “Noncanonical Quantization of Gravity. II. Constraints and the Physical Hilbert Space”, J. Math. Phys. 42, 44404464 (2001). [2] J.R. Klauder, “Quantization of Constrained Systems”, Lect. Notes Phys. 572, 143182 (2001); “Coherent State Quantization of Constraint Systems”, Ann. Phys. 254, 419453 (1997). [3] J.R. Klauder, Beyond Conventional Quantization, Cambridge University Press, Cambridge, 2000. [4] J.R. Klauder, “Overcoming Nonrenormalizability”, Letters in Math. Phys. 63, 229239 (2003); “Overcoming Nonrenormalizability. Part 2”, J. Stat. Phys. 116, 11811187 (2004).
GRAVITATIONAL TUNNELLING OF RELATIVISTIC SHELLS
STEFANO ANSOLDI a
b
a
AND LORENZO SINDONI b
Dipartimento di Matematica e Informatica, Universit` a degli Studi di Udine, via delle Scienze, 206  I33100 Udine (UD), ITALY and I.N.F.N.  Sezione di Trieste Dipartimento di Fisica, Universit` a degli Studi di Trieste, via A. Valerio, 2  I34127 Trieste (TS), ITALY
Abstract Thin shells in general relativity have been used in the past as keystones to obtain realistic models of cosmological and astrophysical situations. A crucial role for these developments was played by the compact description of their dynamics in terms of Israel’s junction conditions. Starting from this geometrical formulation we present a problem related to the WKB regime of shell dynamics and suggest a possible solution.
General relativistic shells are an interesting system in general relativity and because of the simple geometrical description of their dynamics provided by Israel’s junction conditions [1] they became preferred models for many crucial aspects of astrophysical and cosmological situations (see [2] for a more complete bibliography on the subject). Many of these models have been developed under the assumption of spherical symmetry, but (as it happens for instance in the case of gravitational collapse [3]) this does not seem a severe restriction and it is likely that the obtained results can be extended to more general situations. On the other hand, the reduction in the number of degrees of freedom that it is possible to obtain in the spherically symmetric case makes simpler the development of eﬀective models and more transparent the discussion of the interesting subtleties that often appears in the geometrodynamics of shells. Here we are, indeed, going to discuss one of these
69 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 69–73. © 2006 Springer. Printed in the Netherlands.
70
Gravitational Tunnelling of Relativistic Shells
subtleties that already manifests itself in the spherically symmetric case, where the junction conditions reduce to just one equation1 [(R˙ 2 + f (R))1/2 ] ≡ − (R˙ 2 + f− (R))1/2 − + (R˙ 2 + f+ (R))1/2 = M (R)/R,
(1)
a ﬁrst order integral of the second order equation of motion for the shell. In (1) R is the radius of the shell (a function of the proper time τ of an observer comoving with the shell); M (R) describes the matter content of the shell (i.e. it is related to its stressenergy tensor); f± (R) are the metric functions in the two domains of spacetime separated by the shell when the line element is written in the static form adapted to the spherical symmetry; ± are signs (i.e. 0, ±1). Much of the discussion that follows is centered on these last quantities, ± , but, before embarking this program, we also remember that, starting from an eﬀective Lagrangian (the particular form of which is not our concern here), we can compute the second order equation of motion that has (1) as a ﬁrst integral and also obtain the eﬀective momentum [4] conjugated to the only surviving degree of freedom R, Sgn(f (R)) ˙ = R tanh−1 R/( ˙ R˙ 2 + f (R))1/2 P (R, R) . (2) Moreover, equation (1) can be cast in the form of an eﬀective equation [5, 4] for the motion of a unitary mass particle with zero energy in a potential V (R), R˙ 2 +V (R) = 0. Then all the solutions of (1) are solutions of this eﬀective equation and viceversa. This solves the problem of obtaining a qualitative description of how the radius R changes as a function of the proper time τ . Of course this is not the full story, since we still have to build up the global structure of the spacetime in which the shell leaves. It is in this process that we need also the information provided by the functions f± (R) and by the two signs ± . In particular when cutting and pasting the Penrose diagrams to build up the complete spacetime, ± select the sides of the Penrose diagram crossed by the trajectory ± [5]. Expressions for can be obtained with little algebra, ± = Sign M (R) f− − f+ ∓ M 2 (R)/R2 , and the points where ± change from ±1 to ∓1 are the points in which f± (R) are tangent to V (R), if they exist. Since f± (R) ≥ V (R) always, the signs can change i) when the shell is crossing a region with f± (R) ≤ 0 or ii) along a classically forbidden trajectory 2 , where V (R) > 0. It is shown in [4] that integrating the analytic continuation of (2) on the classically forbidden trajectory we can compute WKB transition amplitudes for the tunnelling process through the potential barrier; these amplitudes agree with those already computed by other means3 in [7].
1 Following a standard notation, quantities in the two spacetime regions separated by the shell are identiﬁed by ± subscripts. We use square brackets “[. . . ]” to denote their jump in going from the “−” to the “+” side of the shell and an overdot, “ ˙ ”, to indicate the derivative with respect to the proper time measured by a shellcomoving observer. 2 Our formulation here is far too synthetic and we refer the reader to the literature on the subject (for example [5, 6]) for extended background material. 3 This is a strong argument in favor of an expression for the eﬀective momentum that, when evaluated along a classically forbidden trajectory, diﬀers from (2), also evaluated on a classically forbidden trajectory, by a total derivative of a function of R, at most.
S. Ansoldi and L. Sindoni
Figure 2. Representation of the Euclidean momentum wrapped around a (R, Im(P )) cone that replaces the standard (R, Im(P )) plane (Euclidean phase space). At each value of R the momentum takes values on an S1 with radius proportional to the current value of the radius R. In this way the momentum is both continuous and vanishing at the extrema. The gray straight segment in the picture corresponds to a segment of the horizontal R axis in ﬁgure 1.
71
72
Gravitational Tunnelling of Relativistic Shells
In the cases discussed in [4] the signs ± are constant along the forbidden trajectory, but this is not always the case. We are here interested in a more detailed analysis of those cases in which one of the signs, ± , indeed changes. Let us then ˙ see what happens to the momentum P (R, R). Since on a forbidden trajectory V (R) > 0, then f (R) > 0: we can thus forget the weird exponent in (2). Moreover from the eﬀective equation we obtain that R˙ 2 < 0 i.e. R˙ is purely imaginary and the momentum P (R) also is purely imaginary, since tanh−1 (ı . . . ) = ı arctan(. . . ). ¯ along the forbidden trajectory where, say, − Let us now assume there is an R ¯ ± we have (R˙ 2 + f− (R)) ¯ 1/2 → 0± changes sign. This means that when R → R ∓ (or 0 ) and the argument of the arctan(. . . ) tends to −∞ on one side and to +∞ on the other. Correspondingly, choosing the standard branch of the multivalued function arctan(. . . ), the Euclidean momentum has a discontinuity. We can try to cure this pathology by choosing a diﬀerent branch of arctan(. . . ): but then, following the evolution of the now continuous momentum till the second turning point, the oﬀset introduced by the choice of the new branch makes the momentum nonvanishing there; this seems again a diﬃcult situation to accept. Apparently, we thus face the unpleasant situation of i) having a discontinuous Euclidean momentum that vanishes at both turning points or ii) having a continuous momentum that does not vanish at both turning points (we incidentally point out that if we construct the Penrose diagrams associated to the two spacetimes joined by the shell before and after the transition, some diﬃculties in their interpretation also occur). This situation is pictured in ﬁgure 1 and now, after having stated the problem, we proceed to propose a possible solution, by considering again our Euclidean momentum and following its evolution from the ﬁrst turning point. It ¯ At this point we starts from zero and after some path on the R line it reaches R. enforce its continuity and keep following it until the second turning point, where we impose that it is zero. We said above that this cannot happen, but we implicitly made an assumption, namely that the Euclidean momentum is a function taking values in the real line. Relaxing this assumption we are going to see that not so much remains of the above problem. Figure 2 shows indeed that if we consider the Euclidean momentum as a function that at each point R along the forbidden trajectory takes values in a circle (S1 ) of radius R, then we can make the momentum both continuous and vanishing at both extrema! We end this contribution referring the reader to [9] for an extended discussion from the point of view of Euclidean quantum gravity. References [1] W. Israel, Nuovo Cimento 1 (1966) 1; W. Israel, Nuovo Cimento 2 (1967) 463 (errata); C. Barrabes and W. Israel, Phys. Rev. D, 43 (1991) 1129. [2] S. Ansoldi, Class. Quantum Grav. 19 (2002) 6321. [3] R. Penrose, Phys. Rev. Lett. 14 (1965) 57. [4] S. Ansoldi, A. Aurilia, R. Balbinot and E. Spallucci, Class. Quantum Grav. 14 (1997) 2727; [5] see for instance S. K. Blau, E. I. Guendelman and A. H. Guth, Phys. Rev. D 35 (1987) 1747; A. Aurilia, M. Palmer and E. Spallucci, Phys. Rev. D 40 (1989) 2511.
S. Ansoldi and L. Sindoni
73
[6] V. A. Berezin, V. A. Kuzmin and I. I. Tkachev, Phys. Rev. D 36 (1987) 2919; V. A. Berezin and M. Okhrimenko, Class Quantum Grav. 18 (2001) 2195; P. Hajicek and J. Kijowski, Phys. Rev. D 62 (2000) 044025; L. Sindoni, Graduation Thesis (in Italian) (2004) University of Trieste. [7] S. Coleman and F. de Luccia, Phys. Rev. D 21 (1980) 3305; S. Parke, Phys. Lett. 121B (1983) 313. [8] H. Ishihara, private communication. [9] S. Ansoldi, in preparation.
THREEDIMENSIONAL RELATIVISTIC SIMULATIONS OF ROTATING NEUTRONSTAR COLLAPSE TO A KERR BLACK HOLE
LUCA BAIOTTI a , IAN HAWKE b , PEDRO J. MONTERO b ¨ ,FRANK LOFFLER , LUCIANO REZZOLLA a,c , ´ A. FONT e , NIKOLAOS STERGIOULAS d , JOSE b,c,f ED SEIDEL
a
a
SISSA, International School for Advanced Studies and INFN, Trieste, Italy b MaxPlanckInstitut f¨ ur Gravitationsphysik, AlbertEinsteinInstitut, Golm, Germany c Department of Physics, Louisiana State University, Baton Rouge, LA USA d Department of Physics, Aristotle University of Thessaloniki, Thessaloniki, Greece e Departamento de Astronom´ıa y Astrof´ısica, Universidad de Valencia, Burjassot, Spain f Center for Computation and Technology, LSU, Baton Rouge, LA USA
Abstract We discuss the application of our new threedimensional fully general relativistic hydrodynamics code which uses highresolution shockcapturing techniques and a conformal traceless formulation of the Einstein equations, to the study of the gravitational collapse of uniformly rotating neutron stars to Kerr black holes. We investigate the dynamics of the matter and of the trapped surfaces. We provide precise measurements of the blackhole mass and spin.
75 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 75–82. © 2006 Springer. Printed in the Netherlands.
76 1
Rotating neutronstar collapse to a Kerr black hole Introduction
The numerical investigation of gravitational collapse of rotating stellar conﬁgurations leading to blackhole formation is a long standing problem in numerical relativity. However, it is through numerical simulations in general relativity that one can hope to improve our knowledge of fundamental aspects of Einstein’s theory such as the cosmic censorship hypothesis and blackhole nohair theorems, along with that of current open issues in relativistic astrophysics research, such as the mechanism responsible for gammaray bursts. Furthermore, numerical simulations of stellar gravitational collapse to black holes provide a unique mean of computing the gravitational waveforms emitted in such events, believed to be among the most important sources of detectable gravitational radiation. However, the modelling of blackhole spacetimes with collapsing matter sources in multidimensions is one of the most formidable eﬀorts of numerical relativity. This is due, on one hand, to the inherent diﬃculties and complexities of the system of equations which is to be integrated, the Einstein ﬁeld equations coupled to the generalrelativistic hydrodynamics equations, and, on the other hand, to the immense computational resources needed to integrate the equations in the case of threedimensional (3D) evolutions. In addition to the practical diﬃculties encountered in the accurate treatment of the hydrodynamics involved in the gravitational collapse of a rotating neutron star to a black hole, the precise numerical computation of the gravitational radiation emitted in the process is particularly challenging as the energy released in gravitational waves is much smaller than the total restmass energy of the system. The details of the formulation we use for the hydrodynamics and Einstein equations can be found in [3]. We stress that an important feature of this formulation is that it extends to a general relativistic context the powerful numerical methods developed in classical hydrodynamics, in particular highresolution shockcapturing (HRSC) schemes based on linearized Riemann solvers. Such schemes are essential for a correct representation of shocks, whose presence is expected in several astrophysical scenarios. Another fundamental improvement we implemented is the ability to excise from the evolved grid regions of spacetime within horizons. Such regions are causally disconnected from the rest of the spacetime and so do not have any inﬂuence on the exterior evolution. The ﬁeld and hydrodynamical variables in these regions, though, would get extreme values and so compromise the whole evolution. More details on how the hydrodynamical excision is applied in practice, as well as tests showing that this method is stable, consistent and converges to the expected order are published in [9, 3]. Also for all the details about each of the following sections, which contain only brief summaries of our work on this subject, the reader is referred to [3].
L. Baiotti et al.
77
Figure 1. View of the ﬁnal stages of the collapse of model D1.
2
Initial stellar models
The initial data for our simulations are constructed using a 2D numerical code, that computes accurate stationary equilibrium solutions for axisymmetric and rapidly rotating relativistic stars in polar coordinates [11, 12]. For simplicity, we have focused on initial models constructed with the polytropic EOS, choosing the polytropic exponent Γ = 2 and polytropic constant K = 100 to produce stellar models that are, at least qualitatively, representative of what is expected from observations of neutron stars. In the following we report in particular about two of the models we have studied, one slowly rotating (which we refer to as D1) and one rapidly rotating (D4).
3
Dynamics of the matter
We show in Fig. 1 a representative snapshot of the ﬁnal stages of the evolution of a slowly rotating initial model. Soon after an apparent horizon is found and when this has grown to a suﬃciently large area, the portion of the computational domain containing the singularity is excised (The determination of the apparent horizon is obtained using the fast ﬁnder of Thornburg [13]). This is indicated as an area ﬁlled with squares. Also shown with a thick dashed line is the coordinate location of the apparent horizon and it should be remarked that, because of rotation, this surface is not a coordinate twosphere, although the departures are not signiﬁcant and cannot be appreciated in Fig. 1 (see also Section 5). At t = 0.57 ms, the time which Fig. 1 refers to, most of the matter has already fallen within the apparent horizon and has assumed an oblate shape. Overall, conﬁrming what was already discussed by several authors in the past, the gravitational collapse of the slowly rotating stellar model D1 takes place in an
78
Rotating neutronstar collapse to a Kerr black hole
Figure 2. view of the ﬁnal stages of the collapse of model D4.
almost spherical manner and we have found no evidence of shock formation which could prevent the prompt collapse to a black hole, nor appreciable deviations from axisymmetry. As for the slowly rotating star D1, we show in Fig. 2 a representative snapshot of the evolution of the rapidly rotating model D4. The isocontour levels shown for the restmass density are the same used in Fig. 1. The dynamics is very similar to the one discussed for model D1 up to a time t ∼ 0.49 ms. However, as the collapse proceeds, signiﬁcant diﬀerences between the two models start to emerge and in the case of model D4 the large angular velocity of the progenitor stellar model produces signiﬁcant deviations from a spherical infall. Indeed, the parts of the star around the rotation axis that are experiencing smaller centrifugal forces collapse more promptly and, as a result, the conﬁguration increases its oblateness. In this case the star ﬂattens considerably, all of the matter near the rotation axis has fallen inside the apparent horizon, but a disc of lowdensity matter remains near the equatorial plane, orbiting at very high velocities 0.2 c. The appearance of an eﬀective barrier preventing a purely radial infall of matter far from the rotational axis may be the consequence of the larger initial angular momentum of the collapsing matter and of the pressure wave originating from the faster collapse along the rotational axis. Note that the disc formed outside the apparent horizon is not dynamically stable and, in fact, it rapidly accretes onto the newly formed black hole. This is shown in Fig. 2, where one can notice that the disc is considerably ﬂattened but also has large radial inward velocities which induce it to be accreted rapidly onto the black hole. Note that as the area of the apparent horizon increases, so does the excised region, which is allowed to grow accordingly.
L. Baiotti et al. 4
79
Dynamics of the horizons
In order to investigate the formation of a black hole in our simulations, we have used horizon ﬁnders, available through the Cactus framework, which compute both the apparent horizon and the event horizon. We also adopted several diﬀerent methods for measuring the blackhole mass and spin. The ﬁrst and simplest method of approximating the blackhole mass is ! 2π √ C to use the formula M = 4πeq , where Ceq ≡ 0 gφφ dφ is the proper equatorial circumference. But this would only apply exactly to the Kerr (or Schwarzschild) solution; which is not the case here. A second method for estimating the horizon mass is to look at the distortion of the horizon using the ratio of polar and equatorial proper circumferences, Cr ≡ Cpol /Ceq [6]. For a perturbed Kerr black hole this is expected to oscillate around the asymptotic Kerr value with the form of a quasinormal mode. By ﬁtting to this mode we extract an estimate of the angular momentum parameter a/Mhor from
the relation [5] Mahor = 1 − (−1.55 + 2.55Cr )2 , where Mhor coincides with the blackhole mass M only if the spacetime has become axisymmetric and stationary. A third method of approximating J and hence measuring M is to use the isolated and dynamicalhorizon frameworks of Ashtekar and collaborators [1, 8]. This assumes the existence of an axisymmetric Killing vector ﬁeld intrinsic to a marginally trapped surface such as an apparent horizon. This gives an unambiguous deﬁnition of the spin of the black hole and hence of its total mass. If there is an energy ﬂux across the horizon, the isolatedhorizon framework needs to be extended to the dynamicalhorizon formalism [2]. In practice, our approach to the dynamicalhorizon framework has been through the use of a code by Schnetter which implements the algorithm of [8] to calculate the horizon quantities. The advantage of the dynamicalhorizon framework is that it gives a measure of mass and angular momentum which is accurately computed locally, without a global reconstruction of the spacetime. One possible disadvantage is that the horizon itself is required to be (close to) axisymmetric; so in case it deviates largely from axial symmetry, no information can be found. However, because arbitrarily large distortions are allowed as long as they are axisymmetric, we have not encountered problems in applying the dynamicalhorizon framework to the horizons found in our simulations. A fourth method for computing J only applies if an event horizon is found and if its angular velocity has been measured. In a Kerr background, in fact, the generators of the event horizon rotate with a constant angular velocity ω ≡ −gtφ /gφφ = gtt /gφφ and it can be directly related to the angular momentum 1/2 parameter as a/M = J/M 2 = Aω 2 1 − Aω 2 /4π π , where A is the eventhorizon proper area (The event horizon is located after the simulation has been completed and the data produced is postprocessed using the levelset ﬁnder of Diener [7]). Although the direct comparison of many diﬀerent methods employed here have provided valuable information on the dynamics of the system, we have found the
80
Rotating neutronstar collapse to a Kerr black hole
Figure 3. Evolution of the most relevant surfaces during the collapse for the cases D1 and D4.
dynamicalhorizon framework to be simple to implement, accurate and not particularly aﬀected by the errors from which equivalent approaches seem to suﬀer. As a result, we recommend its use as a standard tool in numerical relativity simulations.
5
Reconstructing the global spacetime
All of the results presented and discussed in the previous Sections describe only a small portion of the spacetime which has been solved during the collapse. In addition to this, it is interesting and instructive to collect all of these pieces of information into a global description of the spacetime and look for those features which mark the diﬀerence between the collapse of slowly and rapidly rotating stellar models. As we discuss below, these features emerge in a very transparent way within a global view of the spacetime. To construct this view, we use the worldlines of the most representative surfaces during the collapse, namely those of the equatorial stellar surface, of the apparent horizon and of the event horizon. For all of them we need to use properly deﬁned quantities and, in particular, circumferential radii. The results of this spacetime reconstruction are shown in Fig. 3, whose left and right panels refer to the collapse of models D1 and D4, respectively. The diﬀerent lines indicate the worldlines of the circumferential radius of the stellar surface (dotted line), as well as of the apparent horizon (dashed line) and of the event horizon (solid line). Note that for the horizons we show both the equatorial and the polar circumferential radii, with the latter being always smaller than the former. For the stellar surface, on the other hand, we show the equatorial circumferential radius only. Note that in both panels of Fig. 3 the event horizon grows from an essentially zero size to its asymptotic value. In contrast, the apparent horizon grows from an initially nonzero size and, as it should, is always contained within the event horizon. At
L. Baiotti et al.
81
late times, the worldlines merge to the precision at which we can compute them. A rapid look at the two panels of Fig. 3 is suﬃcient to appreciate the diﬀerent properties in the dynamics of the collapse of slowly and rapidly rotating models. The worldlines of the stellar equatorial circumferential radius are very diﬀerent in the two cases. In the slowly rotating model D1, in particular, the star collapses smoothly and the worldline always has a negative slope, thus reaching progressively smaller radii as the evolution proceeds (cf. left panel of Fig. 3). By time t 0.59 ms, the stellar equatorial circumferential radius has shrunk below the corresponding value of the event horizon. In the case of the rapidly rotating model D4, on the other hand, this is no longer true and after an initial phase which is similar to the one described for D1, the worldline does not reach smaller radii. Rather, the stellar surface slows its inward motion and, at around t ∼ 0.6 ms, the stellar equatorial circumferential radius does not vary appreciably. Indeed, the right panel of Fig. 3 shows that at this stage the stellar surface moves to slightly larger radii. This behaviour marks the phase in which a ﬂattened conﬁguration has been produced and the material at the outer edge of the disc experiences a stall. As the collapse proceeds, however, also this material will not be able to sustain its orbital motion and, after t ∼ 0.7 ms, the worldline moves to smaller radii again. By time t 0.9 ms, the stellar equatorial circumferential radius has shrunk below the corresponding value of the event horizon.
6
Conclusion
We have implemented a hydrodynamical excision technique within our new 3D generalrelativistic numerical evolution code that combines stateoftheart numerical methods for the spacetime evolution (i.e. the NOK formulation of the Einstein equations with Gammadriver shift conditions) with an accurate hydrodynamical evolution employing several highorder HRSC methods. The evolution of the spacetime and of the hydrodynamics is coupled transparently through the method of lines, which allows for the straightforward implementation of various diﬀerent timeintegrators. As a ﬁrst astrophysical problem for this novel setup, we have here focused on the collapse of rapidly rotating relativistic stars to Kerr black holes. The stars are assumed to be in uniform rotation and dynamically unstable to axisymmetric perturbations. While the collapse of slowly rotating initial models proceeds with the matter remaining nearly uniformly rotating, the dynamics is shown to be very different in the case of initial models rotating near the massshedding limit, for which strong diﬀerential rotation develops. Although the stars become highly ﬂattened during collapse, attaining a disclike shape, the collapse cannot be halted because the speciﬁc angular momentum is not suﬃcient for a stable disc to form. Instead, the matter in the disc spirals towards the black hole and angular momentum is transferred inward to produce a spinning black hole. Several diﬀerent approaches have been employed to compute the mass and angular momentum of the newly formed Kerr black hole. Besides more traditional
82
Rotating neutronstar collapse to a Kerr black hole
methods involving the measure of the geometrical properties of the apparent and event horizons, we have ﬁtted the oscillations of the perturbed Kerr black hole to speciﬁc quasinormal modes obtained by linear perturbation theory. In addition, we have also considered the recently proposed isolated and dynamical horizon frameworks, ﬁnding it to be simple to implement and yielding estimates which are accurate and more robust than those of other methods. This variety of approaches has allowed for the determination of both the mass and angular momentum of the black hole with an accuracy unprecedented for a 3D simulation. Recent progress has been made in using mesh reﬁnement techniques [10] to move the outer boundary suﬃciently far from the source so that important information can be extracted on the gravitational wave emission produced during the collapse. The results of these investigations have been presented in ref. [4].
7
Acknowledgments
It is a pleasure to thank Joachim Frieben, Jos´e Ma . Ib´an ˜ez and Andrea Nerozzi who have participated to the development and testing of the code. We are also very grateful to Peter Diener, Jonathan Thornburg, Erik Schnetter, Badri Krishnan and Frank Herrmann for their help in the study of the trapped surfaces and to Abhay Ashtekar for clarifying some aspects of the dynamicalhorizon framework. We are also grateful to the Cactuscode team, for their eﬀorts in producing an eﬃcient infrastructure for numerical relativity. Financial support for this work has been provided by the MIUR and EU Network Programme (Research Training Network contract HPRNCT200000137). J.A.F. acknowledges ﬁnancial support from the Spanish Ministerio de Ciencia y Tecnolog´ıa (grant AYA 20013490C0201). The computations were performed on the Albert100 cluster at the University of Parma and the Peyote cluster at the Albert Einstein Institute. References [1] A. Ashtekar, C. Beetle, S. Fairhurst, Class. Quantum Grav., 17, 253 (2000) [2] A. Ashtekar, B. Krishnan, Phys. Rev. Lett., 89, 261101 (2002) [3] L. Baiotti, I. Hawke, P.J. Montero, F. L¨ oﬄer, L. Rezzolla, N. Stergioulas, J.A. Font, E. Seidel, Phys. Rev. D71, 024035 (2005) [4] L. Baiotti, I. Hawke, L. Rezzolla, E. Schnetter, Phys. Rev. Lett. in press (2005) [5] S. Brandt, E. Seidel, Phys. Rev. D52, 856 (1995) [6] S. Brandt, J.A. Font, J. M. Ib´ an ˜ ez, J. Mass´ o, E. Seidel, Comp. Phys. Comm., 124, 169 (2000) [7] P. Diener, Class. Quantum Grav., 20, 4901 (2003) [8] O. Dreyer, B. Krishnan, D. Shoemaker, E. Schnetter, Phys. Rev. D67, 024018 (2002) [9] I. Hawke, F. L¨ oﬄer, A. Nerozzi, grqc/0501054 [10] E. Schnetter, S.H. Hawley, I. Hawke, Class. Quantum Grav., 21, 1465 (2004) [11] N. Stergioulas, J.L. Friedman, Astrophys. J., 444, 306 (1995) [12] N. Stergioulas, Living Rev. Relativity, 6 (2003) [13] J. Thornburg, Class. Quantum Grav., 21, 743 (2004)
GRAVITATIONAL WAVE SOURCES
VALERIA FERRARI a
b
a,b
Dipartimento di Fisica, Universit` a di Roma ”La Sapienza”, P.le A. Moro 2, 00185 Roma, Italy INFN, Sezione Roma 1,P.le A. Moro 2, 00185 Roma, Italy
Abstract Gravitational waves may reach the Earth in diﬀerent forms which depend on the nature of the emitting source: they may be short bursts, outcome of a catastrophic event like the gravitational collapse or the coalescence of a binary system; they may be continuous, weak wavetrains emitted by a non axially symmetric, rotating neutron star or by binary systems far from coalescence; or, they may appear as a background created as a polyphonic chorus superposition of diﬀerent voices: those of populations of astrophysical sources born and evolved in the past, or those coming from matter and energy ﬂuctuations in the ﬁrst instant of life of the Universe. In this paper we will shortly review some of the most promising astrophysical sources of gravitational waves to be detected by the the resonant or interferometric antennas actually in operation or under construction.
1
Introduction
The detection of gravitational waves (GWs) is one of the most challenging problems of modern astrophysics. The astronomical observations of binary pulsars like PSR 1913+16 [1], or the more recently discovered PSR J07373039 [2], provide an indirect proof of their existence: indeed, the orbital period of such systems decreases at a rate which is fully explained if one assumes that the variation of
83 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 83–91. © 2006 Springer. Printed in the Netherlands.
84
Gravitational Wave Sources
the orbital energy compensates the energy emitted by the system in GWs, as predicted by General Relativity. However, gravitational waves are very elusive, and have never been observed directly because they are extremely weak, and because they weakly interact with any form of matterenergy. For this reason to detect them directely extremely sophisticated instruments are needed that work at the very edge of modern technology. As an example, few years ago the resonant antenna NAUTILUS in Rome was cooled to a temperature of 100 mK, which is the lowest temperature ever reached for a massive body. At the same time, in studying GW sources theorists have to explore phenomena that occur in regions of spacetime where the known lows of physics have to be applied in some extreme conditions: extreme is indeed the physics of a gravitational collapse and the coalescence of massive bodies, extreme are the densities and pressures reached in the interior of neutron stars, or the state of matter and energy when the universe was born. Thus, in searching for GWs we are certainly operating at the frontiers of fundamental physics. It is instructive to see what is the frequency region that will be explored by the detectors that are in operation, in construction, or planned for the next future. Resonant bar detectors, ALLEGRO, NIOBE, AURIGA, EXPLORER, NAUTILUS, have been operating for years and are now close to the maximum sensitivivity that can be reached by these instruments; they are narrowband detectors, sensitive in a frequency region of a few Hz centered at about 1 kHz. A new generation of spherical, omnidirectional resonant detectors (MINIGRAIL, GRAIL) is presently under study. For instance MINIGRAIL, which is already in operation, is a sphere of 1400 kg, resonant at 2.9 kHz with a bandwidth of 230 Hz. The ground based interferometers are sensitive to a larger frequency region which extends from a few tens of Hz to a few kHz; TAMA, GEO600, LIGO have already started the ﬁrst scientiﬁc runs, whereas the italianfrench experiment VIRGO is in the commissioning phase [3]. Typical sources for these detectors are the gravitational collapse to a neutron star (NS) or to a black hole (BH), stellar pulsations, coalescence of binary systems composed of compact objects like NSs and BHs. At lower frequencies, the sensitivity of ground based detectors is strongly deteriorated by the eﬀect of local disturbances (like wind and atmospheric events, human activities, etc.) and seismic noise that are hard to screen; therefore, spacebased experiments are the only possibility to explore this region, which is rich of expected sources. For instance some of the most interesting sources to be detected are binary systems like white dwarfswhite dwarfs (WDWD) binaries and Low Mass XRay Binaries (LMXRB) with short orbital period T ∈ [10m, 1hour], that are expected to emit gravitational waves at typical frequencies of a few mHz; or, binary systems composed of two black holes with masses ranging from 102 M to 106 M which, during the last phases of coalescence would emit a GWsignal that would span the frequency region νGW ∈ [10−4 , 10−1 ] Hz in a time interval ranging from a tenth of a year to 500 yrs, depending on their mass. LISA (Laser Interferometer Space Antenna) will be the ﬁrst high sensitivity spacebased GW detector, consisting of a constellation of 3 spacecraft in heliocentric orbit, located at the vertices of an equilateral triangle with a 5 · 106 km side. Each spacecraft contains a pair of testmasses in geodesic motion; the test masses are the end
V. Ferrari
85
mirror of a single arm interferometer, the other endmirror being in one of the other two spacecraft. As in all instruments of these kind, gravitational waves will be detected by the strain they induce in the distances among the testparticles. In the following sections we will review some of the most interesting GW sources that emit in the bandwidth of the above mentioned detectors; in particular, we will consider signals that exhibit some well deﬁned features and that last for long enough so that, if detected, will possibly allow the identiﬁcation of the emitting source.
2
Rotating neutron stars
Neutron stars can radiate their rotational energy essentially in two ways. If the star is triaxial, it has a time varying quadrupole moment and emits gravitational waves at twice the rotation frequency, and with an amplitude which can be parametrized as follows 2 r −1 I −24 ms h ∼ 4.2 × 10 P 10 kpc 1045 g cm2 10−6
where I = 1045 g cm2 , is a typical value for the moment of inertia of a neutron star, P is the rotation period, and is the oblateness of the star which shows how much it deviates from axial symmetry, here given in units of 10−6 . These waves could be detected, for instance, by VIRGO with one year of integration, if the amplitude of the signal were of the order of h ∼ 10−26 ; this means that neutron stars outside our Galaxy would practically be out of reach by the interferometers of ﬁrst generation. In addition, much depends on the value of the oblateness, and several studies have tried to constraint its possible range of variation. For instance in ref. [4] observational data on a number of known pulsars have been used to set an upper limit on . Assuming that the observed slowing down of their period is entirely due to the emission of GWs, the authors found ∈∼ [10−2 , 10−9 ]; of course this can only be an upper bound, since the rotational energy of pulsars is dissipated also by other mechanisms, like the electromagnetic emission and/or the acceleration of charged particles in the magnetosphere. Further studies on this problem [5] established that if the triaxial shape is due to strains in the crust of the < neutron star, the oblateness could be ∼ 10−7 10σ−2 ; here σ is the strain needed to break the crust, but it should be said that its value is quite uncertain; for instance σ ∈ [10−2 , 10−1 ] according to [6], and σ ∈ [10−4 , 10−3 ] according to [7]. Thus, in the light of what we presently know we cannot say what is the most likely value of and gravitational waves will probably tell us how much oblate a neutron star can be. A time dependent quadrupole moment can also be due to a precession of the star’s angular velocity around the symmetry axes. In this case the radiation is 1 emitted at a frequency νprec = 2π (ωrot + ωprec ) νrot , but the amplitude of the precessional contribution depends on a further parameter, the “wobbleangle” between the rotation and the symmetry axes which is, however, largely unknown.
86 3
Gravitational Wave Sources Binary systems far from coalescence
In 1975, Hulse and Taylor applied the quadrupole formalism [8] to predict the slowing down of the period of the binary pulsar PSR 1913+16 [1]. They found dP −12 , in excellent agreement with the observed value, dP dt = −2.4 · 10 dt = −(2.3 ± −12 0.22) · 10 , thus providing the ﬁrst indirect evidence of the existence of gravitational waves. Since LISA will be sensitive to the low frequency region, it is now interesting to ask wether these waves could be seen directly. The quadrupole formalism, used to solve Einstein’s equations in the weakﬁeld, low motion limit, shows that when the orbit is circular the radiation is emitted at twice the keplerian orbital frequency. If, as an example, we assume that we are looking at the radiation which emerges in the direction orthogonal to the orbital plane, the wave amplitude, computed in the TT (TransverseTraceless) gauge is given by sin 2ωK t 0 cos 2ωK t 4G2 (M1 + M2 )µ 1 TT sin 2ωK t − cos 2ωK t 0 , hij = − (1) l0 c4 d 0 0 0 where l0 is the orbital separation, ωK = G(M1l3+M2 ) is the orbital frequency, d the 0
1 M2 distance of the source from Earth and µ = MM the reduced mass of the system of total mass M = M1 + M2 . If the orbit is eccentric, waves will be emitted at frequencies multiple of νk = ωk /2π, and the number of equally spaced spectral lines will increase with the eccentricity [9]. For instance, the binary pulsar PSR 1913+16 is composed of two very compact stars with masses M1 = 1.4411 M and M2 = 1.3874 M , revolving around their center of mass with an eccentric orbit (e = 0.617139), and keplerian frequency νk = 3.583 · 10−5 Hz. The system is at a distance d = 5 kpc from Earth. In this case the maximum of the GW emission occurs at νmax = 1.44 · 10−4 Hz, with amplitude hmax ∼ 10−23 . A similar calculation for the recently discovered double pulsar PSR J07373039, in which case M1 = 1.337 M , M2 = 1.250 M , the orbital period is P = 2.4, h d = 5 − 600 pc, e = 0.088, shows that the maximum of the GW emission occurs at νmax = 2.3 · 10−4 Hz, and the amplitude of the corresponding spectral line is hmax ∼ 6 − 7 · 10−22 . Thus in both cases the GW emission is in the bandwidth of LISA, but the expected sensitivity curve of this instrument shows that the signal is too low to be detectable. However, other interesting sources of radiation of this kind exist in our Galaxy; they are, for instance, cataclysmic variables that are semidetached binaries of low mass and very short period, in which the primary star is an accreting degenerate white dwarf, and the secondary is usually a late typestar ﬁlling its Roche lobe and transfering matter on the companion; furthermore, there are doubledegenerate binary systems, WDWD or WDNS binaries; 10 of such systems have been observed so far that have ultrashort periods, smaller than ten minutes. They are typically strong Xray emitters, and they appear very promising for LISA because the emission frequency in GW is well inside the detector bandwidth. In table I the masses, emission frequencies and characteristic wave amplitude are listed for some of these short period systems, that should serve as calibration sources for LISA.
87
V. Ferrari
Table 1. Galactic binaries with component masses M1 and M2 , gravitational wave frequency νGW , and gravitational wave amplitude h.
name WD 0957666 WD 1101+364 WD 1704+481 KPD 0422+4521 KPD 1930+2752 RXJ0806.3+1527 RXJ1914+245 KUV051840939 AM CVn HP Lib CR Boo V803 Cen U Gem IP Peg HU Aqr VW Hyi EX Hya WZ Sge ST LMi SW UMa Z Cha V 436 Cen
4
m1 (M ) 0.37 0.31 0.39 0.51 0.5 0.4 0.6 0.7 0.5 0.6 0.6 0.6 1.26 1.15 0.95 0.63 0.78 0.45 0.76 0.71 0.84 0.7
m2 (M ) 0.32 0.36 0.56 0.53 0.97 0.12 0.07 0.092 0.033 0.03 0.02 0.02 0.57 0.67 0.15 0.11 0.13 0.058 0.17 0.10 0.125 0.17
νGW (Hz) 0.4 · 10−3 0.2 · 10−3 0.2 · 10−3 0.3 · 10−3 0.2 · 10−3 6.2 · 10−3 3.5 · 10−3 3.2 · 10−3 1.9 · 10−3 1.8 · 10−3 1.4 · 10−3 1.2 · 10−3 1.3 · 10−4 1.5 · 10−4 2.7 · 10−4 3.1 · 10−4 3.4 · 10−4 4.1 · 10−4 2.9 · 10−4 4.9 · 10−4 3.1 · 10−4 3.7 · 10−4
log h −21.4 −21.7 −21.4 −21.2 −21.0 −21.4 −21.2 −22.0 −21.7 −21.7 −22.0 −22.0 20.8 20.9 21.3 21.3 21.4 22.1 21.4 21.6 21.5 21.6
Coalescing binary systems
The target of groundbased interferometers is to detect the signal emitted during the latest phases of inspiraling of a binary system. Assuming that the two coalescing bodies are point masses, by using the quadrupole formalism and including the radiation reaction eﬀects it is possible to show that the loss of gravitational energy induces a circularization of the orbit, and that the radius decreases 4 1/4 5 Rin according to the law R(t) = Rin (1 − t/tcoal ) ; where tcoal = 256 µM 2 is the time of the ﬁnal coalescence. The orbital frequency consequently increases, and so does the frequency of the emitted wave, with a time dependence given by " #3/8 5 1 1 ν = π1 256 . Thus, the gravitational signal emitted by a coa2/3 (t −t) µM coal lescing system resembles the chirp of a singing bird. For instance if the binary is composed of two neutron stars, about 108 years after formation the frequency of the emitted signal will enter the bandwidth of the ground based interferometers and in about 1520 minutes (depending on the star masses) will sweep the
88
Gravitational Wave Sources
Figure 1. The strain amplitude of the gravitational signal produced by the coalescence of binary systems. In the left panel we plot the noise strain amplitude of the ﬁrst generation of ground based intrerferometers VIRGO, LIGO and GEO versus frequency, and the strain amplitude of the signal produced by the coalescence of two neutron stars (NSNS) with mass M = 1.4 M each, and two black holes of mass M = 10 M . In both cases the source is assumed to be at the distance of the Virgo cluster (d ∼ 15 Mpc). In the right panel the same plot is done for the spacebased interfereometr LISA and for two cases of BHBH coalescence, respectively of 100 M and 106 M . In both cases the source is assumed to be at a distance d = 3 Gpc. All signals extend up to the frequency corresponding to the ISCO (see text).
frequency region ranging from ∼ 10Hz to ∼ 900Hz. The amplitudes of the two polarizations of the wave emitted during this fast inspiralling phase are 2/3
h+ =
2(1 + cos2 i)µ (πM ν) r
cos(2πνt),
h× = ±
4µ cos i (πM ν) r
2/3
sin(2πνt),
(2) where i is the angle of inclination of the orbit to the line of sight, and the emitted π 2/3 M 2/3 energy per unit frequency is given by dE dν = 3 µ ν 1/3 . In ﬁgure 1 we plot the strain √ amplitude (i.e. νh(ν), where h(ν) is the Fourier transform of h(t)) associated to the GW signal emitted by diﬀerent types of binary systems, compared to the noise strain amplitude (Sh )1/2 of the ﬁrst generation of ground based (left panel) and spacebased (right panel) intereferometric detectors. In all cases the signal extends up to the frequency corresponding to the ISCO (Innermost Stable Circular Orbit RISCO ∼ 6G(Mc12+M2 ) ) after which the two bodies merge, and the waveform given in (2) is no longer appropriate to describe the emitted wave. It should be stressed that near the ISCO the signal (2) has to be corrected to take into account tidal interaction eﬀects and corrections due to the orbital motion. It should also be stressed that the signal described by eq. (2) refers to the case of non rotating bodies. If they are rotating, spincoupling eﬀect may produce signals that are considerably diﬀerent ([10]). From ﬁgure 1 we see that it is quite unlikely that the ﬁrst generation of ground
V. Ferrari
89
based interferometrs will observe the coalescence of neutron stars, while they would be able to detect the coalescence of suﬃciently massive black holes. Conversely, LISA will be a fantastic instrument to detect the coalescence of massive and supermassive black holes out to cosmological distances. It is interesting to mention that given a coalescing source, the larger is the detector bandwidth, the longer will be the time spent by the signal in that bandwidth, and this will increase the detection chances. This is of particular importance for the ground based interferometers: for instance, a signal emitted by two 10M  black holes will stay in the VIRGO bandwidth (∼ [10Hz, 1kHz]) about 38 s, whereas it will sweep LIGO’s bandwidth (∼ [40Hz, 1kHz]) in about 1 s. The evolution of the system when it approaches the ISCO has to be described by fully relativistic numerical simulations. Although these studies are fastly progressing, they are still far from providing reliable waveforms to confront with the detector sensitivity curves. However we have a clue on what the signal, emitted after a single compact object forms, could be; indeed, the newly born object will violently oscillates in its quasinormal modes emitting GWs at the corresponding frequencies, which only depend on the mass and angular momentum if it is a black hole [11], or on the equation of state of matter, if it is a neutron star [12]. Unfortunately, we do not know how much energy can be stored in each mode, and consequently the wave amplitude is unknown. A clue on this crucial information will be available when the numerical studies of the merging phase will progress enough to describe the formation and the subsequent initial evolution of the newly born compact object. However, it should be stressed that it will be the detection of such signals which will provide the most interesting insight into the physics of phenomena about which so little is known.
5
Concluding Remarks
The detection of gravitational waves will allow to test the theory of General Relativity in the strong ﬁeld regime; but beyond that there are many reason why it is of fundamental importance. GWs will allow to test alternative theories of gravity: for example binary pulsars could yield bounds on scalartensor theories because they predict dipole gravitational radiation and violation of the strong equivalence principle. The most famous of such theories is the JordanBransDicke theory which postulates the existence of a scalar ﬁeld which couples only to gravity; ω is the coupling constant, such that the larger the value of ω, the weaker the scalar ﬁeld. The bound from PSR 1913+16 is only ω > 100, because the two stars have nearly equal masses and dipole radiation is suppressed by symmetry. A binary system composed of two compact stars of unequal mass, like a white dwarf and a neutron star, could yield bounds as large as 104 [13]. Through the study of radiation reaction eﬀects we will be able to study how the emission of GWs aﬀects the evolution of the emitting sources, as for instance the rotation rate of a non axisymmetric neutron stars, the process of coalescence in a binary system or the evolution of a newly born neutron star.
90
Gravitational Wave Sources
GWs will provide information on the equation of state of matter at supranuclear densities (unreachable by experiments on Earth) through the frequencies of the quasi normal modes of oscillation of Neutron Stars, which may be excited after the star formation in a gravitational collapse, during a glitch or in the latest phases of binary coalescence [12], [14]. GWs have been produced since the beginning of the Universe, and the existence of relic gravitational waves essentially rely on the validity of General Relativity and on basic principles of quantum ﬁeld theory. Indeed, quantum ﬂuctuations of the gravitational ﬁeld are ampliﬁed by the accelerated expansion of the universe during the inﬂationary stage. In the same way as a sudden change in the electromagnetic ﬁeld produces photons, the rapid change in the spacetime curvature produces gravitons, and consequently a stochastic background of gravitational waves. The detection of such background is the only way to learn about the evolution of the very early Universe, up to the limits of the Planck era and the Big Bang. The stochastic background of gravitational waves has also a component of astrophysical origin. Indeed, bursts and continuous signals of gravitational radiation have been emitted by astrophysical sources since the beginning of star formation, and the cumulative eﬀect of a large number of unresolved and uncorrelated sources generates a stochastic background the features of which depend on the speciﬁc source. The diﬀerent contributions depend also on the rate of events of the selected type occurred per unit comoving volume, as a function of the cosmological redshift, i.e. on the star formation rate history (SFRH). For those sources that emit over a timescale small compared to the typical timescales of evolution of the underlying galaxy population, the emission of radiation can be considered as “instantaneous”, and directly related to the rate of star formation at that particular value of the cosmological redshift when the star formed. Thus, the detection of such background would provide an insight into the formation and evolution of astrophysical sources in the Universe [15, 16, 17] Finally, GWs will allow to discover unknown sources unreachable by electromagnetic search, and will enlighten regions of the Universe dominated by strong gravitational ﬁelds and high velocities, unaccessible in any other way.
References [1] R.A. Hulse, J.H. Taylor, Astrophys. J. 195, L51, 1975 [2] M. Burgay et al, Letter to Nature 426, 531, 2003 [3] For details of the gravitational experiments see the web sites: www.roma1.infn.it/rog/rogmain.html, www.auriga.lnl.infn.it, www.minigrail.nl/index.html, www.virgo.infn.it, www.ligo.caltech.edu, www.geo600.unihannover.de, www.tamago.ntk.nao.ac.jp, lisa.jpl.nasa.gov [4] E. Gourgoulhon, S. Bonazzola, Proceedings of the International Conference on Gravitational Waves, Sources and Detectors, World Scientiﬁc, Singapore 1996, 51
V. Ferrari
91
[5] G. Ushomirsky, C. Cutler, L. Bildsten, Mon. Not. R. Astron. Soc. 319 n. 3, 902, 2000 [6] R. Smoluchowski, Phys. Rev. L. 24, 923 1970 [7] R. Ruderman, Ap. J. 382, 587, 1991 [8] P.C. Peters, J. Mathews, Phys. Rev. 131, 435, 1963 [9] V. Ferrari, M. D’Andrea, E. Berti, Int. J. Mod. Phys. D9 n. 5, 495, 2000 [10] K. Glampedakis, D. Kenneﬁck, Phys. Rev. D 66, 044002, 2002 [11] S.Chandrasekhar The mathematical theory of black holes. (Oxford: Claredon Press 1984) [12] V. Ferrari, G. Miniutti, J.A. Pons Mon. Not. R. Ast. Soc. 342 624, 2003 [13] C. M. Will, in 2001: A Relativistic Spacetime Odyssey, ed. by I. Ciufolini, D. Dominici, L. Lusanna, World Scientiﬁc Pub. , 247, 2003 [14] O. Benhar, V. Ferrari, L. Gualtieri, Phys. Rev. D 70 n.12, 124015, 2004 [15] V. Ferrari, S. Matarrese, R. Schneider, Mon. Not. R. Astron. Soc. 303, 247, 1999 [16] V. Ferrari, S. Matarrese, R. Schneider, Mon. Not. R. Astron. Soc. 303, 258, 1999 [17] R. Schneider, V. Ferrari, S. Matarrese,S.F. Portegies Zwart, Mon. Not. R. Astron. Soc. 324, 797, 2001
MODEL ANALYSIS OF GRAVITATIONAL SHELL COLLAPSES
MASAFUMI SERIU a
a
Department of Physics, Faculty of Engineering, University of Fukui Bunkyo 391, Fukui 9108507, Japan Email: [emailprotected]
Abstract There are two important conjectures concerning gravitational collapses: (A) The cosmic censorship conjecture and (B) The hoop conjecture. If both the conjectures are valid, it should follow that any suﬃciently elongated object (made of reasonably natural matter) evolves regularly without any singularity formation. It implies that the detailed study of the dynamical evolution of elongated matter provides us with valuable information on the validity of the two conjectures. Keeping this point in mind, we here investigate a cylindrical shellcontraction with rotational pressure, accompanying the radiation of gravitational waves and massless particles. The model has been introduced previously, and some authors have claimed the possibility of a singularityforming evolution of the model, which could suggest the invalidity of either (or both) of the above conjectures. Here we analyze this model once again with more rigor and see whether the claimed singular evolution exists. It is rigorously proved that, as far as the weak energy condition is satisfied outside the shell, the collapsing shell bounces back at some point irrespective of the initial conditions, and no singularity is ever formed. This result of bouncing behavior is compatible with the results for the other known cylindrical shellcontracting models, and conﬁrms the essential importance of the energy condition and the rotational eﬀect in the gravitational collapse. The two conjectures (A) and (B) in this manner disclose no ﬂaws in all the analytical, suﬃciently natural models with cylindrical symmetry known today, suggesting their stable validity in wider class of situations.
93 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 93–102. © 2006 Springer. Printed in the Netherlands.
94 1
Model Analysis of Gravitational Shell Collapses Introduction
Physics of gravitation is one of the most actively investigated themes in modern theoretical physics. Though gravitation is universal and can be experienced daily, dynamics involved in it is far from simple and gives rise to various complicated phenomena. Gravitation has so many aspects that it may not be possible to provide a uniﬁed view of the phenomena. In any case, however, at least it can be said that the nonlinearity always plays the key role in any gravitational phenomenon. Problems of gravitational contractions are the kind of topics in which the nonlinear nature of gravity is involved in a particularly explicit manner. Indeed, its highly nonlinear nature is still preventing us from clarifying several issues questioned even more than 30 years ago. Among such unanswered questions, the following two conjectures have been in particular signiﬁcant and have motivated enormous amount of investigations so far: (A) The cosmic censorship conjecture (Penrose 1969 [1]) and (B) The hoop conjecture (Thorne 1972 [2]). The cosmic censorship conjecture [1] is the claim that no naked singularity is allowed in Nature. Since standard physical description breaks down at spacetime singularities, it is not a theoretically desirable situation if such singularities are visible to distant observers. Thus this conjecture seems quite reasonable at least conceptually, setting aside technical details. Another way of stating this conjecture is that any matter in natural, smooth initial conditions should either evolve regularly or evolve into black holes. The hoop conjecture [2], on the other hand, claims that black holes are formed when and only when the inequality C ≤ 2π · 2GM holds (here C is the largest c2 circumference of the matter, while M is the total mass of the matter). Stated diﬀerently, it claims that black holes are never formed unless matter is compressed into a small, compact region with the size comparable to its Schwarzschild radius. The escape velocity v for the matter with mass M and the size R is estimated by the formula v 2 /2 ∼ GM/R (note R ∼ C/2π), so that the above inequality describes the situation in which the matter is compressed so compactly that its escape velocity exceeds even the speed of light. Thus, the hoop conjecture also seems quite reasonable. Now it is very suggestive if the above two conjectures are combined together [3]. Logical consequence of this combination is that any suﬃciently elongated object (made of reasonably natural matter) should evolve regularly without any singularity formation. As a typical example of the elongated matter, gravitational contractions of cylindrical models have been studied quite intensively [2, 5, 6]. We also investigate one of such cylindrical models in this paper. At this stage it should be noted that even spherical collapses reveal surprisingly complicated features, and one can even construct spherical models which show naked singularity formations [4], though their relevance in view of naturalness and generality should be questioned. At least it should not be overlooked that some researchers take those solutions very seriously and that investigations are still going on [4]. It is not surprising, then, that cylindrical models reveal even
M. Seriu
95
richer features than the spherical models, though the former models have actually been less investigated so far. In this paper, we focus on a particular cylindrical shell model discussed by Pereira and Wang [7, 8]. This model can be interpreted as describing a shell ﬁlled with massless particles rotating in the φdirection; those rotating particles give rise to the rotational pressure pφ of the shell. Regarding this model, it is claimed that there is a special class of solutions in which the shell develops into linelike singularity, which could possibly be naked [7, 8]. If this claim were true, it would mean that either (or both) of the two conjectures mentioned above should be invalid. Thus this model needs to be investigated in more detail to see whether the claim is true, since the original argument [7, 8] was just a preliminary one and has been awaiting for detailed analysis. Here we analyze the model once again with much more care: We then rigorously prove that the shell always bounces back and singularities are never formed as far as the weak energy condition is satisﬁed outside the shell [3]. This bouncing behavior is similar to the one for the case of a cylindrical shell of counterrotating dust particles [6]. The two conjectures [1, 2] in this manner disclose no selfinconsistency in all the analytical models with cylindrical symmetry known today (with suﬃciently reasonable matter content and initial conditions), suggesting their validity in wider class of situations.
2
A dynamical shell in a cylindrical spacetime
We start with a metric of a spacetime possessing a cylindrical spatial symmetry, ds2 = −T (t, r)2 dt2 + R(t, r)2 dr2 + Z(t, r)2 dz 2 + Φ(t, r)2 dφ2 ,
(1)
where (t, r, z, φ) is the standard cylindrical coordinatesystem. Let Σ be a timelike surface in the spacetime deﬁned by Eq.(1), given by r = ρ(t) . It is often convenient to introduce the propertime τ for an observer sitting on the shell, determined by R2 dτ 2 = T 2 (1 − 2 ρ˙ 2 ) dt2 =: X −2 dt2 . (2) T
Here the symbols X := dt/dτ and ◦ := ∂τ will frequently appear below. We note that (eτ , ez , eφ ) := (∂τ , 1/Z.∂z , 1/Φ.∂φ ) forms a set of orthonormal bases on the shell Σ. Now the socalled extrinsic curvature Kαβ of Σ serves as a key tool for the analysis below. The extrinsic curvature of a surface Σ describes how a 3dimentional surface Σ is embedded in a 4dimentional spacetime [9, 10]. Roughly speaking, Kαβ is deﬁned as Kαβ ∼ ∂α n ˆ , where n ˆ stands for the unit normal vector of Σ, while ∂α indicates the derivative along Σ. Since n ˆ t is proportional to ρ, ˙ then, Ktt ∼ ∂t n depends on ρ¨. In this manner, we see that Ktt (or Keτ eτ ) contains the dynamical information of the shell Σ. More explicitly, the normal unit vector for
96
Model Analysis of Gravitational Shell Collapses
Σ is given by ◦
n ˆ µ = XT R(− ρ /X, 1, 0, 0)(trzφ) , and it is straightforward to compute the extrinsic curvature of the shell [3]; Keτ eτ
K ez e z
R ◦◦ R ◦ 1 ρ− ρ (ln R2 )◦ + (T 2 X 2 − 1) ∂n (ln R2 ) XT 2XT 2 X 2T 2 2 2 + (X T − 1) ∂n (ln T 2 /R2 ) 2 XT R ◦ ρ (X 2 T 2 − 1) (ln T 2 /R2 )◦ , − 2 = ∂n ln Z , Keφ eφ = ∂n ln Φ , others = 0 . = −
(3)
(4)
As is already remarked, Keτ eτ depends on ρ◦◦ and describes the dynamics of the shell. According to the standard relativistic junctioncondition formulas [10], when the energymomentum density is concentrated on a 3surface Σ, it gives rise to the discontinuity in the extrinsic curvature across Σ. This result is quite analogous to the relation of the electric charge induced on the surface of a conducting ball with the discontinuity in the electric ﬁeld there. In the present case, we get − −
−
1 [Kez ez ] + Keφ eφ = , κ 1 [Keτ eτ ] − Keφ eφ = pz , κ 1 ([Keτ eτ ] − [Kez ez ]) = pφ . κ
(5)
Here , pz and pφ are interpreted as the energy density of the shell, the pressure in the zdirection and pressure in the φdirection, respectively; [K] indicates discontinuity across Σ (i.e. K(outside) − K(inside)); κ := 8πG is the Einstein’s gravitational constant. As we will see soon, the second equation in Eq.(5) yields the dynamical equation for the shell.
3
The cylindrical shell model with rotational pressure
Let us now consider a particular cylindrical shell model discussed by Pereira and Wang [7, 8]. In this model an interior ﬂat spacetime (described by ds− ) and an exterior cylindrical spacetime (ds+ ) are matched together at a timelike shell Σ: ds2− ds2+
= =
−dt− 2 + dr− 2 + dz− 2 + r− 2 dφ− 2 2γ(ξ)
e
2
2
2
(6) 2
(−dt+ + dr+ ) + dz+ + r+ dφ+
2
,
(7)
M. Seriu
97
where γ is a function of ξ := t+ − r+ only. Let the shell Σ be speciﬁed by r+ = ρ+ (t+ ) from the exterior viewpoint1 . We now note that the Cenergy [11] C and the Einstein tensor Gab outside the shell are given by C
=
Gab
=
1 (1 − e−2γ ) , 8κ γ ka kb , 2 α r+
(8) (9)
where kµ = (α, −α, 0, 0)(trzφ) is a nullvector. Equation (9) suggests that the model may be describing a contracting shell accompanied by outward radiation of massless particles and gravitational waves [7]. Postulating the positivity of the Cenergy (see Eq.(8)) and the weak energy condition (see Eq.(9)), then, the function γ should satisfy γ ≥ 0 , γ ≥ 0 . (10) The quantity X (see Eq.(2)) becomes X :=
2
dt+ e−γ+ = dτ 1 − ρ˙ +2
so that ρ˙ +2 < 1. With the metric Eq.(7), it implies that the radial velocity of the shell is always less than the lightvelocity, as it should be. Now the junction conditions Eq.(5) yield κ =
κpz
=
e−γ+
∆−1 ρ+ 1 − ρ˙ 2+
e−γ+ { (∆ − 1)¨ ρ+ − γ+ ρ˙ + (1 − ρ˙ + )[(∆ + 1)ρ˙ + + 1] ∆(1 − ρ˙ 2+ )3/2
(11)
(12)
1 − ρ˙ 2+ } ρ+ $ % e−γ+ (∆ − 1)¨ ρ+ − γ + ρ˙ + (1 − ρ˙ + )[(∆ + 1)ρ˙ + + 1] , (13) 2 3/2 ∆(1 − ρ˙ + )
−∆(∆ − 1)
κpφ
=
with ∆ :=
$ %1/2 dt− = (1 − e2γ+ )ρ˙ +2 + e2γ+ . dt+
(14)
It should be noted that γ+ > 0 (Eq.(10)) implies ∆ > 1 (Eq.(14)), so that it follows > 0 (Eq.(13)).
1 The suﬃx “+” is for the exterior quantities (ds , t , r + + + and so on), while “” is for the interior quantities. 2 Here the suﬃx + in e.g. ρ + or γ+ implies that the quantity is evaluated on the shell from the viewpoint of the exterior observer.
98
Model Analysis of Gravitational Shell Collapses
Figure 1. Typical evolutions of the shellradius ρ+ (t+ ) (thinner curve) and the rotational pressure pφ (thicker curve). We have set γ+ (r+ − t+ ) = 1016 (r+ − t+ + 100)3 . Initial conditions are ρ+ (0) = 0.1 and ρ˙ + (0) = −0.999. The vertical line indicates 100ρ+ and pφ , while the horizontal line indicates t+ .
Let us set P0 := κpz (constant) in Eq.(13), then, we ﬁnally get the dynamical equation for the shell, ρ¨+
4
=
1 − ρ˙ 2+ ρ˙ + + γ+ (1 − ρ˙ + )[(∆ + 1)ρ˙ + + 1] ρ+ ∆−1 ∆ + (1 − ρ˙+ 2 )3/2 eγ+ P0 . ∆−1 ∆
(15)
Bouncing of the shell without singularityformation
As is mentioned in Introduction, the problems of gravitational contractions inevitably involve highly nonlinear dynamics. Equation (16) is indeed extremely nonlinear in ρ+ (t+ ), considering how ρ+ (t+ ) is included in ∆ and γ+ = γ+ (t+ − ρ+ (t+ )). Before studying the shell dynamics Eq.(16) in detail, it is then helpful to get a rough idea for the dynamics by showing typical numerical results. Figure 1 shows the evolutions of the shellradius ρ+ (t) and the rotational pressure pφ for the case γ+ (r+ − t+ ) = 1016 (r+ − t+ + 100)3 and P0 = 0 with the initial conditions ρ+ (0) = 0.1 and ρ˙ + (0) = −0.999 (i.e. very close to the lightvelocity). The ﬁgure indicates that the rotational pressure prevents the shell from collapsing and that the shell bounces back and its velocity approaches to the lightvelocity after the bouncing. On the other hand, Figure 2 indicates the behavior of the same model as
M. Seriu
99
Figure 2. The same model as Figure 1 with initial conditions, ρ+ (0) = 0.1 and ρ˙ + (0) = −0.1. The vertical line indicates 10ρ+ (for the thinner curve) and pφ (for the thicker curve), while the horizontal line indicates t+ .
Figure 1 but with diﬀerent (milder) initial conditions ρ+ (0) = 0.1 and ρ˙ + (0) = −0.1. The bouncing behavior is again observed. Even though we change initial conditions in various ways, the bouncing behavior similar to the above typical numerical results always appears. These results lead us to doubt the claim by the preceding authors [7, 8] that a linelike singularity (possibly naked) could form in this model. Indeed, as we see below, it is rigorously proved that the shell always bounces back and never forms a singularity in this model, irrespective of initial conditions. Let us set V := −ρ˙ + which takes the value 0 < V < 1 in the contracting phase. Instead of studying Eq.(16), then, it is more convenient to investigate the following; ρ¨+ = ∆
1−V2 V ∆ + γ+ (1 + V )[(∆ + 1)V − 1] + (1 − V 2 )3/2 eγ+ P0 , (16) ρ+ ∆−1 ∆−1
1/2 . where ∆ = V 2 + e2γ+ (1 − V 2 ) Let [[1]], [[2]] and [[3]] be the ﬁrst, the second and the third term on the righthand side of Eq.(16), respectively. Then we observe the following: (i) The ﬁrst term [[1]] produces a strong repulsive force, which prevents the shell from collapsing. The only possible exception may arise when V approaches to 1 faster than ρ+ → 0. Only in this case the term [[1]] can be negligible. Let us call, then, the phasepoint (ρ+ , V ) = (0, 1) the singularityprone point. However, near the singularityprone point, the other two terms ([[2]] + [[3]]) become dominant and they produce a strong repulsive force:
100
Model Analysis of Gravitational Shell Collapses
(ii) Indeed, the second term [[2]], which is an attractive force when V is small, turns to a repulsive force as V → 1. This is easily seen by noting that the function f (V ) := [(∆ + 1)V − 1] is a monotonic, increasing and continuous function with f (0) = −1 and f (1) = 1. Thus f (V ) has only zero in (0,1), which means f (V ) turns from negative to positive as V → 1. (iii) The term [[3]] is nonnegative when P0 ≥ 0 (pressure in zdirection), while it is negative when P0 < 0 (tension). In either case, however, it turns out that the term [[3]] is not important for the whole dynamics. (iv) Setting V = 1 − δ (δ > 0), it is easy to see that the three terms behave as [[1]] ∼ ρδ+ , [[2]] ∼ 1δ and [[3]] ∼ δ 1/2 . Thus even near the singularityprone point, the shell always feels a very strong repulsive force and never collapses.
The above argument is reﬁned to prove Theorem shown below [3]. Deﬁnition We deﬁne the coreregion as a connected region in the phasespace (for the dynamical variable ρ+ ) which contains a neighborhood of the singularityprone point (ρ+ , V ) = (0, 1) and where [[2]] + [[3]] > 0 is satisﬁed [3]. Now we ﬁrst show the following Lemma: Lemma Once the shell enters the coreregion at t = t∗ , it bounces back without reaching zeroradius. Indeed, ρ+ (t) is bounded from below as 2 ρ+ (t) > ρ+ (t∗ ) 1 − ρ˙ + (t∗ ) .
Proof: The claim follows from the fact that ρ¨ =
1 − ρ˙ 2 , with ρ(0) = a , ρ(0) ˙ = −b (a > 0, 1 > b > 0) ρ
is√exactly solved as ρ(t) = ((t − ab)2 + +a2 (1 − b2 ))1/2 , so that ρ+ (t) > ρ(t) ≥ a 1 − b2 > 0.
Then we immediately arrive at Theorem The shell never reaches ρ+ = 0 irrespective of its initial conditions. Proof: If the shell could have ever reached ρ+ = 0, it should have been through the coreregion. However, Lemma indicates that this never happens.
M. Seriu
101
The above Theorem reveals that the singularityformation “solution” claimed by the other authors is actually not a solution and should be discarded. The reader is referred to Ref. [3] for more detailed arguments. Here it suﬃces to give a simple example which is analogous to the present case: Let us consider a diﬀerential equation, y¨ = 1 , (17) the general solution of which is the one indicating a uniform positive acceleration. However, if we artiﬁcially rewrite Eq.(17) as y¨ = β/y˙ ,
(18)
β = y˙ ,
(19)
with a formal contracting solution for Eq.(18) emerges as y(t) = y0 −
√ 2 2 3/2 βt , 3
(20)
provided that β is assumed to be a constant. However, the solution Eq.(20) is in √ any case inconsistent: Noting that β := y˙ = − 2βt, it follows that β = 2t so that it contradicts with the starting assumption of β being a constant. The singularityformation “solution” is obtained by the same trick as Eqs.(18) and (19). Though much more sophisticated arguments are required than this simple example, it turns out that the singularityformation “solution” should be discarded in a more or less similar manner [3].
5
Summary
In this paper, we have investigated the dynamics of a cylindrical shellcontraction model with rotational pressure. As far as the weak energy condition is satisﬁed outside the shell, the collapsing shell always bounces back at some point irrespective of the initial conditions, and escapes from the singularity formation. This result reveals that the singularityformation solution claimed by the preceding authors is not a relevant solution and that it should be discarded. This bouncing behavior is compatible with the results of other cylindrical shellcollapse models and conﬁrms once again the essential importance of the rotational eﬀect in the gravitational collapse. The two conjectures (“the cosmic censorship conjecture” and “The hoop conjecture”) combined together imply that any suﬃciently elongated object (made of reasonably natural matter) should evolve regularly without any singularity formation. The present result is compatible with this statement, and along with other known models with cylindrical symmetry, suggests the stable validity of the two conjectures for wider class of situations.
102
Model Analysis of Gravitational Shell Collapses
This work has been supported by the Japan Ministry of Education, Culture, Sports, Science and Technology with the grant #14740162. References [1] R. Penrose, Riv. Nuovo Cimento 1 (Numero Special), 252 (1969). [2] K. S. Thorne, in Magic Without Magic: John Archibald Wheeler (J. Klauder (ed.), Freeman, San Francisco, 1972), p.231. [3] M. Seriu, Phys. Rev. D69, #124030 (2004). [4] For basic accounts on the spherical models of gravitational contractions, see e.g. P.S. Joshi, Global Aspects in Gravitation and Cosmology (Clarendon, Oxford, 1993). [5] S. L. Shapiro and S. A. Teukolsky, Phys. Rev. Lett. 66, 994 (1991). [6] T. A. Apostolatos and K. S. Thorne, Physical Review D46, 2435 (1992). [7] P. R. C. T. Pereira and A. Wang, Physical Review D62, #124001 (2000). [8] P. R. C. T. Pereira and A. Wang, Physical Review D67, #129902(E) (2003). [9] For a standard account of the extrinsic curvature, see e.g. R. M. Wald, General Relativity (University of Chicago Press, Chicago, 1984), Appendix E.2. Note, however, that we are now concerned with a timelike surface and not a spacelike surface, and some care is needed for dealing with a timelike surface. [10] W. Israel, Il Nuovo Cimento B44, 1 (1966); B48, 463(E) (1967). [11] K. S. Thorne, Phys. Rev. 138, B251 (1965).
THE DEUTERON AND THE BIG BANG
MASATAKA MIZUSHIMA a
a
Department of Physics, University of Colorado, Boulder, Colorado 80309 U.S.A. [emailprotected]
Abstract Following Einstein’s theory of general relativity, the gravitational acceleration of deuteron is shown to be less than that given by Newton’s theory. During the primordial nucleosynthesis, deuterons were more important than proposed by the original BigBang theory. But these deuterons escaped out of the gravitational ﬁelds of the solar system.
1
Gravitational Acceleration of the Deuteron
Newton, in his equation of motion, assumed that the inertial mass is equal to the gravitational mass. Thus everything located at a distance r from a gravitational source with mass M is accelerated toward the source with a gravitational acceleration g, d2 r = g. (1) dt2 Following Einstein’s general relativistic equation of motion [1], using the Schwarzschild metric, we obtain [2] an expression of this gravitational acceleration as 2 dr g = 1− gN , (2) cdt
neglecting a correction factor on the speed of light c. Because Einstein assumed that (ds)2 = g00 (cdt)2 − gαα (dxα )2 , (3)
103 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 103–106. © 2006 Springer. Printed in the Netherlands.
104
The Deuteron and the Big Bang
that is,
2 dxα g00 2 ≤ c ∼ c2 , (4) dt gαα where the correction factor, on gN in eq. (2), is a natural consequence of his theory, because, without this correction factor, the radial speed of a test particle can exceed the speed of light soon, as it falls toward the gravitational source. Inside a deuteron nucleus, one proton and one neutron are orbiting around each other. However, the ground state of the deuteron is mostly 3 S, which means that the quantum mechanical expectation value of dr/dt is zero. Because of the tensor component of the nuclear force, it is shown that a small contribution of a 3 Dstate exists. But the correction factor due to this contribution, in eq. (2), is negligible. The resonating excited state of the deuteron, d∗ , must be an (isospin = 1)state, which means that its energy is about the same as the ground states of a diproton and a dineutron nucleus, both of which are unstable. In the diproton, the Coulomb repulsive force is too strong, and the binding energy of the n+n system is expected to be less than 0.05 MeV. Even though no dineutron system is observed, we know a neutron star exists as a pulsar. We thus expect that an excitedstate deuteron would exist. Actually, a deuteron coupled with a nonresonating γ ray would be virtually excited. We call such a state a dressed deuteron, d∗ . The virtually excited (isospin = 1)state should be a 3 P state, in which nucleons are orbiting around each other with an extremely high speed [3]. d + γ ↔ d∗ .
(5) iωt
The dressed deuteron is described by a wave function ψP (r)e + ψS (r), where ψP (r) and ψS (r) are the space parts of the wave functions for the 3 P and 3 S states, respectively, and ω is the excitation energy in . When the intensity of the γ ray is very high the contributions of the 3 S and the 3 P states will be about the same to each other. Thus the expectationvalue of dr/cdt, the correction term of eq. (2), is about Rω/c where R is the average radius of the nucleon orbits. If the excited 3 P state is at the dissociative level, ω must be 2.73 MeV = 4.15×1021 rad/s. Assuming R = 2 × 10−15 m, we obtain the expectationvalue of dr/cdt to be 0.03 when the nucleon orbits are in the radial (vertical) plane.
2
The Big Bang
Gamow, Alpher, and others [4,5,6,7,8] proposed that nuclei were formed from primordial neutrons in a few minutes and were then frozen, as an unﬁnished buildingup process. A neutron decays into a proton and an electron, n → p + el, with a lifetime of 10.28 minutes [9]. But before all neutron decay, the newborn proton interacts with the surviving neutron, forming a deuteron, p + n → d, in about 3 minutes. Reactions p + d → 3 He, n + 3 He → α, and d + d → α follow very quickly. Thus, the helium/proton ratio is expected to be about 0.3 at this stage of the primordial nucleosynthesis. This ratio is close to what we actually ﬁnd in the universe [10].
M. Mizushima
105
For the abundance of heavy nuclei, the theory implies a relation [8] n(A)σ(A) = n(A + 1)σ(A + 1),
(6)
where σ is the cross section for neutron capture and n is the abundance. The relative abundance of heavy nuclei in meterorites [11,12] can be explained by this simple formula, including the maxima at mass numbers A = 50, 82, and 126. However, an almost fatal objection to this theory lies in the fact that nuclei with mass numbers 5 and 8 are all shortlived with lifetimes 3 × 10−22 (5 Li), 7 × 10−22 (5 He), 2 × 10−20 (8 C), 0.77(8 B), 7 × 10−17 (8 Be), 0.845(8 Li), and 0.119(8 He) in sec, respectively. Bethe, in his famous theory of the production of solar energy, pointed out [13] that reactions 3 He + α → 7 Be and 7 Be + el → 7 Li are possible. Therefore, the gap at the mass number 5 is no trouble to Gamow’s theory. We propose that a large enough number of deuterons or/and dressed deuterons existed during the period of primodial nucleosynthesis so that a reaction d∗ +
7
Li →
9
Be,
(7)
was possible. After 9 Be, synthesis of heavier nuclei follows. It is known that deuterons exist on Earth (as ocean water) with the abundance ratio, d/p, of 1.5×10−4 . However, this abundance ratio must be far below this number in the sun [14]. This isotope ratio in the upper atmosphere of Venus is 0.016 (which is about 100 times the corresponding number in ocean water), although the isotope ratios for N e, Ar, C, and O are almost the same as those on Earth [15]. On Jupiter, the d/p ratio is 2.8 × 10−5 to 7.5 × 10−5 as CH3 D and is 2.1 × 10−5 as HD, and in meteorites, it is 1.3 × 10−4 to 2.0 × 10−4 [16]. Even on Earth, the d/p ratio varies from 3.2 × 10−5 to 1.85 × 10−4 as HD. The temperature of the sun is so high that some deuterons dissociate. But after the primordial period, the outside region of the solar system cooled down so that some deuterons remained on the planets. The large variations in the isotope ratio, however, suggest that some deuterons have escaped from the gravitational ﬁelds of planets (and the sun) by now, because the gravitational acceleration of the dressed deuteron is less than the standard acceleration gN , as shown above. The big bang may have taken place at the time of the galactic nuclear explosions that have created galactic spiral arms [17,18]. The Milky Way galaxy has four pairs of spiral arms. Therefore, there were four big bangs in our galaxy during the past. Near the black hole, the gravitational ﬁeld was very strong, and that situation produced the yelm, the primordial compressed neutron gas, from which the big bang process started. References [1] A. Einstein, Ann. Phys. 49, 769 (1916). [2] G. C. McVittie, General Relativity and Cosmology, section 51, John Wiley & Sons Inc. New York (1956).
106
The Deuteron and the Big Bang
[3] M. Mizushima, Hadronic J. 26, 5 (2003). [4] G. Gamow, Phys. Rev. 70, 572 (1946). [5] R. A. Alpher, H. Bethe, and G. Gamow, Phys. Rev. 73, 803 (1948). [6] G. Gamow, Phys. Rev. 74, 505 (1948). [7] R. A. Alpher, R. Herman, and G. Gamow, Phys. Rev. 74, 1198 (1948). [8] R. A. Alpher and G. Gamow, Proc. Natl. Acad. Sci. U.S. 61, 363 (1968). [9] W. Mampe, P. Ageron, C. Bates, J. M. Pendlebury, A. Steyerl, Phys. Rev. Lett. 63, 2173 (1989). [10] C. E. Rolfs and W. S. Rodney, Cauldrons in the Cosmos, The Univ. Chicago Press, Chicago (1988). [11] H. E. Suess and H. C. Urey, Revs. Modern Phys. 28, 53 (1956). [12] A. G. W. Cameron (1963) as introduced in Nucleosynthesis by W. D. Arnett et al., Gordon and Breach, New York (1965). [13] H. A. Bethe, Phys. Rev. 55, 434 (1939). [14] V. A. Krat and L. M. Pravdjuk, Solar Phys. 61, 279 (1979). [15] T. M. Donahue and J. B. Pollack 29. Origin and Evolution of the Atmosphere of Venus in Venus ed. by D. M. Hunten et al, Univ. of Arizona Press, Tuscon AZ (1983). [16] D. N. Scramm and R. V. Wagoner, What Can Deuterium Tell Us? Phys. Today, Dec. 41 (1974). [17] M. Mizushima, Gravitational Wave and Spiral Galaxy (GravitoRadiative Force. p. 179 in Frontiers of Fundamental Physics 4, ed. by Sudharth and Altaisky, Kluwer Academic, New York (2001). [18] M. Mizushima, GravitoRadiative Force and the Spiral Galaxy in 4th International Conference on Dynamic Systems and Applications, May 2124, 2003, Atlanta GA.
ASTROPHYSICAL APPLICATIONS OF THE THEORY OF SCALE RELATIVITY
LAURENT NOTTALE a
a
CNRS, LUTH, Observatoire de ParisMeudon, F92195 Meudon Cedex, France
Abstract In the framework of the theory of scale relativity, one considers the possibility that spacetime be not only curved but also fractal at large scales. Then one can show that the equation of dynamics in such a space (i.e. the geodesics equation), takes the form of a generalized Schr¨ odinger equation. This approach allows one to suggest new solutions to the problems of the formation and evolution of gravitational structures and of the ’missing mass’. Indeed, such a description leads to a natural selforganization process and involves the apparition of a new potential energy which is a manifestation of the fractal geometry (in similarity with the Newtonian potential being a manifestation of curvature). We suggest that this ’dark potential’ is the cause of the various eﬀects that have up to now been attributed to a postulated nonbaryonic dark matter. We conclude by a brief survey of various observational eﬀects that come in support of such a proposal.
1
Introduction
In its present acceptance, gravitation is understood as the various manifestations of the geometry of spacetime at large scales. Up to now, in the framework of Einstein’s theory [1], this geometry was considered to be Riemannian, i.e. curved. However, in the new framework of scale relativity, the geometry of spacetime is assumed to be characterized not only by curvature, but also by fractality [2] beyond a new relative timescale and/or spacescale of transition, which is an horizon of predictibility for the classical deterministic description. While the concept of
107 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 107–114. © 2006 Springer. Printed in the Netherlands.
108
Applications of scale relativity
fractal spacetime has been ﬁrst introduced in connection with the microscopic quantum theory [3, 4, 5, 6, 7], it can indeed be also applied to the macroscopic realm, but with a diﬀerent interpretation. Now, fractality manifests itself, in the simplest case, in terms of the appearance of a new scalar ﬁeld. We have suggested that this new ﬁeld leads to spontaneous selforganization and may also be able to explain [11, 18], without additional nonbarionic matter, the various astrophysical eﬀects which have been, up to now, attributed to unseen “dark” matter .
2
Gravitational Schr¨ odinger equation
Let us ﬁrst brieﬂy recall the basics of the scalerelativistic theoretical approach. Under three general conditions, namely, {(i) inﬁnity of geodesics (which leads to introduce a nondeterministic velocity ﬁeld), (ii) fractal dimension DF = 2 of each geodesic, on which the elementary displacements are described in terms of the sum dX = dx+dξ of a classical, diﬀerentiable part dx and of a fractal, nondiﬀerentiable ﬂuctuation dξ, (iii) twovaluedness of the velocity ﬁeld, which is a consequence of time irreversibility at the inﬁnitesimal level issued from nondiﬀerentiability}, one can construct a complex covariant derivative that reads d¯ ∂ = + V.∇ − iD∆ , dt ∂t
(1)
where D is a parameter that characterizes the fractal ﬂuctuation, which is such that < dξ 2 >= 2Ddt, and where the classical part of the velocity ﬁeld, V is complex as a consequence of condition (iii) (see [8] for a recent complete demonstration). Then this covariant derivative, that describes the nondiﬀerentiable and fractal geometry of spacetime, can be combined with the covariant derivative of general relativity, that describes the curved geometry. We shall brieﬂy consider in what follows only the Newtonian limit. In this case the equation of geodesics keeps the form of Newton’s fundamental equation of dynamics in a gravitational ﬁeld, ¯ ¯ DV dV φ = +∇ = 0, (2) dt dt m
where φ is the Newtonian potential energy. Introducing the action S, which is now complex, and making the change of variable ψ = eiS/2mD , this equation can be integrated under the form of a generalized Schr¨ odinger equation [3]:
D2 ∆ψ + iD
∂ φ ψ− ψ = 0. ∂t 2m
(3)
Since the imaginary part of this equation is the equation of continuity (Sec. 3), and basing ourselves on our description of the motion in terms of an inﬁnite family of geodesics, P = ψψ † naturally gives the probability density of the particle position [8]. This result is supported by Hermann’s numerical simulations [25]. Even though it takes this Schr¨ odingerlike form, equation (3) is still in essence an equation of gravitation, so that it must come under the equivalence principle [13,
109
L. Nottale
19], i.e., it is independent of the mass of the testparticle. In the Kepler central potential case (φ = −GM m/r), GM provides the natural lengthunit of the system under consideration. As a consequence, the parameter D reads: D=
GM , 2w
(4)
where w is a constant that has the dimension of a velocity. The ratio αg = w/c actually plays the role of a macroscopic gravitational coupling constant [19, 17].
3
Formation and evolution of structures
Let us now compare our approach with the standard theory of gravitational structure formation and evolution. By separating the real and imaginary parts of the Schr¨ odinger equation we obtain, after a new change of variables, respectively a generalized EulerNewton equation and a continuity equation, namely, m(
∂ + V · ∇)V = −∇(φ + Q), ∂t
∂P + div(P V ) = 0, ∂t
(5)
where V is the real part of the complex velocity ﬁeld V. In the case when the density of probability is proportional to the density of matter, P ∝ ρ, this system of equations is equivalent to the classical one used in the standard approach of gravitational structure formation, except for the appearance of an extra potential energy term Q that writes: Q = −2mD
√ P √ . P
2∆
(6)
The existence of this potential energy, (which amount to the Bohm potential in standard quantum mechanics) is, in our approach, readily demonstrated and understood: namely, it is the very manifestation of the fractality of space, in similarity with Newton’s potential being a manifestation of curvature. In the case when actual particles achieve the probability density distribution (structure formation), we have ρ = m0 P ; then the Poisson equation (i.e., the ﬁeld equation) becomes ∆φ = 4πGmm0 ψψ † and it is therefore strongly interconnected with the Schr¨ odinger equation (which is here a new form for the equation of motion). Such a system of equations is similar to that encountered in the description of superconductivity (Hartree equation). We expect its solutions to provide us with general theoretical predictions for the structures (in position and velocity space) of selfgravitating systems at multiple scales [10, 18]. This expectation is already supported by the observed agreement of several of these solutions with astrophysical observational data [3, 13, 17, 14, 15, 20, 21, 16].
110
Applications of scale relativity
Figure 1. Distribution of the semimajor axis of Kuiper belt objects (KBO) and scattered Kuiper belt objects (SKBO), compared with the theoretical predictions (arrows) of probability density peaks for the outer solar system [18] (see text). The existence of probability density peaks for the Kuiper belt at ≈ 40, 55, 70, 90 AU, etc..., has been theoretically predicted before the discovery of these objects [22], and it is now supported by the observational data.
Figure 2. Observed distribution of the semimajor axes of recently discovered exoplanets and inner solar system planets, compared with the theoretical prediction. One predicts the occurence of peaks of probability density for semimajor axes an = GM (n/w0 )2 , where n is integer, M is the star mass and w0 = 144.7 ± 0.7 km/s is a gravitational coupling constant (see text). The probability to obtain such an agreement by chance is P = 4 × 10−5 .
4
Examples of applications to astrophysics
The theory has been able to predict in a quantitative way several new eﬀects in the domain of gravitational structuring [18]. Moreover, these predictions have been successfully checked in various systems on a large range of scales and in terms of a common gravitational coupling constant (or one of its multiples or submultiples) whose value averaged on these systems was found to be w0 = c αg = 144.7 ± 0.7 km/s [13]. Indeed, new structures have been theoretically predicted, then checked by the observational data in a statistically signiﬁcant way, for our solar system, including distances of planets [3, 14] and satellites [16], sungrazer comet peri
L. Nottale
111
Figure 3. Observed distribution of the eccentricities of exoplanets. The theory predicts that the product of the eccentricity e by the quantity n ˜ = 4.83(a/M )1/2 , where a is the semimajor axis and M the parent star mass, should cluster around integers. The data support this theoretical prediction at a probability level P = 10−4 [18].
helions [18], obliquities and inclinations of planets and satellites [20], exoplanets semimajor axes [13, 17] (see Fig. 2) and eccentricities [18] (see Fig. 3), including planets around pulsars, for which a high precision is reached [13, 21], double stars [15], planetary nebula [18], binary galaxies [9] (see Fig. 4), our local group of galaxies [18], clusters of galaxies and large scale structures of the universe [15, 18]. Let us brieﬂy consider the application of the theory to the formation of planetary systems. The standard model of formation of planetary systems can be reconsidered in terms of a fractal description of the motion of planetesimals in the protoplanetary nebula. On lengthscales much larger than their mean free path, we have assumed [3] that their highly chaotic motion satisfy the three conditions upon which the derivation of a Schr¨ odinger equation is based (large number of trajectories, fractality and time symmetry breaking). This description applies to the distribution of planetesimals in the protoplanetary nebula at several embedded levels of hierarchy. Each hierarchical level (k) is characterized by a lengthscale deﬁning the parameter Dk (and therefore the velocity wk ) that appears in the generalized Schr¨ odinger equation describing this subsystem. This hierarchical model has allowed us to recover the mass distribution of planets and small planets in the inner and outer solar systems [14]. It is generally supported by the structure of our own solar system, which is made of several subsystems embedded one in another, namely: *The Sun. Through Kepler’s third law, the velocity w = 3 × 144.7 = 434.1 km/s is very closely the Keplerian velocity at the Sun radius (R = 0.00465 AU corresponds to w = 437.1 km/s). Moreover, one can also apply our approach to the organization of the sun surface itself. One expect the distribution of the various relevant physical quantities that characterize the solar activity at the Sun surface (sun spot number, magnetic ﬁeld, etc...) to be described by a wave function whose stationary solutions read ψ = ψ0 eiEt/2mD . The energy E results from the rotational velocity and, to be complete, should 2 2 also include the turbulent velocity, so that E = (vrot + vturb )/2. This means that
112
Applications of scale relativity
we expect the solar surface activity to be subjected to a fundamental period: τ=
2πmD 4πD = 2 , 2 E vrot + vturb
(7)
The parameter D at the Sun radius is D = GM /2w , then we obtain: τ=
2πGM . 2 + v2 w (vrot turb )
(8)
The average sideral rotation period of the Sun is 25.38 days, yielding a velocity of 2.01 km/s at equator [23]. The turbulent velocity has been found to be vturb = 1.4 ± 0.2 km/s [24]. Therefore we ﬁnd numerically τ = (10.2 ± 1.0) yrs.
(9)
The observed value of the period of the Solar activity cycle, τobs = 11.0 yrs, supports this theoretical prediction. We shall in future works test this proposal by a more detailed study of the activity of the Sun and of other stars. *The intramercurial system, organized on the constant w = 3 × 144 = 432 km/s. The existence of an intramercurial subsystem is supported by various stable and transient structures observed in dust, asteroid and comet distributions (see [18]). We have in particular suggested the existence of a new ring of asteroids, the ‘Vulcanoid belt’, at a preferential distance of about 0.17 AU from the Sun. *The inner solar system (earthlike planets), organized with a constant wi = 144 km/s (see Fig. 2). *The outer solar system (Jovian planets), organized with a constant wo = 144/5 = 29 km/s (see Fig. 1), as deduced from the fact that the mass peak of the inner solar system lies at the Earth distance (n = 5). The recently discovered Kuiper and scattered Kuiper belt objects (Fig. 1) show peaks of probability at n = 6 to 9 [18], as predicted before their discovery [22]. We have suggested more than ten years ago [3, 22], before the discovery of exoplanets, that the theoretical predictions from this approach should apply to all planetary systems, not only our own solar system. Meanwhile more than 120 exoplanets have been discovered, and the observational data support this prediction in a highly statistically signiﬁcant way (see [13, 17, 18] and Figs. 2 and 3). A full account of this new domain would be too long to be included in the present contribution. We have given here only few typical examples of these eﬀects (see the ﬁgures) and we refer the interested reader to the review paper Ref. [18] and references therein for more detail.
5
Possible solution to the “dark matter” problem
In the case (ii) of isolated test particles, the density of matter ρ may be nearly zero while the probability density P does exist, but only as a virtual quantity that determines the potential Q. In this situation, even though there is no or few
L. Nottale
113
Figure 4. Deprojection of the intervelocity distribution of galaxy pairs (Tricottet and Nottale, reported in [18]) from the SchneiderSalpeter catalog with precision redshifts. The main probability peak is found to lie at 144 km/s (plus secondary peaks at 72=144/2 km/s and 24=144/6 km/s), in agreement with the exoplanet and inner solar system structuring (see Fig. 2).
matter at the point considered (except the test particles that are assumed to have a very low contribution), the eﬀects of the potential Q are real (since it is the result of the structure of the geodesics twoﬂuid). We have therefore suggested [11, 18, 12] that this extra scalar ﬁeld, which is a manifestation of the fractality of space, may be responsible for the various dynamical and lensing eﬀects which are usually attributed to unseen “dark matter”. This interpretation is supported by the fact that, for a stationary solution of the gravitational Schr¨ odinger equation, one gets the general energy relation E = φ + Q + 21 mV 2 , where E/m can take only quantized values. This result can be applied, as an example, to the motion of bodies in the outer regions of spiral galaxies. In these regions there is practically no longer any visible matter, so that the Newtonian potential (in the absence of additional dark matter) is Keplerian. While the standard Newtonian theory predicts for the velocity of the halo bodies v ∝ φ1/2 , i.e. v ∝ r−1/2 , we predict v ∝ (φ + Q)/m1/2 , i.e., v = constant. More speciﬁcally, assuming that the gravitational Schr¨ odinger equation is solved for the halo objects in terms of the fundamental level wave function, one ﬁnds Qpred = −(GM m/2rB )(1 − 2 rB /r), where rB = GM/w02 . This is the result systematically observed in spiral galaxies (i.e., ﬂat rotation curves) which has motivated (among other eﬀects) the assumption of the existence of dark matter.
6
Conclusion
We have recalled in the present contribution how the theory of scale relativity is able to yield equations that describe a natural tendency to make structures, as a very consequence of the fractality of space.
114
Applications of scale relativity
Moreover, we have suggested that the eﬀects tentatively attributed to unseen matter are simply the result of this geometry of spacetime. In this proposal, spacetime is not only curved but also fractal beyond some given relative time and spacescales. While the curvature manifests itself in terms of the Newton potential, fractality would manifest itself in terms of the new scalar potential Q, and then ﬁnally in terms of the anomalous dynamics and lensing eﬀects. References [1] A. Einstein, Annalen der Physik 49, 769 (1916), translated in The Principle of Relativity (Dover, 1923, 1952) [2] B. Mandelbrot, The Fractal Geometry of Nature (Freeman, San Francisco, 1982) [3] L. Nottale, Fractal SpaceTime and Microphysics: Towards a Theory of Scale Relativity (World Scientiﬁc, Singapore, 1993) [4] L. Nottale and J. Schneider, J. Math. Phys. 25, 1296 (1984) [5] G. N. Ord, J. Phys. A: Math. Gen. 16, 1869 (1983) [6] L. Nottale, Int. J. Mod. Phys. A 4, 5047 (1989) [7] M. S. El Naschie, Chaos, Solitons & Fractals 2, 211 (1992) [8] M.N. C´el´ erier & L. Nottale, J. Phys. A: Math. Gen. 37, 931(2004) [9] L. Nottale, Chaos, Solitons & Fractals 7, 877 (1996) [10] L. Nottale, Astron. Astrophys. 327, 867 (1997) [11] L. Nottale, in Frontiers of Fundamental Physics, Proceedings of Birla Science Center Fourth International Symposium, 1113 december 2000, edited by B. G. Sidharth and M. V. Altaisky, (Kluwer Academic, 2001) p. 65 [12] L. Nottale, Chaos, Solitons and Fractals 16, 539 (2003) [13] L. Nottale, Astron. Astrophys. Lett. 315, L9 (1996) [14] L. Nottale, G. Schumacher and J. Gay, Astron. Astrophys. 322, 1018 (1997) [15] L. Nottale and G. Schumacher, in Fractals and beyond: complexities in the sciences, edited by M. M. Novak (World Scientiﬁc, 1998) p. 149 [16] R. Hermann, G. Schumacher, and R. Guyard, Astron. Astrophys. 335, 281 (1998) [17] L. Nottale, G. Schumacher, E.T. Lef`evre, Astron. Astrophys. 361, 379 (2000) [18] D. Da Rocha, L. Nottale, Chaos, Solitons & Fractals 16, 565 (2003) [19] A.G. Agnese, R. Festa, Phys. Lett. A227, 165 (1997) [20] L. Nottale, Chaos, Solitons & Fractals 9, 1035 (1998) [21] L. Nottale, Chaos, Solitons & Fractals 9, 1043 (1998) [22] L. Nottale, in “Chaos and Diﬀusion in Hamiltonian Systems”, ed. D. Benest and C. Froeschl´ e, (Fronti` eres, 1994), p. 173 [23] J.C. Pecker and E. Schatzman, Astrophysique g´en´ erale (Masson, Paris, 1959) [24] K.R. Lang, Astrophysical Formulae (SpringerVerlag 1980) [25] R. Hermann, J. Phys. A: Math. Gen 30, 3967 (1997)
MAKING MAPS OF THE REESSCIAMA EFFECT
` MARIUS J. FULLANA I ALFONSO ´ ´ d DIEGO P. SAEZ MILAN a b c d
a,b,c
,
Institut de Matem` atica Multidisciplinar Univ. Polit`ecnica de Val`encia, 46022 Val`encia Departament de Matem` atica Aplicada, Univ. Polit`ecnica de Val`encia, 46022 Val`encia email adress:[emailprotected] Departament d’Astronomia i Astrof´ısica, Univ. de Val`encia, 46100 Burjassot, Spain
Abstract The anisotropies of the Cosmic Microwave Background are caused by several physical phenomena. Those anisotropies produced by cosmological inhomogeneities placed far away from (close to) the last scattering surface are called secondary (primary). We focus our attention on one of the secondary anisotropies: the ReesSciama (RS) eﬀect due to the time variation of the peculiar gravitational potential of nonlinear cosmological inhomogeneities. A full description of a distribution of these structures requires Nbody simulations with appropriate resolutions and box sizes. CMB photons cross a periodic universe formed by repeated Nbody boxes. Periodicity leads to a magniﬁcation of the RS eﬀect. A method to avoid this magniﬁcation is described and, then, it is applied to simulate maps of the RS eﬀect which are being statistically analysed. RS is a subdominant eﬀect which appears superimposed to other ones in nature. Each one of these eﬀects must be studied in detail to make possible an eventual separation of components in maps of future CMB experiments as PLANCK.
115 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 115–122. © 2006 Springer. Printed in the Netherlands.
116 1
Making Maps of the RS effect Introduction
The CMB radiation was discovered in 1965 [1]. A few years before, it had been teoretically predicted in the framework of the Big Bang cosmological model (see [2] and references therein). This radiation plays a crucial role in modern cosmology and, consequently, many experiments have been designed to measure its spectrum, polarisation, and anisotropy. These measurements are to be compared with theoretical predictions based on a cosmological background containing some type of cosmological structures. Now, a few words about the nature and importance of the CMB are presented. In the primordial universe, baryonic matter was strongly coupled to radiation. Protons and electrons were not conﬁned in atoms, in other words, these particles were free in the socalled primeval plasma. In that situation, photons could not travel freely because they frequently interacted with the plasma free electrons and, consequently, the universe was opaque; afterwards, as a result of expansion, the universe cooled and became transparent. In fact, when the temperature was low enough ( 3500 K ), protons captured electrons and formed stable hydrogen atoms (recombination). The absence of free electrons led to the decoupling between matter and radiation at redshift Z ≈ 103 . The recombinationdecoupling process lasted for a time corresponding to a redshift increment ∆Z ≈ 80. This short duration allows us to consider a Last Scattering Surface (LSS) separating the opaque ionized Universe from the transparent neutral one. From that period on, CMB photons travelled almost freely. Astrophysical phenomena could have produced a reionization, which would have altered the CMB radiation at low redshifts ( Z 20 ). Now, some comments about the information supplied by the study of the CMB are made. This radiation appears to be very isotropic (see for example [3]) and, furthermore, it has a blackbody spectrum at temperature T ≈ 2.73K, and a low level of linear polarisation. Small angular temperature ﬂuctuations (anisotropies) have been detected in the CMB (see for instance [4]). In a fully homogeneous universe, the CMB would be absolutely isotropic. The anisotropies are due to the fact that photons coming along distinct directions undergo diﬀerent frequency shifts, it occurs because these photons cross distinct inhomogeneous regions. Hence, CMB anisotropies give information about the distribution of inhomogeneities (large scale structure) evolving in the universe. On account of these considerations, it is obvious that the observation of the CMB anisotropies allows us to look for the best theory explaining structure formation in the framework of the Big Bang model.
2
The anisotropies of the CMB
The small angular dependence of the CMB temperature can be understood taking into account the following formula for the temperature contrast: ∆T ∆T 1 ∂φ (n) = + n · (vO − vE ) − (φO − φE ) − 2 (n) dt (1) T T 3 ∂t E
´ez M.J. Fullana, D.P. Sa
117
where n is the unit vector of the line of sight, v is the peculiar velocity, and φ the peculiar gravitational potential. E stands for emission and O for observation. The terms of the r.h.s. of Eq. (1) have the following interpretation: • ∆T T E is the initial anisotropy due to temperature ﬂuctuations on the LSS produced by previous evolution.
• n · vE is a Doppler eﬀect due to peculiar motions on the LSS. • n ·vO is another Doppler eﬀect due to the motion of the observer with respect to the background universe. This component has dipolar structure. • − 31 (φO ) is a monopole due to the peculiar gravitational potential at the position observer. •
1 3
(φE ) is the SachsWolfe eﬀect due to the peculiar gravitational potential created by the inhomogeneities located close to the LSS. ! • − 2 ∂φ ∂t dt is the contribution due to time variations of φ, this term is called the integrated SachsWolfe (ReesSciama) eﬀect when the potential is created by large scale linear structures (nonlinear structures). Hence, the anisotropy given by Eq. (1) is the superimposition of various eﬀects. Among them, the RS eﬀect is a secondary subdominant anisotropy produced by the time variation of the potential φ created by nonlinear structures at Z ≤ 50. It is given by the following formula: t0 ∆T ∂φ(x, t) ≈ −2 dt (2) T ∂t t50
The evolution of strongly nonlinear structures is a hard problem. Only two methods are known to deal with it: in the ﬁrst one, only one symmetric structure is described using exact solutions of Einstein equations, and in the second method, a realistic statistical distribution of asymmetric structures is evolved using Nbody simulations. Several studies of the anisotropy generated by a unique symmetric cosmological inhomogeneity (Great Attractor, voids) have been done. Two main models have been used for this purpose: the LemaˆıtreTolmanBondi model (see for example [2]) or the SwissCheese model (see for instance [5]). Although these models led to important qualitative ideas about the ReesSciama eﬀect with a moderated computational cost, only Nbody simulations describing a set of inhomogeneities can lead to a realistic prediction of the RS eﬀect. These predictions are based on a periodic Universe ﬁlled by identical Nbody boxes where CMB photons move along null geodesics. By this reason raytracing through Nbody simulations is ﬁrst used to get maps of the RS eﬀect and, then, these maps are statistically analysed to estimate deviations from gaussianity; see for example [6, 7]; in this last paper, the raytracing procedure used here was described in detail. It is based on the existence of a preferred direction minimising the magniﬁcation of the RS eﬀect in a periodic universe.
118
Making Maps of the RS effect
Figure 1. APS of the dominant anisotropies. Model parameters: Ω0 = 1.0, ΩΛ = 0.73, Ωb = 0.04, Ωc = 0.23, h = 0.71 and reionization at Z = 17.
In this paper, the Angular Power Spectrum (APS) of the CMB is calculated using the code CMBFAST [8] for the model suggested by WMAP observations [9]. This APS is displayed in Fig. 1 and no subdominant anisotropies as RS, lens deformations, SunyaevZel’dovich eﬀect and so on have been considered at all.
3
Raytracing methods
In standard raytracing procedures, photons move along null geodesics in a periodic universe. Identical Nbody boxes are used to tile the space at any given time (periodicity). All the boxes evolve in the same way according to Nbody predictions. Equation (2) must be integrated along each null geodesic (direction) in order to estimate the RS eﬀect. This integration must be performed along a set of appropriate directions to get a RS map. The main problems with this type of raytracing is that CMB photons can be affected by the same structures in successive boxes and, consequently, a wrong magniﬁcation of the resultingwidth=9.0cm,height=9.0cm,angle=270 eﬀect appears. This magniﬁcation depends on the propagation direction and it is particularly great for directions parallel to the box edges. In [10] one of these raytracing procedures involving multiple plane projection is described. It is the classical method. In order to avoid the above problem, the boxes could be moved (rototraslations) when the photon crosses from a box to the next one. Thus, a given photon would move through separated independent regions in successive boxes. Nevertheless, this method introduces unavoidable discontinuities in the crossing points, which aﬀect the integral of the r.h.s. of Eq. (2). Another procedure was proposed by [11], it is the tiling raytracing method, in which a diﬀerent Nbody simulation describes structure evolution in each box crossed by the CMB photon. In this way, the above periodicity eﬀect is surmounted, but discontinuities at crossing points appear as it occurs for rototraslations. Furthermore, many Nbody simulations –large computational cost– must be
´ez M.J. Fullana, D.P. Sa
119
performed. Recently, another method based on a unique Nbody simulation has been proposed [7]. It is based on the existence of preferred directions. A photon moving along these directions crosses independent regions of the simulation box through thousands of Megaparsecs and, consequently, no magniﬁcation of the RS eﬀect appears; at least, if an appropriate cutoﬀ is performed to avoid scales larger than a certain spatial scale Smax satisfying the following conditions: (a) scales smaller than Smax are well evolved in the Nbody simulation, and (b) the scale Smax is greater than the distance, L, between the regions crossed by the CMB photon in two neighbouring boxes. No discontinuities appear in the crossing points because only a simulation is used and no rototraslations are performed. Some comments about our simulations are now necessary. A PM (particlemesh) Nbody code designed by V. Quilis and D. S´ aez (see [12] and references cited therein) is used. The initial redshift: zin = 5 is assumed. The box size is: Lbox = 256 M pc. The preferred direction is: θ = 77.2◦ , φ = 12.6◦ . The separation, L, between crossed regions in neighbouring boxes is: 52 M pc. Spatial scale larger than Smax are eliminated in the peculiar gravitational potential (not in the Nbody simulation). Various scales Smax ≤ 60 M pc have been tried. The photon crosses ∼ 30 boxes (∼ 8000 M pc) before reentering a box by the zone where it was placed at the initial position (at zin = 5). Two types of simulations have been performed, in the ﬁrst one (hereafter poor resolution simulations) the cell size is of 2 M pc and 1283 particles of dark matter are used, whereas in the second type (hereafter medium resolution simulations) the cell size is 1 M pc and the number of particles is 2563 . Higher resolution simulations are being developed but the computational cost is large. A set of directions ensures a uniform coverage of ∼ 2◦ × 2◦ in the sky. A RS map is generated using these directions.
4
Results
Two particular RS simulations (preferred directions plus cutoﬀ) are now displayed. Fig. 2 corresponds to a low resolution simulation with the following characteristics: (1) spatial resolution 2 M pc, (2) maximum spatial scale Smax = 60 M pc, (3) number of particles 1283 , (4) 1.82◦ × 1.82◦ map, (5) maximum temperature contrast 1.92 × 10−6 , and (6) minimum temperature contrast −1.85 × 10−6 . Fig. 3 is a particular middle resolution simulation with: (1) spatial resolution 1 M pc, (2) maximum spatial scale 128 M pc, (3) number of particles 2563 , (4) 1.82◦ × 1.82◦ map, (5) maximum temperature contrast 1.70 × 10−6 , and (6) minimum temperature contrast −1.16 × 10−6 . The simulations look well, no rare structures appear in them, but what about their APS? In the left panel of Fig.4, we present the APS of the low resolution simulation of Fig. 2, whereas in the right panel of the same Figure, we show the ratio between this APS and that of Fig. 1. Since this ratio is a few times 10−3 in the wide
120
Making Maps of the RS effect
Figure 2. Low resolution RS simulation.
Figure 3. Middle resolution RS simulation.
Figure 4. Left panel: APS of Fig. 2 simulation. Right panel: Ratio between APS of left panel and APS of Fig. 1.
´ez M.J. Fullana, D.P. Sa
121
Figure 5. Left panel: APS of Fig. 3 simulation. Right panel: Ratio between APS of left panel and APS of Fig. 1.
Figure 6. Deﬁning mdirection correlation functions in maps.
interval under consideration, we conclude that the RS eﬀect is much smaller than the dominant anisotropy due to other eﬀect (see Section 2 and Fig. 1). Fig. 5 has the same structure as Fig. 4, but it corresponds to the medium resolution simulation of Fig. 3. Although the resolution is greater than in the case of Figs. 2 and 4, results are similar. The amplitude of the spectrum (ratio) presented in the left (right) panel of Fig. 5 is smaller than that of the left panel of Fig. 4, but it is only due to the fact that we are comparing two arbitrary simulations and the amplitude (ratio) varies from simulation to simulation for a given resolution. We have veriﬁed that the average spectra corresponding to twenty simulations are rather similar for low and medium resolutions. Simulations with higher resolution are in progress.
5
Conclusions and Perspectives
A raytracing procedure based on a preferred direction and an appropriate cutoﬀ has been designed. It has been used to get maps of the RS eﬀect, which appears to be a subdominant contribution to the total anisotropy of the CMB. The maps are being statistically analysed in order to estimate deviations from Gaussianity due to RS. mdirection correlation functions of the form Cm =
122
Making Maps of the RS effect
ζ(n1 )ζ(n2 ) · · · ζ(nm ) are being calculated from our maps (for m ≤ 6), where ζ = ∆T /T . The chosen sets of m directions draw –on the LSS– the ﬁgures displayed in Fig. 6. In Gaussian statistics, Cm functions vanish for odd m values, whereas in the even case, all these functions can be written in terms of function C2 . For Gaussian statistics and sets of directions with √ the above relative positions one obtains the relation: C4 (α) = 2[C2 (α)]2 + [C2 ( α)]2 , which implies C4 (0) = 3[C2 (0)]2 . The relation corresponding to C6 (α) is more complicated and it is not written. For α = 0, it leads to C6 (0) = 15[C2 (0)]3 . NonGaussianity implies either non vanishing correlations for odd m or violations of the above relations between the even correlations and C2 . We are computing Cm in order to distinguish the RS eﬀect, which is subdominant and nonGaussian, from the dominant eﬀects, which are Gaussian. This work is in progress. We are preparing a new code based on the meshreﬁned AP3 M Nbody algorithm designed and used by the HYDRA Consortium [13]. Great resolutions will be considered with the new Nbody.
6
Acknowledgments
This work has been partially supported by the Generalitat Valenciana under the projects GRUPOS03/062 and GRUPOS03/170, by the Spanish DGI grant BFM20012783, and by the Spanish MCyT (project AYA200308739C0202 partially funded with FEDER). One of us, M.J. Fullana i Alfonso, wants to dedicate this work to his mother, Eug`enia M. Alfonso i N` acher, who died recently. References [1] A.A. Penzias and R.W. Wilson, ApJ Lett. 142 (1965) 419 [2] M.J. Fullana, J.V. Arnau and D. S´ aez, Vistas in Astronomy 41 (1998) 467 [3] J.C. Mather et al., ApJ 420 (1994) 439 [4] B.E. Corey and D.T. Wilkinson, Bull. Am. Astron. Soc. 8 (1976) 351 [5] M.J. Rees and D.W. Sciama, Nature 217 (1968) 511 [6] A.M. Aliaga, V. Quilis, J.V. Arnau, and D. S´ aez, MNRAS 330 (2002) 625 [7] P. Cerd´ aDur´ an, V. Quilis, and D. S´ aez, Phys. Rev. 69D (2004) 043002 [8] U. Seljak and M. Zaldarriaga, ApJ 469 (1996) 437 [9] C.L. Bennet et al., ApJS 148 (2003) 1 [10] B. Jain, U. Seljak and S. White, ApJ 530 (2000) 547 [11] M. White and W. Hu, ApJ 537 (2001) 1 [12] V. Quilis, J.M. Ib´ an ˜ ez, and D. S´ aez, ApJ 502 (1998) 518 [13] H.M.P. Couchman, ApJ Lett. 368 (1991) 23
STATIONARY POINTS OF SCALAR FIELDS COUPLED TO GRAV ITY
a ¨ , G. MELKONYAN a , F. PARADIS a , H. KROGER b,c S.G. RUBIN a
b c
D´epartement de Physique, Universit´e Laval, Qu´ebec, Qu´ebec G1K 7P4, Canada Moscow State Engineering Physics Institute, Kashirskoe sh., 31, Moscow 115409, Russia Center for Cosmoparticle Physics “Cosmio”, Moscow, 125147, Russia
Abstract We investigate the dynamics of gravity coupled to a scalar ﬁeld using a noncanonical form of the kinetic term. It is shown that its singular point represents an attractor for classical solutions and the stationary value of the ﬁeld may occur distant from the minimum of the potential. In this paper properties of universes with such stationary states are considered. We reveal that such state can be responsible for modern dark energy density.
1
Introduction
Scalar ﬁelds play an essential role in modern cosmology. A realistic scenario of the origin of our universe is based on the inﬂationary paradigm and a vast majority of inﬂationary models use the dynamics of scalar ﬁelds. Here we show in a natural way how to produce a class of eﬀective potentials of the scalar ﬁeld. It is achieved by invoking the simplest form of a potential but noncanonical kinetic terms. The drawback of using scalar ﬁelds is the occurence of potentials with unnatural forms. For example, potentials have to be extremely ﬂat to be consistent with the standard inﬂationary scenario [1].
123 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 123–130. © 2006 Springer. Printed in the Netherlands.
124
Stationary Points in Scalar Fields
We consider an action which couples gravity to a scalar ﬁeld. The latter has a nontrivial kinetic term K(ϕ) = 1. By supposition, it contains a singular point of the following form K(ϕ) = M n /(ϕ − ϕs )n , n = −1, 1, 2 ,
(1)
and investigate their eﬀect on the scalar ﬁeld dynamics, see also [6]. Here M is some model parameter. The existence of the singular kinetic term opens a rich variety of possibilities for the construction of cosmological models. The well known Brans  Dicke model [4] is one of the particular case. It is known that appropriate change of the ﬁeld variable, leads to the standard form of kinetic term, i.e. K = ±1 what can be done during inﬂationary stage. The situation becomes much more complex when the ﬁeld ﬂuctuates around a singular point. The equation of motion for a uniform ﬁeld distribution has the form n ϕ¨ + 3H ϕ˙ − ϕ˙ 2 + V (ϕs ) (ϕ − ϕs )n /M n = 0 . 2(ϕ − ϕs )
in the FriedmannRobertsonWalker universe, H is the Hubble parameter and expression (1) is taken into account. The ﬁeld value ϕs is a stationary solution for any smooth potential V and n > 0 provided that ϕ˙ = o(ϕ − ϕs ). The cosmological energy density of the vacuum is connected usually with one of its potential minima. Here the situation is diﬀerent  the vacuum state is connected with the singular point of the kinetic term K(ϕ). To prove this statement, we consider the simplest form of the potential V (ϕ) = V0 + m2 ϕ2 /2 . In the following we will only consider the class of models characterized by the set of parameters m, V0 , M . The stationary state ϕs is chosen in a way such that it ﬁts the cosmological Λterm (see review [5]), V0 + m2 ϕ2s /2 = V (ϕs ) = Λ .
(2)
The energy density ∼ Λ in a modern epoch is small compared to any scale during the inﬂationary stage, which allows us to neglect it whenever this is possible and obtain the relation ϕs ∼ (3) = 2 V0 /m. To proceed, an auxiliary variable χ will be taken into account. We suggest the substitution of variables ϕ → χ in the form dχ = ± K(ϕ)dϕ, K(ϕ) > 0 , (4)
what leads to the action in terms of the auxiliary ﬁeld χ √ R 1 S = d4 x −g + sgn(χ) ∂µ χ∂ µ χ − U (χ) , 16πG 2
(5)
where the potential U (χ) ≡ V (ϕ(χ)) is a ’partly smooth’ function. Its form depends on the form of the initial potential V (ϕ), the form of the kinetic term and the position of the singularities at ϕ = ϕs . Now let us consider some particular cases of K(ϕ), [7].
¨ ger et al. H. Kro
125
Figure 1. Potential in terms of auxiliary ﬁeld χ for the case n = 1. If χ < 0 the auxiliary ﬁeld behaves like a phantom ﬁeld moving classically to the local extremum at χ = 0.
2
Eﬀective potentials
The case n = 1 In this case formulas (1,4) give the action (5) with the potential 1 χ2 2 U (χ) ≡ V (ϕ(χ)) = V0 + m2 (ϕs + sgn(χ) ) for ϕs > 0. 2 4M
(6)
Here and below we keep the one  to  one correspondence between the physical variable ϕ and auxiliary variable χ in the intervals: ϕ < ϕs → χ < 0; ϕ > ϕs → χ > 0. If the auxiliary ﬁeld starts from χ > 0, it ﬁnally approaches the singular point χ = 0 (see Fig.[1]). If the ﬁeld obeys χ < 0, than the auxiliary ﬁeld behaves like a phantom ﬁeld, which climbs up to the top of the potential and hence tends to the singular point as well. Finally, the ﬁeld settles down in the vicinity of the singular point χ = 0 (ϕ = ϕs ). One concludes that this point is the stationary point and the vacuum energy density equals to V (ϕs ), (see Eq.(2)) rather than to V0 . The value of parameters can be estimated if we interpret the auxiliary ﬁeld as the inﬂaton which in addition is responsible for the dark energy. In the course of inﬂation, a slow roll condition [1] should hold. This happens if the parameters take the values M ∼ MP ; V0  ∼ MP4 ; m ∼ 10−12 MP . (7) The parameter m is small in order to ﬁt data of large scale temperature ﬂuctuations [8]. The problem of smallness of the vacuum energy density, Λ = 10−123 MP4 remains topical in this approach although the situation has changed. As mentioned above,
126
Stationary Points in Scalar Fields
the smallness of the vacuum energy density is usually connected with the smallness of a potential minima. In the case considered here the modern energy density is determined by the singular point ϕs of the noncanonical kinetic term (see Eq.(2)). The smallness of Λ may be realized if the singular point ϕs is placed very close to the zero point ϕ0 of the potential (V (ϕ0 ) = 0). A suitable interval is Λ ϕs ∈ [ϕ0 , ϕ0 +∆ϕ], ∆ϕ ≡ −2V0 /m2 + 2Λ/m2 − −2V0 /m2 ∼ . (8) = m 2 V0 
This interval is extremely small, making its explanation still diﬃcult. The next section is devoted to a discussion of this subject matter. We will show that a probabilistic approach may help to obtain a selfconsistent picture. The case n = 2 Now formulas (1, 4) give the action (5) with the potential " #2 1 U (χ) = m2 ϕ2s 1 + sgn(ϕs ) · sgn(χ) · eχ/M + V0 . (9) 2
In the case ϕs < 0; ϕ > ϕs the potential (9) is highly asymmetric, and the behavior of the inﬂaton is rather diﬀerent at χ < 0 from that at χ > 0. If we suppose that the inﬂation starts with χ = χin > 0, the picture is similar to the improved quintessence potential [10]. It is free of problems with the description of the radiationdominated stage during Big Bang nucleosynthesis which could explain the modern distribution of chemical elements [9]. The chosen parameter values M ∼ MP , m ∼ MP , V0  ∼ 10−14 MP4 (10) permit a suitable inﬂationary stage and they are in agreement with observations of temperature ﬂuctuations [8]. The case n = −1 A nontrivial situation occurs when the kinetic function has not a pole but a root at some point, K(ϕ) = (ϕ−ϕs )/M . Let the initial ﬁeld value obey ϕ = ϕin > ϕs ∼ MP , which gives rise to the inﬂation in early universe. Then the potential of the auxiliary ﬁeld χ becomes U (χ) =
1 2 m (ϕs + sgn(χ) · γχ2/3 )2 + V0 . 2
(11)
U (χ) is ﬁnite at χ = 0 but its derivative is singular. Classically, the situation looks very strange  the singular point attracts the solution, but forbids it to stay there forever. It looks is similar to quantum mechanics, in particular to the case of an electron in the Coulomb ﬁeld. The potential (11) behaves like χ4/3 at large ﬁeld values. It leads to standard inﬂation with moderate ﬁne tuning of the parameters. Namely M ∼ MP ,
m ∼ 10−6 MP ,
V0 ∼ 10−12 Mp4 .
(12)
If ϕs > 0, the ﬁeld ϕ will ﬂuctuate around some critical point with energy density (2). This motion never attenuates completely because classical stationary points are absent in this region.
¨ ger et al. H. Kro 3
127
Probabilistic approach to the form of action
Here we have investigated several speciﬁc forms of eﬀective potentials. Many other potentials and kinetic terms have been discussed in the literature. A substantial number of them does not contradict observational data. In this context the question can be raised and need to be answered: Why is it that particular shape of potential and kinetic term is realized in nature? What are the underlying physical reasons? Some theoretical hints on the form of the potential have been given in the context of supergravity, which predicts an inﬁnite power series expansion in the scalar ﬁeld potential [11]. Its minima, if they exist, correspond to stationary states of the ﬁeld. The potential, due to an inﬁnite number of terms in a power series could correspond to a function with an inﬁnite set of potential minima. This assumption with randomly distributed minima appears to be selfconsistent [2]. In the low energy regime it is reasonable to retain only a few terms (lowest powers in the Taylor expansion) of the scalar ﬁeld [12]. In the vicinity of each of those minima the potential has a particular form. A similar behavior may hold also for the kinetic term. If the scalar ﬁeld is responsible for the inﬂation, each local minimum produces an individual universe, diﬀerent from any other universe. Our own universe is associated with a particular potential minimum, not necessarily located at ϕ = 0. The observed smallness of the value of the Λterm is explained usually in terms of a more fundamental theory like supergravity or the anthropic principle. Our point of view is that we have to merge these approaches. The more fundamental theory supplies us with an inﬁnite set of minima of the potential. These minima having an individual shape are responsible for the formation of those universes used in the anthropic consideration. Practically, it could be performed in the framework of the random potential [2, 3] and the kinetic term of the scalar ﬁeld discussed in sect.(1). A part of such potential and the kinetic term in a ﬁnite region of the ﬁeld ϕ are represented in Fig.[2]. Fluctuations of the scalar ﬁeld being generated at high energies in the inﬂationary stage move classically to stationary points. Those of them who reach stationary points with appropriate energy density could form a universe similar to our Universe. This energy density (∼ 10−123 MP4 ) is the result of a small value of the concrete potential minimum or a small value of the diﬀerence ϕs − ϕm , where ϕm is a zero of the potential (V (ϕm ) = 0). The fraction of such universes is relatively small, but nevertheless is inﬁnite because of an inﬁnite number of stationary states. How could one decide which of the stationary points is most promising? To get an idea we should recall that the main defect of the inﬂationary scenario is the smallness of some intrinsic parameter compared to unity. It is the value of selfcoupling λ ∼ 10−13 for the potential V4 = λϕ4 or the smallness of the mass of the inﬂaton ﬁeld in Planck units, m/MP ∼ 10−6 for the potential V2 = m2 ϕ2 /2. Let us consider an inﬁnite set of potential wells corresponding to inﬁnite set of its minima [2] as discussed above. Then we can use the concept of probability
128
Stationary Points in Scalar Fields
to ﬁnd a potential well with speciﬁc properties. To estimate the relative number of speciﬁc universes, let us suppose that if there are no observational data on the value of a parameter g, the probability density W for any parameter g is distributed by a random uniform distribution in the range (0, 1) in Planck scale. An immediate conclusion is that the probability of a potential λϕ4 is about 10−13 while the probability of a potential m2 ϕ2 /2 is about 10−6 . It means that the latter is realized 106 times more frequently. In fact the probability is much smaller due to smallness of the cosmological Λterm. So the probability to ﬁnd a universe with such small vacuum energy is PΛ = 10−123 . Recall that the set of potential minima is inﬁnite. It means that the set of universes with an appropriate vacuum energy density is relatively small but still inﬁnite. So the probability to ﬁnd an appropriate potential V4 is P (V4 ) = 10−13 PΛ ,
V4 ∼ ϕ4 ,
(13)
V2 ∼ ϕ 2 .
(14)
while the same for the potential V2 is P (V2 ) = 10−6 PΛ ,
The lowest stationary state could be a singular point of the kinetic term, rather than a potential minimum. Thus we could expect that singular point(s) ϕs may be found near some minima ϕm of the potential. Now the problem is reformulated as follows: “which part of inﬁnite number of minima contains singular points located closely to them? ” This part is very small, but not zero, due to inﬁnite number of the minima. Only this part is important  it represents those vacua where galaxies could be formed due to extremely small value of Λ− term [13]. Following the way discussed above we can compare the probability of realization of such potentials. Their common factor is connected with the probability to ﬁnd the singular point of the kinetic term in a small interval Eq.(8), P0 = ∆ϕs /MP ∼ =
Λ M3 = PΛ √ P . m 2V0 MP m 2 V0 
(15)
For the case n = 1 the only additional smallness is dictated by expression (7) and the probability for such universes to occur is P1 ∼
m M3 P0 = PΛ √P ≈ PΛ . M M 2V0
(16)
Universes with the properties described in the case n = 2 are distributed with probability √ V0 2V0 P2 ∼ 4 P0 PΛ ∼ 10−7 PΛ , (17) MP mMP if the inﬂation starts at the right branch of the potential. Here we assumed m ∼ MP , V0 ∼ MP4 . The last case considered, n = 2, has a probability by an order of magnitude larger √ m V0 2V0 P−1 ∼ P0 PΛ ∼ 10−6 PΛ . (18) 4 M MP MP2
¨ ger et al. H. Kro
129
Figure 2. Random potential and kinetic term. Dots denote stationary states of the ﬁeld ϕ.
An important conclusion from this consideration is that the model with kinetic −1 term K ∼ (ϕ − ϕs ) is much more probable (at least by a factor 106 ) comparing with other models discussed above, including the models with a standard kinetic term and potentials ∼ ϕ2 and ∼ ϕ4 , see expressions (14), (13). It means that our −1 Universe is likely governed by the model with kinetic term K ∼ (ϕ − ϕs ) . In conclusion we have discussed several inﬂationary models having common features like the occurrence of singular points in noncanonical kinetic terms. We have shown that the existence of such points where the kinetic term changes its sign or tends to inﬁnity opens new possibilities for scalar ﬁeld dynamics. It takes place even for the simplest form of the potential. Depending on a position of the singular point of the kinetic term, speciﬁc forms of the potential of the auxiliary ﬁeld have been obtained. One of the main results is that the stationary value of scalar ﬁeld could occur at singular points of kinetic term rather than at minima of the potential. We estimated the parameter values for three type of new inﬂationary models. The probabilities to ﬁnd universes with speciﬁc values of parameters have been estimated. It was shown that the probability is much greater for the model −1 with kinetic term K ∼ (ϕ − ϕs ) than for the other models including the most promising model of chaotic inﬂation with the quadratic potential. Another interesting result is that if the singular point is a root of the kinetic term, the ﬁnal state is intrinsically a quantum state.
130
Stationary Points in Scalar Fields
The work was partially performed in the framework of Russian State contract 40.022.1.1.1106, RFBR grant 02−02−17490. H.K. has been supported by NSERC Canada. References [1] A. D. Linde, The Largescale Structure of the Universe. London: Harwood Academic Publishers, 1990. [2] S. Rubin, Gravitation & Cosmology, vol. 9, pp. 243–248, 2003, hepph/0309184. [3] S. Rubin, H. Kr¨ oger and G. Melkonian, astroph/0310182, 2003. [4] C. Brans and R. Dicke, Phys. Rev, vol. 124, no. 3, pp. 925–935, 1961. [5] V.Sahni and A.Starobinsky, Int.J.Mod.Phys., vol. D9, no. 4, pp. 373–444, 2000. [6] K. Bronnikov, J. Math. Phys., vol. 43, no. 12, pp. 6096–6115, 2002. grqc/0204001. [7] S. Rubin, H. Kr¨ oger, G. Melkonian, Gravitation & Cosmology, Supplement, vol. 8, pp. 27– 31, 2002; hepph/0207109. [8] C.L.Bennett, Astrophys.J.Lett., vol. 464, no. 1, pt. 2, pp. L1–L4, 1996. [9] J.P. Kneller and G. Steigman, astroph/0210500 [10] A. Albrecht and C. Skordis, , Phys. Rev. Lett., vol. 84, no. 10, pp. 2076–2079, 2000. [11] H.P.Nilles, Phys.Rept., vol. 110, no. 12, pp. 1–162, 1984. [12] D.H.Lyth and E. D. Stewart, Phys.Rev., vol. D 54, no. 12, pp. 71867190, 1996. [13] S. Weinberg, astroph/0005265.
THE DENSITY MATRIX DEFORMATION IN QUANTUM AND STATISTICAL MECHANICS OF THE EARLY UNIVERSE
ALEXANDER E. SHALYTMARGOLIN a , VASSILII I. STRAZHEV a a
National Centre of Particles and High Energy Physics, Bogdanovich Str. 153, Minsk 220040, Belarus
Abstract In this paper a new approach to investigation of Quantum and Statistical Mechanics of the Early Universe (Planck scale)  density matrix deformation  is proposed. Quantum Mechanics of the Early Universe is chosen as Quantum Mechanics with Fundamental Length. Here Quantum Mechanics with Fundamental Length is obtained as a deformation of wellknown Quantum Mechanics. The proposed approach allows to describe dynamics. It is demonstrated that Statistical Mechanics of the Early Universe is a deformation of the conventional Statistical Mechanics. The statisticalmechanics deformation is constructed by analogy to the earlier quantum mechanical results. Some implications of the obtained results are discussed. In particular, the problem of singularity, possible improvement of the deﬁnition of statistical entropy and the problem of information loss in black holes are considered. It is noted that the obtained results enable the derivation of the BekensteinHawking’s formula for the black hole entropy in a semiclassical approximation in a simple and natural way.
As known, Quantum Mechanics of the Early Universe (Planck scale) diﬀers from the conventional one [1]. The main motivation for this diﬀerence is the presence of the General Uncertainty Relations (GUR) [2] appropriated to describe a physical behavior of the Early Universe and unavoidably leading to the fundamental length concept. The derivation of GUR proceeding from the basic principles has been performed relatively recently [3]. By this means the Quantum Mechanics of the
131 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 131–134. © 2006 Springer. Printed in the Netherlands.
132
The Density Matrix Deformation In Quantum
Early Universe should be treated as a deformation of the wellknown Quantum Mechanics (QM). The deformation is understood as an extension of a particular theory by inclusion of one or several additional parameters in such a way that the initial theory appears in the limiting transition. The deformation in Quantum Mechanics at Planck scale is realized by the commutator deformation, or more precisely by deformation of the respective Heisenberg algebra [2],[4]. However, this approach suﬀers from two serious disadvantages:1)the deformation parameter is a dimensional variable κ with a dimension of mass [2] and 2)in the limiting transition to QM this parameter goes to inﬁnity, whereas ﬂuctuations of other values are hardly sensitive to it. In this paper it is proposed to study the quantum mechanics of the Early Universe proceeding from the density matrix deformation approach [5][7]. First starting from the existence of a minimum length lmin ∼ Lpl , we give strong validation for the dependence of the density matrix on the addi2 tional dimensionless parameter α = lmin /x2 (where x is the scale of measurement) varying over the interval 0 < α ≤ 1/4. An exact deﬁnition and the principal properties of the characteristic deformed density matrix ρ(α) referred to as the density promatrix may be found in [5]. Of particular importance is the fact that in the lowenergy limit lim ρ(α) = ρ α→0
transforms to the conventional density matrix in Quantum Mechanics. Besides, the following condition must be fulﬁlled: Sp[ρ(α)] − Sp2 [ρ(α)] ≈ α.
(1)
As a matter of fact, the deformation parameter α should assume the value 0 < α ≤ 1. As seen from (1), however, Sp[ρ(α)] is well deﬁned only for 0 < α ≤ 1/4. That is if x = ilmin and i ≥ 2, then there is no problem at all. At the point of x = lmin there is a singularity related to the complex values following from Sp[ρ(α)] , i.e. to the impossibility of obtaining a diagonalized density promatrix at this point over the ﬁeld of real numbers. In [5, 7] it has been noted that this singularity is directly associated with the description problem of spacetime singularities, at any rate in case of Schwarzschild black holes. We consider possible solutions for (1). For instance, one of the solutions of (1), at least to the ﬁrst order in α, is & ρ∗ (α) = αi exp(−α)i >< i, (2) i
where all αi > 0 are independent of α and their sum is equal to 1. In this way Sp[ρ∗ (α)] = exp(−α) We can easily verify that Sp[ρ∗ (α)] − Sp2 [ρ∗ (α)] = α + O(α2 )
A.E. ShalytMargolin, V.I. Strazhev
133
Note that in the momentum representation α ∼ p2 /p2pl , where ppl is the Planck momentum. When present in the matrix elements, exp(−α) can damp the contribution of great momenta in a perturbation theory. The applications of the obtained results may be as follows [5, 7]: 1) An additional component in the deformed Liouville’s equation arises in two cases: at transfer out of inﬂation to low energies; on absorption of the matter by a black hole and its motion to singularity [8]. Note that the appropriate additional component in the righthand side of Schrodinger equation may be obtained in similar way [5, 7]. 2)The approach used allows for introduction of the following matrix value that is a complete analog for the statistical entropy in case under consideration and is meaning the entropy density over a minimum unit area [5, 7]: Sαα21 = −Sp[ρ(α1 ) ln(ρ(α2 ))] = − < ln(ρ(α2 )) >α1 ,
(3)
where 0 < α1 , α2 ≤ 1/4. Sαα21 has a clear physical meaning: the entropy density is computed on the scale associated with the deformation parameter α2 by the observer who is at a scale corresponding to the deformation parameter α1 . Note that with this approach the diagonal element Sα = Sαα of the described matrix Sαα21 is the density of entropy measured by the observer, who is at the same scale as the measured object associated with the deformation parameter α. In [5, 7] such a construction and exponential ansatz (2) were used in derivation of a semiclassical BekensteinHawking formula [9] for the Black Hole entropy from the basic principles. 3) Besides, value Sαα21 may be used in analysis of the Information Paradox Problem [8]. In [5, 7] it has been demonstrated that for any observer in the vicinity of a black hole the information loss close to the Big Bang initial singularity and close to the ﬁnal singularity of a black hole is comparable. But then in both cases the entropy density should be the same: S(in) = S α1 1 = S(out) 4
and in fact there is no any information loss. Proceeding to the Statistical Mechanics [10], we further assume that an internal energy of any ensemble U could not be in excess of Emax ∼ Ep which appears from the generalized uncertainty relations for the ”energy  time” pair and hence temperature T could not be in excess of Tmax = Emax /kB ∼ Tp . On this basis the density matrix deformation in Statistical Mechanics at Planck scale is constructed in the [6, 7]. In this way ρstat at very high T 0 becomes dependent on the param2 eter τ = T 2 /Tmax , i.e. in the most general case ρstat = ρstat (τ ),Sp[ρstat (τ )] < 1 and for τ 1 we have ρstat (τ ) ≈ ρstat in the wellknown Statistical Mechanics [10].The following condition must be fulﬁlled in the same way as in case of Quantum Mechanics of of the Early Universe: Sp[ρstat (τ )] − Sp2 [ρstat (τ )] ≈ τ.
(4)
134
The Density Matrix Deformation In Quantum
Again similar to QMFL, as a possible solution for (4) we have an exponential ansatz [6, 7] with the help of which the deformation of a canonical Gibbs distribution at Planck scale (up to factor 1/Q) takes an elegant and completed form: ωn (τ ) = exp(−τ )ωn = exp(−
T2 − βEn ) T2max
(5)
References [1] L. Garay, Int.J.Mod.Phys.A10(1995)145, D.V. Ahluwalia, Mod.Phys.Lett.A17(2002)1135. [2] M. Maggiore, Phys.Lett.B319(1993)83, A. Kempf,G. Mangano,R.B. Mann, Phys.Rev.D52(1995)1108. [3] R.J. Adler,D.I. Santiago,Mod.Phys.Lett.A14(1999)1371. [4] M. Maggiore,Phys.Rev.D49(1994)5182 [5] A.E. ShalytMargolin,J. Suarez, Intern.Journ.of Mod.Phys.D.12(2003)1265, A.E. ShalytMargolin, Mod.Phys.Lett.A19(2004)391, A.E. ShalytMargolin, Intern.Journ. of Mod.Phys.D.13 (2004)853. [6] A.E. ShalytMargolin,A.Ya. Tregubovich,Mod. Phys. Lett. A19(2004)71. [7] A.E. ShalytMargolin, The Density Matrix Deformation in Physics of the Early Universe and Some of its Implications, will be published in the book ”Progress in General Relativity and Quantum Cosmology Research” (Nova Science Publishers, Inc., Hauppauge, NY) [8] S. Hawking,Phys.Rev.D14(1976)2460, A. Strominger, Les Houches Lectures on Black Holes, [hepth/9501071]. [9] J.D. Bekenstein,Phys.Rev.D7(1973)2333, S. Hawking, Phys.Rev.D13(1976)191. [10] R.P. Feynman,Statistical Mechanics,A Set of Lectures,California, Institute of Technology W.A.Benjamin,Massachusets 1972.
Part II
Foundations of Physics
DARK ENERGY, CHAOTIC FIELDS, AND FUNDAMENTAL CONSTANTS
CHRISTIAN BECK School of Mathematical Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS, UK
Abstract To explain the currently observed accelerated expansion of the universe, a large number of diﬀerent theoretical models are presently being discussed. In one way or another, all of these contain ‘new physics’, though at diﬀerent levels. The big question is how to select out of inﬁnitely many possible models the right one. We here discuss a possibility that has so far been somewhat neglected, namely that the new physics underlying dark energy arises out of a gravitationally active amendment of the electroweak and strong sector of the standard model. This amendment basically consists of a rapidly ﬂuctuating gravitationally active dynamics of vacuum ﬂuctuations with a cutoﬀ of the order of the neutrino mass scale. We consider a concrete model for this based on secondquantized selfinteracting scalar ﬁelds, which evolve in a chaotic way. It is shown that expectations with respect to the chaotic dynamics yield statements on the observed numerical values of the electroweak coupling constants with amazing precision, thus providing evidence for the physical relevance of this model.
1
Introduction
To understand the fact that the universe is currently in a phase of accelerated expansion [1, 2, 3, 4], an enormous amount of theoretical and experimental work is currently being performed. The favoured explanation for the acceleration is the existence of dark energy, though other possibilities might exist as well. An amazing number of models has been developed in the mean time, and basically
137 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 137–146. © 2006 Springer. Printed in the Netherlands.
138
Dark energy, chaotic fields, and fundamental constants
every week one ﬁnds in the preprint archives some new idea concerning the nature of dark energy. The most popular models are currently quintessence models of various kinds [5], phantom ﬁelds [6], BornInfeld quantum condensates [7], the Chaplygin gas [8], ﬁelds with nonstandard kinetic terms [9], to name just a few. All of these approaches contain ‘new physics’ in one way or another, though at diﬀerent levels. However, it should also be clear that the number of possible dark energy models that are based on new physics is inﬁnite, and in that sense there is much more to study than the above, currently most popular, models. When trying to select the most suitable theoretical model for dark energy out of many possibilities, the most straightforward idea would be to compare the various predictions of diﬀerent theoretical models with the experimental observations (supernovae, cosmic microwave background, large scale structure, etc.) to single out the most relevant model. However, there is a huge degeneracy in the sense that completely diﬀerent theoretical models cannot be distinguished with the currently available observational data, they often make the same or indistinguishable predictions (e.g. for the equation of state as a function of redshift) or no prediction at all. This situation of degeneracy is not expected to improve signiﬁcantly in the near future, though the precision of the observational data will increase. We are thus lead to single out good dark energy models by other, more theoretical criteria. Let us here suggest the following checklist applicable to any dark energy model: 1. Does the model explain why the current value of dark energy density is so small (the cosmological constant problem) ? 2. Does the model explain why the current value of dark energy density is of the same order as the matter density (the cosmological coincidence problem) ? 3. Does the model have a reasonable quantum ﬁeld theoretical background, or does it represent a reasonable extension of quantum ﬁeld theory? 4. Does the model ﬁt into Einstein’s gravity, or does it represent a reasonable extension of it? 5. Is the model compatible with observations? 6. Does the new physics contained in the model explain some further phenomena that are so far unexplained, for example why the fundamental constants of nature (coupling constants, mass ratios, mixing angles) take on the values we observe and not some other values? 7. Does the model give some sense to dark energy (for example, by explaining it as a relic of inﬂation), rather than letting it look like an unnecessary curiosity of the universe? 8. Besides dealing with dark energy, does the model also explain why there is dark matter?
C. Beck
139
9. Is the theory aesthetic and accessible to many physicists, or is it just so complicated that hardly anybody understands it? 10. Are there any laboratory experiments that can verify or disprove this theory? Every theoretical model builder may rank his or her favourite dark energy model on a scale from 0–10, depending on how much of the above criteria are satisﬁed. The perfect theory, from which we are still far away, reaches a mark of 10. Current models, perhaps, reach something in the region 1–5, at best. As mentioned before, with the assumption that new physics is relevant there is an enormous number of possible models. In the following I will restrict myself to a model introduced in [10], which scores relatively high on points 3,6,7,10. With some additional assumptions described in [10], it also scores high on points 1,2 and 4. The basic idea is that there is a rather ‘sterile’ amendment of the standard model of electroweak and strong interactions which just consists of a scalar dynamics of vacuum ﬂuctuations with a ﬁnite cutoﬀ. The expectation of the underlying potentials produces the currently observed dark energy. Amazingly, the above model seems to distinguish the observed values of the electroweak coupling parameters as local minima in the dark energy landscape. It is well known that quintessence ﬁelds can produce a very slow time variation of fundamental constants of nature, e.g. of the ﬁne structure constant [11]. Here we go a step further and show that not only a possible variation of the ﬁne structure constant but also its currently observed equilibrium value can be understood by a suitable scalar ﬁeld dynamics underlying dark energy.
2
Amending the standard model by gravitationally active vacuum ﬂuctuations
Let us consider the standard model of electroweak and strong interactions. It is a secondquantized ﬁeld theory and it allows for vacuum ﬂuctuations. The vacuum energy density associated with a particle of mass m and spin j is given by 1 d3 k 2 ρvac = (−1)2j (2j + 1) k + m2 (1) 2 (2π)3
in units where√ = c = 1. Here k represents the momentum and the energy is given by E = k2 + m2 . Unfortunately, the above integral is divergent. One has to introduce a suitable upper cutoﬀ. Choosing as an upper cutoﬀ the Planck mass mP l , one gets an enormous amount of vacuum energy density of the order m4P l , larger than the currently observed dark energy density by a factor of ∼ 10120 . This is the famous cosmological constant problem. To circumvent it, the common view is that the absolute value of the above vacuum energy (e.g. in QED) is not observable, it is ‘renormalized away’, which in a sense means that one adds an inﬁnite constant to get rid of the vacuum energy. This works as long as one does not consider gravity.
140
Dark energy, chaotic fields, and fundamental constants
However, ultimately we have to unify the standard model with gravity. Also note that almost all particles in the standard model do have mass, so they know what gravity is. It looks a bit like a ‘dirty trick’ to say that gravity is decoupled from the standard model if the particles have mass. So let us for the moment assume that the vacuum energy in the standard model is cancelled by some kind of symmetry, for example some kind of supersymmetry. This still doesn’t explain why we do observe some tiny positive amount of vacuum energy density in the universe, corresponding to the currently measured dark energy density, which is more of the order m4ν rather than m4P l , where mν is a typical neutrino mass scale. Hence let us assume that there is something more to the standard model: In addition to the ordinary standard model ﬁelds (whose vacuum energies are cancelled by some symmetry) there could be other ﬁelds that just show up in a rather sterile way in form of vacuum ﬂuctuations, with a rather small cutoﬀ scale of the order of the neutrino mass. The vacuum energy of these ﬁelds is not cancelled by symmetry, there is a symmetry breaking towards positive vacuum energy, at least at the current stage of the universe. Of course, the above assumption represents new physics, but any decent dark energy model seems to require new physics, in one way or another. The advantages of the above idea are straightforward: • Since we associate dark energy with a broken symmetry in some sector of this extended standard model, it is not too surprising that the relevant scale of the dark energy density is of the order of some typical particle mass to the power 4 in this model, in this case a neutrino. • There is increasing experimental evidence [12] for the existence of sterile neutrinos in addition to the known three ordinary neutrino ﬂavours, so apparently there is something more to the standard model than we know. Sterile neutrinos may have something to do with the above gravitationally active sector of the standard model. • Since our dark energy model deals with vacuum ﬂuctuations that are part of the electroweak sector, there is a chance to measure the eﬀects of these ﬂuctuations in laboratory experiments on the earth, such as in Josephson junction experiments, which do probe the spectrum of vacuum ﬂuctuations near the neutrino mass scale due to a nonlinear mixing eﬀect in the junction [13]. In the following, we want to consider a concrete model for vacuum ﬂuctuations with a small cutoﬀ, as introduced in [10]. For quantum ﬁeld theories with a cutoﬀ, a particular quantization scheme is very convenient to choose, namely the stochastic quantization scheme introduced by Parisi and Wu [14]. This scheme is based on a stochastic diﬀerential equation, which naturally embeds various kinds of cutoﬀs and is by far simpler to deal with than the canonical quantization procedure. For that reason, our gravitationally active amendment of the standard model will be formulated in terms of stochastic quantization. Amazingly, the model will turn
C. Beck
141
out to distinguish the numerical values of the electroweak coupling constants as corresponding to local minima in the dark energy landscape. This can be seen as an indication that one is on the the right track with these kinds of models.
3
Chaotic model of vacuum ﬂuctuations
Let us consider a homogeneous selfinteracting scalar ﬁeld ϕ with potential V (ϕ) that forms the basis for our gravitationally active amendment of the standard model. Our amendment should have rather ‘sterile’ properties, so it is in good approximation suﬃcient to look at the scalar ﬁeld equations of this sector on its own, rather than coupling them to the ordinary standard model ﬁeld equations. We also need the amended sector to consist mainly of vacuum ﬂuctuations with a suitable cutoﬀ, rather than containing stable observable particles, with the possible exception of sterile neutrinos and/or dark matter. In fact, the only connection to the ordinary standard model is that the virtual particles underlying the vacuum ﬂuctuations could potentially interact with the same electroweak and strong coupling constants as in the ordinary standard model. We quantize the scalar ﬁeld underlying the steril sector using the ParisiWu approach of stochastic quantization. The 2nd quantized equation of motion is ∂ (2) ϕ = ϕ¨ + 3H ϕ˙ + V (ϕ) + L(s, t), ∂s where H is the Hubble parameter, t is physical time, s is ﬁctitious time (just a formal coordinate to do quantization) and L(s, t) is Gaussian white noise, δcorrelated both in s and t. The ﬁctitious time s is just introduced as a formal tool for stochastic quantization, it has dimensions GeV −2 . Quantum mechanical expectations can be calculated as expectations of the above stochastic process for s → ∞. The simplest way to introduce a cutoﬀ is by making t and s discrete (as in any numerical simulation). Hence we write
s = t =
nτ iδ,
(3) (4)
where n and i are integers and τ is a ﬁctitious time lattice constant, δ is a physical time lattice constant. Note that the uncertainty relation ∆E∆t = O() always implies an eﬀective lattice constant ∆t for a given ﬁnite energy ∆E. We also introduce a dimensionless ﬁeld variable Φin depending on i and n by writing ϕin = Φin pmax , where pmax is some (so far) arbitrary energy scale. The discretized scalar ﬁeld dynamics (2) can be written as the following discrete dynamical system [15, 3, 17] 3 α Φin+1 = (1 − α)T (Φin ) + Hδα(Φin − Φni−1 ) + (Φi+1 + Φi−1 n ) + τ · noise, 2 2 n where the local map T is given by τ T (Φ) = Φ + V (pmax Φ) pmax (1 − α)
(5)
(6)
142
Dark energy, chaotic fields, and fundamental constants
and α is deﬁned by 2τ . δ2 For old universes, one can neglect the term proportional to H, obtaining α Φin+1 = (1 − α)T (Φin ) + (Φi+1 + Φi−1 n ) + τ · noise. 2 n α :=
(7)
(8)
We now want to construct a ﬁeld ϕin that is diﬀerent from ordinary ﬁelds: Rather than evolving smoothly it should exhibit strongly ﬂuctuating behaviour, so that we may be able to interpret it in terms of vacuum ﬂuctuations. As a distinguished example of a ϕ4 theory generating such behaviour, let us consider the map Φn+1 = T−3 (Φn ) = −4Φ3n + 3Φn
(9)
on the interval Φ ∈ [−1, 1]. T−3 is the negative thirdorder Tchebyscheﬀ map, a standard example of a map exhibiting strongly chaotic behaviour. It is conjugated to a Bernoulli shift, and is distinguished as generating the strongest possible chaotic behaviour possible for a smooth lowdimensional deterministic dynamical system [18]. The corresponding potential is given by ' ( 1−α 1 2 4 V−3 (ϕ) = ϕ − 2 ϕ + const, (10) τ pmax
or, in terms of the dimensionless ﬁeld Φ, V−3 (ϕ) =
1−α 2 pmax (Φ2 − Φ4 ) + const. τ
(11)
The important point is that starting from this potential we obtain by second quantization a ﬁeld ϕ that rapidly ﬂuctuates in ﬁctitious time on some ﬁnite interval, choosing initially ϕ0 ∈ [−pmax , pmax ]. Since these chaotic ﬂuctuations are bounded, there is a natural cutoﬀ. The idea is now that the expectation of the potential of this and similar chaotic ﬁelds (plus possibly kinetic terms) underlie the measured dark energy density in the universe. Expectations · · · can be easily numerically determined by iterating the dynamics (8) for random initial conditions. One has V−3 (ϕ) =
1−α 2 pmax (Φ2 − Φ4 ) + const, τ
(12)
which for α = 0 can be analytically evaluated [10] to give V−3 (ϕ) =
1 p2max + const. 8 τ
(13)
Alternatively, we may consider the positive Tchebyscheﬀ map T3 (Φ) = 4Φ3 − 3Φ. It is easy to show that this generates vacuum energy of opposite sign. Symmetry considerations between T−3 and T3 suggest to take the additive constant const as 1−α 2 1 2 const = + pmax Φ . (14) τ 2
143
C. Beck In this case one obtains the fully symmetric equation ' ( 1−α 2 3 2 4 V±3 (ϕ) = ± pmax − Φ + Φ , τ 2
which for α → 0 reduces to V±3 (ϕ) = ±
p2max τ
3 − . 8
(15)
(16)
To reproduce the currently measured dark energy, we only need to ﬁx the ratio of the parameters τ and pmax as 3 p2max (17) = ρΛ ∼ m4ν . 8 τ This is the simplest model of steril vacuum ﬂuctuations one can think of, a 2nd quantized ﬁeld theory underlying the cosmological constant Λ. It is easy to show [10] that for α = 0 the equation of state of this ﬁeld is w = −1. For small α, it is close to w = −1. More complicated models, with w = −1, as well as symmetry breaking breaking between T+3 and T−3 can be worked out in detail [10]. These can produce tracking behaviour of dark energy during the evolution of the universe, and mimic some of the properties of quintessence ﬁelds.
4
Electroweak couplings as local minima in the dark energy landscape
Let us now give a heuristic argument why the coupling constant α in the above chaotic ﬁeld equations could have the physical meaning of a gauge coupling. Consider two charges of opposite sign, say, a virtual electronpositron pair, which exists for a short time interval due to a vacuum ﬂuctuation. If the charges are at distance r, the Coulomb potential between them is given by 1 (18) r (in units where = c = 1), where α is the ﬁne structure constant. Now for any vacuum ﬂuctuation the inverse distance 1/r is certainly a ﬂuctuating random variable. Motivated by our interpretation of vacuum ﬂuctuations of the ﬁeld ϕin we may choose 1 = , ϕin − ϕi−1 (19) n r which has the right dimension and is allowing for both attracting and repelling deforces. The above choice basically means that the ﬁeld diﬀerence ϕin − ϕi−1 n termines the inverse interaction distance to neighbours in this chaotically evolving discrete model of vacuum ﬂuctuations. Eq. (19) just represents the uncertainty relation ∆p∆r = O(), interpreting ϕin − ϕni−1 as a momentum uncertainty. Combining eq. (18) and (19), the ﬂuctuating Coulomb potential can thus be written as i i i−1 VC (ϕi−1 (20) n , ϕn ) = αpmax (Φn − Φn ). VC (r) = α
144
Dark energy, chaotic fields, and fundamental constants
Summing the two contributions of the pair (i, i − 1) and (i, i + 1) we just get the linear interaction terms of the nearest neighbours in the discrete dynamics (8) (for more details on this model, see [16], chapter 5). The remarkable thing is that in this interpretation α has now the physical meaning of a gauge coupling. Of course, a similar consideration applies to all kinds of coupling constants (electroweak and strong) in the standard model, not only the ﬁne structure constant. Our central hypothesis is thus that the chaotic ﬁelds, though being diﬀerent from the ordinary standard model ﬁelds, interact with the same coupling constants as in the ordinary standard model. There may be various degrees of freedom of the chaotic ﬁelds underlying dark energy. For example, from a dynamical systems point of view, it makes sense to generalize the chaotic ﬁeld dynamics (8) to α b i+1 Φin+1 = (1 − α)T (Φin ) + σ (T b (Φi−1 n ) + T (Φn )) 2
(21)
(the noise term can usually be neglected for chaotic maps). The case σ = +1 is called ‘diﬀusive coupling’, the case σ = −1 ‘antidiﬀusive coupling’. Chaotic ﬁelds with b = 1 are called to be of ‘type A’ ( T 1 (Φ) =: T (Φ)), chaotic ﬁelds with b = 0 to be of ‘type B’ (T 0 (Φ) =: Φ). There are two diﬀerent types of vacuum energies 3 2 p2 4 for the chaotic ﬁelds, namely the self energy V (α) := max τ 2 Φ − Φ and the p2
i i+1 interaction energy W (α) := max 2τ Φn Φn (see [17] for more details). Fig. 1 shows that the self energy V (α) of the chaotic ﬁelds indeed distinguishes electroweak coupling constants known from the standard model. Everybody can easily reproduce this plot, by simply iterating the dynamics (8) for random initial conditions Φi0 ∈ [−1, 1] for a long time on a large lattice and averaging the variable 1.5(Φin )2 − (Φin )4 . We observe that V (α) has local minima at a1 = 0.000246(2), a2 = 0.00102(1), a3 = 0.00220(1) (a1 and a3 are actually small local minima on top of the hill). On the other hand, in the standard model of electroweak interactions the weak coupling constant is given by
αweak = αel
(T3 − Q sin2 θW )2 sin2 θW cos2 θW
(22)
Here Q is the electric charge of the particle (Q = −1 for electrons, Q = 2/3 for ulike quarks, Q = −1/3 for dlike quarks), and T3 is the third component of the isospin (T3 = 0 for righthanded particles, T3 = − 21 for eL and dL , T3 = + 21 for νL and uL ). Consider righthanded fermions fR . With sin2 θW = s¯2l = 0.2318 (as experimentally measured) and the running electric coupling αel (E) taken at energy scale E = 3mf we obtain from eq. (22) the numerical values dR cR eR (3md ) = 0.000246, αweak (3mc ) = 0.001013, αweak (3me ) = 0.00220. There αweak is an amazing numerical coincidence between the local minima a1 , a2 , a3 of V (α) and the experimentally measured weak coupling constants of fR = uR , cR , eR , respectively. The factor 3 of the energy scale can be related to the index of the Tchebyscheﬀ polynomial [16].
C. Beck
145
Figure 1. Self energy V (α) of the typeA chaotic ﬁeld in the lowcoupling region. There are local minima at couplings ai that coincide with the weak coupling constants of righthanded fermions in the standard model.
Now regard the ﬁne structure constant αel and the Weinberg angle sin2 θW as a priori free parameters. Suppose these parameters would change to slightly diﬀerent values. Then immediately this would produce larger vacuum energy V (α) in our sterile amendment of the standard model, since we get out of the local minima. The system is expected to be driven back to the local minima, and the fundamental parameters are ﬁxed and stabilized in this way. Further coincidences of this type have been observed for various other observables associated with the chaotic ﬁelds, allowing for a ﬁxing of further fundamental constants such as mass ratios and strong couplings at bosonic mass scales. See [3, 17] for details. All these numerically observed coincidences are not explainable as a random coincidence. Rather, they suggest to interpret the coupling constant α of our secondquantized chaotic ﬁelds ϕ as a running gauge coupling. The chaotic ﬁelds are most naturally associated with an additional, sterile sector of the standard model, which just consists of vacuum ﬂuctuations of a scalar ﬁeld with a cutoﬀ. This sector generates dark energy, and its sense is to ﬁx and stabilize the fundamental constants of nature.
References [1] A.G. Riess et al., Astron. J. 116, 1009 (1998), S. Perlmutter et al., Astrophys. J. 517, 565 (1999), J.L. Tonry et al., astroph/0305008 [2] N.W. Halverson et al., Astrophys. J. 568, 38 (2002), C.B. Netterﬁeld et al., Astrophys. J. 571, 604 (2002), [3] D.N. Spergel et al., Astrophys. J. Suppl. 148, 175 (2003) (astroph/0302209), C.L. Bennett et al., Astrophys. J. Suppl. 148, 1 (2003) (astroph/0302207)
146
Dark energy, chaotic fields, and fundamental constants
[4] S. Caroll, astroph/0004075, S. Weinberg, astroph/0005265, T. Padmanabhan, Phys. Rep. 380, 235 (2003) (hepth/0212290) [5] P.J.E. Peebles and B. Ratra, Astrophys. J. 325, L17 (1988), C. Wetterich, Nucl. Phys. B312, 668 (1988), M.S. Turner and M. White, Phys. Rev. D56, 4439 (1997), J.A. Frieman and I. Waga, Phys. Rev. D57, 4642 (1998), R.R. Caldwell, R. Dave, and P.J. Steinhardt, Phys. Rev. Lett. 80, 1582 (1998), P.J.E. Peebles and A. Vilenkin, Phys. Rev. D59, 063505 (1999) [6] R.R. Caldwell, Phys. Lett. B545, 23 (2002), S.M. Caroll, M. Hoﬀmann, and M. Trodden, astroph/0301273, J. Hao and X Li, hepth/0306033 [7] A.A. Tseytlin, hepth/9908105, E. Elizalde, J.E. Lidsey, S. Nojiri, and S.D. Odintsov, hepth/0307177 [8] A.Y. Kamenshchik, U. Moschella, V. Pasquier, Phys. Lett. B511, 265 (2001), M.C. Bento, O. Bertolami, A.A. Sen, Phys. Rev. D66, 043507 (2002) [9] S. Mukohyama and L. Randall, Phys. Rev. Lett. 92, 211302 (2004) (hepth/0306108), T. Padmanabhan, T. Roy Choudhury, Phys. Rev. D66, 081301 (hepth/0205055) [10] C. Beck, Phys. Rev. D69, 123515 (2004) (astroph/0310479) [11] C. Wetterich, Nucl. Phys. D302, 645 (1988), H.B. Sandvik, J.D. Barrow and J. Magueijo, Phys. Rev. Lett. 88, 031302 (2002) (astroph/0107512), C. Wetterich, Phys. Lett. B561, 10 (2003) (hepph/0301261), M. Doran, astroph/0411606, N.J. Nunes and J.E. Lidsey, Phys. Rev. D69, 123511 (2004) (astroph/0310882) [12] See, e.g., neutrino reviews at pdg.lbl.gov [13] C. Beck and M.C. Mackey, Phys. Lett. B605, 295 (2005) (astroph/0406505) [14] G. Parisi, Y. Wu, Sci Sin. 24, 483 (1981), P.H. Damgaard, H. H¨ uﬀel (eds.), Stochastic Quantization, World Scientiﬁc, Singapore (1988) [15] K. Kaneko, Progr. Theor. Phys. 72, 480 (1984), R. Kapral, Phys. Rev. A31, 3868 (1985), C. Beck, Phys. Lett. A248, 386 (1998) [16] C. Beck, Spatiotemporal Chaos and Vacuum Fluctuations of Quantized Fields, World Scientiﬁc, Singapore (2002) (50page summary at hepth/0207081) [17] C. Beck, Physica D171, 72 (2002) (hepth/0105152) [18] C. Beck and F. Schl¨ ogl, Thermodynamics of Chaotic Systems, Cambridge University Press, Cambridge (1993), C. Beck, Nonlinearity 4, 1131 (1991), A. Hilgers, C. Beck, Physica D156, 1 (2001)
HOW FUNDAMENTAL IS GRAVITATION?
B.G. SIDHARTH International Institute for Applicable Mathematics, and Information Sciences,Hyderabad(India) and Udine(Italy), B.M. Birla Science Centre, Adarsh Nagar, Hyderabad  500 063 (India)
Abstract We consider a model in which the Universe has an underpinning of oscillators in the Quantum Vacuum (or dark energy) at the Planck scale and deduce a number of otherwise inexplicable Large Number relations which have been considered to be empirical accidents. The analysis shows that the gravitational energy is the residual energy of the Planck oscillators constituting the Universe at large on the one hand, and elementary particles on the other. This explains a mysterious puzzle ﬁrst pointed out by Weinberg several years ago, in a formula relating the pion mass to the Hubble Constant, a puzzle which has remained unexplained ever since.
1
Introduction
More than ﬁve thousand years ago, the Rig Veda repeatedly raised the question: “How is it that though unbound the sun does not fall down?” This was a question that puzzled thinking man over the millennia. Indian scholars right up to Bhaskaracharya who lived about a thousand years ago believed in some attractive force which was responsible for keeping the celestial bodies from falling down. The same problem was addressed by Greek thinkers about two thousand ﬁve hundred years ago. They devised transparent material spheres to which each of the celestial objects were attached  the material spheres prevented them from falling down.
147 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 147–154. © 2006 Springer. Printed in the Netherlands.
148
How Fundamental is Gravitation?
Unfortunately it was this answer to the age old question, which held up further scientiﬁc progress during the time of Kepler, for even Copernicus accepted the transparent material spheres. Kepler had a powerful tool in the form of the accurate observations of Tycho Brahe. He also had the advantage of the Indian numeral system, which via the Arabs reached Europe just a few centuries earlier. These lead him to his famous laws of elliptical orbits with deﬁnite periods correlated to distances from the Sun. This was the beginning of modern science. The important point was that the Greek answer to the problem of why heavenly objects do not fall down was now demolished. Kepler himself speculated about some type of a magnetic force between the Sun and the Planets, rather on the lines of speculations in India. It was Newton who provided the next chapter in the story. To quote Hawking [1] “The Philosophiae Naturalis Principia Mathematica by Isaac Newton, ﬁrst published in Latin in 1687, is probably the most important single work ever published in the physical sciences. Its signiﬁcance is equalled in the biological sciences only by The Origin of Species by Charles Darwin. The original impulse which caused Newton to write the Principia was a question from Edmund Halley as to whether the elliptical orbits of the planets could be accounted for on the hypothesis of an inverse square force directed towards the Sun. This was something that Newton had worked out some years earlier but had not published, like most of his work on mathematics and physics. However, Halley’s challenge, and the desire to refute the suggestions of others such as Hooke and Descartes, spurred Newton to try to write a proper account of this result.” Newton using Galileo’s ideas of Mechanics, thus stumbled upon the universal Law of Gravitation. This held sway for nearly two hundred and twenty ﬁve years, before Einstein came out with his own theory of gravitation. This was no force in the sense that Newton’s and preceding scholars had envisaged it to be. Rather it was due to the curvature of spacetime itself. Einstein’s bizarre ideas have had some experimental veriﬁcation while there are some other experimental consequences, such as gravitational waves, which need to be conﬁrmed. After Einstein’s formulation of gravitation a problem that has challenged and deﬁed solution has been that of providing a uniﬁed description of gravitation along with other fundamental interactions. One of the earliest attempts was that of Hermann Weyl  the gauge geometry [2] , which though elegant was rejected on the grounds that in the ﬁnal analysis, it was not really a uniﬁcation of gravitation with electromagnetism. Modern approaches to this problem have lead to the abandonment of a smooth spacetime manifold. Instead, the Planck scale is now taken to be a minimum fundamental scale. We had already argued from diﬀerent points of view to arrive at the otherwise empirically known equations [3, 4, 5] √ √ R = N lP = N l
B.G. Sidharth l=
√
nlP
149
(1)
where lP , l and R are the Planck length, the pion Compton wavelength and the ¯ and n are certain large numbers. Some of these radius of the universe and N, N ¯ ∼ 1080 being the number of elementary are well known empirically for example N particles, which typically are taken to be pions in the literature, in the universe. One way of arriving at the above relations is by considering a series of N Planck mass oscillators which are created out of the Quantum Vacuum. In this case (Cf. also ref.[6]) we have √ r = N a2 (2) In (2) a is the distance between the oscillators and r is the extent. Equations (1) follow from equation (2). There is another way of arriving at equations (1) (Cf.ref.[5]). For this, we observe that the position operator for the KleinGordan equation is given by [7], 2 op = xop − ıc p X 2 E2
Whence we get 2m3 c4 2 2m3 c6 2 p2 2 ˆ op X ≡ X = x + op 2 2 2m
(3)
2 ˆ op deﬁnes the Harmonic osIt can be seen that purely mathematically (3) for X cillator equation, this time with quantized, what may be called space levels. It 2 turns out that these levels are all multiples of ( mc ) . This Compton length is the Planck length for a Planck mass particle. Accordingly we have for any system of extension r, r 2 ∼ N l2
which gives back equation (1). It is also known that the Planck length is also the Schwarzschild radius of a Planck mass, that is we have lP = GmP /c2
(4)
Using equations (1) and (4), we will now deduce a few new and valid and a number of otherwise empirically known relations involving the various microphysical parameters and large scale parameters. Some of these relations are deducible from the others. Many of these relations featured (empirically) in Dirac’s Large Number Cosmology. We follow Dirac and Melnikov in considering l, m, , lP , mP and e as microphysical parameters [8, 9]. Large scale parameters include the radius and the mass of the universe, the number of elementary particles in the universe and so on. In the process we will also examine the nature of gravitation. It must also be observed that the Large Number relations below are to be considered in the Dirac sense, wherein for example the diﬀerence between the electron and pion (or proton) masses is irrelevant [10].
150 2
How Fundamental is Gravitation? Interrelationships
We will use the following well known equation which has been obtained through several routes: GM =R (5) c2
For example in an uniformly expanding ﬂat Friedman spacetime, we have [10] 8πGρR2 R˙ 2 = 3 If we substitute R˙ = c at the radius of the universe in the above we recover (5). We now observe that from the ﬁrst two relations of (1), using the Compton wavelength expression we get √ m = mP / n (6)
Using also the second relation in (1) we can easily deduce ¯n N =N Using (1) and (5) we have M=
√
(7)
N mP
(8)
Interestingly (8) can be obtained directly, without recourse to (5), from the energy of the Planck oscillators (Cf.ref.[4]). Combining (8) and (6) we get √ M= Nn m (9)
Further if we use in the last of equation (1) the fact that lP is the Schwarzchild radius that is equation (4), we get, G=
lc2 nm
(10)
We now observe that if we consider the gravitational energy of the N Planck masses (which do not have any other interactions) we get, Gravitational Energy =
GN m2P R
If this is equated to the inertial energy in the universe, M c2 , as can be easily veriﬁed we get back (5). In other words the inertial energy content of the universe equals the gravitational energy of all the N Planck oscillators. Similarly if we equate the gravitational energy of the n Planck oscillators constituting the pion we get Gm2P n (11) = mc2 R
B.G. Sidharth
151
Using in (11) equation (4) we get lP m P n =m R Whence it follows on using (7), (6) and (1), n3/2 =
√
N, n =
¯ N
(12)
Substituting the value for n from (12) into (10) we will get lc2 G= √ ¯m N
(13)
¯m M =N
(14)
If we use (12) in (9) we will get
Alternatively we could use (14) which expresses the fact that the mass of the ¯ elementary particles in it and deduce universe is given by the mass of the N equations (11), (12) and (13). Indeed a rationale for this is the fact that the Universe at large is electrically neutral and so it is the gravitational force which predominates, and this very weak in comparison. Using the expressions for the Planck length as a Compton wavelength and equating it to (4) we can easily deduce Gm2 =
e2 e2 =√ ¯ n N
(15)
wherein we have also used c ∼ e2 and (6). Equation (15) is another empirically well known equation which was used by Dirac in his Cosmology. Interestingly, as we have deduced (15), rather than use it empirically, this points to a uniﬁed description of electromagnetism and gravitation. Interestingly also rewriting (13) as G=
l2 c2 Rm
wherein we have used (1) and further using the fact that H = c/R, where H is the Hubble constant we can deduce 1 H2 3 m≈ (16) Gc
Equation (16) is the so called mysterious Weinberg formula, known empirically [10]. As Weinberg put it, “...it should be noted that the particular combination of , H, G, and c appearing (in the formula) is very much closer to a typical elementary particle mass than other random combinations of these quantities; for instance, from , G, and c alone one can form a single quantity (c/G)1/2 with
152
How Fundamental is Gravitation?
the dimensions of a mass, but this has the value 1.22 × 1022 M eV /c2 , more than a typical particle mass by about 20 orders of magnitude! “In considering the possible interpretations (of the formula), one should be careful to distinguish it from other numerical “coincidences”... In contrast, (the formula) relates a single cosmological parameter, H, to the fundamental constants , G, c and m, and is so far unexplained.” We will come back to this point but remark that (13) brings out gravitation in a diﬀerent light– somewhat on the lines of Sakharov. In fact it shows up gravitation as the excess or residual energy in the universe. Finally it may be observed that (13) can also be rewritten as 2 2 c l ¯ N= ∼ 1080 (17) mG
and so also (10) can be rewritten as 2 lc n= ∼ 1040 Gm It now immediately follows that N ∼ 10120 Looking at it this way, given G and the microphysical parameters we can deduce ¯ and n. the numbers N, N
3
Discussion
Thus the many so called large number coincidences and the mysterious Weinberg formula can be deduced on the basis of a Planck scale underpinning for the elementary particles and the whole universe. This was done from a completely diﬀerent point of view, namely using fuzzy spacetime and ﬂuctuations in a 1997 model that successfully predicted a dark energy driven accelerating universe with a small cosmological constant [5, 11]. However the above treatment brings out the role of the Planck scale particles in the Quantum Vacuum. It resembles, as remarked earlier the SakharovZeldovich metric elasticity of space approach [12]. Essentially Sakharov argues that the renormalization process in Quantum Field Theory which removes the Zero Point energies is altered in General Relativity due to the curvature of spacetime, that is the renormalization or subtraction no longer gives zero but rather there is a residual energy similar to the modiﬁcation in the molecular bonding energy due to deformation of the solids. We see this in a little more detail following Wheeler [13]. The contribution to the Lagrangian of the Zero Point energies can be given in a power series as follows L(r) = A k 3 dk + B(4) r kdk
153
B.G. Sidharth +[C((4) r)2 + Drαβ rαβ ]
k −1 dk
+(higherorder terms).
(18)
where A, B, C etc. are of the order of unity and r denotes the curvature. By renormalization the ﬁrst term in (18) is eliminated. According to Sakharov, the second term is the action principle term, with the exception of some multiplicated factors. (The higher terms in (18) lead to corrections in Einstein’s equations). Finally Sakharov gets c3 ! G= (19) 16πB kdk
Sakharov then takes a Planck scale cut oﬀ for the divergent integral in the denominator of (19). This immediately yields G≈
c3 lP2
(20)
In fact using relations like (1), (6) and (12), it is easy to verify that (20) gives us back (10) (and (13)). According to Sakharov (and (20)), the value of G is governed by the Physics of Fields and Particles and is a measure of the metrical elasticity at small spacetime intervals. It is a microphysical constant. However in our interpretation of (13) (which is essentially the same as Sakharov’s equation (20)), G appears as the expression of a residual energy over the entire universe: The entire universe has an underpinning of the N Planck oscillators and ¯ elementary particles, which again each have an underpinning of is made up of N n Planck oscillators. It must be reiterated that (20) obtained from Sakharov’s analysis shows up G as a microphysical parameter because it is expressed in their terms. This is also the case in Dirac’s cosmology. This is also true of (10) because n relates to the micro particles exclusively. ¯ , that is However when we use the relation (12), which gives n in terms of N links up the microphysical domain to the large scale domain, then we get (13). With Sakharov’s equation (20), the mysterious nature of the Weinberg formula remains. But once we use (13), we are eﬀectively using the large scale character of G – it is not a microphysical parameter. This is brought out by (17), which is another form of (13). If G were a microphysical parameter, then the number of elementary particles in the universe would depend solely on the microphysical parameters and would not be a large scale parameter. The important point is that G relates to elementary particles and the whole universe [14]. That is why (13) or equivalently the Weinberg formula (16) relate supposedly microphysical parameters to a cosmological parameter. Once the character of G as brought out by (13) is recognized, the mystery disappears. Finally it may be remarked that attempts to unite gravitation with other interactions have been unsuccessful for several decades. However, it is possible to get a description of gravitation in an extended gauge ﬁeld formulation using noncommutative geometry (to take account of the fact that the graviton is a spin 2 particle)
154
How Fundamental is Gravitation?
[15, 16]. It is important to recognize what Witten has said, ”The existence of gravity clashes with our description of the rest of physics by quantum ﬁelds” [17]. This sentiment was echoed much earlier by Pauli who even went so far as to say that we should not try to unite what God had intended to be separate. References [1] S.W. Hawking, and Werner Israel, ”300 Years of Gravitation”, Cambridge University Press, Cambridge, 1987. [2] H. Weyl, ”SpaceTime Matter”, Denver Publications Inc., New York, 1962, 282ﬀ. [3] B.G. Sidharth, Found.Phys.Letters., 15 (6), 2002, 577583. [4] B.G. Sidharth, Found.Phys.Letters., 17 (5), 2004, 503506. [5] B.G. Sidharth, Int.J.Mod.PhysA., 13 (15), 1998, p.2599ﬀ. [6] Y. Jack Ng and H. Van Dam, Mod.Phys.Lett.A., 9 (4), 1994, p.335340. [7] S.S. Schweber, “An Introduction to Relativistic Quantum Field Theory”, Harper and Row, New York, 1961. [8] J.V. Narlikar, “Introduction to Cosmology”, Cambridge University Press, Cambridge, 1993, p.57. [9] V.N. Melnikov, Int.J.Th.Phys., 33 (7), 1994, 15691579. [10] S. Weinberg, “Gravitation and Cosmology”, John Wiley & Sons, New York, 1972, pp.619ﬀ. [11] B.G. Sidharth, ”Chaotic Universe: From the Planck to the Hubble Scale”, Nova Science Publishers, Inc., New York, 2001. [12] A.D. Sakharov, Soviet Physics  Doklady, Vol.12, No.11, 1968, p.10401041. [13] C.W. Misner, K.S. Thorne and J.A. Wheeler, “Gravitation”, W.H. Freeman, San Francisco, 1973, p.819ﬀ. [14] B.G. Sidharth, “A Note on the Characterization of Gravitation”, To appear in Foundation of Physics Letters. [15] B.G. Sidharth, ”Gravitation from a Gauge like Formulation”, to appear in Annales de la Fondation Louis De Broglie. [16] B.G. Sidharth, ”Universe of Fluctuations”, Springer (under publication). [17] W. Witten, Physics Today, April 1996, pp.2430.
SCALEDEPENDENT STOCHASTIC QUANTIZATION
MIKHAIL ALTAISKY a b
a,b
Joint Institute for Nuclear Research, Dubna, 141980, Russia Space Research Institute, Moscow, 117997, Russia
Abstract Based on the waveletdeﬁned multiscale random noise proposed in [2], a multiscale version of the stochastic quantization procedure is considered. A new type of the commutation relations emerging from the multiscale decomposition of the operatorvalued ﬁelds is derived.
1
Introduction
A highly original method of stochastic quantization of gauge ﬁelds proposed by G.Parisi and Y.Wu [1] have been attracting attention for more than 20 years. Let SE [φ] be the action Euclidean ﬁeld theory in Rd . Then, instead of direct calculation of the Green functions from the generation functional of the ﬁeld theory, it is possible to introduce a ﬁctitious time variable τ , make the quantum ﬁelds into stochastic ﬁelds φ(x) → φ(x, τ ), x ∈ Rd , τ ∈ R and evaluate the moments φ(x1 , τ1 ) . . . φ(xm , τm ) by averaging over a random process φ(x, τ, ·) governed by the Langevin equation ∂φ(x, τ ) δS + = η(x, τ ). ∂τ δφ(x, τ )
(1)
The Gaussian random force is δcorrelated in both the Rd coordinate and the ﬁctitious time:
η(x, τ )η(x , τ ) = 2D0 δ(x − x )δ(τ − τ ),
η(x, τ ) = 0.
155 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 155–161. © 2006 Springer. Printed in the Netherlands.
(2)
156
Scaledependent stochastic quantization
The physical Green functions are obtained by taking the steady state limit G(x1 , . . . , xm ) = lim φ(x1 , τ ) . . . φ(xm , τ ). τ →∞
Following [3] we extend the method of stochastic quantization by introducing the scaledependent random processes W (a, b, ·), where b ∈ Rd is a spatial coordinate, and a is the spatial resolution. For a squareintegrable function f (x, ·) the wavelet coeﬃcients are x−b W (a, b, ·) = a−d ψ (3) f (x, ·)dd x. a
Hereafter they will be referred to as the scale components of f with respect to the basic wavelet ψ. The reconstruction of a function from its scale components is given by the inverse wavelet transform ∞ ˜ 2 da x−b ψ(k)2 d d f (x, ·) = d bψ = W (a, b, ·), C d k, (4) ψ d+1 Cψ 0 a a Sd kd
with Sd being the are of the unit sphere in d dimensions, is the normalization for the isotropic wavelets. Performing the wavelet transform (in spatial coordinate) of the ﬁelds and the random force in the Langevin equation, we get the possibility to substitute the white noise (2) by a scaledependet random force ˜ η (a1 , k1 , τ1 )˜ η (a2 , k2 , τ2 ) = Cψ (2π)d δ d (k1 +k2 )δ(τ1 −τ2 )a1 δ(a1 −a2 )D(a1 , k1 ) (5) In case the spectral density of the random force is a constant D(a1 , k1 ) = D0 , the inverse wavelet transform ∞ 2 da dd k dω ˜ ˜ k, ω), φ(x) = exp(ı(kx − ωτ ))ψ(ak) φ(a, (6) Cψ 0 a (2π)d 2π
drives the process (5) into the white noise (2). In case of arbitrary functions φ(a, x, ·) we have more possibilities. In particular, we can deﬁne a narrow band forcing that acts at a single scale D(a, k) = a0 δ(a − a0 )D0 .
(7)
The contribution of the scales with the wave vectors apart from the the typical scale a−1 is suppressed by rapidly vanishing wings of the compactly supported 0 ˜ wavelet ψ(k). Here we present two examples of the divergence free stochastic perturbation expansion: (i) the scalar ﬁeld theory φ3 , (ii) the nonAbelian gauge ﬁeld theory.
2
Scalar ﬁeld theory
Let us turn to the stochastic quantization of the φ3 theory with the scaledependent noise [3]. The Euclidean action of the φ3 theory is 1 m2 2 λ 3 SE [φ(x)] = dd x (∂φ)2 + φ + φ . (8) 2 2 3!
157
M.V. Altaisky The corresponding Langevin equation is written as
∂φ(x, τ ) λ + −∆φ + m2 φ + φ2 = η(x, τ ). ∂τ 2!
(9)
Substituting the scale components in representation (6) we get the integral equation for the stochastic ﬁelds
2 ! 2 ˜ (−ıω + k 2 + m2 )φ(a, k, ω) = η(a, k, ω) − λ2 ψ(ak) Cψ ˜ 2 (k − k1 ))φ(a1 , k1 , ω1 )φ(a2 , k − k1 , ω − ω1 ). ˜ 1 k1 )ψ(a ψ(a
dd k1 dω1 da1 da2 (2π)d 2π a1 d+1 a2 d+1
(10) Starting from the zeroth order approximation φ0 = G0 η with the bare Green function G0 (k, ω) = 1/(−ıω + k 2 + m2 ) and iterating the integral equation (10), we get the oneloop correction to the stochastic Green function
dd q dΩ 2∆(q)G0 (q, Ω)2 G0 (k−q, ω−Ω)+. . . , (2π)d 2π (11) where ∆(k) is the scale averaged eﬀective force correlator G(k, ω) = G0 (k, ω)+λ2 G20 (k, ω)
2 ∆(k) ≡ Cψ
∞
da ˆ ψ(ak)2 D(a, k). a
(12)
In the same way the other stochastic momenta can be evaluated. Thus the common stochastic diagram technique is reproduced with the scaledependent random force (5) instead of the standard one (2). The 1PI diagrams corresponding to the stochastic Green function decomposition (11) are shown in Fig. 1.
Figure 1. Diagram expansion of the stochastic Green function in φ3 model
It can be easily seen that for a singleband forcing (7) and a suitably chosen wavelet the loop divergences are suppressed. For instance, the use of the Mexican hat wavelet ˆ ψ(k) = (2π)d/2 (−ık)2 exp(−k 2 /2), Cψ = (2π)d (13) for the single band random force (7) gives the eﬀective force correlator 2
∆(q) = (a0 q)4 e−(a0 q) D0 .
(14)
158
Scaledependent stochastic quantization
Table 1. Stochastic diagrams for the nonAbelian gauge ﬁelds. Redrawn from [4]
Diagram
Notation Gab µν (k, τ − τ )
ab Dµν (k, τ − τ )
Formula
" # 2 k k k k δ ab θ(τ − τ ) δµν − µk2 ν e−k (τ −τ ) + µk2 ν " 2 2 k k δ ab δµν − µk2 ν e−k τ −τ  − e−k (τ +τ ) # k k k k +2min(τ, τ ) µk2 ν + µk2 ν
g abc 2 Vµκλ (k1 , k2 , k3 )
2
g 6
abcd Wµνκλ
ıg abc (k1 − k2 )λ δµκ 2f +(k3 − k1 )κ δµλ
+ (k2 − k3 )µ δκλ
2 − g6 f xab f xcd (δµκ δνλ − δµλ δνκ ) +f xac f xbd (δµν δκλ − δµλ δνκ ) +f xad f xbc (δµν δκλ − δµκ δνλ )
The loop integrals taken with this eﬀective force correlator (14) can be easily seen to be free of ultraviolet divergences ∞ dd q dΩ 1 G2 (k, ω) = G20 (k, ω) 2∆(q) 2 + (q 2 + m2 )2 (2π)d 2π Ω −∞ 1 × (15) −ı(ω − Ω) + (k − q)2 + m2
3
NonAbelian gauge theory
The Euclidean action of a nonAbelian ﬁeld is given by 1 a a a S[A] = (x)Fµν (x), Fµν (x) = ∂µ Aaν (x)−∂ν Aaµ (x)+gf abc Abµ (x)Acν (x). dd xFµν 4 (16) The Langevin equation for the gauge theory (16) can be written as
∂Aaµ (x, τ ) (17) + −δµν ∂ 2 + ∂µ ∂ν Aaν (x, τ ) = ηµa (x, τ ) + Uµa (x, τ ), ∂τ where ηµa (x, τ ) is the random force and Uµa (x, τ ) is the nonlinear interaction term g 0 g2 V (A, A) + W 0 (A, A, A). 2 6 The stochastic diagram technique for the gauge ﬁeld Langevin equation (17) is summarized in the Table 1. The two terms standing in the free ﬁeld Green function correspond to the transversal and the longtitudal mode propagation: U [A] =
Gab µν (k) =
Tµν (k)δab Lµν (k)δab + , −ıω + k 2 −ıω
Tµν (k) = δµν −
kµ kν kµ kν , Lµν (k) = 2 . k2 k
M.V. Altaisky
159
(Here we are concerned with divergences and do not touch any gauge ﬁxing.) Similarly the scalar ﬁeld theory, we can use the scaledependent forcing (18) in the Langevin equation (17). Since there is no dynamic evolution for the longtitudal modes in the Langevin equation (17), it is natural to use the transversal scaledependent random force ηµa (a1 , k1 , τ1 )ηνb (a2 , k2 , τ2 ) = (2π)d δ d (k1 + k2 )δ(τ1 − τ2 )Tµν (k1 ) × Cψ a1 δ(a1 − a2 )D(a1 , k1 ).
(18)
Let us consider a gluon loop with two cubic vertices. Summing up over the gauge 2 2 group indices 2ı g f abc δbd f der δcr = g4 δae C2 , with C2 = N for SUN groups, we can wright the gluon loop as a sum of two diagrams – those with the transversal and the longtitudal stochastic Green functions
Gab 2µν (k, ω)
2
= g δab C2 G0 (k, ω)
2
& dΩ dd q I N I (k, ω, q, Ω)lµν (k, q)2∆(q) 2π (2π)d
I=T,L
(19) where
) )2 1 ) ) 1 2 −ı(ω−Ω)+(k−q) ) N (k, q) = )) 1 −ıΩ + q 2 ) −ı(ω−Ω) Tκσ (k − q) lµν (k, q) = Vµκλ (k, k − q, q)Tλγ (q)Vσνγ (k − q, k, −q) Lκσ (k − q) T L and lµν , As it can be observed after explicit evaluation of the tensor structures lµν and integration over dΩ, the wavelet factor in the eﬀective force correlator ∆(q) will suppress the divergences for a narrowband forcing (7). The power factor ˜ k n of the basic wavelet ψ, that provides ψ(0) = 0, also makes the IR behavior softer. In this !respect the wavelet regularization is diﬀerent from the continuous regularization dd yRΛ (∂ 2 )η(y, τ ), see e.g. [6], that makes UV behavior softer by k2
the factor e− Λ2 , but do not aﬀect the IR behavior.
4
Commutation relation
The stochastic quantization with a forcing localized at a given scale a0 is in some way similar to the lattice regularization with the mesh size of order a0 . However there is a question what is the physical sense of the scale components, and what are the implications for canonical quantization of these ﬁelds? The answer to the ﬁrst question stems from the deﬁnition of wavelet transform: the scale component φ(a, x) is a projection of the state vector φ to a certain multiresolution space [7], where ψ is a basis, i.e., the basic wavelet stands for the apperture of the microscope by which we perceive the system φ. To clarify the second question one can use the
160
Scaledependent stochastic quantization
wavelet decomposition ∞ 2 da dd k ˜ φ(x) = ψ(ak) eıkx u(a, k) + (−1)d e−ıkx u(a, −k) , Cψ 0 a k0 >0 (2π)d
u+ (a, k) ≡ u(a, k)k0 >0 ,
u− (a, k) ≡ u(a, −k)k0 >0 ,
(20)
where the positive and the negative energy components (20) are summed up into the known plane wave components ∞ 2 da ˜ u± (k) = ψ(ak)u± (a, k). Cψ 0 a
The canonical quantization of a scalar massless ﬁeld, the implies the commutation relations (21) [u+ (k1 ), u− (k2 )] = (2π)d δ(k1 − k2 ), that can be maintained if we set [5] Cψ (22) a1 δ(a1 − a2 ). 2 √ For a massive ﬁeld, with the given energy of the free particle ωk = k 2 + m2 , the commutation relations for creation and annihilation operators [u+ (a1 , k1 ), u− (a2 , k2 )] = (2π)d δ(k1 − k2 )
[b(a1 , k1 ), b+ (a2 , k2 )] = (2π)d−1 ωk δ d−1 (k1 − k2 )Cψ a1 δ(a1 − a2 ).
(23)
To keep the Lorentz invariance at all scales the basic wavelet ψ can depend only on Lorentz scalars, such as kµ k µ = m2 . Being compactly supported in both x and ˜ k spaces the wavelet ﬁlter ψ(ak) ≡ µ(a2 m2 ) suppresses the contribution of the scale components which are far from the typical scale am = m−1 . It should be emphasized that the commutation relations for scale components (22,23) are not unique: there may be constructed some other commutation relations is wavelet space that maintain the same canonical commutation relations in wavenumber space. As it concerns the causality and operator ordering, the introduction of the scale argument in operatorvalued functions implies the operators should be ordered in both the time and the scale. Extending the causality in this way it was suggested [5] to arrange the operator products by decreasing scale from right to left; so that the rightmost operator should correspond to the largest outermost object A(∆x , x)B(∆y , y) y0 < x0 ±B(∆y , y)A(∆x , x) x0 < y0 T (A(∆x , x)B(∆y , y)) = (24) A(∆x , x)B(∆y , y) ∆y > ∆x , y0 = x0 ±B(∆y , y)A(∆x , x) ∆y < ∆x , x0 = y0 .
M.V. Altaisky
161
References [1] G.Parisi and Y.S.Wu. Scientica Sinica, v.24, p.483 (1981) [2] M.V.Altaisky. Doklady Physics, v.48, p. 478 (2003) [3] M.V.Altaisky. in Group 24: Physical and mathematical aspects of symmetries pp.893897, IoP, (2002) [4] M. Namiki. Stochastic quantization, Springer (1992) [5] M.V.Altaisky. Renormalization group and geometry. Proc. Int. Conf. FFP5, Hyderabad (2003) [6] M.B.Halpern. Progr. Theor. Phys. Suppl. v.111, p.163 (1993) [7] S. Mallat, A theory for multiresolution signal decomposition: wavelet transform. Preprint GRASP Lab. Dept. of Computer an Information Science, Univ. of Pensilvania. (1986)
PRINCIPIA GEOMETRICA PHYSICAE
JAIME KELLER Department of Physics and Theoretical Chemistry, Facultad de Qu´ımica and Facultad de Estudios Superiores Cuautitl´ an, Universidad Nacional Aut´ onoma de M´exico 04510 ´ M´exico D.F., apartado 70528, MEXICO
Abstract We show that quadratic spaces are a fundamental clue to understand the structure of theoretical physics. Classical Physics is embedded within a 5D ﬂat quadratic space, 3 Space, one Time and one Action like basis manifolds (Lorentz signature 1,4) faithfully providing a Relativistic Theory (START) describing Newtonian, Maxwell, geometrical optics and General Relativity as particular linear and quadratic forms of this (ﬂat) START space. The 5D space has a quadratic form which maps into the real quadratic form of a (hyperboliccomplex) 4D spacetime dS 2 = dS • dS † . Otherwise the understanding of a many electron quantum mechanical (QM) system, with all its QM intricacies, is presented considering the QM ”density” as a quadratic form and the QM ”wave function” as its corresponding linear form.
SpaceTimeAction, General Relativity, Geometric Optics. 83D05, 83Exx, 11E88, 15A66(67). Email: [emailprotected] and [emailprotected]
1
Physics and Quadratic Spaces
Physics is understood as the science describing nature as a whole in a useful way. For the Scientiﬁc Method (SM) the usefulness requires that two observers will ﬁnd
163 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 163–173. © 2006 Springer. Printed in the Netherlands.
164
Principia Geometrica Physicae
in their experiments similar phenomena and describe them with similar logical structures. Theoretical Physics provides a mathematical, SM acceptable, framework for this purpose. In this paper we show that a ﬂat 5D Lorentz Geometry is a useful formulation to consider frames of reference. Additionally we show the relation between quadratic spaces and Quantum Mechanics. • The central purpose of the here formulated START Theory is twofold: ﬁrst to have a description of nature, the way we perceive it with our senses and experiments, which could be useful in accordance with the Scientiﬁc Method, that is which could be a sound basis for physics. Second START is aimed to be a valid general mathematical theory for all the fundamental physical objects and the frame of reference we use for their description. The physical objects are in general aspects of matter, this concept enlarged to englobe also what previously was known as radiation or particles. 2
• The quadratic form of the Minkowski space s2 = (ct) − (x2 + y 2 + z 2 ) describes kinematics (motion) in Space–Time. To faithfully include matter and interaction (dynamics) we propose a generalization of this quadratic form S 2 = s2 − w2 , where w = κ(0) a is an equivalent distance expressing the physical Action corresponding to a time interval t during which a distance l = (x2 + y 2 + z 2 ) has been covered. The deﬁnitions being such that energy–momentum corresponds to (Planck constant times) the space–time derivatives of w. We show that the resulting space s2 → S 2 = guv xu xv can also be pictured as a curved space–time. The corresponding linear form (in Cliﬀord algebra a 5dimensional vector) is S = eu xu , where the ﬁrst four components correspond to the (Lorentz)Minkowski vector s = eµ xµ . The usefulness, and in fact the motivation, of this geometry is illustrated through the analysis of light propagation in a medium and in a gravitational ﬁeld. We adopt the Poincar´e’s principle of Relativity for the formulation of the Theory [2] (set of initials: START), in all cases the base space for the description is ﬂat. In our presentation in the present paper optics was taken as an example to guide the reader.
• The 5D START geometry is presented as a ﬂat simple connected Lorentzian manifold timelike oriented in which all light rays correspond to null lines. Massive objects are described as bundles of null trajectories in START. Observers to 4D timelike trajectories. • START provides: ﬁrst a uniﬁed presentation of the main structures of theoretical physics, second a common mathematical language for seemingly different branches as geometric optics, optics in a medium, or general relativity. 1.1 THE QUADRATIC FORM IN PHYSICS We claim that quadratic spaces are the fundamental clue to understand the structure of theoretical physics. We show several examples: optics in a medium, general
165
J. Keller
relativity and, on a diﬀerent level, the understanding of a many electron quantum mechanical (QM) system, with all its QM intricacies, considering the QM ”density” as a quadratic form and the QM ”wave function” as its corresponding linear form. Our frames of reference formulation (see [1][2][3][4][5]) claims that there is a useful generalization of the quadratic form which historically started with the Pythagorean formulation: Quadratic form 2
2
2
l =x +y +z
2
Dim/diﬀ. Op (t)
s2 = (ct)2 − (x2 + y 2 + z 2 ) S 2 = (ct)2 − x2 + y 2 + z 2 − w2 d c w = κ(0) a, κ(0) = (0) h = E(0)
2
3D
∇, ∇
4D
D, 2
5D K, K 2 + (a)2 = µ a2µ
Group Statics Galileo Kinetics Poincar´e Dynamics START
where l, (x, y, z), c, h, w, a, E(0) are distance, distance components, vacuum speed of light, Planck’s constant, distance equivalent to action, action of a system and characteristic energy of the system respectively. The development of the theory follows this scheme in such a form to obtain a comprehensive theory: SP ACE + T IM E + ACT ION + F U N DAM EN T AL P RIN CIP LES and P OST U LAT ES (Carriers) ⇓ 2 ST ART GEOM ET RY were dS = dS • dS † and carriers were ρ = Ψ • Ψ† General Relativity Quantum M echanics Standard M odel In our formulation [2] of the Standard Model all symmetries are SpaceTimeAction symmetries (no need for “internal” symmetries). In the START space we introduce physics through the START Relativity Principle “All trajectories are null for all observers” “The vacuum Speed of Light is c for all observers” The set of principles we have introduced are [2]: START Relativity (5D Poincar´e group and 5D Lorentz transformations); Existence (physical objects are represented by energy densities); Least Action (null, optimal possible, trajectories in START); Quantized Exchange of Action (deﬁnes systems or subsystems as those among a quanta of action can be exchanged) and, Choice of Descriptions (allows all useful physical models to be employed). 1.2
MULTIVECTOR REPRESENTATION
The base space R5 corresponds to the real variables set {ct, x, y, z, κ0 α} ↔ {xu ; u = 0, 1, 2, 3, 4} that is: time, 3D space and action (in units of distance introducing
166
Principia Geometrica Physicae
the universal speed of light in vacuum c and the system under observation dewith energy mc2 /h = 1/m0 c). For physics time is usually pendent κ0 = λsystem Compton an independent evolution coordinate and action (matter and interaction) is distributed in space, then we consider the functions x(t), y(t), z(t) and w(t, x, y, z) = κ0 α(t, x, y, z). The four (spacetime) partial derivatives of w(t, x, y, z) gives to dS 2 an structure similar to a, modiﬁed, special relativity ds2 world line element. The nested vectors dS =
& µ
&
dxu eu ; u = 0, 1, 2, 3, 4
ds = dxµ eµ ; µ = 0, 1, 2, 3 µ & dx = dxi ei ; i = 1, 2, 3 ; ei = e0 ei i
5−D 4−D 3−D
are members of a Cliﬀord algebra generated by the deﬁnition of a quadratic form dS 2 ST ART guv
2
≡ (dS) =
& µ
dxu eu
=
& µν
ST ART guv dxu dxv ,
= diag (1, −1, −1, −1, −1) , eu ev = −ev eu
e = e0 e1 e2 e3 e4 = −e†
2
2
eu e = eeu
5D Formulation of OPTICS
Geometric optics in a medium is used as a guiding concept to enlarge the representation space of physical phenomena to 5dimensions. Optics was used in the XX century to create a logical 4dimensional geometric representation of the frame of reference to describe events in nature, the basic consideration was that of “free space” light rays. We show that considering the concept of light propagation in the medium, where the speed of light is lower, a geometry in 5dimensions appears as a natural frame of reference. This 5D space, a quadratic space, is constructed from a set of 5 variables: one timelike variable, three spacelike variables and one actionlike variable. The Lorentz Geometry fundamental quadratic form of this space is dS 2 = ds2 − (κ0 da)2 , here ds2 is the Poincar´e–EinsteinMinkowski 4D space–time quadratic form. Below we show that General Relativity is also faithfully represented in this quadratic space. 2.1 FROM SPACE–TIME GEOMETRIC OPTICS TO START The trajectory of a light ray in a medium can be represented in two equivalent forms. Consider dl = ve dt = (c/η)dt, for light propagation in a medium (refraction index η) and use the quadratic form dl2 = dx2 + dy 2 + dz 2 for the elementary trajectory in space dl2 = (c/η)2 dt2 , or (cdt)2 = η 2 dl2 = (1 + 4πχ)dl2 ,
J. Keller
167
to deﬁne 4πχ 2 2 c dt = 0, 1 + 4πχ 2 = 1 − η η−1 c2 dt2 − dx2 + dy 2 + dz 2 2
2 = (cdt)2 − dl2 − 4πχdl2 = (cdt)2 − dl2 − dSlight−η
2 dSlight−η
The relations above allow a new interpretation of the light propagation in a medium as a propagation with the vacuum speed of light but in a 5dimensional quadratic manifold. In the 5D form the ﬁfth term η 2 − 1 /η 2 c2 dt2 represents a distance equivalent to the interaction of the medium, with refraction index η, upon the light ray. We now propose the following 5dimensional construction, including the action variable, in the space–timeaction quadratic form dS 2 = ds2 − (κ0 da)2 = (cdt)2 − (dx2 + dy 2 + dz 2 ) − (κ0 da)2
(1)
where in the case of light the constant κ0 = c/hν the inverse of the momentum associated (in vacuum) with the light (one carrier’s, one photon) energy hν. If we associate then the last term above with the last term in the description of the light ray propagation (κ0 da)2 =
η2 − 1 2 2 c2 c dt = ∆ E 2 dt2 , 2 2 η (hν)
we can deﬁne the equivalent interaction potential as (here E represents light→media interaction energy) ∆ E2 η2 − 1 = 2 , η2 (hν)
therefore the action Edt equivalent to the interaction between light and the medium. In this discussion it is fundamental that these processes take place when the carrier of light is in interaction with a medium, the medium being described by an index of refraction η. 2.2
THE ACTION DIFFERENTIAL TERM
The quadratic form which is more relevant for Physics considers that observable objects are extended in space and then an action density α in spacetime is required. Then, deﬁning m(x, t)c2 = εtotal (x, t), and the (Cliﬀord algebra valued) inverse of the spacetime and the spacetime Lapla+ volume e0 e1 e2 e3 /xyzt, + cian operator = +µ eµ ∂µ such that along b = µ bµ eµ the directional change operator is db · = µ dbµ ∂µ (we apply four times for b = cte0 , xe1 , ye2 , ze3 ), we
168
Principia Geometrica Physicae
can obtain the sum of the directed changes of the density of w: 2
m(x,t)c t a(x, t)e4 = κ0 α(x, t)e4 = κ0 xyzt e=
m(x,t)c2 t 1 m0 c xyzt e
w(x,t) 0 )ct a(x, t)e4 = (m(x,t)/m xyzt e = xyzt e = w(x, t)e + edw = µ [(∂µ w(x, t))dxµ ] eµ e 2 † 2 (dS) = (dS) (dS) = 1 − (κ0 p0 ) (cdt)2 +
−
=
(m(x,t)/m0 )ct xyzt e
2 2 2 2 2 2 1 − (κ0 p1 ) (dx) + 1 − (κ0 p2 ) (dy) + 1 − (κ0 p3 ) (dz)
here pµ = ∂µ α(x, t) is a momentum density. Notice that w(x, t) is the distance equivalent to a reduced action density, this makes the approach universal for all systems.
3
The photon as a general relativity test particle
General Relativity is considered a comprehensive theory, the best known solutions are developed for the so called matterfree space and a test particle. We show that (1) corresponds to a description of the action distribution which agrees with the conceptual development of General Relativity (GR), this last theory itself being based on the physical postulate that all observers have the right to consider their measurements equally valid.. 3.1 THE BASIC START SOLUTION There are two fundamental (energy)carrier structures: the massless (as the photon) and the massive ﬁelds with basic relation 2
E 2 = (E0 + ∆E) , E 2 − E02 = (pc)2 ,
(2)
where ∆E is any gaugefree energy contribution and E0 = m0 c2 =⇒ hν (for a photon). The concept of test particle (at position {r, θ, φ}) in general relativity is compatible with the Newtonian limit for the interaction gravitational energy ∆E (r) = −m0
GM , r
(3)
where M is the total mass of ‘the external system’ (conﬁned within a radius rs ) which we are exploring with the test particle. START uses the action square diﬀerence, writing E = E0 + ∆E for large (classical limit) values of r > rs 2
2
E 2 − E02 = E02 + 2E0 ∆E + (∆E) − E02 = (pc)
2 GM GM 2 2 = 2E0 ∆E + (∆E) → −2m0 c m0 + m0 , r r
(4)
169
J. Keller
this corresponds both to the energy (diﬀerence square) and radial momentum terms in (dA)2 − (dA )2 if (dA )2 = (m0 c2 dt)2 , and according to the GR basic description principles, substituting (the negative of (4)) in (1) using κ0 = 1/m0 c and space–time spherically symmetric coordinates t, r, θ, φ we obtain 2 " # , GM GM 2 2 2 2 2 2 2 2 (dS) = 1 − 2 2 + c (dθ) , (dt) − (dr) + r + sin θ (dφ) c r c2 r
with the same physical consequences for the so called “tests of GR” as those of the Schwarzschild [1916] metric in the limit of r GM/c2 . The speed of light in this metric remains to be c. 3.2
GENERAL RELATIVITY VIEWED AS A LORENTZ TRANSFORMATION
The gravitational ﬁeld induces a coordinate transformation where it can be required that the vacuum speed of light is c for all observers. It is shown that this transformation has a formal structure similar to, and includes, the Lorentz transformation. The new (LorentzKeller) transformation simpliﬁes General Relativity (GR), giving to this theory a Special Relativity structure. It also removes some conceptual diﬃcult points of GR and simpliﬁes the computation of solutions to the problem of ﬁnding the corresponding quadratic form. In the Lorentz transformation, written as a ds2 = (cdt´)2 − (dx´)2 form 2 2 vdx vdx vcdt vcdt c2 2 2 2 ds = (cdt) [1 − 2 + ] − (dx) [1 − 2 + ] ccdt ccdt cdx cdx c2 − v 2
consider the case where dx = dl = cdt, the case for light propagation, to write (cdt)2 [1 − 2
v c
+
v 2 c
] − (dl)2 [1 − 2
v c
+
v 2 c
]=0
a transformation symmetric in t and x, with a form similar to a solution of the Einstein’s equations. Notice that v is an independent variable 0 ≤ v ≤ c. Our examples above have the same form and the addition of the action term dw2 has imposed that dS 2 should be null: dS 2 = ds2 − dw2 = 0.
4
Action Carriers and Quantum Mechanics
We proceed in the following logical sequence: formally deﬁne a carrier as a distribution ﬁeld, decompose this ﬁeld into the product of an action per unit density and a density ρ, impose an observability condition to this density, as a result an auxiliary amplitude function Ψ is required, show that Ψ is a vector (multivector) to be deﬁned as a locally dependent sum of basis vectors (multivectors).
170
Principia Geometrica Physicae
4.1 DEFINITION OF CARRIER FIELDS Within our fundamental formulation we will have to deﬁne properties of the ﬁelds we call “carriers” (see Keller and Weinberger [4] ). A carrierdomain B is a connected open set whose elements can be put into bijective correspondence with the points of a region (domain in some instances) B of a Euclidean point space E. Use X to denotes a representative element of B and x the position relative to an origin 0 of the point x occupied by X in B. At each X a scalar quantity is given, called carrier density (X ), such that if x = φ(X , t) then (X ) → ρ(x,t) deﬁnes a scalar ﬁeld called local carrier density. A carrier will have physical signiﬁcance through its set of properties. The value of the density of a carrier ﬁeld can be deﬁned through a set of fundamental scalar constants (main examples: mass, electric charges, weak charge, strong charge, spin) such that the integral of the product of these constants and the density gives the experimentally attributed value of a property for that carrier. The density might become an indirect observable through the repeated measurement of those properties, but it is not an observable in itself. We will use an example. A carrier ﬁeld identiﬁed with an electron will have a density ρ(x, t) and if the property is Q we will obtain the deﬁnition ! ! Q = V q(x, t)dx = V Qρ(x, t)dx for all t in the system’s volume V, which deﬁnes that Q is a constant property (in space and time) for that ﬁeld (otherwise the variable quantity q(x, t) can be called “the density of Q ”) The set of properties {Q} characterizes a carrier ﬁeld and in turn establishes the conditions for a density ﬁeld to correspond to an acceptable carrier. 4.2 COMPOSITE, DECOMPOSABLE, AVERAGE AND AVERAGE DESCRIPTION OF CARRIERS There are several forms of analyzing the density. Each one allows a physical interpretation. For example: • A composite carrier is deﬁned as one for which the density & ρC (x, t) = Ac ρc (x, t),
(5)
c
with the deﬁnition of each of the ρc (x, t) being ! meaningful as a description of a carrier itself. In particular we can choose V ρc (x, t)dx = 1 and Ac =Nc . • Similarly a nondecomposable, (nondecomposable) elementary, average, etc. carrier can be deﬁned. 4.3 THE DENSITY To properly describe systems in nature the conditions to be obeyed by the analytical function carrier density ρc (x, t) are:
171
J. Keller
D1. ρc (x, t) is a real quantity ρc (x, t) ⊂ R. D2. The density 0 ≤ ρc (x, t) < ∞ in order to represent a ﬁnite amount of action. D3. The derivatives of the density −∞ < ∂µ ρc (x, t) < +∞ in order to represent a ﬁnite amount of energy–momentum. Theorem (Keller and Weinberger 2002) If Ψ(x, t) is an analytical quadratic integrable complex or multivector function, conditions D1, D2 and D3 are 2 2 fulﬁlled identically if ρc (x, t) = Ψc (x, t) . Here f  means the real quadratic form of any more general function f , even if f itself is not necessarily a real 2 2 function and we deﬁne: if f  = f + f then ∂µ f  = (∂µ f + )f + f + (∂µ f ). 2 Condition D1 is fulﬁlled by the deﬁnition ρc (x, t) = Ψc (x, t) , D2 by the 2 requirement of quadratic integrability, D3 by the deﬁnition ∂µ f  = (∂µ f + )f + f + (∂µ f ) and the !analytical properties of Ψ(x, t). It is seen that the conditions D1, D2, D3 and V ρc (x, t)dx = Nc correspond to the Ψ(x, t) being quadratic integrable Hilbert functions. Thus, even if we selected the density (current in the general formulation) as the basic function, the auxiliary amplitude function Ψ is unavoidably necessary. 4.4
THE LINEAR FORM CHARACTER OF THE MANY ELECTRON AUXILIARY AMPLITUDE FUNCTION
Besides all the conditions mentioned above that the auxiliary amplitude function should fulﬁll, in the case of a many carriers system there are some additional conditions. Those conditions arise from two fundamental principles: equivalent carriers in the same system should correspond to the same density and, second, the statistics which the carriers should obey among themselves are to be included in the auxiliary amplitude function. We consider the case of electrons, fermions, then the statistics are the FermiDirac statistics and require: • The density for the N equivalent carriers system is to be+constructed as a M ≥N sum of M independent alternative contributions ρ(x, t) = i=1 ρi (x, t, si ). • There should be at least one linearly independent function (pseudocarrier amplitude) contributing to the density for each of the N equivalent carriers in the system ψc (x, t) =
aic aic
=
M ≥N 1 & i √ ac ϕi (x, t, si ) aic = bc αi N i=1 &M . .2 . i .2 .ac . ≤ 1 ; .aic . = N −aic aic
(6)
i=1
• The total amplitude function should be a sum of single (pseudo)carrier amplitude functions ψc (x, t, s), such that the exchange among two carrier’s of the space–time–spin (x, t, s) descriptions, in fact the exchange of the indexes
172
Principia Geometrica Physicae c and c , should correspond, to avoid two descriptions being identical, to a change in sign of the many carrier amplitude function Ψ & & † Ψ= ω c ψc (x, t, s) and Ψ = c ψc (x, t, s) (7) c
c
This deﬁnes Ψ as a vector, linear form, expressed in the basis {ω c }. $ % • The Ψ, Ψ are deﬁned and the products ordered to obtain 2 Ψ dx = ΨΨdx = N
(8)
The conditions which the ϕi (x, t, s) and the ω c should obey are simply those which correspond to the two fundamental principles above: ϕi (x, t, si )ϕi (x, t, si )dx =δi i ; ω c ω c = −ω c ω c ; also c ω c = δ cc (9) The 1st condition is double: ﬁrst the orthonormality among the ϕi functions (requiring them to be eigenfunctions of the same diﬀerential equation operator) to fulﬁll the condition of making linear independent combinations ψc and second the Grassmann character of the aic coeﬃcients, to make the local density per carrier )2 ) correspond tho the sum of the squares )aic ϕi (x, t, si )) , and the 2nd condition, equivalent to the Pauli principle, deﬁnes the ω c as Grassmann variables and the c as their Grassmann conjugates. Full presentation see [4]. Acknowledgments: The technical assistance of Mrs. Irma Vigil de Arag´on is gratefully acknowledged. This project was additionally supported by the UNAM program PAPIIT number IN1131023. References [1] J. Keller, The Foundations of Density Functional Theory and Wave Quantum Mechanics, Rev. Soc. Quim. Mex. 44(1) 2000, 22–28 [2] J. Keller, The Theory of the Electron A Theory of Matter from START, From the series: Fundamental Theories of Physics, Kluwer Academic Publishers, Dordrecht, Vol. 115 (2001), ebook http://www.wkap.nl/prod/b/0792368193 [3] J. Keller, General Relativity from START, in: Cliﬀord Analysis and Related Topics, K. G¨ urlebeck X. Ji, W. Spr¨ ossig Eds., Adv. in Applied Cliﬀord Algebras 11(S2) (2001), 183204. Also J. Keller, START: 4D to 5D generalization of the (Minkowski)Lorentz Geometry, Proceedings of the Meeting Lorentzian GeometryMurcia 2003 in the series: Publicaciones de la Real Sociedad Matem´ atica Espa˜ nola, vol. 6 (2004). [4] J. Keller, General Relativity as a Symmetry of a Uniﬁed SpaceTimeAction Geometrical Space, Proceedings of Institute of Mathematics of NAS of Ukraine, 43(2) also J. Keller, A Derivation of Quantum Mechanics from START, idem 50, Editors A.G. Nikitin, V.M. Boyko and R.O. Popovych (and I.A.Yehorchenko), Kyiv, Institute of Mathematics, (2002). 557568 and (2004).811820, also J. Keller and Weinberger P., A Formal Deﬁnition of Carriers , Advances in Applied Cliﬀord Algebras, 2002, 12(1), 3962. http://www.imath.kiev.ua/activities.php
J. Keller
173
[5] J. Keller, Uniﬁcation of Electrodynamics and Gravity from START, Annales de la Fond. Louis de Broglie, 27(3) (2002) 359. http://www.ensmp.fr/aﬂb/AFLB272/aﬂb272p359.htm
DO REAL NUMBERS OBSCURE REAL PHYSICS?
G. N. ORD, a b
a,b
Perimeter Institute of Theoretical Physics, Waterloo, Ont. Canada Permanent address: Mathematics Department, Ryerson University, Toronto Ont. Canada
Abstract The real number line has served classical physics well as a model continuum for both space and time. Quantum mechanics preserves this continuum picture of space and time while extending the range of functions on spacetime to complex numbers. The resulting theory is precise, elegant, experimentally veriﬁed and universally accepted. However, since the existence of a ‘realistic’ basis for quantum mechanics remains an open question, there is as yet no consensus on a satisfactory interpretation of the theory. The step in the transition from classical to quantum physics that characteristically impedes attempts to construct realistic models is the special extension of functions from R to C. In quantum mechanics this step is formal and algebraic in nature. This talk discusses an alternative in which the geometry of a spacetime path automatically builds the extension from Z to C, with all spacetime functions simply counting contributions of a single path. The model produces the correct continuum propagator in the continuum limit, but the realistic origin of the propagator is only apparent when the smallest scale in the system (Planck scale) remains ﬁnite. This type of model is consequently missed by current formulations of quantum mechanics because the continuum limit is assumed at the outset. This suggests that it may be fruitful to look to discrete formulations of quantum mechanics in the quest for an ontological version of the theory.
175 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 175–182. © 2006 Springer. Printed in the Netherlands.
176 1
Do Real Numbers Obscure Real Physics? Introduction
In Classical physics, the correspondence between the physical world we describe, and the mathematics we use, is usually very close. When we write an equation 2 such as F = m ddt2x , we are tracking the position of an object x as a function of time t, and here the diﬀerential equation seems a very natural language to use. The extra assumption brought in by the language is that the trajectory of our physical object is smooth in space and time. In terms of classical physics this seems harmless enough. However there is reason to question the smoothness of space and time in quantum physics. In quantum mechanics, both momentum and energy vary with an inverse length, be it space or time, so the concept of localization over small distances or times is problematic. The language we use, for example Schr¨ odinger’s equation, maintains the classical paradigm of a smooth background continuum by changing from a description of pointlike objects to wave propagation. This has worked spectacularly well and quantum mechanics provides a theory with unequalled empirical accuracy and generality. The price we pay is that we no longer know exactly what it is that the mathematics is describing. Whereas we can easily believe that the x(t) in Newton’s law above describes the centre of mass of an object. It is not easy to say what is described by a solution of Schr¨ odinger’s equation. When we look at how we quantize a classical system, we can see in the prescription where the break with classical ontology occurs. For example, for the one p2 dimensional free particle the energy is related to the momentum by E = 2m . This ∂ ∂ system is quantized by replacing E and p by operators: E → i ∂t and p → −i ∂x 2
2
∂ ψ to give the wave equation i ∂ψ 2 . Here the main harbinger of quantum ∂t = − 2m ∂t√ mechanics is the explicit presence of i = −1 in the partial diﬀerential equation. Without the i, the above equation is just a diﬀusion equation and may be derived directly from a discrete random walk model. The presence of i in Schroedinger’s equation and the resulting extension of the number system from R to C is responsible for waveparticle duality and all the interference eﬀects associated with quantum mechanics. What is unclear is what physical process is behind the algebra. Note that whatever the process itself, it is unlikely to be directly related to the conception of a continuum of phase. That is, if we relate the i in Schr¨ odingers equation to a real phase φ via i = eiφ , the number φ = π/2 would have to be extremely accurately maintained by nature in order for solutions of the Schroedinger equation to survive domination by diﬀusion. Whereas we can think of the ’diﬀusion constant’ 2m as a real number because our solutions to Schroedinger’s equation would be stable under a variation of the constant to say one part in 1020 , we cannot say the same for φ. If it is varied by one part in 1020 the solution of the resulting equation changes qualitatively. It would seem that, in spite of the solutions of Schr¨ odinger’s equation having continuous phase, the origin of phase in nature is more likely to be discrete than continuous. This means that if we are to ﬁnd a physical origin for the complex algebra of quantum mechanics, it makes sense to look for it in a discrete space prior to taking a continuum limit.
177
G. N. Ord
This talk gives an example of the above scenario. We construct a realistic subquantum dynamic in 1+1 dimensions through a completely deterministic discrete process. We do this by essentially drawing a picture on spacetime with a single path using a severely constrained geometry. The constrained geometry leads to the required algebra in a fairly transparent way. A free particle propagator may be constructed so that, in the continuum limit, it is appropriate for a Dirac particle. However the continuum limit itself essentially erases the evidence for the underlying dynamical process. The technical details of the model are not diﬃcult but will appear elsewhere. [1] What I want to do is to sketch the ideas involved. First of all we might ask: Q. Is i really necessary? There are many famous statements stressing the necessity of complex numbers in QM but in the end, surely i is just a notational convenience? A. Yes, but because the empirical accuracy of QM is so convincing, the algebra of quantum mechanics is something that is clearly recognized by nature. What I would like to argue is that in terms of number systems, the origin of Complex numbers in QM is based on integer counting and geometry. The idea is that, for example, in the Euler formula:
eimt
= cos(mt) + i sin(mt) ∞ ∞ & & (mt)2k (mt)2k+1 = (−1)k (−1)k + i (2k)! (2k + 1)! k=0
(1)
k=0
Nature does not construct the alternating signs in the series expansion by having an algebraic analog of i speciﬁcally, but through geometry she does have a way of implementing the alternating signs in the series. That is, if Nature can add and subtract integers, she can build the even and odd components of eimt through geometry, thus imitating a continuum of phase. Q. If the alternating signs in the above series expansion have a geometric signiﬁcance, where does the addition and subtraction come from. A. We can at least imagine a scenario that would appear to observers moving forward through time that Nature is able to add and subtract. (This is a wellknown example due to Feynman.) Consider Fig. 1 where we sketch a spacetime trajectory that progresses from past to future but contains a single closed loop. An observer moving forward in time sees a single particle before time tA . At time tA two new particles appear and move away from each other. At time tB they both reverse direction and at time tC they both disappear. Notice that continuous closed loops in spacetime, if they occur in Nature, have the appearance of particle pair creation and annihilation. We would say that the pair of particles created consist of a particle and its antiparticle. The creation and annihilation then give us a physical analog of addition and subtraction of integers. In the ﬁgure, if we want to count the number of continuous trajectories in spacetime, we would associate a +1 with forward moving path segments and a 1 for reversed time segments. The total count for the ﬁgure, for any t, is then just +1, even though there appears to be more than one particle in the system between tA and tC . That
178
Do Real Numbers Obscure Real Physics?
Figure 1. On the left is a continuous spacetime trajectory that has a single closed loop. When viewed by an observer moving forward in time (right sketch) the corner in the path at A looks like a pair creation event with a particle moving oﬀ to the right and an antiparticle moving oﬀ to the left. Particle and antiparticle annihilate at C. Antiparticles provide a geometric mechanism for subtraction and can be thought of as a discrete precursor of phase.
is, the one continuous trajectory in the system is represented by the integer 1 for t < tA and by 1 + 1 − 1 for tA < t < tC . Notice that the natural subtraction of integers signalled by this simple spacetime loop is not present in spacetime paths that are constrained to have no loops. Thus, for example, the Brownian paths of diﬀusion have no loops and are incapable of imitating subtraction. Simply put, we can use this integer counting of particles and antiparticles to simulate continuous phase and ultimately the ’quantum mechanical’ extension from R to C. All we need is this natural addition–subtraction mechanism combined with a simple spacetime geometry. To see how this works consider Fig. 2. Here a
Figure 2. An entwined path with two loops appears on the left. The particle travels from t = 0 to t = 4 on the zigzag path that starts oﬀ moving to the right(dark line in ﬁgure). At t = 4 the particle zigzags back to the origin. The time reversed portions of the trajectory contribute to a counting process with opposite sign to forwardint segments. We count loops using ‘envelopes’ one of which is pictured to the right of the path. The lower two segments of this envelope contribute positively to a density, the upper two negatively. The resulting ’density of right moving particles on the envelope’ is pictured to the right. The density is 1 for 0 < t < 1 because the envelope counts a single right moving particle on that interval. The density is −1 for 2 < t < 3 because the envelope counts a single rightmoving antiparticle in the interval. The density is 0 on the two remaining intervals because there are no rightmoving particles on those intervals.
G. N. Ord
179
Figure 3. Evolution of the trigonometric functions. The left frame shows a single entwined path with the associated right and left densities. The second frame shows the path whose densities form a periodic array of alternating delta functions. The third frame shows the emergent trigonometric functions with the associated path. The details of the path are below the resolution of the ﬁgure at this point. In conventional quantum mechanics, the continuum language we use assumes the smooth version of the trigonometric functions on the right of the ﬁgure. That being the case, a path that forms the function would be below the resolution of the language itself.
single spacetime path is sketched in which a particle zigzags in space while moving forward in time, and then zigzags on the return journey to the origin in such a way that the forward and reversed paths are entwined. An observer moving forward in t would see this as a pair creation event at t = 0, an annihilation/creation event at t = 2 and an annihilation event at t = 4. However, since it is really particle pairs that always appear from closed loops, let us just count pair contributions by agreeing to count particles to the right of the line x = 0. We shall also count by spacetime direction. Since we shall have the perspective of an observer moving forward in time we shall associate a +1 with particles moving in a +t direction and a −1 for particles moving in a −t direction. Note that in the ﬁgure the density of rightmoving particles alternates in sign. In Fig. 3 we ﬁrst see a single path that has traversed the tdirection four times instead of just two times as in the previous ﬁgure. This time the regular geometry has been shifted so that the blank parts of the particle density have been ﬁlled in to produce square wave densities. Here both left and right moving densities are plotted. Note that the regular geometry forces the left and right densities to be automatically shifted by 1/4 wavelength. We can repeat the construction shifting the trajectories in t slightly at each return in just such a way as to form periodic alternating delta functions as in the centre frame of the ﬁgure. Finally we can repeat the process to form the two trigonometric functions pictured in the last frame. Notice that although the path on the left looks complicated, it is still only a single path and the geometry is completely regular except for shifts of ’phase’ at the origin. Notice also that if we are thinking of this as a model for the way in which a particle could write phase on spacetime, the smallest scale given by shifts of phase could be identiﬁed with the Planck scale. The wavelength of the trig function density that we wish to construct is the Compton length. The two densities are then the two components of eimt .
180
Do Real Numbers Obscure Real Physics?
Figure 4. On the left, a few velocity eigenpaths are shown with velocities ranging from −.9 c to +.9 c. These diﬀer from a v = 0 eigenpath by a Lorentz transformation. On the right is a scan with a denser coverage of velocities. There is still a single path here but the details are below the resolution of the ﬁgure. Note the alternating bands corresponding to particle or antiparticlerich areas. The oscillation between the regions is the origin of phase in this model.
Q. It does appear that you could construct the real and imaginary components of eimt this way, but what relevance would this have to quantum mechanics? A. In all current versions of quantum mechanics, our language starts in the continuum. In Fig. 5 we can imagine taking the continuum limit (’Planck length’ goes to zero) so that the discrete approximations to the trig functions on the right approach their continuum version. The path on the left would then be inﬁnitely long, and we would not be able to infer its presence with the language of the continuum. This is similar to the relationship between kinetic theory and thermodynamics. We can infer the latter from the former, but the language is wrong for going in the opposite direction. In this case if we start with the assumption of a continuum phase, we shall not be able to see that there is a way that nature could, in principle, approximate continuum phase using only a single continuous spacetime path for each physical ’particle’ in the system. Q. The process outlined gives a method for showing how we get eimt from simple geometry and a continuum limit. How would we get a quantum propagator from this? A. This was essentially the content of Feynman’s ’Chessboard model’ which started out as a problem in Feynman’s book on path integrals [2], and subsequently developed into a small industry in a series of papers including [3, 4, 6, 5] and [7, 8]. All we really have to do is to notice that the paths we have generated correspond to a particle that has propagated with a macroscopic velocity v = 0. We can get the Dirac propagator from this model by applying a Lorentz transformation to get other velocities. We can still traverse all these ’velocity eigenpaths’ with a single trajectory because they may all be joined serially at the origin. Fig. 4 shows a series of ‘velocity eigenpaths’ which are just our v = 0 path of the previous ﬁgure viewed from a moving frame. The dense scan to the right illustrates the hyperbolic geometry through the alternating light/dark patterns. It is the ﬂuctuation of particle/antiparticle rich areas that imitates the phase of quantum mechanics and
G. N. Ord
181
Figure 5. Here the velocity eigenpaths are chosen to correspond to the energy eigenstates of a particleinabox. The initial condition is established by choosing the correct velocity eigenpaths, and moving the origin of the eigenpaths along the xaxis, shifting the tcoordinate in such a way that the propagator adds to the appropriate eigenfunction of the box. Here the typical standing wave pattern is still the manifestation of a single continuous trajectory. There is no canonical quantization, analytic continuation or wave mechanics in this picture.
allows us to construct the Dirac propagator from a single entwined path. Since the propagator can be constructed in this way we can simulate onedimensional quantum phenomena with this ontological model. Fig. 5 is an illustration of this. The ﬁgure appears to be the standingwave pattern corresponding to an excited state of a particleinabox. Within numerical error this is exactly what it is. However it was generated by a single entwined path. The path is a concatenation of two velocity eigenpaths corresponding to the required energy, shifted in such a way that the path densities at t = 0 correspond to the eigenfunctions of the box. The single spacetime path then just propagates the pattern in t through its intrinsic geometry. Physically this is interesting because to get Fig. 5 we have not had to resort to quantization, analytic continuation or the introduction of waves. We are still solidly within the particle paradigm and the ‘waves’ appearing in the ﬁgure are ultimately a manifestation of simple spacetime geometry.
2
Conclusions
The thesis of this paper is that it is possible that the power and convenience of a language can sometimes be an impediment to its use as a tool to help understand nature. The use of real numbers as a representation of space and time is useful and intuitive. However, it may be the case that when we build the continuum limit into our language, we may not be able to see features hidden by that limit, and as a result we may have to build structures after the limit when they have a more natural appearance before the limit. We illustrated a case in point. Quantum mechanics
182
Do Real Numbers Obscure Real Physics?
typically assumes real numbers represent space and time and as a result has to propose a formal analytic continuation in the replacement of dynamical variables by operators. So the transition R → C is formal. The above demonstration shows that the transition Z → C can be mediated by simple spacetime geometry. This onlyrequires that we leave the continuum limit to the last in our construction.
3
Acknowledgments
I would like to thank Professor Siddarth for his kind invitation to the meeting. It is also a pleasure to thank the FFP6 organizers for running an excellent conference! This research was funded by NSERC. References [1] G. N. Ord, J. A. Gualtieri, and R. B. Mann . [2] R. P. Feynman and A. R. Hibbs, Quantum Mechanics and Path Integrals (New York: McGrawHill, 1965). [3] H. A. Gersch, Int. J. Theor. Physics 20, 491 (1981). [4] T. Jacobson and L. Schulman, J. Physics A 17, 375 (1984). [5] G. N. Ord, J. Stat. Phys. 66, 647 (1992). [6] B. Gaveau, T. Jacobson, M. Kac, and L. S. Schulman, Phys. Rev. Lett. 53, 419 (1984). [7] L. H. Kauﬀman and H. P. Noyes, Phys. Lett. A 218, 139 (1996). [8] G. N. Ord and J. A. Gualtieri, Phys. Rev. Lett 89 (2002).
CANTORIAN SPACE IN NATURE AND DYNAMICAL SYSTEMS
GERARDO IOVANE a , PAOLA GIORDANO a , SAVERIO SALERNO a a
Dipartimento di Ingegneria dell’Informazione e Matematica Applicata, Universit` a di Salerno, via Ponte Don Melillo, 84084 Fisciano, Italy
Abstract In this paper we consider the motion of pointlike massive systems on a Cantorian path. The results is that an harmonic material support with an external force is equivalent, under some assumptions, to a classical or quantum harmonic system on a continuous support. The idea that we want stress in this paper is that a Cantorian space could explain some relevant stochastic and quantum processes, if the space acts as an harmonic oscillating support, such as happen in Nature. This means that a quantum process, in some cases, could be explained as a classical one, but on a non continuous and fractal support. We consider the validity of this point of view, that in principle could be more realistic, since it describe the real nature of the matter and space, which does not only exist in Euclidean space or curved one, but in a Cantorian one.
1
Introduction
Nature shows us structures with scaling rules, where clustering properties from cosmological to nuclear objects reveals a form of hierarchy. Moreover, many systems shows an oscillatory behaviour. In the previous papers, the authors consider the compatibility of a Stochastic SelfSimilar, Fractal Universe with the observation and the consequences of this model. In particular, it was demonstrated that the observed segregated Universe is the result of a fundamental selfsimilar law, which
183 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 183–190. © 2006 Springer. Printed in the Netherlands.
184
Cantorian Space in Nature and Dynamical Systems
generalizes the Compton wavelength relation (see [1], [2], [3]). Starting from an universal scaling law, the author showed its agreement with the well–known Random Walk equation or Brownian motion relation [4], [5]. The main results by Iovane in [2], can be summarized as follows. Theorem 1. The structures of the Universe appears as if they were a classically selfsimilar random process at all astrophysical scales. The characteristic scale length has a selfsimilar expression R(N ) =
h h N 1+φ = N φ, Mc mn c
where the mass M is the mass of the structure, mn is the mass of a nucleon, N is the number of nucleons in to the structure and φ is the Golden Mean value. In terms of Plankian quantities the scale length can be recast in
lP c (1+φ) RP (N ) = . N mP G
The previous expression reﬂects the quantum memory of the Universe at all scales, which appears as hierarchy in the clustering properties. Theorem 2. The mass and the extension of a body are connected with its quantum properties, trough to the relation EE,N (N ) = EP N 1+φ , that links Plank’s energy EP = hν and Einstein’s one EE = mc2 . The quantum memory is reﬂected at all scales and it manifests itself trough a clusterization principle of the mass and extension of the body. In the present paper we study some relevant force ﬁeld on Cantorian space and analyze the diﬀerences with respect to the analogous case of a continuum, from a classical and quantum point of view. The idea that we want to stress in this paper is that a Cantorian space could explain some relevant stochastic and quantum processes, if the space acts as an harmonic oscillating support, such as happen in Nature. In other word, the vision is that an apparent indetermination, linked with a fractal support rather than a continuous one, can produce an indetermination on the motion of a physical object, which is explained via a stochastic or quantum process. This means that a quantum process, in some cases, could be explained as a classical one, but on a non continuous and fractal support. Consequently, an external observer looking at the motion of a particle under a ﬁxed solicitation can measure an unusual behaviour with respect to a continuous material support, that is obvious with respect to the knowledge of the fractal support behaviour. In this case, he can make the hypothesis of an indetermination or a stochasticity in the process (motion), while there is just really only ignorance with respect to the support on which the motion take place.
185
Gerardo Iovane, et al.
Let us introduce some mathematical tools. In descriptive set theory and the theory of polish spaces we ﬁnd the following deﬁnitions [6]. Deﬁnition 1. When a space AN is viewed as the product of inﬁnitely many copies of A with discrete topology and is completely metrizable and if A is countable, then the space is said to be polish. In particular, when A = {0, 1} , A = 2, then we call C = 2N Cantor space. For −1 −1 A deﬁned in an√interval: A ∈ ]0, 1[, then CF = AN is called a fuzzy Cantor −1 space. If A = ( 5−1)/2 and N = n−1, where −∞ ≤ n ≤ ∞, then CF = ε(n) is the Einﬁnity Cantorian space. Mohamed El Naschie in [7] showed the relationship between the Cantor space C and ε(∞) . As He reports: ”the relationship comes from the cardinality problem of a Borel set in polish spaces Thus we call a subset of a topological space a Cantor set if it is homeomorphic to the Cantor space”. $ % Let us consider a set of intervals A = A(1) , A(2) , ...., A(n) with
A(0) A
(1)
A(n)
= [a, b]
(1)
= [a, a + (b − a)/3] ∪ [a + 2(b − a)/3, b] .... = [a, a + (b − a)/3n ] ∪ ... ∪ [a + 2(b − a)/3n , b] .
where a, b are real numbers. If w is the level corresponding to A(w) the number of extreme points is 2w . For a = 0 and b = 1 we obtain the Cantor space C = ∩ Cn with the following n∈N
well know properties: • C is compact, with null Reinmannian measure; • There are no intervals in C; • C has the cardinality of continuum. For our purpose we consider the set A. In particular at the level w the length of a segment is kw+1 = (b − a)/3w with w = 0, 1, 2, .., n − 1. To evaluate the extremes for each a level w without using an iterative procedure we show the following algebraic method. Let us introduce the vector gm = (gm,1 , gm,2 , ..., gm,2n ) with m = 1, ..., n − 1 and with 2n components, whose values are: 2 if z = 2m (1 + 2s) + h 1 if z = 2m (1 + 2s) gm,z = (2) 0 elsewhere with s = 0, ..., 2n−m−1 − 1 and h = 1, ..., 2m − 1. For m = n we have a vector with all components equal to zero except the component at the place 2n , that is equal
186
Cantorian Space in Nature and Dynamical Systems
to 1, i.e., gn = (0, 0, ..., 0, 1). Consequently the coordinate of the extremes at a ﬁxed level n is given by the following vector A(n) z
=a+
n−1 &
kn−p gp+1,z .
(3)
p=0
Clearly appear how the previous results can be also used on the set A, which generalizes the Cantor space on the interval [a, b] instead of [0, 1].
2
An elastic force ﬁeld on Cantorian space
As we said a Cantorian space could explain some relevant stochastic and quantum processes, if the space acts as an harmonic oscillating support, such as happen in nature. In other words the vision is that an apparent indetermination, linked with a fractal support rather than a continuous one, can produce an indetermination on the motion of a physical object, which is explained thanks to stochastic or quantum process. This means that a quantum process in same case could be explained as a classical one, but on a non continuous and fractal material support. To show this result we have to consider a classical harmonic oscillator on a fractal support and then we have to consider it not as the subject of our study, but as the scenario where a process could happen. (1) Let us " suppose #that an elastic force ﬁeld f (q) = −αq is on the interval A = (1) (1) [a, b] = A1 , A2 , moreover α is a nonnegative elastic constant; it is well known the motion equation for a material point of mass m coming from the Newtonian equation f = ma is, 2 ··(1) q + ω (1) q = 0, (4) where ω (1) = α/m and with the solution (1) q (1) (t) = A2 cos ω (1) t ,
(5) (1)
· (1)
where we have assumed the initial condition q (1) (t = 0) = A2 = b and q (t = 0) = 0. This can be seen as the level 1 of a recursive procedure, where we consider a fractal support. In fact, at the level 2 we have A(2)
= [a, a + (b − a)/3] ∪ [a + 2(b − a)/3, b] # " # " (2) (2) (2) (2) = A1 , A1 + k2 ∪ A1 + 2k2 , A4 " # " # (2) (2) (2) (2) = A1 , A 2 ∪ A 3 , A 4 ,
187
Gerardo Iovane, et al. (1)
(1)
where k2 = (A2 − A1 )/3 = (b − a)/3; consequently we consider the force ﬁeld # " (2) (2) −(2) αq ∀q ∈ A1 , A2 , " # (2) (2) (2) 0 ∀q ∈ A2 , A3 , (6) f (q) = # " −(2) αq ∀q ∈ A(2) , A(2) , 3 4 where (2) is a parameter to take into account the reduction of the material support and to preserve the total energy. We will determine it in the following. The total path of the level 2 can be see as composed by three subpath: the ﬁrst and the third with an oscillation motion, while the second where we ﬁnd a uniform motion. In fact, it is easy to ﬁnd for the third subpath (2) (2) q3 (t) = A4 cos ω (2) t , with
ω (2)
2
2 = (2) ω (1) , where we have considered the initial conditions
(2)
(2)
q3 (t = 0) = A4 ond subpath
· (2)
= b and q3
(t = 0) = 0. Analogously, we get for the sec
(2)
(2)
(2)
q2 (t) = β2 t + A3 , (2)
(2)
(2)
where we have used the following initial condition q3 (t1 ) = q2 (t1 ) = A3 , and · (2) (2) (2) (2) (2) 1 so A3 = A4 cos ω (2) t1 → t1 = ω(2) and q 2 (t = t1 ) = cos−1 A3 /A4 · (2) (2) (2) (2) (2) q 3 (t = t1 ) ≡ β2 = −A4 ω (2) sin cos−1 A3 /A4 . On the ﬁrst path, we ﬁnd again an oscillation motion (2) (2) (2) q1 (t) = A2 cos ω (2) t + B2 ,
with the following initial condition:
/ · (2) (2) (2) (2) β2 and q 2 (t = t2 ) = → t 2 = A2 − A 3 / · (2) (2) (2) (2) (2) (2) (2) − A2 − A 3 β2 . To sumq 1 (t = t2 ) and so B2 = tan−1 −β2 /A2 marize at the level 2 we obtain " # (2) (2) (2) (2) (2) (2) ∀q ∈ A , q (t) = A cos ω t + B , A 1 2 2 1 2 # " (2) (2) (2) (2) (2) (7) q (2) (t) = ∀q ∈ A2 , A3 , q2 (t) = β2 t + A3 # " (2) (2) q (2) (t) = A(2) cos ω (2) t ∀q ∈ A3 , A4 . 3 4 (2)
(2)
(2)
q2 (t2 ) = q1 (t2 ) = A2
With respect to the level 1 we have diﬀerent amplitude, frequency and phase, but not casual ones since they are in agreement with the fractal model inside them.
188
Cantorian Space in Nature and Dynamical Systems
Consequently, an external observer looking at the motion of a particle under a ﬁxed solicitation can measure an unusual behaviour with respect to a continuous material support, that is obvious respect to the knowledge of the support behaviour. In this case he can make the hypothesis of an indetermination or a stochasticity in the process (motion), while there is just an ignorance respect to the material support on which the motion happen. More in general at the level n the initial conditions for the velocity are (n) (n) (n) βz(n) = −Az+2 ω (n) sin cos−1 Az+1 /Az+2 , with z = 0, ..., 2n − 1. (8) and for the phases factor we can get / (n) (n) (n) βz(n) Bz+1 = tan−1 −βz(n) /Az+1 − Az+1 − A(n) z
(9)
Consequently, by taking into account Hausdorﬀ ℵ(C) measure the asymptotic behaviour gives us f (∞)
=
k∞ (∞) 2 ω
=
(∞) βz (∞)
= =
Bz
=
(∞) Az
∈
αq lim f (n) = − ℵ(C) ,
n→∞
lim kn+1 = 0, 2 2 (ω(1) ) lim ω (n) = ℵ(C)
n→∞ n→∞
(n)
lim βz
n→∞
(n)
lim Bz
n→∞
(1)
= β1
(10)
=0
=0
C
(1)
(1)
With respect to β1 , here we have used the initial condition β1 = 0, but it is (1) obvious that there are no changes if β1 = v0 . Thanks to the previous relation we obtain 0 (∞) (∞) (∞) qodd (t) = Aodd+1 cos ω (∞) t + Bodd , (∞) q (11) (t) = (∞) (∞) qeven (t) = v0 t + Aeven+1 . The same considerations can be done for a (∞) El Naschie Cantorian space to obtain similar results. It is interesting to note that if we have an external solicitation F = αq the motion equation on C becomes ··
q + 2ω 2 q = F, and so
··
q + ω 2 q = 0,
(12) (13)
that is the traditional motion equation for a massive point in an elastic force ﬁeld. In other word, what we consider an external force for the support could be
Gerardo Iovane, et al.
189
the classical elastic solicitation on a massive point moving on a continuous material support1 . This example is just a toy model when we deal with macroscopic systems, since the frequency of oscillation of the material support could be very diﬀerent with respect to the system. But this toy model could be very realistic with respect to microscopic and quantum processes and systems. For this reason in [8] we considered oscillating force ﬁelds in quantum mechanics to show how the Heisenmberg uncertainty principle can be translated from the processes and systems to the material support, where we have classical dynamics. We showed the validity of this point of view, that in principle could be more realistic, since it describe the real nature of the matter and space, which does not exist in Euclidean space or curved one, but in a Cantorian one. A more realistic model can be the linear chain (see [8]).
3
Conclusions
In this paper we studied the eﬀect of a stochastic selfsimilar and fractal material support on some physical quantities and relations. In particular, we present an algebraic uniterative relation to ﬁnd the extremes of a Cantor segmentation at any level of fragmentation. We studied the motion under an elastic force ﬁeld on a Cantorian space, from classical point of view. The results of this work can be summarized as follows. • A Cantorian space could explain some relevant stochastic and quantum processes, if the space acts as an harmonic oscillating support, such as happen in Nature (see [9]). • An apparent indetermination, linked with a fractal support rather than a continuous one, can produce an indetermination on the motion of a physical object, which is explained thanks to stochastic or quantum process. This means that a quantum process, in some cases, could be explained as a classical one, but on a non continuous and fractal support. Consequently, an external observer looking at the motion of a particle under a ﬁxed solicitation can measure an unusual behaviour with respect to a continuous material support, that is obvious with respect to the knowledge of the fractal support behaviour. In this case, he can make the hypothesis of an indetermination or a stochasticity in the process (motion), while there is just an ignorance respect to the material support on which the motion happen.
1 To be more correct there is a change in the sign of the force, but this is linked with the subject of the problem (material support or massive point). In other words the sign minus of the elastic source becomes plus if we have it as an external source for the material support.
190
Cantorian Space in Nature and Dynamical Systems
References [1] G.Iovane, E.Laserra, F.S.Tortoriello, Stochastic SelfSimilar and Fractal Universe, Chaos, Solitons and Fractals, 20, 3, 415, 2004. [2] G.Iovane, Varying G, accelerating Universe, and other relevant consequences of a Stochastic SelfSimilar and Fractal Universe, Chaos, Solitons and Fractals, 20, 4, 657, 2004. [3] G.Iovane, E.Laserra, P.Giordano, Fractal Cantorian structures with spatial pseudospherical symmetry for a possible description of the actual segregated universe as a consequence of its primordial ﬂuctuations, Chaos, Solitons & Fractals, 2004, 22, 4, 521. [4] M.S. El Naschie, On the uniﬁcation of the fundamental forces and complex time in the ε(∞) space, Chaos, Solitons and Fractals, 11, 11491162, 2000. [5] B.J.Sidharth, Chaos Solitons Fractals 11, 2155, 2000; Chaos Solitons Fractals 12, 795, 2001. [6] T.Jech, Set Theory, Berlin: Springer, 2003. [7] M.S. El Naschie, Quantum gravity, Cliﬀord algebras, fuzzy set theory and fundamental constants of nature, Chaos, Solitons and Fractals, 20, 437, 2004. [8] G.Iovane, P.Giordano, S.Salerno, Dynamical System on El Naschie’s (∞) Cantorian spacetime, accepted to Chaos Solitons & Fractals, 2004. [9] G.Iovane, SelfSimilar and Oscillating solutions of Einstein’s equation and other relevant consequences of a Stochastic SelfSimilar and Fractal Universe via El Naschie’s (∞) Cantorian spacetime, in press on Chaos, Solitons & Fractals, 2004.
MATHEMATICAL STRUCTURE OF INDIVIDUAL QUANTUM STATES
S.N. MAYBUROV a
a
Lebedev Institute of Physics, Leninsky Prospect 53, Moscow, 117924, Russia
Abstract The possible structure of quantum states is analyzed in relation to their preparation and the information available to observer; it’s demonstrated that beside pure and mixed states the states of other kind  primordial states can exist. In particular, the states of environment are supposedly to be primordial, because they can’t be related to the standard way of preparation admitted in QM.
1
Introduction
The development of quantum information systems revives the interest to the different aspects of Quantum Mechanics (QM) foundations. In this note we analyze the general structure of quantum states which can be especially important for the study of decoherence eﬀects induced by the Environment interaction with the information system (detector) D [1]. Remind that QM deals with two sorts of states: an individual state ζ which characterizes the object state in the individual events representing so the most detailed object description, and the statistical state Ξ which describes the average properties of the objects ensemble. According to QM axiomatics any object, in principle, posseses the individual state which formally expressed by the timedependent density matrix ζ = ρ(t) [2]. It’s well known that the structure of states in QM depends of their production conditions; in particular, the pure state ψi of some object is prepared by the experimentalist O according to the particular preparation procedure P Ri . It deﬁnes unambiguously
191 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 191–194. © 2006 Springer. Printed in the Netherlands.
192
Mathematical Structure of Individual Quantum States
the produced state properties resulting in ρ(t) = ψi ψi  which is the extremal positive functional (PF) on {Q} set (algebra) of object observables [3]. The states mixture prepared by use of {P Rj } set with the frequencies {Pjp }, in this case at any t ρ(t) is one of ψj ψj  with the probabilities Pjp ; here and below it assumed that the diﬀerent objects from the same ensemble don’t intersect on t axe. Due to statistical character of QM measurements their results are normally averaged over large time interval T which are expressed via some functions of ζ(t). In particular, the ensemble state  Gemenge W = {Pip , ψi } reproduces ζ statistical content, ! the statistical state (statistical operator)  Ξ = ζ¯ = ρ(t)dt averaged over T [4]. Meanwhile we observe around us many objects and systems which aren’t prepared in the described way and a strict sense weren’t prepared at all, yet they are undoubtly possess the individual quantum state of some kind, and here we attempt to describe their structure for an arbitrary system. To illustrate our main hypothesis let’s consider the toy model of particle production in cosmological Big Bang [5]. Despite its exotics, it is the simplest example of such unprepared system with high degree of symmetry. We consider the states of massive particles e with spin 21 and assume that they are produced directly from the initial singularity. Obviously for any e production mechanism their statistical state Ξb possess the fundamental symmetries of Universe, in particular, the rotational invariance, so that Un Ξb Un−1 = Ξb , where Un is the rotation in n direction. state components are factorized, then Assuming that the momentum p and spin S for spin component the only solution is Ξb = σb where:
σb =
1 (s1 s1  + s2 s2 ) 2
(1)
here s1,2 are u, d e spinors. Because of the invariance of e production process relative to the time shift (at least at not very large time scale) we can assume that e individual state ζb = ρb (t) doesn’t depend on t and therefore ρb (t) = σb is constant. Such state called here the primordial state ξb = ρb (t) isn’t extremal PF on {Q} as the pure states are, but in general is an arbitrary PF on {Q}. Because of the assumed time invariance it isn’t the probabilistic mixture of pure states, and has completely uncertain spin ns direction in each event. It can be characterized as the quantum state of third kind which is produced without participation of any observer. Meanwhile the same statistical state Ξb can be dispatched by ρ(t) described by the mixture of pure states in case of the spontaneous breaking of rotational symmetry, For example, it can be the ensemble of pure states with random spin orientation ρ(tj ) = s(nr ) ; yet currently no suitable randomization mechanism was proposed which can result in such eﬀect [5].. The same results can be derived for the collective system e production in N → ∞ limit with total angular momentum J 2 = 0. It means that all N particles ej are entangled and the collective state ρc (t) = ψc ψc  where: ψc =
M & i
ai
N 1 j
s1l(1,i) ...sjl(j,i) ...sN l(N,i)
(2)
S.N. Mayburov
193
+N ! here M = 4N , l(j, i) = 1, 2  the discrete function with j l(j, i) = 23 N , ai are 2N 1 the complex parameters with ai  = M . For N = 2 it gives EPR state ψc = 1 1 2 1 2 √ (s s − s s ) The partial ei state for such system ηi (t) = T rρc = σb , 1 2 2 1 2 where the trace taken over all j = i. The spin correlation for any particles pair j S i = − 1 and is negligible for N → ∞, consequently e in such state is: S 4N states can be regarded as factorized in this limit and therefore are equivalent to the primordial state ξb . This example assumes that e primordial states can be produced in lab. conditions within some approximation. Considering the states of environment E [1], it’s plausible to assume that E states are also primordial states. Really, in some models E elements AE j are atoms (of atmospheric air, for example) which in lab conditions can’t be prepared by O one by one in some pure state. Supposedly AE interact with D independently of each other and therefore change its individual state ζD . If AE atoms structure is similar to e, and there is no external inﬂuence on E state, then assuming that the rotational invariance holds also in lab. conditions, it follows that E statistical spin state ΞE = Ξb . If the lab. conditions are time invariant, then AE individual spin states ζE = ρE (t) are equivalent to Big Bang e spin states ζE = σb so they are the primordial states ξE = ζE . They diﬀer from E individual states used in some decoherence models; in particular, in Zurek model it assumed that E states are pure and has the random orientation in space: E = s(nr ) [6]. This diﬀerence can lead to the important eﬀects but we don’t discuss them here. In general E states properties can depend on some classical lab. parameters like the temperature τ , pressure P, etc.; such ensembles are studied by Quantum Statistics (QS). According to its postulates the quantum ensemble of any objects is described by the statistical operator R [3]. For E embedded in thermostat its spin component is equal to: & F −Λ ∗ RE (si , sj ) = e− kτ ψΛ (si )ψΛ (sj ) Λ
Here F, Λ is AE free and total energy, ψΛ are its eigenstates. If spin and energy AE state components are factorized then RE = σb . In our approach the statistical state ΞE supposedly coincides with RE , but in addition the individual E states ρE can be deﬁned; from the mentioned t invariance it follows that ρE (t) = RE and AE states are primordial states with ξE = ρE (t). In this case also we don’t see any dynamical mechanism which can make such individual states stochastic. Moreover if E atoms can interact with each other, then their collisions result in their states entanglement. Consequently, even if some AE j initial state was pure with the deﬁnite spin direction ns it will become entangled with other AE states and its partial state will approximate ηi (t). The primordial states has the analogy with the so called algebraic states ϕ arising in Algebraic QM formalism [3]. In this theory the set of states ϕ ∈ Ω is the linear space dual to Observable Algebra U. The states ϕ, regarded as the individual states, are also PFs, but aren’t necessarily extremal on U observables, and due to it their physical meanings is still questionable and put some doubts on
194
Mathematical Structure of Individual Quantum States
Algebraic QM consistency [2]. Now it seems that nonextremal ϕ can correspond to the primordial states ξ, at least for the situations regarded here. Therefore the introduction of primordial states supports Algebraic QM premises, and in particular, evidences in favor of the feasibility of algebraic states. For the conclusion we argued here that the primordial quantum states, which diﬀer from the pure and mixed states, can exists in the nature. Their appearance is deﬁned by the fact that they are produced without direct involvement of any observer, in distinction from QM deﬁnition of the pure and mixed states preparation. It seems that the introduction of primordial states in Quantum Theory can help to ﬁll the existing gap between QM and QS axiomatics. References [1] D.Guilini et al., ’Decoherence and Appearance of Classical World’, (SpringerVerlag, Berlin, 1996) [2] J.M.Jauch ’Foundations of Quantum Mechanics’ (AdisonWesly, Reading, 1968) [3] G.Emch, ’Algebraic Methods in Statistical Physics and Quantum Mechanics’, (Wiley, NY, 1972) [4] (1996) P.Busch, P.Lahti, P.Mittelstaedt, ’Quantum Theory of Measurements’ (SpringerVerlag, Berlin,1996) [5] S.Dodelson ’Modern Cosmology’ (Academic Press, London, 2003) [6] W.Zurek, Phys Rev, D26,1862 (1982)
SPACE AND TIME PHYSICS WITH THE LORENTZ ETHER: THE CLOCK PARADOX
FRANCO SELLERI a
a
Universit` a di Bari  Dipartimento di Fisica INFN  Sezione di Bari
Abstract The description of natural phenomena by observers in motion is a problem that many consider solved by the Lorentz transformations, but that actually was left open. Consequences of my alternative “inertial” transformations are: (i) an explanation of the empirical data better than provided by the Theory of Special Relativity (TSR); (ii) the elimination of those features of the TSR giving rise to paradoxes thanks to the recovery of a privileged inertial frame in which the Lorentz ether is at rest. The example of the “clock paradox” is discussed and a complete resolution is obtained by giving an exhaustive uniﬁed description of all possible situations. Velocity (and nothing else) is thus seen to be responsible for the diﬀerential retardation eﬀect.
1
The equivalent transformations
According to Poincar´e [1], Reichenbach [2], Jammer [3], Mansouri and Sexl [4] the clock synchronization in inertial systems is conventional and the choice of the invariance of the one way velocity of light made in the TSR is only based on simplicity. Following the same line of thought, I introduced in the transformations of the time variable a suitable parameter e1 , describing synchronization [5]. The TSR is obtained for a particular nonzero value of e1 . In this way an inﬁnite set of theories empirically equivalent to the TSR was developed. Given the inertial frames and one can set up Cartesian coordinates and make the following standard assumptions:
195 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 195–208. © 2006 Springer. Printed in the Netherlands.
196 Space and time physics with the Lorentz ether: the clock paradox (i) Space is homogeneous and isotropic and time homogeneous, at least if judged by observers at rest in S0 ; (ii) In the isotropic system S0 the velocity of light is “c” in all directions, so that clocks can be synchronized in S0 and one way velocities relative to S0 can be measured; (iii) The origin of S, observed from S0 , is seen to move with velocity v < c parallel to the +x0 axis, that is according to the equation x0 = v t0 ; (iv) The axes of S and S0 coincide for t = t0 = 0; The system S0 turns out to have a privileged status in all theories satisfying the assumptions (i) and (ii), with the exception of the TSR. Two further assumptions based on direct experimental evidence can be added: (v) The two way velocity of light is the same in all directions and in all inertial systems; (vi) Clock retardation takes place with the usual velocity dependent factor when clocks move with respect to S0 . This assumption is the same as A2 of the second section. These conditions were shown [5] to lead to the following transformations of the space and time variables from S to S0 x0 − v t0 x= R y = y0 ; z = z0 (1) t = R t0 − e1 (x0 − v t0 )
where R=
1 − v 2 /c2
(2)
and e1 is the synchronization parameter. By using again (ii) and Eq.s (1) one can ﬁnd the one way velocity of light relative to the moving system S for light propagating at an angle θ from the velocity v of S relative to S0 [5]: c1 (θ) =
c 1 + Γ cos θ
with Γ = e1 R c +
v c
(3)
(4)
The TSR is a particular case, obtained for e1 = −
v c2 R
giving Γ = 0 and c1 (θ) = c and reducing (1) to their Lorentz form.
(5)
F. Selleri
197
These results are enough for the needs of the present paper. We should mention, however, that we also found that the choice e1 = 0 is the only one allowing for a treatment of accelerations rationally connected with the physics of inertial systems S0 . The theory was applied to the rotating platform and to the Sagnac eﬀect [6] with the result that only the choice e1 = 0 could give a satisfactory explanation [7]. A consequence of this research was the discovery of a relativistic discontinuity between inertial systems and slowly accelerated systems. The discontinuity disappears only if e1 = 0 [8]. Its existence in the TSR is the root of the diﬃculties met by Langevin [9], Post [10], LandauLifschitz [11], Anandan [12] in explaining the physics on a rotating platform. There are several other good reasons to adopt e1 = 0: (a) Einstein in his 1905 article assumed that aberration depends on the relative velocity starEarth, but this relativistic description works poorly, given that empirically the aberration angle is the same for all stars in the sky [13]. (b) The growing evidence for the existence of superluminal signals can easily be accommodated in the theory with e1 = 0, while it is incompatible with relativity due to the presence of a causal paradox in which events belonging to the future of an observer can actively modify the past of the same observer [14]. (c) Practically all paradoxes of the special theory of relativity disappear in a theory based on the inertial transformations. For these reasons a satisfactory theory of the physics of space and time has to be based on absolute simultaneity (e1 = 0). Assuming the inertial transformations, the S0 system is initially considered to be privileged, and the velocity of light relative to it isotropic. Other inertial systems are described as ”moving” and relative to them the observers detect an anisotropic velocity of light. In ref. [15] I described a resynchronization of clocks (ROC) and showed that it is uniquely determined by the new inertial frame S chosen to replace S0 as “privileged” and by the requirement that absolute simultaneity should be preserved. Thus a weak form of relativity principle is restored. One can add, however, that from the point of view of the inertial transformations the validity of relativity appears accidental, more than fundamental. It would be enough to discover a very small noninvariance of the two way speed of light to make the whole game of ROC impossible.
2
The clock retardation formula
In 1905 Einstein [16] presented the clock retardation prediction as follows: Imagine one of the clocks which mark the time t0 when at rest in the “stationary” inertial system S0 , and the time t when at rest in the “moving” inertial system S, to be located at the origin of the coordinates of S, where it marks the time t. What is the rate of this clock, if viewed from S0 ? The quantities x0 , t0 , and t, which refer to the position of the clock, satisfy x0 = v t0 and the Lorentz transformation
198 Space and time physics with the Lorentz ether: the clock paradox of time t=
where R =
1 − v 2 /c2 . Therefore
1 (t0 − v w0 /c2 ) R
t = t0 R = t0 − (1 − R)t0 whence it follows that the time t marked by the clock is slow by 1 − R seconds per second with respect to the S0 time t0 . From this Einstein deduced another consequence, which became known as “clock paradox”. If at the points A and B of S0 there are two synchronous stationary clocks; and if the clock at A is moved with the velocity v along the line AB to B, then on its arrival at B the two clocks no longer show the same time, but the clock moved from A to B lags behind the other which has remained at B by (1 − R)t0 , t0 being the duration of the journey from A to B. Einstein considered evident that this result still holds good if the clock moves from A to B in any polygonal line, and also when the points A and B coincide. Furthermore, he assumed that the result obtained for a polygonal line holds also for a continuously curved line. That is all. Clearly Einstein went beyond what he could say by considering only inertial systems, as he introduced also accelerated motions in the vertices of the polygonal line and along the continuous curve. He did so in an intuitive way, without solid foundations, and yet he was much closer to the correct result than with the 1918 paper (as we shall see). Today, the retardation of moving clocks and its independence of acceleration are well established empirical facts. In a CERN experiment [17] muons with a velocity of 0.9994 c, corresponding to R = 0.0346, were circling in a ring with diameter of 14 m, with a centripetal acceleration 1018 g. The lifetime τ0 of the circling muons, measured in the laboratory, was in agreement with the formula τ = τ0 R, where τ is the lifetime measured in the muon rest system. No eﬀect of the huge acceleration on the lifetime was observed. The experiment by Hafele and Keating [18] compared six synchronized caesium atomic clocks. Two were carried by ordinary commercial jets in an eastbound tour around the planet; another two were carried in a westbound tour; the last two remained on the ground. It was observed that with respect to the time shown by the latter clocks, those of the westbound trip had lost 59±10 ns, while those on the eastbound trip had advanced 273±7 ns. These results were in agreement with the usual formula t = t0 R if one used three diﬀerent R’s for the three pairs of clocks. The largest (smallest) R was that of eastward (westward) clocks, for which the Earth rotation velocity added to (subtracted from) the jet velocity. One had to include the eﬀect of the Earth gravitational potential, variable with altitude, which modiﬁes the rates of travelling clocks diﬀerently from those on the ground. Similar conclusions have been obtained with the GPS (Global Positioning System) network of 24 satellites [19]. With an orbital radius of about four Earth radii and an orbital speeds of about 3.9 km/sec, each satellite has on board four atomic
F. Selleri
199
clocks marking time with an error of a few ns/day. The gravitational eﬀect implies that the atomic clocks on board the satellites tick faster by about 45.900 ns/day because they are in a larger gravitational potential than atomic clocks on the Earth surface. The velocity eﬀect makes atomic clocks moving at GPS orbital speeds tick slower by about 7.200 ns/day. Therefore the global prediction is a gain of about 38.700 ns/day. Rather than having clocks with such large rate diﬀerences, the satellite clocks were reset in rate before launch slowing them down by 38.700 ns/day. The rich data show that the on board atomic clock rates do indeed agree with ground clock rates. The experimental evidence, in full agreement with Einstein’s 1905 statements, points to the validity of the following clock retardation formula. If a clock U , marking the time t0 when at rest in the isotropic inertial system S0 , is set in motion with arbitrarily oriented and possibly variable velocity u(t0 ) relative to S0 , the rate of the time marked by U at S0 time t0 is given by dτ = dt0 1 − u2 (t0 )/c2 (6)
This τ is exactly what an observer travelling with U reads on the clock itself. Therefore dτ is the “proper time” variation of U . The idea behind Eq. (6) is that only instantaneous velocity (and not acceleration) ﬁxes the clock rate, in agreement with observations. In physics one can recognize the cause of a phenomenon by varying it and verifying the existence of corresponding variations of the eﬀect. Viceversa, if arbitrary variations of a physical quantity Q do not modify the eﬀect E, one can exclude that Q is among the causes of E. Let us apply this criterion to Eq. (6). If u(t0 ) is varied a corresponding variation of the proper time rate arises: therefore velocity can be claimed to be a cause of the proper time rate variation. On the contrary, if the acceleration is modiﬁed at time t0 while u(t0 ) remains the same dτ will not change. Therefore the acceleration has no eﬀect on dτ and cannot be counted among the causes of its variation. This reasoning is important for determining the real physical roots of the “clock paradox”.
3
Absolute motion resolution of the clock paradox
As stressed by S. Prokhovnik [20] the resolution of the clock paradox in terms of absolute motion was found by G. Builder [21] who showed that the diﬀerential retardation eﬀect between two clocks which separate and reunite can be validly considered in respect to a single inertial reference frame. Such an eﬀect (read on the clocks) appears obviously to be the same to observers in all states of motion, and in this sense is absolute (“invariant”). Builder concluded that the emergence of an absolute eﬀect consequence of velocity implies the existence of a privileged inertial frame, in the sense that motion relative to this frame assumes an absolute signiﬁcance and is quantitatively related to absolute eﬀects. Builder’s paper is mostly qualitative and in some points diﬃcult to understand.
200 Space and time physics with the Lorentz ether: the clock paradox My hopefully clearer reformulation of his argument is based on two assumptions: A1. The velocity of light relative to an inertial system, S0 , is “c” in all directions, so that clocks can be synchronized in S0 with the Einstein method and one way velocities relative to S0 can be measured; A2. A clock moving with speed u(t0 ) relative to S0 during the S0 time interval dt0 marks a (proper) time increase dτ given by Eq. (6). The TSR is well known to satisfy the above assumptions in all inertial systems. The theory of the “equivalent transformations” [5] accepts S0 as the privileged system relative to which A1 and A2 are satisﬁed as well. Therefore all the consequences deduced below from A1 and A2 are valid both in the TSR and in the theory of the equivalent transformations. In this paper we consider only one spatial dimension. According to A2 the proper time increase T marked by a clock moving from the point a at time t0a to the point b at time t0b , both ﬁxed in S0 , (see Fig. 1) is t0b T = dt0 1 − u2 (t0 )/c2 (7) t0a
where u(t0 ) =
dx0 dt0
(8)
and x0 = x0 (t0 ) is the equation of motion of the clock on some “trajectory” in the (x0 , t0 ) plane connecting the points a, b of Fig. 1. We consider a second (“varied”) trajectory, very near to the original one, as follows →
x0 (t0 )
x0 (t0 ) + δx0 (t0 )
(9)
with δx0 (t0a ) = δx0 (t0b ) = 0
(10)
According to Eq. (10) the clock on the varied trajectory occupies the points a and b at the same times t0a and t0b as on the unvaried trajectory. This clearly corresponds to the situation in which two clocks separate at point a at time t0a , follow diﬀerent trajectories and reunite again in point b at time t0b . Also the velocity undergoes a variation →
u(t0 )
u(t0 ) + δu(t0 )
(11)
with u(t0 ) given by Eq. (8) and δu(t0 ) ≡
d δx0 (t0 ) dt0
The proper time integral will correspondingly become t0b 2 T + δT = dt0 1 − [u(t0 ) + δu(t0 )] /c2 t0a
(12)
(13)
201
F. Selleri
Figure 1. A space and time diagram showing a “trajectory” between two points, and a second varied trajectory between the same points.
From (12) and (13) it follows, for small variations 1 δT = − 2 c
t0b t0a
u(t0 ) d dt0 δx0 (t0 ) 2 2 dt 1 − u(t0 ) /c 0
(14)
Integrating by parts one gets
t0b
t0b 1 u(t0 )δx0 (t0 ) d u(t0 ) δT = − 2 − dt0 δx0 (t0 ) c dt0 1 − u(t0 )2 /c2 t 1 − u(t0 )2 /c2 t0a 0a
(15) Due to (10) the ﬁrst term in the right hand side vanishes. The derivative in the second term gives 1 δT = 2 c
t0b
t0a
u (t0 )
3/2
dt0 1 − u(t0 )2 /c2
δx0 (t0 )
(16)
where u (t0 ) = du(t0 )/dt0 . Clearly, δT = 0 for arbitrary δx0 (t0 ) satisfying (10) if and only if u (t0 ) = 0 at all times. This is like saying that the extremum proper time T of all motions is the uniform one with the constant velocity u1 =
x0b − x0a t0b − t0a
for which the proper time integral (7) takes the value T1 = (t0b − t0a ) 1 − u21 /c2
(17)
(18)
202 Space and time physics with the Lorentz ether: the clock paradox
Figure 2. Space and time diagram showing a constant velocity connection between two points a and b, and a second varied trajectory between the same points.
Among all motions connecting a and b this extremum is unique, as it can be obtained for u (t0 ) = 0 only. Therefore it gives either the maximum or the minimum proper time of all possible motions from a to b, not only of those obtained with an inﬁnitesimal deformation of the straight line. That the extremum is actually a maximum can be seen as follows. One has 2 1 u1 δu 1 (δu)2 1 − u1 + δu /c2 ∼ − 2 (19) = 1 − u21 /c2 − 2 c 2c 1 − u2 /c2 3/2 1 − u21 /c2 1
But u1 is constant and δu satisﬁes (12) so that, after integration, the ﬁrst order variation of T1 arising from (19) vanishes due to (10). The second order variation is 2 t0b δu(t0 ) 1 2 δ T =− 2 dt0 (20) 3/2 2c t0a 1 − u2 /c2 1
2
Clearly, δ T < 0 for all possible velocity variations. Therefore, moving away from the constant velocity line of Fig. 2 to a diﬀerent line connecting a and b implies in all cases a decrease of the elapsed proper time. Then the found extremum is a maximum. Comparing the motion with velocity u1 with any diﬀerent motion from a to b with velocity u2 (t0 ) one has t0b 2 2 2 2 ∆T = T1 − T2 = dt0 1 − u1 /c − 1 − u2 (t0 ) /c > 0 (21) t0a
As a diﬀerence of proper times, ∆T is exactly what two observers who traveled with the clocks ﬁnd by direct comparison of the clocks readings. The clock moving with rectilinear uniform motion can be considered at rest in a diﬀerent inertial system. Therefore the previous argument can be taken to
F. Selleri
203
describe all possible “clock paradox” situations from the point of view of the particular inertial system S0 we have considered. The description is the same in the TSR and in all other theories satisfying the assumptions A1 and A2 above. But the principle of relativity, the metrics of Minkowski space and/or the gravitational potential of the ﬁctitious forces have nothing to do with the essence of the matter. Only velocities are able to inﬂuence the working of a clock. In a forthcoming paper we will show that the previous argument can be extended to three dimensional space [22].
4
Most general clock retardation formula
We now generalize the obtained results by substituting assumption A2 with the following one: If during the S0 time interval dt0 a clock is moving with velocity u(t0 ) relative to S0 , it marks a (proper) time change dτ given by dτ = dt0 F u(t0 ) (22) where F is an arbitrary function of its argument. Thus the proper time T spent by a clock moving between any two points a, b ﬁxed in S0 is given by t0b T = dt0 F u(t0 ) (23) t0a
where
dx0 (24) dt0 and x0 = x0 (t0 ) is the equation of motion of the clock on some “trajectory” connecting the points a, b of Fig. 1 in the (x0 , t0 ) plane. We consider a variation of this trajectory as in eq. (9), with the conditions (10) satisﬁed. Thus the proper time integral becomes t0b T + δT = dt0 F u(t0 ) + δu(t0 ) (25) u(t0 ) =
t0a
From (12) and (25) it follows, for small variations t0b d δT = dt0 F u(t0 ) δx0 (t0 ) dt0 t0a
(26)
where F indicates the u derivative of F . An integration by parts now gives , t0b t0b % d $ δT = F u(t0 ) δx0 (t0 ) − dt0 F u(t0 ) δx0 (t0 ) (27) dt0 t0a t0a
Due to (10) the ﬁrst term in the right hand side vanishes and from the second term one gets t0b δT = − dt0 F u(t0 ) u (t0 ) δx0 (t0 ) (28) t0a
204 Space and time physics with the Lorentz ether: the clock paradox Clearly, δT = 0 for arbitrary δx0 (t0 ) if, at all times t0 F u(t0 ) u (t0 ) = 0
(29)
Thus we see that at least one of the possible solutions is u (t0 ) = 0 at all times, the uniform motion. Whether it corresponds to a maximum or to a minimum has to be seen case by case. Many functions F give rise to a maximum of the proper time integral for u (t0 ) = 0, e.g. ' ( 1 u2 F u(t0 ) = exp − (30) 2 c2
Therefore there is nothing typical of a relativistic theory in the maximum of the proper time integral provided by the rectilinear uniform motion.
5
Clock retardation in general relativity
The 1905 formulation of the clock paradox had implications that probably Einstein did not like. The delay is an absolute eﬀect, as all observers agree that the clock moving with variable velocity marks a smaller time. They disagree, however, on the numerical value of this variable velocity. In relativity all potential observers (forming an inﬁnite set) are completely equivalent, so that, in a sense, one can say that the clock velocity assumes at any time all conceivable values. But a quantity having inﬁnitely many values is totally undeﬁned. In this way the presumed cause of the diﬀerential retardation eﬀect seems to vanish into nothingness. But the cause of a real physical eﬀect should be concrete as well, in spite of the evasive description given by the theory. Therefore velocity itself should be well deﬁned, that is, relative to a physically active reference background (ether) deﬁning at the same time the privileged reference frame. It is no surprise, then, that the original formulation was completed with a later one based on the theory of general relativity (TGR) [23], whose essential points we will now review. Let S be an inertial reference system. Further, let U1 and U2 be two exactly similar clocks working at the same rate when at rest in S. If one of the clocks  let us say U2  is in a state of uniform translatory motion relative to S, then, according to the TSR it works more slowly than U1 , which is at rest in S. At this point Einstein adds a remark: “This result seems odd in itself. It gives rise to serious doubts when one imagines the following thought experiment.” In the thought experiment A is the origin of S, and B a diﬀerent point of the positive xaxis. The two clocks, initially at rest at A, work at the same rate and their readings are the same. Next, a constant velocity in the direction +x is imparted to U2 , so that it moves towards B. At B the velocity is reversed, so that U2 returns towards A. When it arrives at A its motion is stopped, so that it is again at rest near U1 . Since U2 works more slowly than U1 during its motion along the line AB, U2 must be behind U1 on its return. Now comes the problem, says Einstein. According to the principle of relativity the whole process must surely take place in exactly the same way if it is considered
F. Selleri
205
in a reference frame S sharing the movement of U2 . Relatively to S it is U1 that executes the toandfro movement while U2 remains at rest throughout. From this it would seem to follow that, at the end of the process, U1 must be behind U2 , which contradicts the former result. But, Einstein adds, the TSR is inapplicable to the second case, as it deals only with inertial frames, while S is at times accelerated. Only the TGR deals with accelerated frames. From the point of view of the TGR, one can use the coordinate system S just as well as S. But in describing the whole process, S and S are not equivalent as the following comparison shows. Description relative to the S Reference System
1. The clock U2 is accelerated by an external force in the direction +x until it reaches the velocity v. U1 is at rest, now as in all the subsequent steps; 2. U2 moves with constant velocity v to the point B on the +xaxis; 3. U2 is accelerated by an external force in the direction −x until it reaches the velocity v in the direction −x; 4. U2 moves with constant velocity v in the direction −x back to the neighbourhood of U1 ; 5. U2 is brought to rest by an external force very near to U1 . Description relative to the S Reference System
1. A gravitational ﬁeld, oriented along −x, appears, in which the clock U1 falls with an accelerated motion until it reaches the velocity v. When U1 has reached the velocity v the gravitational ﬁeld vanishes. An external force applied to U2 prevents U2 from falling in the gravitational ﬁeld; 2. U1 moves with constant velocity v to a point B on the xaxis. U2 remains at rest; 3. A homogeneous gravitational ﬁeld in the direction +x appears which accelerates U1 in the direction +x until it reaches the velocity v, whereupon the gravitational ﬁeld vanishes. U2 is kept at rest by an external force; 4. U1 moves with constant velocity v in the direction +x into the neighbourhood of U2 . U2 remains at rest; 5. A gravitational ﬁeld in the direction −x appears, which brings U1 to rest. The gravitational ﬁeld then vanishes. U2 is kept at rest by an external force.
The second description is based on the principle of equivalence between ﬁctitious
206 Space and time physics with the Lorentz ether: the clock paradox and gravitational forces. According to both descriptions, at the end of the process the clock U2 is retarded by a deﬁnite amount compared with U1 . With reference to S this is explained by noticing that during the stages 2 and 4, the clock U1 , moving with velocity v, works more slowly than U2 , which is at rest. But this retardation is overcome by the quicker working of U1 during stage 3. For, according to the GTR, a clock works faster the higher is the gravitational potential in the point where it is placed, and during stage 3 U1 is indeed placed in a region of higher gravitational potential than U2 . A calculation made with instantaneous acceleration shows that the consequent advancement amounts to exactly twice as much as the retardation during stages 2 and 4 [24]. After this conclusion Einstein states: “This completely clears up the paradox.” The prediction of the TGR that a clock works faster the larger is the gravitational potential φ in the region in which it is placed is fully conﬁrmed by the experiments in the gravitational ﬁeld of the Earth. At ﬁrst sight the 1918 reasoning could seem to be a consequence of the empirical facts. The mathematical treatment of the clock paradox situation given by the TGR leads to the right result by describing the retardation of U2 as a consequence of the action of φ on U1 [24]. Yet the theory shows its weakness in several ways. It is unreasonable that something happens to U1 , as the eﬀect should be objective and during the (short) times of acceleration some change in U1 should be seen also by some independent observers. For the sake of clarity let us compare two diﬀerent experiments, E1 and E2 below, both starting in the same way, as follows. Let A be a point of the inertial system S. Let U1 be the clock constantly at rest in A and U2 the mobile clock, initially at rest in A. At time t = 0 we give U2 a constant acceleration in the direction +x until it reaches the constant velocity v as in Einstein’s thought experiment. After this U2 moves with velocity v until, at time t = t1 , two alternative developments can start. E1. In the time interval (t1 , t2 ), with t2 > t1 > 0, U2 experiences a constant acceleration, exactly reversing its velocity. The 1918 formulation introduces a gravitational potential φ in the rest system of U2 , during (t1 , t2 ). The role of φ is very important as it must give rise to a variation in the time marked by U1 opposite and twice as large as that arising in the long stretches of uniform motion during which U1 is delayed with respect to U2 . E2. In the time interval (t1 , t2 ) U2 continues its rectilinear and uniform motion in the direction +x. The two clocks do not reunite anymore. No gravitational potential can arise, as there is no acceleration. Therefore in this second case there is no retardation of U1 with respect to U2 . In comparing E1 and E2, we note that the observers at rest with respect to U1 see in both cases the same identical situation, namely no modiﬁcation of the rate of time keeping of U1 . This can also be checked from a distance by observers in arbitrary states of motion, e.g. by monitoring U1 on a TV screen. The fundamental
F. Selleri
207
fact is always the same: nothing ever happens to U1 , in particular nothing happens in the time interval (t1 , t2 ), when U2 accelerates. Therefore φ cannot be the cause of any change of U1 , as in passing from E1 to E2 the presumed cause varies from a position dependent φ to φ = const. without any variation of the eﬀect. The criticism becomes even stronger if one considers not just one but several clocks U1 , U1 , U1 , . . .
at rest in diﬀerent points of the line AB of the inertial reference frame S. When U2 accelerates these clocks should be inﬂuenced diﬀerently, as φ depends on distance from U2 . But in reality nothing happens: observers can check in particular that the delays with which a light signal originated near U1 touches U1 , U1 , . . . are the same before and after the time interval during which U2 was accelerated. We can conclude that the gravitational potential of the ﬁctitious forces exerts no action on the clocks, contrary to Einstein’s 1918 opinion. The gravitational ﬁelds in the accelerated systems are not ordinary static ﬁelds, but arise from the accelerations of bodies [25]. Einstein assumed that these ﬁelds had on clocks the same action as ordinary ﬁelds, but we can now conclude that on this particular point he was not right, in spite of the very probable correctness of the general idea of equivalence between ﬁctitious and gravitational forces. One ﬁnds an analogy in the magnetic ﬁeld, which can be considered a dynamical manifestation of the electric ﬁeld, but has quite diﬀerent interaction properties. We can claim, ﬁnally, that concerning the nature of the diﬀerential retardation of separating and reuniting clocks Einstein was much nearer to the truth in 1905 than in 1918.
References [1] H. Poincar´e, La mesure du temps, Rev. M´ etaphys. Morale 6, 113 (1898). [2] H. Reichenbach, The Philosophy of Space & Time, Dover, New York, (1958). [3] M. Jammer, Some fundamental problems in the special theory of relativity, in: G. Toraldo di Francia, ed., Problems in The Foundations of Physics, pp. 202236, Societ` a Italiana di Fisica, Bologna and North Holland, Amsterdam, (1979). [4] R. Mansouri and R. Sexl, A Test Theory of Special Relativity, I, II and III, General Relat. Gravit., 8, 497513, 515524, 809814 (1977). [5] F. Selleri, Noninvariant OneWay Velocity of Light, Found. Phys. 26, 641664 (1996); Noninvariant OneWay Velocity of Light and Particle Collisions, Found. Phys. Lett. 9, 4360 (1996). [6] M. G. Sagnac, Eﬀet tourbillonnaire optique. La circulation de l’´ ether lumineux dans un interf´ erographe tournant, J. de Phys. 4, 177 (1914). [7] F. Selleri, Sagnac eﬀect: end of the mystery, in Relativity in rotating frames, pp. 5777, G. Rizzi and M. L. Ruggiero, eds., Kluwer, Dordrecht, (2004). [8] F. Selleri, Zero acceleration discontinuity and absolute simultaneity, to be published in the book: Absolute simultaneity, W. L. Craig, ed., Routledge (2005). [9] P. Langevin, Sur l’exp´ erience de Sagnac, Compt. Rend. 205, 304 (1937).
208 Space and time physics with the Lorentz ether: the clock paradox [10] E. J. Post, Sagnac eﬀect, Rev. Mod. Phys. 39, 475 (1967). [11] L. D. Landau and E. M. Lifshitz, The classical theory of ﬁelds, ButterworthHeinemann, Oxford (1996), §§ 8287. [12] J. Anandan, Sagnac eﬀect in relativistic and nonrelativistic physics, Phys. Rev. D 24, 338 (1981). [13] G. Puccini and F. Selleri, Doppler eﬀect and aberration of light from the point of view of absolute motion, Nuovo Cim. B, 117, 283 (2002). [14] F. Selleri, Superluminal signals require absolute simultaneity, Bull. Soc. Sciences et Lettres de Lodz, LII, 63 (2002) [15] F. Selleri, The inertial transformations and the relativity principle, in print on Found. Phys. Letters (2005). [16] A. Einstein, Zur Elektrodynamik bewegter K¨ orper, Ann. der Phys. 17, 891921 (1905). English translation: On the electrodynamics of moving bodies, in A. Einstein, H. A. Lorentz, . . . , The principle of relativity, pp. 3765, Dover, New York, (1952). [17] J. Bailey, et al., Measurements of relativistic time dilatation for positive and negative muons in a circular orbit, Nature 268, 301304 (1977). [18] J. C. Hafele and R. E. Keating, Around the World Atomic Clocks, Science 177, 166170 (1972). [19] T. Van Flandern, What the Global Positioning System Tells us about Relativity, in: F. Selleri, ed., Open questions in relativistic physics, pp. 8190, Apeiron, Montreal, (1998). [20] S. J. Prokhovnik, An introduction to the neoLorentzian relativity of Builder, Spec. Science and Techn. 2, 225229 (1979). [21] G. Builder, The resolution of the clock paradox, Aust. J. Phys. 10, 246262 (1957). [22] F. Selleri, Absolute velocity resolution of the clock paradox, to appear in the book: The Aether: Poincar´ e and Einstein, V. Dvoeglazov & C. Roy Keys, eds., (2005). [23] A. Einstein, Dialog u ¨ber Einw¨ ande gegen die Relativit¨ atstheorie, Die Naturwissenschaften 6, 697702 (1918). [24] C. Møller, The Theory of Relativity, Clarendon Press, Oxford (1957), § 98. [25] D. W. Sciama, On the Origin of Inertia, Monthly Notices Roy. Astron. Soc., 113, 3442 (1953). A. Ghosh, Origin of Inertia, Apeiron, Montreal, (2000).
Part III
Nuclear and HighEnergy Particle Physics and Astrophysics
UNDERSTANDING THE NUCLEON SPIN
FRANCO BRADAMANTE a
b
a,b
Dipartimento di Fisica, Universit` a degli Studi di Trieste, via A. Valerio 2, 34100 Trieste, Italy INFN, Sezione di Trieste, via A. Valerio 2, 34100 Trieste, Italy
Abstract A major eﬀort is undergoing at DESY, at CERN, and at BNL to investigate the spin structure of the nucleon. Following the original EMC discovery in 1988 that the quarks contribute only a small amount to the nucleon spin, a direct determination of ∆G, the gluon polarization in a polarized nucleon, has become the primary goal of many presently running experiments. After reviewing our present knowledge of the spin of the nucleon, I brieﬂy describe COMPASS, a new ﬁxed target experiment at CERN, and present some preliminary physics results.
1
The nucleon spin puzzle
The original EMC discovery [1] that the singlet axial vector current matrix element ∆Σ is smaller than the value predicted by the EllisJaﬀe sum rule is by now ﬁrmly established. In the simple quark model the spin of the proton is carried by its three valence quarks, so that ∆Σ = ∆u + ∆d = 34 − 31 = 1. In the usual notation, the quantities
∆q = 0
1
$
q(x)
↓↑
+ q¯(x)
↓↑
↑↑
− q(x)
+ q¯(x)
↑↑
%
1
∆q(x) dx
dx = 0
are the diﬀerences of the integrated quark densities of a given ﬂavour q, for quark spin antiparallel or parallel to the proton direction. In general terms one writes
221 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 221–230. © 2006 Springer. Printed in the Netherlands.
222
Understanding the nucleon spin
the spin equation for the proton (or neutron) as 1 1 = ∆Σ + ∆G + < Lz > 2 2
where ∆Σ = ∆u + ∆d + ∆s is the contribution of the quarks spins, ∆G is the contribution of the gluons, and < Lz > is a possible contribution from the gluons and quarks angular momentum. Since ﬁfteen years, i.e. since the EMC discovery, we know that this picture does not correspond to reality, and that the contribution of the quarks to the spin of the nucleons is much smaller. Polarized lepton  polarized nucleon deep inelastic scattering (DIS) provided a way to access the polarized quark densities ∆q. It can be shown [2] that the crosssection diﬀerence ∆σ for parallel and antiparallel spins depends on two structure functions, g1 and g2 , ∆σ = a·g1 (x, Q2 )+b·g2 (x, Q2 ), very much like the unpolarized crosssection depends on two structure functions, F1 (x, Q2 ) and F2 (x, Q2 ). The coeﬃcients a and b are kinematical factors, and can be calculated in QED. In the case of longitudinally polarized beam and target the quantity b is small, and a measurement of ∆σ allows g1 (x, Q2 ) to be measured. Conversely, if the target spin is perpendicular to the beam direction, a is smaller than b and g2 (x, Q2 ) can be measured. In the quark and parton model the structure function g1 and F1 are simply related to the quark densities ∆q(x, Q2 ) : % + 2 + 2 $ g1 (x, Q2 ) = 21 F1 (x, Q2 ) = 21 eq · ∆q(x, Q2 ); eq · q(x, Q2 ) + q¯(x, Q2 ) .
The summations extend over the quark ﬂavours. By measuring ∆σ over a large x and Q2 domain one can derive the ﬁrst moment of g1 , i.e. ! 1 p(n) p(n) Γ1 (Q2 ) = 0 g1 (x, Q2 )dx . To leading order in QCD we can neglect scaling violation, identify ∆q(x, Q2 ) with p(n) as a linear combination of ∆q : ∆q(x), and write the moment Γ1 p(n)
Γ1
1 (Q2 ) = +(−) 12 a3 +
1 36 a8
+ 91 a0 .
In this expression the ai are the diagonal combinations in the SU(3) nonet of the axial matrix elements, and are related to the quantities ∆q by the relations: a3 = ∆u − ∆d = gA , a8 = ∆u + ∆d − 2∆s, a0 = ∆u + ∆d + ∆s = ∆Σ. ¯ q γ µ γ5 Ψq P, S >, where S µ is the spin From the identity ∆q · 2M S µ =< P, SΨ of the proton, one can see that the quantities ∆q are also related to the weak decay of the baryons. Using the values of a3 and a8 derived from the neutron and hyperon decay, and the measured value of Γ1 coming from polarized DIS one can derive ∆Σ. When this was done by the EMC Collaboration [1] the measured value of Γp1 implied that ∆Σ = 0.12 ± 0.09 ± 0.14, i.e. only a small part (compatible with zero!!) of the spin of the proton is carried by the quarks.
F. Bradamante
223
In order to understand the picture which emerged from the EMC experiment, a number of new experiments were executed (SMC at CERN [3], E142, E143, E154 and E155 at SLAC [4], and HERMES at DESY [5], the only one which is still running), and a much larger number of theoretical papers were dedicated to the subject. On the experimental side, the ﬁnding of EMC for the proton was conﬁrmed, and a similar result was obtained for the neutron. Today the compilation of all the available measurements for g1 for the proton, the deuteron and the neutron is impressive [6]. On the theoretical side, much progress was done to go beyond the quarkparton model and analyze the polarized DIS data within the framework of QCD. Perturbative QCD analysis in nexttoleading order have been performed by several groups to determine the polarized parton distribution functions ∆q and ∆G using all the available data sets. By now it is understood that i) the singlet axial charge a0 receives a contribution from ∆G due to the Adler  Bell  Jackiw anomaly, thus the ∆Σ and ∆G contributions to a0 cannot be separately determined; ii) the results depend on the renormalisation scheme which is used; iii) although in principle ∆G could be determined from the Q2 evolution of g1 (given the large span of energies of the incident lepton in the various experiments, a broad range of Q2 is covered by the data in most bins of xBjorken), it turns out that the present inclusive DIS data do not put a strong constraint on the polarized gluon distribution. The results of various ﬁts suggest a large positive value (1  2) for ∆G (see f.i. [7]), but the uncertainties in these determinations are ∼ 100%. A direct measurement of ∆G seems essential to progress in this ﬁeld.
2
The spin puzzle, present activities
To understand the nucleon spin today three lines of attack are being pursued: i) a direct measurement of ∆G. This is the main goal of both the COMPASS [8] experiment at CERN and of the RHIC spin program [9] at BNL, which all have recently started data taking. ii) to measure transversity. These parton distributions represent a new territory and there is a broad interest in their properties. COMPASS and RHIC have deﬁnite plans to measure them, and at lower energy transversity is an important part of the physics objectives of the HERMES experiment at DESY and of the experiments at JLab. iii) on a longer term there are ideas to derive < Lz > from measurements of Generalized Parton Distributions (GPDs) via the Ji sum rule [10]. First measurements of exclusive reactions (in particular deeply virtual Compton scattering) have been done by HERMES and at JLab, but the full program
224
Understanding the nucleon spin
is terribly ambitious and anyway necessitates of a new high energy and high intensity CW electron machine. Particularly interesting is the physics case for transversity. Large transverse spin eﬀects at high energy have been discovered since many years, but their explanation in terms of QCD has always been lacking. Well known examples are the Λ polarization in N N scattering, the impressive Cnn values in elastic pp scattering at 90o c.m. measured at BNL, and the asymmetry of pion production in polarized protonproton scattering measured at FNAL. Undoubtedly the awareness of these phenomena played a major role in launching the polarized proton collider project at BNL and has been an important stimulus in developing a theory of transversity. Although the basic ideas are almost 30 years old, the rigorous treatment of transversity is relatively recent, and has gained a high momentum only the past few years. As originally shown by Jaﬀe and Ji [11], to completely specify the quark state at the twisttwo level, to the momentum distribution q(x) and to the helicity distribution ∆q(x) one has to add the transverse spin distributions ∆T q(x). Its measurement gives access to new information related to relativistic eﬀects for bound quark states, the study of new evolution in QCD, the knowledge of the tensor charge of the nucleon, and predictions for other processes involving transversity. The transversity distributions ∆T q(x) have never been measured, since they are chirallyodd functions and do not contribute to inclusive deep inelastic scattering. They may instead be extracted from measurements of the spin asymmetries in crosssections for semiinclusive deep inelastic scattering between leptons and transversely polarized nucleons, in which a hadron is also detected in the ﬁnal state. In such processes the measurable asymmetry is due to the combined eﬀect of ∆T q(x) and another chirallyodd function, which contributes to the fragmentation of the transversely polarized quark. In the case in which the observed ﬁnal hadron is a pion or, in general, a scalar particle, this new fragmentation function is the socalled Collins function ∆T Dqh that describe the hadronization of a transversely polarized quark q in a hadron h [12], as yet unmeasured, which in its own right merits serious study. In the case in which the observed ﬁnal particle is for example a Λ◦ , the chirallyodd function is a transverse fragmentation function, also unknown and interesting. Other channels for accessing ∆T q require the detection of a vector particle or two pions in the ﬁnal state. An important sideproduct of the study of transversity is, therefore, the determination of a certain number of fragmentation functions.
3
Spin physics with COMPASS
Inclusive DIS does not allow to solve the spin puzzle, and to progress a direct determination of ∆G is necessary. A new experimental approach is required, namely semiinclusive DIS with a full reconstruction of the hadronic current jet. A ﬂavour tagging procedure allows to identify the struck quark in the DIS process. A suggestion to isolate the photongluon fusion process and directly measure ∆G was put
F. Bradamante
225
Figure 1. The photongluon fusion diagram, dominant mechanism for charm production at COMPASS energies.
forward already in 1988 [13, 14], and implied measuring the crosssection asymmetry of open charm in DIS. A new experiment, with full hadron identiﬁcation and calorimetry, seemed to be necessary: COMPASS (COmmon Muon and Proton Apparatus for Structure and Spectroscopy) was proposed in 1996 and approved one year later. The main goal of COMPASS is indeed a direct measurement of ∆G by meac¯ c suring the crosssection asymmetry AµN c¯ c AµN =
∆σ µN →c¯cX . σ µN →c¯cX
(1)
At COMPASS energies the production of charm goes predominantly via photongluon fusion (PGF), according to the diagram shown in Fig. 1, and the quantities σ µN →c¯cX and ∆σ µN →c¯cX can be expressed as a convolution of the elementary photongluon crosssection with the gluon distributions G and ∆G. The most promising additional way to measure ∆G in COMPASS uses the asymmetry of charged hadron pairs at high pt [15]. Originally developed for the COMPASS experiment, the method has been recently applied also to the HERMES data [16]. The basic diagram is still the PGF, γg → q q¯ → h+ h− X, and the hardness of the process is guaranteed by the large pt . The background from the leading order process γq → q, and the QCDCompton process, γq → γq, is in general dominating the PGF creation of a light q q¯ pair, but suitable kinematic cuts can enhance considerably this process and allow for a statistically precise measurement. Apart from ∆G, the COMPASS spectrometer is measuring ∆q and ∆T q from the relevant identiﬁed hadron asymmetries, in semiinclusive polarized muon polarized nucleon DIS, in the longitudinal and in the transversal mode respectively. COMPASS has also an important physics programme using hadron beams, in particular to address the long standing question of exotic states. A ﬁrst pilot run will take place at the end of the 2004 run, and the measurement will be resumed after the technical stop of the CERN accelerators in 2005.
226
Understanding the nucleon spin
COMPASS Layout 2002
Hadron Calorimeter2
Hadron Calorimeter1
(Topview)
Muon Filter 2
MiddleTrigger Ladder Trigger
SM2
Outer Trigger
Straws
Inner Trigger
SM1
Inner Trigger
Silicons Veto Trigger
Target
µBeam
Ladder Trigger
MWPC GEM SciFi
SciFi
Middle Trigger
Leadwall
MicroMeGas SciFi SDC x
Gem Scifi SDC
MWPC GEM SciFi
RICH
MWPC MW2 (DriftTubes)
W4/5
MW1 (IarocciTubes )
z
Outer Trigger
MuonFilter 1
10
20
30
40
50 m
Figure 2. Top view of the layout of the spectrometer for the COMPASS experiment in 2002. The labels and the arrows refer to the major components of the tracking, trigger, and PID systems.
4
The COMPASS spectrometer
A common requirement of all the measurements foreseen by COMPASS is the detection and identiﬁcation of particles over a large angular (±200 mrad) and dynamical (up to ∼ 150 GeV) range. To perform these measurements an International Collaboration of 31 Institutes, presently consisting of more than 200 physicists from 11 diﬀerent Countries, including a strong group from Trieste, has built COMPASS, a new stateoftheart spectrometer, capable of standing beam intensities of up to 2 · 108 particles/spill. The COMPASS spectrometer comprises two magnetic stages. Both stages are complemented with charged particle identiﬁcation with fast RICH detectors, electromagnetic calorimetry, hadronic calorimetry, and muon identiﬁcation via ﬁltering through thick absorbers. The layout of the spectrometer which was on the ﬂoor in the year 2002 is shown in Fig. 2. The experiment has been run at a muon energy of 160 GeV. The beam is naturally polarized by the π−decay mechanism. The beam polarization was measured by the SMC Collaboration to be about 80% at 190 GeV. The triggering system and the tracking system of COMPASS have been designed to stand the associated rate of secondaries, and use stateoftheart detectors. Also, fast frontend electronics, multibuﬀering, and a large and fast storage of events are essential. We use the polarized target system of the SMC experiment, which allows for two oppositely polarized target cells, 60 cm long each. The PT magnet can provide both a solenoidal ﬁeld (2.5 T) and a dipole ﬁeld (0.5 T), for adiabatic spin rotation. The target polarization can then be oriented either longitudinally or transversely to the beam direction. Use of two diﬀerent target materials, NH3 as proton target and 6 LiD as deuteron target, is foreseen. Polarizations of 85 % and 50 % have
F. Bradamante
227
been reached, respectively. In so far we have used 6 LiD: its favourable dilution factor of 0.4 is of the utmost importance for the measurement of ∆G. To match the expected particle ﬂux in the various locations along the spectrometer, COMPASS uses very diﬀerent tracking detectors, in particular the experiment has been the ﬁrst to use on a large scale novel detectors like Micromegas’s and tripleGEM detectors. The charged particle identiﬁcation relies on the RICH technology. Presently, only RICH1 (the RICH in the ﬁrst magnetic spectrometer) exists. As VUV photon detectors we use a novel technique, developed at CERN, i.e. MWPC’s with a CsI photocathode (segmented in 8 × 8 mm2 pads) which detect photons with wave length shorter than 200 nm, i.e. in the far UV domain. The total active area of the photon detectors is 5.6 m2 and the total number of pads is about 80,000. The frontend electronics uses a modiﬁed version of the Gassiplex chip, and the readout cards constitute a major project of the Trieste ICTP Microprocessor Laboratory, which utilises hundreds of DSP’s. The readout system uses a modern concept, involving highly specialized integrated circuits. The readout chips are placed close to the detectors and the data are concentrated at a very early stage via high speed serial links. At the next level high bandwidth optical links transport the data to a system of readout buﬀers. The event building system is based on PCs and Gigabit or Fast Ethernet switches and is highly scalable. This high performance network is also used to transfer the assembled data to the computer center for database formatting, reconstruction, analysis and mass storage. The data are sent via an optical link from the Hall 888 directly to the Computer building for Central Data Recording (CDR). To handle the huge amount of data (the collected raw data size is ∼300 TB/year) we used Objectivity/DB until the end of 2002, and Oracle since. The power needed to process COMPASS data is about 100 kSI2k. In the oﬀline farm, the data servers handle the network traﬃc from the CDR, distribute the raw data to the CPU clients (where they are put in the data base), receive them back from the PCs, and ﬁnally send them to a hierarchical storage manager (HSM) system. In parallel, the data servers receive the data to be processed from the HSM, send them to the PCs for processing, collect the output (DST or mDST), and send it to the HSM. Data processing is performed on the farm at CERN while DST’s data analysis is done on satellite farms in the major home institutes (including Trieste). A major eﬀort was devoted to writing from scratch the oﬀline programs (CORAL, the new COmpass Reconstruction and AnaLysis program) using objectoriented technology and C++ language.
5
First physics results
Over the 80 days of the 2002 run, a total of about 6000 millions of events have been collected, corresponding to a data size of 260 TB. Similar numbers have been taken in the 2003 run and are presently being collected in the 2004 run. Many diﬀerent physics channels are presently being investigated,
228
Understanding the nucleon spin
Figure 3. Preliminary results for the asymmetry AU T for positive (a) and negative (b) leading hadrons produced by 160 GeV µ+ on a transversely polarized deuterium target. The data refer to the year 2002 COMPASS run, while the curves are from Ref. [17].

∆G from open charm and high pT hadron pair, ALL to extract g1 , vector meson (ρ, φ, J/Ψ) production to test Schannel helicity conservation, Λ physics, transversity (single hadron, hadron pairs, Λ), Chan asymmetries, search of exotics (Θ+ , Ξ−− , ...). A ﬂavour of the physics results is given below.
5.1 COLLINS ASYMMETRY AND TRANSVERSITY SemiInclusive Deep Inelastic Scattering (SIDIS) provides the possibility to measure the transverse polarized parton distribution function ∆T q(x) via the azimuthal dependence of the momentum of the leading hadron. This measurement requires to operate with a transversely polarized target: about 20% of the COMPASS running time has been devoted to this measurement. Figure 3 shows preliminary values of the ﬁrst ever measured Collins asymmetry on a deuteron target, separately for positive and negative leading hadrons. The data are compared with a model calculation [17] of the asymmetry AU T , which includes the transversity distribution function ∆T q(x) through the following linear combination of quark ﬂavours: Σe2q · ∆T q(x) · ∆T Dqh AU T (x) = . Σe2q · q(x) · Dqh
The small values of the measured asymmetries at all x might imply either a cancellation between the proton and the neutron asymmetries, or a small Collins eﬀect in the fragmentation. 5.2 ∆G/G FROM CHARM PHOTOPRODUCTION AND HIGH pT HADRON PAIRS As mentioned in Section 3, the gluon polarization will be obtained from the measured crosssection asymmetry of open charm events. Opencharm events are
F. Bradamante
229
Figure 4. Left: D∗ produced by requiring the invariant mass of the Kπ pair to be in a 60 MeV window around the D0 peak and a soft pion being detected. Cutting the events around the D∗ mass, the D 0 peak in the invariant mass spectrum of Kπ is very clear (ﬁgure at the right).
¯ ◦ , and D∗± mesons from they decay products, identiﬁed by reconstructing D◦ , D i.e. D◦ → K − π + and D∗+ → D◦ π + → K − π + π + and charge conjugate. In the ∗ ﬁrst case, cuts on the K direction in the D◦ rest frame (cos(θK ) < 0.5) and on the D◦ energy fraction (zD = ED /Eγ ∗ > 0.25) are needed to reduce the background contamination. Preliminary signals of the D mesons are shown in ﬁg. 4 Kaonpion ∗ pairs are selected by asking: zD > 0.2; cos(θK ) < 0.85; 10 < pK < 35 GeV in order to be in the RICH K identiﬁcation region. A soft pion (< 10 GeV) is also required. This measurement is statistically limited. The D◦ signal in our data is at the level of 10−7 , and other decay channels are also being investigated presently. From the present analysis the projected error on ∆G/G using the data from 2002, 2003 and 2004 should be about 0.24. From the same data, using unidentiﬁed hadron pairs at high pT , as mentioned in Section 3, we should be able to estimate ∆G/G with an error of 0.16 when requiring Q2 > 1. Releasing the cut on Q2 (as HERMES did) the hadron pair statistics increases by one order of magnitude, both the theoretical interpretation of the result becomes more diﬃcult.
6
Conclusions and outlook
I have brieﬂy summarized our present understanding of the nucleon spin. Following the EMC discovery in 1988, the importance of this issue has been broadly recognized, and the ﬁeld has enjoyed a true “Renaissance”. In this context one has witnessed the approval and the starting up of the most ambitious project spin physics could ever dream of, namely the polarized protonproton collider at BNL,
230
Understanding the nucleon spin
as well as of the COMPASS experiment at CERN, to which I have largely contributed and to which I have dedicated a good fraction of my talk. The physics goals of COMPASS and of RHIC largely overlap, although the measurements are complementary and both necessary to assess a coherent picture of the QCD structure of the nucleon, Both COMPASS and the RHIC experiments have just started to produce physics results, which presumably will be ﬁrst shown in three weeks from now in Trieste, at the 16th International Spin Physics Symposium. And, as it has always been the case for spin physics, many more surprises are expected in the years to come. References [1] The European Muon Collaboration, J. Ashman et al., Phys. Lett. B 206, 364 (1988); Nucl. Phys. B 328, 1 (1989). [2] V.H. Hughes and J. Kuti, Ann. Rev. Nuclear Particle Science 331983611. [3] The Spin Muon Collaboration, B. Adeva et al., Phys. Rev. D 58, 112001 (1998); Phys. Rev. D 60, 072004 (1999). [4] SLAC E142, D.L. Anthony et al., Phys. Rev. Lett. 71, 959 (1993); SLAC E143, K. Abe et al., Phys. Rev. Lett. 75, 25 (1995); SLAC E154, K. Abe et al., Phys. Rev. Lett. 79, 26 (1997); SLAC E155, P.L. Anthony et al., Phys. Lett. B 436, 339 (1999). [5] The HERMES Collaboration, K. Ackerstaﬀ et al., Phys. Lett. B 404, 383 (1997). [6] Particle Data Group, Phys. Rev. D 66, 010001 (2002). [7] The Spin Muon Collaboration, B. Adeva et al., Phys. Rev. D58, 112002 (1998). [8] COMPASS, A Proposal for a COmmon Muon and Proton Apparatus for Structure and Spectroscopy, CERN/SPSLC 9614, SPSLC/P297, 1 March 1996. [9] see, for instance, G. Bunce, N. Saito, J. Soﬀer, and W. Vogelsang, Annual Rev. Nucl. Part. Sci. 50, 525 (2000). [10] X. Ji, Phys. Rev. Lett. 78, 610 (1997);Phys. Rev. D55, 7114 (1997). [11] R.L. Jaﬀe and X. Ji, Phys. Rev. Lett. 67, 552 (1991). [12] J. Collins, Nucl. Phys. B 396, 161 (1993). [13] G. Altarelli and G.G. Ross, Phys. Lett. B 212, 391 (1988). [14] M. Gl¨ uck and E. Reya, Z. Phys. C 39, 569 (1988). [15] A. Bravar, D. von Harrach, and A. Kotzinian, Phys. Lett. B 421, 349 (1998). [16] A. Airapetian et al., DESY9907, July 1999. [17] A.V. Efremov et al. Eur. Phys. J. C32, 346 (2003)
THE MASS AND SPIN OF THE MESONS, BARYONS, AND LEPTONS
E. L. KOSCHMIEDER Center for Statistical Mechanics, The University of Texas at Austin, Austin, TX 78712, USA email: [emailprotected]
Abstract The rest masses of the stable mesons and baryons and the rest masses of their antiparticles, as well as the rest masses of the µ± and τ ± leptons can be explained, within 1% accuracy, with the standing wave model, which uses only photons, neutrinos, charge and the weak nuclear force. And we can explain the spin of the stable mesons and baryons and the spin of the µ± and τ ± leptons without any additional assumption. We can also determine the rest masses of the e, µ and τ neutrinos.
1
Introduction
When elementary particles are discussed usually the socalled “Standard Model” is invoked and it is implied that this solves the problem. However, it is an obvious fact that in the forty years since the introduction of the quark concept we have not arrived at precise theoretical values for the masses of the mesons and baryons and for the masses of the leptons. That means that neither the mass of the fundamental proton nor the mass of the fundamental electron have been explained so far. Many other attempts to explain the elementary particles have been made. For example, El Naschie [1] has proposed a topological theory from which the masses of the proton and the electron, as well as of the pions, can be determined with great accuracy. And, as another example, Sidharth [2] has suggested a model of the π 0 meson in which an electron and a positron circle their center of mass.
231 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 231–237. © 2006 Springer. Printed in the Netherlands.
232
The mass and spin of the mesons, baryons, and leptons
In the following I will show that one can explain the masses of the mesons, baryons and leptons with the “Standing Wave Model”. This model uses only photons, neutrinos, charge and the weak nuclear force. We can also explain the spin of the mesons and baryons without any additional assumption. A complete description of the standing wave model has been posted on the internet [3].
2
The integer multiple rule of the particle masses
The spectrum of the socalled “stable” mesons and baryons consists of the γbranch and the νbranch as we have shown in [4]. The masses of the π 0 , η, Λ, Σ0 , Ξ0 , 0 0 0 Ω− , Λ+ c , Σc , Ξc , and Ωc particles, i.e. the masses of the γbranch, are in a ﬁrst approximation integer multiples of the mass of the π 0 meson. A least square analysis shows them to follow the formula m(N )/m(π 0 ) = 1.0065 N − 0.0043
N ≥ 1,
(1)
with the correlation coeﬃcient 0.999. The letter N stands for the integer number closest to the ratio of any particle of the γbranch to the mass of π 0 meson. The principal decays, i.e. decays with a percentage >1%, of the γbranch particles occur always with the emission of γrays or π 0 mesons, that means without the emission of neutrinos. The neutrino branch, or νbranch, consists of the π ± , K±,0 , n, p, D±,0 and D± s particles. Their masses follow the formula m(N )/0.853m(π ± ) = 1.000 N + 0.00575
N ≥ 1,
(2)
with a correlation coeﬃcient 0.998. The principal decays of the νbranch particles occur always with the emission of neutrinos. That means: The masses of the γbranch particles are, in a ﬁrst approximation, integer multiples of the mass of π 0 . The masses of the νbranch particles are, in a ﬁrst approximation, integer multiples of π ± times a factor ≈ 0.85.
3
The γbranch particles
The characteristic particle of the γbranch is the π 0 meson which is created in the process γ + p → π 0 + p and decays via π 0 → γγ (98.8%). A γray impinges on a proton and creates a π 0 meson. Fourier analysis dictates that a continuum of frequencies of electromagnetic waves must be present in the wave packet created by the high energy collision. The wave packet is the π 0 meson. We assume that the wave packet is organized in a cubic lattice, because the continuous Fourier frequency spectrum of the high energy collision can be absorbed by the continuous frequency spectrum of the lattice oscillations. The relevance of cubic lattices for particle theory was ﬁrst suggested by Wilson [5]. According to lattice theory the lattice oscillations are standing waves. A rest mass can only be formed by standing waves.
E.L. Koschmieder
233
The frequencies of longitudinal standing waves in a cubic lattice are according to Born and v. Karman [6] given by ω = ± 2 α/m sin(φ/2) , (3)
where α is the constant of the force acting between the particles of mass m within the lattice, and φ is given by φ = 2πa/λ. There is a continuum of frequencies with a cutoﬀ at φ = π. However, the increase of the frequencies is limited by the requirement that the group velocity cannot exceed the velocity of light. That means that the frequencies of the oscillations must follow the formula ν = ν0 φ .
(4)
The rest mass of the packet of standing waves is the sum of the energy of all oscillations. The second mode of the oscillations has frequencies twice the frequencies of the basic mode and twice as many oscillations, that means 4 × as much energy as the basic mode. That conﬁrms the integer multiple rule: The particles of the γbranch consist of the basic mode (the π 0 meson), or of higher modes (the η meson m(η) ∼ = 4m(π 0 )), or of superpositions of modes, for example two higher modes make the Λ baryon m(Λ) ∼ = 2m(η). Summing up: The π 0 meson and the other particles of the γbranch are like cubic black bodies ﬁlled with standing electromagnetic waves. A conservative explanation of the γbranch particles which does not use hypothetical particles. It appears that the π 0 meson can be derived from the theory of electromagnetic waves, that means from ﬁrst principle.
4
The νbranch particles
The characteristic particles of the νbranch are the π ± mesons which can be created in the reaction γ + p → π + + π − + p. A γray impinges on a proton and is converted into a π + and a π − meson. The π ± mesons decay as π ± → µ± + νµ (¯ νµ ) (99.98%), followed by the decay of the µ± mesons, e.g. µ+ → e+ + ν¯µ +νe (≈100%). It appears that the π ± mesons consist of νµ , ν¯µ , νe , ν¯e neutrinos which are held together in a cubic lattice by the weak nuclear force. The existence of the neutrinos and of the antineutrinos and of the weak nuclear force is unquestionable. Since the weak nuclear force has a range of only 10−16 cm and the radius of the π ± mesons is ≈ 0.8·10−13 cm, the weak force can hold the π ± mesons together only if the neutrinos and antineutrinos are arranged in a lattice. The lattice points must oscillate because the lattice has been created in a high energy collision. The size of the nucleon and the range of the weak nuclear force determine the number of the lattice points. From the measured radius of the proton rp = 0.88·10−13 cm and the lattice constant which is equal to the range of the weak nuclear force rw = 1·10−16 cm follows that the number of the lattice points is N = 2.854 · 109 .
(5)
234
The mass and spin of the mesons, baryons, and leptons
The energy in the rest mass of the π ± mesons is the energy of all standing lattice oscillations plus the energy in the rest masses of all neutrinos in the lattice m(π ± )c2 = Eν (π ± ) + Σ m(neutrinos)c2 .
(6)
Using Eq.(4) we can calculate the energy Eν (π ± ) = 67.82 MeV ≈ 0.5 E(m(π ± )) of the lattice oscillations. If the number of the neutrinos and their rest masses is also known, then m(π ± ) is known. Assuming that m(νe ) m(νµ ) we obtain from Eq.(6) a ﬁrst approximation for the mass of the muonneutrino m(νµ ) ≈ 50 milli eV/c2 ,
(7)
Summing up: The π ± mesons can be explained by an oscillating cubic lattice of νµ , ν¯µ , νe , ν¯e neutrinos plus an electric charge. The primary decay of the K± mesons K± → µ± + νµ (¯ νµ ) (63.5%) leads to the same products as the decay of the π ± mesons π ± → µ± + νµ (¯ νµ ) (99.98%). Therefore it appears that the K± mesons should consist at least partially of the same νµ , ν¯µ , νe , ν¯e neutrinos as the π ± mesons. However the K± mesons with a mass of 0.8843·4m(π ± ) cannot be solely the second mode of the π ± mesons because the second mode of π ± does not have the energy 4 times of π ± , because the rest masses of the neutrinos which make up part of the energy of π ± do not change when the energy of the lattice oscillations is increased fourfold in the second mode of the oscillations. The diﬀerence between the second mode of π ± and the real mass of the K± mesons is made up by a π 0 meson, whose presence in K± is conﬁrmed by the appearance of π 0 among the decay products of K± , in particular in the decay K± → π ± π 0 (21.13%). So the K± mesons are the state (2.)π ± + π 0 . The K0 and K0 mesons are the state (2.)π ± + π ∓ , they consist of neutrinos only. The neutron is a superposition of a K0 and a K0 meson and must have a neutrino lattice. The structure of the proton follows from the βdecay of the neutron. The proton must have a neutrino lattice as well. The characteristic feature of the particles of the νbranch is their neutrino lattice of νµ , ν¯µ , νe , ν¯e neutrinos. We do not use hypothetical particles.
5
The mass of the µ± mesons
Surprisingly the mass of the µ± mesons can be explained with the standing wave model as well. The µ± mesons are leptons which do not interact strongly with the mesons and baryons. Leptons make up 1/2 of the number of stable elementary particles. The µ± mesons originate from the decay π ± → µ± + νµ (¯ νµ ) (99.98%). From the experiments we learn that m(µ± )/m(π ± ) = 0.757027 = 1.00936 · 3/4 ∼ = 3/4 .
(8)
E.L. Koschmieder
235
With m(νµ ) from Eq.(7) or from Eq.(12) we ﬁnd that m(π ± ) − m(µ± ) ≈ N/4 · m(νµ ) ,
(9)
where N is the number of all neutrinos in the π ± lattice as given by Eq.(5). We ﬁnd also that the oscillation energies Eν are Eν (π ± ) ≈ Eν (µ± ) .
(10)
Eq.(9) says that the energy in the rest masses of all muon (respectively) antimuon neutrinos is consumed in the π ± decay. Eq.(10) says that the oscillation energy of all neutrinos in the π ± mesons is conserved in the π ± decay. Consequently there is no other source for the kinetic energy of the decay products of the π ± mesons than the energy in the rest masses of either all muon or antimuon neutrinos in π ± . That means: The µ± mesons have a neutrino lattice which diﬀers from the neutrino lattice of the π ± mesons by the absence of the muon or antimuon neutrinos. This contradicts the common belief that the µ mesons are point particles. However, since neutrinos do not interact, in a good approximation, with neither mass nor charge the ﬁnite size of the neutrino lattice of the µ mesons cannot be detected by conventional scattering experiments. Summing up: The mass of the µ± mesons can be explained with a neutrino lattice which contains the remains of the π ± lattice and an electrical charge. Since either all muon neutrios or all antimuon neutrinos have been removed from the π ± lattice to form the µ± mesons it must be that m(µ± ) ≈ 3/4 m(π ± ) as the experiments found. The mass of the τ ± mesons can be explained along similar lines.
6
The neutrino masses
The rest mass of the electron neutrino can be determined from the energy released in the decay of the neutron n → p + e− + ν¯e . Likewise the rest mass of the antielectron neutrino follows from the decay of the antineutron. We ﬁnd that m(νe ) = m(¯ νe ) = 0.365 meV/c2 .
(11)
Inserting Eq.(11) into Eq.(6) we obtain an accurate value for the mass of the muon neutrino νµ ) = 49.91 meV/c2 . (12) m(νµ ) = m(¯ Since the same considerations apply for either the π + or π − meson it must be that m(νµ ) = m(¯ νµ ). The ratio of the neutrino masses is then m(νµ )/m(νe ) = 136.74 ,
(13)
or 99.8% of the inverse of the ﬁne structure constant αf = 1/137.036. It does not seem likely that this is just a coincidence.
236
The mass and spin of the mesons, baryons, and leptons
We also ﬁnd from the decay of the D± s mesons that the mass of the τ neutrino is m(ντ ) = m(¯ ντ ) ∼ = 0.54 eV/c . ± ∼ Since N/4·m(νe ) = 0.51 m(e ) = 0.5 m(e± ) we arrive with Eq.(13) at 2
m(µ± )/m(e± ) ∼ = 3/2αf + 2 ,
(14)
(15) ±
whereas the empirical formula for the ratio of the mass of the µ mesons to the mass of the electron is m(µ± )/m(e± ) = 3/2αf + 1 = 206.55. The experimental ratio is 206.768. Similarly we arrive at m(π ± )/m(e± ) ∼ = 2/αf + 2 , ±
(16)
±
whereas the empirical formula is m(π )/m(e ) = 2/αf − 1 = 273.07. The experimental ratio is 273.13. Finally we ﬁnd for the long sought for ratio of the mass of the proton to the mass of the electron the value m(p)/m(e− ) ∼ = 0.9426[14/αf + 14] = 1821.5 ,
(17)
which is 99.2% of the experimental value 1836.16. For details see [3]. Summing up: We can explain the rest masses of the µ± and τ ± mesons and we can determine the rest masses of the e, µ, τ neutrinos. In other words, we have found the masses of all leptons exempting the electron. We will deal with the electron later.
7
The spin of the particles
It appears to be a necessary condition for the validity of a model of the particles that the model can also explain the spin of the particles. We have proposed an explanation of the spin of the particles in [7]. The spin, or the intrinsic angular momentum, of a particle is, of course, the sum of the angular momentum vectors of all components of the particle. The π 0 and η mesons do not have spin because their O(109 ) standing electromagnetic waves must be linearly polarized and hence do not have an angular momentum. The longitudinal oscillations of the π 0 or η lattices do not have an angular momentum either because for longitudinal oscillations r × p = 0. So s(π 0 , η) = 0 as it must be. The Λ baryon appears to be the superposition of two η mesons, m(Λ) = 1.019·2 m(η). The superposition of two perpendicular oscillations of the same amplitude and frequencies produces a circular oscillation whose angular momentum is /2, see [7]. The angular momentum vector of each of the O(109 ) circular oscillations is canceled by the angular momentum vector of the circular oscillation at the mirror position in the lattice. Only the angular momentum /2 of the circular oscillation at the center is not canceled. That means that the intrinsic angular momentum of the Λ baryon is /2, or s(Λ) = 1/2 as it must be. The other baryons of the γbranch are composites of the Λ baryon with spin 1/2 and one or two π mesons which do not have spin. Consequently their spin is 1/2, exempting the Ω− particle.
E.L. Koschmieder
237
The explanation of the spin of the νbranch particles is diﬀerent because each of the O(109 ) neutrinos which are in the νbranch particles has spin 1/2. However, the spin vectors of the neutrinos in the lattice cancel because the spin vector of each neutrino has an opposite spin vector at the mirror position in the lattice, but for the central neutrino. That makes the intrinsic angular momentum of the neutrino lattice of the π ± mesons equal to /2, however s(π ± ) is zero. This problem is solved by the spin of the electric charge e± carried by the π ± mesons. The spin 1/2 carried by one of the electric charges cancels the spin 1/2 of the neutrino lattice, so s(π ± ) = 0 as it must be. Similar considerations apply for K± with s(K± ) = 0. The explanation of s(K0 ) = 0 is more diﬃcult because there is no charge to cancel the spin of the neutrino lattice. However, the K0 lattice does not contain single neutrinos, but neutrino pairs, because the K0 and K0 mesons are superpositions of the second mode of the π ± mesons and the ﬁrst mode of the π ∓ mesons, K0 = (2.)π ± + π ∓ . The spin of the neutrino pairs is zero, therefore the spin of the entire neutrino lattice is zero. The K0 meson also contains a pair of opposite electrical charges whose spin cancels likewise. That means that the intrinsic angular momentum of K0 is zero. The spin of the neutron whose mass is ≈ 2 m(K0 ) is s(n) = 1/2. The spin of the neutron is caused by the angular momentum of circular oscillations which result from the superposition of two perpendicular lattice oscillations from a K0 and a K0 meson, similar to the superposition of two η mesons in the Λ baryon. For details see [7]. The spin of the proton s(p) = 1/2 is determined by the decay of the neutron, details are in [7].
8
Conclusions
The standing wave model solves a number of problems for which an answer heretofore has been hard to come by. Only photons, neutrinos, charge and the weak nuclear force are needed to explain the masses of the stable mesons and baryons and of the leptons. We can also explain the spin of the mesons and baryons and of the µ± and τ ± mesons. References [1] El Naschie MS. Chaos,Sol.Frac. 22,494 (2004). [2] Sidharth BJ. The Chaotic Universe p.49, Nova Science Publishers (2001). [3] Koschmieder EL. http://arXiv.org/physics/0408070 (2004). [4] Koschmieder EL. Chaos,Sol.Frac. 18,1129 (2003). [5] Wilson KG. Phys.Rev.D 10,2445 (1974). [6] Born M. and v.Karman Th. Phys.Z. 13,297 (1912). [7] Koschmieder EL. Hadr.J. 26,471 (2003).
CLUSTERS OF MATTER AND ANTIMATTER  A MECHANISM FOR COLD COMPRESSION
WALTER GREINER a
a
Frankfurt Institute for Advanced Studies, Johann Wolfgang Goethe Universit¨ at, D–60054 Frankfurt am Main, Germany
Abstract In this talk I ﬁrst present the vacuum for the e+ e− ﬁeld of QED and show how it is modﬁed for baryons in nuclear environment. Then I discuss the possibility of producing new types of nuclear systems by implanting an antibaryon into ordinary nuclei. The structure of nuclei containing one antiproton or antilambda is investigated within the framework of a relativistic meanﬁeld model. Selfconsistent calculations predict an enhanced binding and considerable compression in such systems as compared with normal nuclei. I present arguments that the life time of such nuclei with respect to the antibaryon annihilation might be long enough for their observation.
It is generally accepted that physical vacuum has nontrivial structure. This conclusion was ﬁrst made by Dirac on the basis of his famous equation for a fermion ﬁeld which describes simultaneously particles and antiparticles. The Dirac equation in the vacuum has a simple form (iγ µ ∂µ − m)Ψ(x) = 0 ,
(1)
where γ µ = (γ 0 , γ) are Dirac matrices, m is the fermion mass and Ψ(x) is a 4component spinor ﬁeld. For a plane wave solution Ψ(x) = e−ipx up this equation is written as (2 p − m)up = 0 , (2)
211 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 211–220. © 2006 Springer. Printed in the Netherlands.
212
Matter and Antimatter − Mechanism for Cold Compression
Figure 1. Lowest bound states of the Dirac equation for nuclei with charge Z. While the Sommerfeld ﬁnestructure energies (dashed line) for ξ = 1 (s states) end at Z = 137, the solutions for extended Coulomb potentials (full line) can be traced down to the negativeenergy continuum reached at the critical charge Zcr for the 1s state. The bound states entering the continuum obtain a spreading witdth as indicated.
where p2 = γ 0 E − γp. Multiplying by (2 p + m) and requiring that up = 0 one obtains the equation E 2 − p2 − m2 = 0 which has two solutions E ± (p) = ± p2 + m2 . (3)
Here the + sign corresponds to particles with positive energy EN (p) = E + (p), while the − sign corresponds to solutions with negative energy. To ensure stability of the physical vacuum Dirac has assumed that these negativeenergy states are occupied forming what is called now the Dirac sea. Then the second solution of eq. (3) receives natural interpretation: it describes holes in the Dirac sea. These holes are identifed with antiparticles. Their energies are obviously given by EN (p) = −E − (p) = p2 + m2 . Unfortunately, the Dirac sea brings divergent contributions to physical quantities such as energy density, and one should introduce a proper regularization scheme to rid oﬀ these divergences. This picture has received numerous coﬁrmations in quantum electrodynamics and other ﬁelds. One of the most fascinating aspects is the structure of the vacuum in QED and its change into charged vacuum states under the inﬂuence of strong (supercritical) electric ﬁelds [1]. I shortly remind of this phenomenon.
W. Greiner
213
Figure 2. Time dependence of the quasimolecular energy levels in a supercritical heavy ion collision. The arrows denote various excitation processes which lead to the production of holes and positrons.
Fig. 1 shows the diving of the deeply bound states into the lower energy continuum of the Dirac equation. In the supercritical case the dived state is degenerate with the (occupied) negative electron states. Hence spontaneous e+ e− paircreation becomes possible, where an electron from the Dirac sea occupies the additional state, leaving a hole in the sea which escapes as a positron while the electron’s charge remains near the source. This is a fundamentally new process, whereby the neutral vacuum of QED becomes unstable in supercritical electrical ﬁelds. It decays within about 10−19 s into a charged vacuum. The charged vacuum is now stable due to the Pauli principle, that is the number of emitted particles remains ﬁnite. The vacuum is ﬁrst charged twice because two electrons with opposite spins can occupy the 1s shell. After the 2p1/2 shell has dived beyond Zcr = 185, the vacuum is charged four times, etc. This change of the vacuum structure is not a perturbative eﬀect, as are the radiative QED eﬀects (vacuum polarization, selfenergy, etc.). The timedependence of the energy levels in a supercritical heavyion collision is depicted in Fig. 2. An electron (or hole) which was in a certain molecular eigenstate at the beginning of the collision can be transfered with a certain probability into diﬀerent states by the dynamics of the collision. This can lead to the hole production in an inner shell by excitation of an electron to a higher state and/or hole production by ionization of an electron to the continuum. Further possibilities are induced positron production by excitation of an electron from the lower continuum to an empty bound level and direct pair production [2].
214
Matter and Antimatter − Mechanism for Cold Compression
Figure 3. Positron energy spectra measured in collisions of Th+Ta, Th+Th, and U+Cm at energies of about 6 MeV per nucleon. The QED predictions (dashed lines) and the experimentally determined background from nuclear pair conversion (dotted lines) add up to the full lines which are in close agreement with experiment.
A comparison of the theoretical predictions and expectations and experimental data is shown in ﬁgure 3. Sharp positron peeks can be expected if there were a mechanism in the heavy ion collision leading to a time delay. This may be caused by a pocket in the potential between the two ions. Spontaneous pair production should then be enhanced in supercritical systems. Until now, however, the situation remains inconclusive [2]. It has been noticed already many years ago (see e. g. ref. [3]) that nuclear physics may provide a unique laboratory for investigating the Dirac picture of vacuum. The basis for this is given by relativistic meanﬁeld models which are widely used now for describing nuclear matter and ﬁnite nuclei. Within this approach nucleons are described by the Dirac equation coupled to scalar and vector meson ﬁelds. Scalar S and vector V potentials generated by these ﬁelds modify planewave solutions of the Dirac equation as follows E ± (p) = V ± p2 + (m − S)2 (4)
Again, the + sign corresponds to nucleons with positive energy EN (p) = V + p2 + (m − S)2 , and the − sign corresponds to antinucleons with energy EN (p) = −E − (−p) = −V + p2 + (m − S)2 . It is remarkable that changing sign of the vector potential for antinucleons is exactly what is expected from the Gparity transformation of the nucleon potential. As follows from eq. (4), in nuclear environment the spectrum of singleparticle states of the Dirac equation is modiﬁed in two ways. First, the mass gap between positive and negativeenergy states, 2(m − S), is reduced due to the scalar potential and second, all states are shifted upwards due to the vector potential. These changes are illustrated in Fig. 4.
W. Greiner
215
Figure 4. Schematic spectrum of Dirac equation in vacuum (upper panel) and in a nucleus of radius R (lower panel). A divergent contribution of negativeenergy states is often regularized by introducing a cutoﬀ momentum Λ.
It is well known from nuclear phenomenology that good description of nuclear ground state is achieved with S 350 MeV and V 300 MeV so that the net potential for nucleons is V − S −50 MeV. Using the same values one obtains for antinucleons very a deep potential, −V − S −650 MeV. Such a potential would produce many strongly bound states in the Dirac sea. However, if these states are occupied they are hidden from the direct observation. Only creating a hole in this sea, i.e. inserting a real antibaryon into the nucleus, would produce an observable eﬀect. If this picture is correct one should expect the existence of strongly bound states of antinucleons with nuclei. Below I report on our recent study of antibaryon doped nuclear systems [4].
216
Matter and Antimatter − Mechanism for Cold Compression
Unlike some previous works, we take into account the rearrangement of nuclear structure due to the presence of a real antibaryon. The structure of such systems is calculated using several versions of the relativistic mean ﬁeld model (RMF): TM1 [5], NL3 and NLZ2 [6]. Their parameters were found by ﬁtting binding energies and charge formfactors of spherical nuclei from 16 O to 208 Pb. The general Lagrangian of the RMF model is written as & L= ψ j (iγ µ ∂µ − mj )ψj j=B,B
1 1 b c + ∂ µ σ∂µ σ − m2σ σ 2 − σ 3 − σ 4 2 2 3 4 1 1 d − ω µν ωµν + m2ω ω µ ωµ + (ω µ ωµ )2 4 2 4 1 µν 1 2 µ − ρ µν + mρ ρ ρ ρ µ 4 2 & + ψ j (gσj σ + gωj ω µ γµ + gρj ρ µ γµτj )ψj
(5)
j=B,B
+ Coulomb part Here summation includes valence baryons B, in fact the nucleons forming a nucleus, and valence antibaryons B inserted in the nucleus. They are treated as Dirac particles coupled to the scalarisoscalar (σ), vectorisoscalar (ω) and vectorisovector ( ρ) meson ﬁelds. The calculations are carried out within the meanﬁeld approximation where the meson ﬁelds are replaced by their expectation values. Also a “nosea” approximation is used. This implies that all occupied states of the Dirac sea are “integrated out” so that they do not appear explicitly. It is assumed that their eﬀect is taken into account by nonlinear terms in the meson Lagrangian. Most calculations are done with antibaryon coupling constants which are given by the Gparity transformation (gσN = gσN , gωN = −gωN ) and SU (3) ﬂavor symmetry gσΛ = 32 gσN , gωΛ = 32 gωN ). In isosymmetric static systems the scalar and vector potentials for nucleons are expressed as S = gσN σ and V = gωN ω 0 . Following the procedure suggested in Ref. [7] and assuming the axial symmetry of the nuclear system, we solve eﬀective Schr¨odinger equations for nucleons and an antibaryon together with diﬀerential equations for mean meson and Coulomb ﬁelds. We explicitly take into account the antibaryon contributions to the scalar and vector densities. It is important that antibaryons give a negative contribution to the vector density, while a positive contribution to the scalar density. This leads to increased attraction and decreased repulsion for surrounding nucleons. To maximize attraction, nucleons move to the center of the nucleus, where the antiproton has its largest occupation probability. This gives rise to a strong local compression of the nucleus and leads to a dramatic rearrangement of its structure. Results for the 16 O nucleus are presented in Fig. 5 which shows 3d plots of nucleon density distributions. The calculations show that inserting an antiproton into the 16 O nucleus leads to the increase of central nucleon density by a factor 2–4 depending on the parametrization. Due to a very deep antiproton potential
W. Greiner
217
Figure 5. Sum of neutron and proton densities for 16 O (top), 16 O with p (bottom right) and with Λ (bottom left) calculated with the parametrization NLZ2.
16 O
the binding energy of the whole system is increased signiﬁcantly as compared with 130 MeV for normal 16 O. The calculated binding energies of the p–16 O system are 830, 1050 and 1160 MeV for the NLZ2, NL3 and TM1, respectively. Due to this anomalous binding we call such systems super bound nuclei (SBN). In the case of antilambdas we rescale the coupling constants with a factor 2/3 that leads to the binding energy of 560÷700 MeV for the Λ–16 system.
218
Matter and Antimatter − Mechanism for Cold Compression
Figure 6. Contour plot of nucleon densities for 8 Be without (left) and with (right) antiproton calculated with the parametrization NL3.
As a second example, we investigate the eﬀect of a single antiproton inserted into the 8 Be nucleus. The normal 8 Be nucleus is not spherical, exhibiting a clearly visible 2α structure with the ground state deformation β2 1.20. As seen in Fig. 6, inserting an antiproton in 8 Be results in a much less elongated shape (β2 0.23) and disappearance of its cluster structure. The binding energy increases from 53 MeV to about 700 MeV. Similar, but weaker eﬀects have been predicted [8] for the K − bound state in the 8 Be nucleus. The calculations have been performed also with reduced antinucleon coupling constants as compared to the Gparity prescription. We have found that the main conclusions about enhanced binding and considerable compression of pdoped nuclei remain valid even when coupling constants are reduced by factor 3 or so. The crucial question concerning possible observation of the SBNs is their life time. The main decay channel for such states is the annihilation of antibaryons on surrounding nucleons. The energy available for annihilation of a bound antinucleon equals Q = 2mN − BN − BN , where BN and BN are the corresponding binding energies. In our case this energy is at least by a factor 2 smaller as compared with the vacuum value of 2mN . This should lead to a signiﬁcant suppression of the available phase space and thus to a reduced annihilation rate in medium. We have performed detailed calculations assuming that the annihilation rates into diﬀerent channels are proportional to the available phase space. All intermediate states with heavy mesons like ρ, ω, η as well as multipion channels have been
W. Greiner
219
considered. Our conclusion is that decreasing the Q value from 2 GeV to 1 GeV may lead to the reduction of total annihilation rate by factor 20÷30. Then we estimate the SBN life times on the level of 525 fm/c which makes their observation feasible. This large margin in the life times is mainly caused by uncertainties in the overlap integral between antinucleon and nucleon scalar densities. Longer life times may be expected for SBNs containing antihyperons. The reason is that instead of pions more heavy kaons must be produced in this case. We have also analyzed multinucleon annihilation channels (Pontecorvolike reactions) and have found their contribution to be less than 40% of the singlenucleon annihilation. We believe that such exotic nuclear states can be produced by using antiproton beams of multiGeV energy, e.g. at the future GSI facility. It is well known that lowenergy antiprotons annihilate on the nuclear periphery (at about 5% of the normal density). Since the annihilation cross section drops signiﬁcantly with energy, a highenergy antiproton can penetrate deeper into the nuclear interior. Then it can be stopped there in an inelastic collision with a nucleon, e.g. via the reaction A(p, N π)p A , leading to the formation of a pdoped nucleus. Reactions like A(p, Λ)Λ A can be used to produce a Λdoped nucleus. Fast nucleons or lambdas can be used for triggering such events. In order to be captured by a target nucleus ﬁnal antibaryons must be slow in the lab frame. Rough estimates of the SBN formation probability in a central pA collision give the values 10−5 − 10−6 . With the p beam luminocity of 2 · 1032 cm−2 s−1 planned at GSI this will correspond to the reaction rate of a few tens of desired events per second. Several signatures of SBNs can be used for their experimental observation. First, annihilation of a bound antibaryon can proceed via emission of a single photon, pion or kaon with an energy of about 1 GeV (such annihilation channels are forbidden in vacuum). So one may search for relatively sharp lines, with width of 10÷40 MeV, around this energy, emitted isotropically in the SBN rest frame. Another signal may come from explosive disintegration of the compressed nucleus after the antibaryon annihilation. This can be observed by measuring radial collective velocities of nuclear fragments. It is interesting to look at the antibaryonnucleus system from somewhat diﬀerent point of view. An antibaryon implanted into a nucleus acts as an attractor for surrounding nucleons. Due to the uncompensated attractive force these nucleons acquire acceleration towards the center. As the result of this inward collective motion the nucleons pile up producing local compression. If this process would be completely elastic it would generate monopolelike oscillations around the compressed SBN state. The maximum compression is reached when the attractive potential energy becomes equal to the compression energy. Simple estimates show that local baryon densities up to 5 times the normal nuclear density may be obtained in this way. It is most likely that the deconﬁnment transition will occur at this stage and a highdensity cloud containing an antibaryon and a few nucleons will appear in the form of a multiquarkantiquark cluster. One may speculate that the whole 4 He or even 16 O nucleus can be transformed into the quark phase by this mechanism. As shown in ref. [9], an admixture of antiquarks to cold quark matter is energetically favorable. The problem of annihilation is now transferred
220
Matter and Antimatter − Mechanism for Cold Compression
to the quark level. But the argument concerning the reduction of available phase space due to the entrancechannel nuclear eﬀects should work in this case too. Thus one may hope to produce relatively cold droplets of the quark phase by the inertial compression of nuclear matter initiated by an antibaryon. I am grateful to T. B¨ urvenich, I.N. Mishustin and L.M. Satarov for fruitful discussions and help in the preparation of this talk. References [1] W. Greiner, B. M¨ uller, J. Rafelski, Quantum electrodynamics of strong ﬁelds, Springer Verlag, 2nd edition, December 1985 [2] J. Reinhardt and W. Greiner, Quantum electrodynamics, Springer Verlag, 3rd edition, February 2003 [3] N. Auerbach, A.S. Goldhaber, M.B. Johnson, L.D. Miller, A. Picklesimer, Phys. Lett. B182 (1986) 221. [4] T. B¨ urvenich, I.N. Mishustin, L.M. Satarov, H. St¨ ocker, W. Greiner, Phys. Lett. B542 (2002) 261. [5] Y. Sugahara and H. Toki, Nucl. Phys. A579 (1994) 557. [6] M. Bender, K. Rutz, P.–G. Reinhard, J.A. Maruhn, and W. Greiner, Phys. Rev. C60 (1999) 34304. [7] G. Mao, H. St¨ ocker, and W. Greiner, Int. J. Mod. Phys. E8 (1999) 389. [8] Y. Akaishi and T. Yamazaki, Phys. Rev. C65 (2002) 044005. [9] I.N. Mishustin, L.M. Satarov, H. Stoecker, W. Greiner, Phys. Rev. C 59 (1999) 3343.
EQUALITY AND IDENTITY AND (IN)DISTINGUISHABILITY IN CLASSICAL AND QUANTUM MECHANICS FROM THE POINT OF VIEW OF NEWTON’S NOTION OF STATE
PETER ENDERS a
a
Fischerinsel 2, D10179 Berlin, Germany
Abstract The notion of state is a central notion for all branches of physics. Surprisingly enough, Newton’s notion diﬀers from the nowadays notion. Our review of the beneﬁts of Newton’s notion comprises Gibbs’s paradox, Einstein’s derivation of the classical and quantum distribution laws from the energetic spectrum (serving to remove anthropomorphic elements), the diﬀerence between ‘identical’ and ‘indistinguishable’ (being a property of states rather than of particles), a new physical content of ψ(x, t) (the invariance of ψ(x, t) rather than ψ(x, t) against permutations yields not only fermions and bosons, but also anyons), and a novel classiﬁcation of forces (leading eventually to a derivation of the MaxwellLorentz equations from classical mechanics).
1
Introduction
Classical mechanics is the safest (if not the only safe) ground we can move on. For this, we will analyze the implications of Newton’s notion of state diﬀering considerably from the contemporary one for the notions ‘equality’, ‘identity’ and ‘(in)distinguishability’ playing a paramount role in statistics and in quantum mechanics. Newton’s notion allows for considering them within classical point mechanics, what frees the discussion from anthropomorphic elements. Bach’s (1997) fundamental results are obtained within an elementary dynamical framework.
239 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 239–245. © 2006 Springer. Printed in the Netherlands.
240 2
Equality and Identity and (In)distinguishability Newton’s notion of state
Newton’s First Law “Every body perseveres in its state of being at rest or of moving uniformly straight forward, except insofar as it is compelled to change its state by forced impressed.” – Newton’s state corresponds to nowadays’ stationary state. Newton’s Second Law “A change in motion is proportional to the motive force impressed and takes place along the straight line in which that force is impressed.” – ∆ p ∼ F : Newton’s state variable is the momentum (the conserved quantities). Laplace’s demon “A suﬃciently powerful intelligence knowing all loci and velocities in a mechanical systems at one time is able to calculate the loci and velocities at all later times.” – Laplace’s state is nowadays’ state, including both stationary and nonstationary ones, the state variables are a complete set of independent dynamical variables. Advantages of Newton’s notion: • Easily generalized to classical and quantum systems (not tied to orbits); • Amount of conserved quantities ˜ amount of quantum (state) numbers; • Symmetry of state = symmetry of state function → gauge symmetry, geometric phases, . . . Advantages of modern (Laplace’s) notion: • Complete description of motion; • Identiﬁcation of phase space points diﬀering only by interchanging equal bodies → multiply connected spaces → appropriate topology for anyons. Disadvantages of modern (Laplace’s) notion: • The state changes even in absence of causes; • State at rest is ignored; • Interchange of equal bodies changes state → Gibbs’s paradox; • Inapplicable to quantum systems.
P. Enders 3
241
Equal bodies in Newtonian states 1) Equal / identical bodies / particles
Equality We call two classical bodies or quantum particles equal, when their interchange does not change the properties / state / motion of a system. (cf Helmholtz, §10) Identity “Particles are called identical, if they agree in all their intrinsic (i.e. state independent) properties.” (Bach, p.15) Remark: The restriction to the intrinsic properties circumvents the conﬂict with the logical notion ‘identical’ (= equal in all properties). 2) Permutation symmetry of Newtonian states All conserved quantities of a classicalmechanical system: total energy / momentum / angular momentum / . . . , are invariant w.r.t. the permutation of equal parameters, ie, w.r.t. the permutation of (labels of) equal bodies, ie, w.r.t. bodies with equal properties concerning the system considered, ⇒ • A Newtonian state is invariant against interchanging (labels of) equal bodies, • Equal bodies cannot be distinguished or identiﬁed by means of the conserved quantities (Newtonian/ stationarystate variables), • Anthropomorphic arguments like ‘particle can be marked or not’ are not relevant points of view (interchanging or marking two resting red balls in a snooker game does not interfere the game); • Indistinguishable classical particles have no trajectories (Bach provides probabilistic proof). 3) Comparison with Laplace’s notion of state (continued) The locus of a body is that part of space it occupies. Euler’s exclusion principle (not to be interchanged with Pauli’s exclusion principle!) states, that no body can occupy more than one locus, and no part of space can be occupied by more than one body. ⇒ • Equal bodies can, at least in principle, always be distinguished and identiﬁed by means of their locus. • Equal bodies are distinguishable within Laplace’s notion of state. • (In)distinguishability is a property of states, not of particles/bodies. (cf Bach, p.15)
242 4
Equality and Identity and (In)distinguishability Classical and quantum distribution laws
Einstein (1907) has derived the classical and quantum distribution laws using just the energetic spectra of a classical (continuous spectrum) and a quantum harmonic oscillator (discrete spectrum): !∞ 0
Ee− kT dE E
Eclass = ! ∞ 0
Equant =
e− kT dE E
+∞
− ωn kT n=0 ωne +∞ − ωn kT n=0 e
=
ω e
ω kT
−1
= kT
(1)
= Eclass
ω kT
e
ω kT
−1
(2)
(In)distinguishability does not play any role (cf Bach’s “BoseEinstein statistics as invented by Boltzmann”).
5
New physical content of ψ(x, t)
FFP5: Quantummechanical systems are conservative systems which may assume conﬁgurations for which V (x) > E ⇒ There is an eﬀective potential energy Vnkl (x) = VEnkl (x) = FEnkl (x) · V (x) ≤ Enkl ;
−∞ <x < +∞
(3)
2
FEnkl (x) ∼ ψEnkl (x) is a limiting function such, that Vnkl (x) ≤ Enkl even if V (x) > Enkl (→ tunnel eﬀect demystiﬁed). Progress since FFP5: • Common principles of state change for classical and quantum systems → Eulerean derivation of timedependent Schr¨ odinger equation from timeindependent one; • FEnkl is dimensionless ⇒ FEnkl = FEnkl (x/x0 ): All quantum systems exhibit characteristic length x0 ; 2
• FE(nkl) (x/x0 ) ∼ ψE (x)
→ gauge symmetry;
• Novel classiﬁcation of ﬁelds: A) Fields being related to total energy, E (accelerating ﬁelds like electric ﬁeld, E), B) Fields being related not to E, but to x0 (refracting ﬁelds like magnetic ﬁeld, B), C) Fields being related neither to E, nor to x0 (gauge ﬁelds, see AharonovBohm eﬀect); Remark: The MaxwellLorentz equations turn out to be just compatibility and B. conditions for such ﬁelds E
243
P. Enders 6
Newtonian state variable ψ(x, t)
1) By virtue of their deﬁnition, limiting functions are invariant against permutations of equal particles (omitting x0 ): FE (x2 , x1 ) = FE (x1 , x2 ) ≥ 0
(4)
⇒ the most general representation of FE (x1 , x2 ) reads (mE entire) FE (x1 , x2 ) = γ,m (x1 , x2 ) = ψE
2
γ,m ψE (x1 , x2 ) ; " # 1 √ eiγE (x1 ,x2 ) ψ˜E (x1 , x2 ) + eimE π ψ˜E (x2 , x1 ) 2
(5)
(6)
2) Wigner’s theorem: γ,m If ψE (x1 , x2 ) is eigenfunction, then ˆ γ,m (x1 , x2 ) Rψ E
(1,2)
= eiρE
(1,2) iρE
=
e
γ,m ψE (x2 , x1 )
(7)
γ,m eimE π ei[γE (x2 ,x1 )−γE (x1 ,x2 )] ψE (x1 , x2 )
(8)
(1,2)
is also eigenfunction ⇒ [γE (x2 , x1 ) − γE (x1 , x2 )] can be absorbed into ρE
.
3) 2nd permutation: (1,2)
m ˆ 2 ψE ˆ iρE R (x1 , x2 ) = Re
(1,2)
γ,m ψE (x2 , x1 ) = eiρE
(2,1)
eiρE
m ψE (x1 , x2 )
(9)
Standard case: (1,2)
ˆ 2 = eiρE R
(2,1)
eiρE
= 1;
(2,1)
ρE
(1,2)
= −ρE
;
γE (x1 , x2 ) = 0
(10)
⇒ Wave functions are either symmetric (bosons, +) or antisymmetric (fermions, −): # 1 " ± ψE (x1 , x2 ) = √ ψ˜E (x1 , x2 ) ± ψ˜E (x2 , x1 ) (11) 2
Nonstandard, ‘anyonic’ case: (1,2)
ˆ 2 = eiρE R
(2,1)
eiρE
= 1;
(2,1)
ρE
(1,2)
= −ρE
;
γE (x2 , x1 ) − γE (x1 , x2 ) = 0
(12)
⇒ If there are topologically inequivalent paths, the wave function is neither symmetric, nor antisymmetric, but can exhibit any intermediate behaviour → anyons (the clue to the fractional quantum Hall eﬀect).
244 7
Equality and Identity and (In)distinguishability About the meaning of ‘identical’ and ‘indistinguishable’
1) Some rigorous deﬁnitions Equal means ‘equal in some welldeﬁned properties such as mass, density, shape, charge . . . , but not in all’ Example: the 2 electrons in the ground state of He (they diﬀer in sz ) ‘equal’ depends on the view, ie, which properties one is looking at. Congruent means ‘equal in all essential (geometric) properties, but not in locus’ Example: 2 red snooker balls of high quality Identical means ‘one and the same’, ie, equal in all properties (strictly speaking, no exception at all) Example: 2 squares of equal side length on the same place of a sheet Indistinguishable means, that there is no mean (no one diﬀering property/attribute) for discrimination. ⇒ Indistinguishable things are identical (Leibniz) 2) Questions If we weaken the deﬁnition of ‘identical’, there may be a weakening of ‘distinguishable’ to ‘identiﬁable’, so that we are led to the question • Are there nonclassical indistinguishable bodies/particles not being identical? • Are there principally distinguishable (ie, not identical) bodies/particles not being identiﬁable? 3) Observations • Quantum particles are identical w.r.t. intrinsic properties and ‘almost identical’ w.r.t. state properties: All electrons (protons, . . . ) exhibit the same mass at rest, electrical charge, modulus of spin, etc.; • Pauli’s exclusion principle: 2 electrons diﬀer in at least one quantum number – however: it does not say, which electron is in which state (entanglement); • The quanta occupying an oscillator loose their individuality: Say, 12 quanta in state E12 occupy all together the one 12quanta state, not 12 singlequantum states, ⇒ they have got no individual properties (parameter values) (in contrast to electrons, these quanta – Planck’s “energy elements” – occupy not singleparticle, but singlesystem states); – nevertheless: these 12 quanta are not one and the same (one thing) as we are thinking them as 12 particles. • Cluster law: Wave functions of distinct systems need not to be entangled ↔ There are distinguishable equal quantum particles.
P. Enders
245
4) Conclusions • The notions ‘identical’ or ‘distinguishable’ as used in logics play almost no role, in fact: the actual physical meaning of ‘distinguishability’ as property of states is the identiﬁability; • ‘Identical’ is meaningful in the sense of Bach (including only intrinsic properties, eg, spin s – but not sz ); • There is no principal diﬀerence between classical bodies, bosons and fermions w.r.t. these properties.
8
Summary
Newton’s notion of state is an addition to, though not a complete replacement of Laplace’s notion. Our treatment of equality, identity and (in)distinguishability accounting for Newton’s notion of state reveals the following advantages and new results. • Common treatment of classical bodies and quantum particles; • Nonprobabilistic classiﬁcation of (bodies/particles in) states; • The limiting function FE gives ψE  a new physical meaning as relative space occupation; both exhibit the same symmetry as Newtonian (stationary) state variables (quantum numbers) w.r.t. external ﬁelds (→ gauge invariance) and permutations (→ fermions, bosons, and anyons); • Distribution functions can be related to energetic spectra and occupation ⇒ they are independent of (in)distinguishability; • The meaning of ‘identity’ and ‘indistinguishability’ in physics is partly at variance with their meaning in logics, hence, careful restrictions are necessary; • There are particles exhibiting diﬀerent extrinsic properties (eg, spin direction in EPR), but not being identiﬁable: Equal bodies in symmetric states cannot be identiﬁed, be there diﬀerent attributes or not.
References [1] A. Bach, Indistinguishable Classical Particles (Lecture Notes in Physics m44) Springer, Berlin Heidelberg (1997) [2] P. Enders, From Classical to Quantum Physics. A Deductive Approach, Springer, Berlin Heidelberg (2005) (in German) [3] H. v. Helmholtz, Introduction to the Lectures on Theoretical Physics, Barth, Leipzig (1903) (in German)
NUMERICAL MODELLING OF QUANTUM STATISTICS IN HIGHENERGY PHYSICS
OLEG V. UTYUZH a , GRZEGORZ WILK b , ZBIGNIEW W lODARCZYK b a
b
The Andrzej Soltan Institute for Nuclear Studies, Nuclear Theory Department, ul. Ho˙za 69; 00681 Warszawa, Poland; emails: [emailprotected], [emailprotected] ´ etokrzyska Academy, ul. Swi¸ ´ etokrzyska Institute of Physics, Swi¸ 15, 25406 Kielce, Poland, email: [emailprotected]
Abstract Numerical modelling of quantum eﬀects caused by bosonic or fermionic character of secondaries produced in high energy collisions of diﬀerent sorts is at the moment still far from being established. In what follows we propose novel numerical method of modelling BoseEinstein correlations (BEC) observed among identical (bosonic) particles produced in such reactions. We argue that the most natural approach is to work directly in the momentum space of produced secondaries in which the Bose statistics reveals itself in their tendency to bunch in a speciﬁc way in the available phase space. Fermionic particles can also be treated in similar fashion.
The multiparticle production processes consist substantial part of the high energy collisions and are of considerable theoretical interest. Unfortunately their description is so far available only by means of numerical Monte Carlo codes based to some extent on modern theoretical ideas but otherwise remaining purely phenomenological [1]. They are build in such manner as to describe as close as possible the complicated ﬁnal state of such reaction, cf., Fig. 1. However, using classical (positive deﬁned) probabilities as their basic tool such MC codes cannot directly describe
247 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 247–254. © 2006 Springer. Printed in the Netherlands.
248
Numerical modelling of quantum statistics
some observed features, which are connected with BoseEinstein (BE) (or FermiDirac (FD)) statistics of produced secondaries (in case they are identical and occupy almost the same parts of the phase space deﬁned by the uncertainty relation, see Fig. 1). When two (or more) identical particles of the same kind are observed, their common wave function should be symmetrized (for BE statistics) or antisymmetrized (for FD statistics) what results in characteristic shapes of (two particle, for example) correlation function C2 (Q = p1 − p2 ) = N (p1 , p2 )/[N (p1 ) · N (p2 )], cf., Fig. 2.
Figure 1. Schematic view of high energy collision resulting in production of many particles of diﬀerent statistics.
Figure 2. Example of BE and FD statistics (left panel) and two particle correlation functions they lead to (right panels).
Referring for details of BoseEinstein correlations (BEC) to the literature (cf., for example, [2] and references therein) let us concentrate here directly on the problem of their proper numerical modelling, i.e., such in which the bosonic character of secondaries produced in hadronization process are going to be accounted
O.V. Utyuzh et al.
249
for from the very beginning. This problem was so far considered only in [3] (using statistical approach based on information theory approach, cf., however, also [4]). All other approaches, which claim to model BEC numerically [5], simply add to the outcomes of existing MC codes [1] some afterburners, which modify them in a suitable way to be able ﬁt the BEC data. Such approach inevitably leads to such unwanted features as violation of energymomentum conservation or changes in the original (i.e., obtained directly form MC code) multiparticle spectra. In [6] we have proposed afterburner free from such unwanted eﬀects. It was based on diﬀerent concept of introducing quantum mechanical (QM) eﬀects in the otherwise purely probabilistic distributions then those proposed in [7]. Namely, each MC code provides us usually with a given number of particles, each one endowed with either (+) or (−) or (0) charge and with well deﬁned spatiotemporal position and energymomentum. But experiment provides us information on only the ﬁrst and last characteristics. The spatiotemporal information is not available directly. In fact, the universal hope expressed in [2, 5] is that precisely this information can be deduced from the previous two via the measured BEC. Our reasoning was as follows: (i) BEC phenomenon is of QM origin therefore one has to introduce in the otherwise purely classical distributions provided by MCG the new element mimicking QM uncertainties; (ii) it cannot be done with energymomenta because they are measured and therefore ﬁxed; (iii) the next candidate, i.e., spatiotemporal characteristics can be changed but it was already done in [7, 5]; (iv) one is thus left with charges and in [6] we have simply assigned (on eventbyevent basis) new charges to the particles from MCG conserving, however, the original multiplicities of (+/ − /0). This has been done in such way as to make particles of the same charge to be located maximally near to each other in the phase space exploring for this natural ﬂuctuations in spatiotemporal and energymomentum characteristic of the outcome of MCG. The advantages of such approach are: (a) energymomentum is automatically conserved and multiparticle distributions are not modiﬁed and (b) it is applicable already on the level of each event provided by MCG (not only, as some of propositions of [5] only to all events). However, the new assignment of charges introduces a profound change in the structure of the original MCG. Generally speaking (cf. [6] for details) it requires introduction of bunchings of particles of the same charge. This observation will be the cornerstone of our new proposition. Let us ﬁrst remind that idea of bunching of particles as quantum statistical (QS) eﬀect is not the new one [8]. It was used in connection with BEC for the ﬁrst time in [9] and then was a basis of the so called clan model of multiparticle distributions leading in natural way to their negative binomial (NB) form observed in experiment [10]. It was then again introduced in the realm of BEC in [11] and [3, 4]. Because our motivation comes basically from [3] let us outline shortly its basic points. It deals with the problem of how to distribute in a least biased way a given number of bosonic secondaries, n = n(+) +n(−) +n(0) , n(+) = n(−) = n(0) . Using information theory approach (cf., [12]) their rapidity distribution was obtained in form of grand partition function with temperature T and chemical potential µ. In addition, the rapidity space was divided into cells of equal size δy each (it was
250
Numerical modelling of quantum statistics
ﬁtted parameter). It turned out that whereas the very fact of existence of such cells was enough to obtain reasonably good multiparticle distributions, P (n), (actually, in the NBlike form), their size, δy, was crucial for obtaining the characteristic form of the 2−body BEC function C2 (Q = pi − pj ) (peaked and greater than unity at Q = 0 and then decreasing in a characteristic way towards C2 = 1 for large values of Q, see Fig. 2) out of which one usually deduces the spatiotemporal characteristics of the hadronization source [2] (see [3] for more details). The outcome was obvious: to get C2 peaked and greater than unity at Q = 0 and then decreasing in a characteristic way towards C2 = 1 for large values of Q one must have particles located in cells in phase space which are of nonzero size. It means then that from C2 one gets not the size of the hadronizing source but only size of the emitting cell, in [3] R ∼ 1/δy, cf. [13]. In the quantum ﬁeld theoretical formulation of BEC this directly corresponds to the necessity of replacing delta functions in commutator relations by a well deﬁned peaked functions introducing in this way same dimensional scale to be obtained from ﬁts to data [14]. This fact was known even before but without any phenomenological consequences [15]. Let us suppose now that we have mass M and we know that it hadronizes into N = n bosonic particles (assumed to be pions of mass m) with equal numbers of (+/ − /0) charges and with limited transverse momenta pT . Let the multiplicity distribution of these pions follows some NBlike form, broader than Poissonian one. Suppose also that the twoparticle correlation function of identical particles, C2 (Q), has the speciﬁc BEC form mentioned above. How to model such process from the very beginning, i.e., in such way that bosonic character of produced particles is accounted for from the very beginning and not imposed at the end? We propose the following steps (illustrated by comparison to some selected LEP e+ e− data [16]): (1)
(1) Using some (assumed) function f (E) select a particle of energy E1 and charge Q(1) . The actual form of f (E) should reﬂect somehow our a priori knowledge of the particular collision process under consideration. In what follows we shall assume that f (E) = exp (−E/T ), with T being parameter (playing in our example the role of ”temperature”). (2) Treat this particle as seed of the ﬁrst elementary emitting cell (EEC) and add to it, until the ﬁrst failure, other particles of the same charge Q(1) selected according to distribution P (E) = P0 · f (E), where P0 is another parameter (actually it plays here the role of ”chemical potential” µ = T · ln P0 ). This assures that the number of particles in this ﬁrst EEC, k1 , will follow geometrical (or BoseEinstein) distribution, and in precisely this way one accounts for the bosonic character of produced pions. This results in C2 (Q) > 1 but only at one point, namely for Q = 0. (3) To get the observed spread out of C2 (Q) one has to allow that particles in this EEC have (slightly) diﬀerent energies from energy of the particle being its seed. To do it allow that each additional particle selected in point (2)
251
Clan2
Clan3
correlated
correlated
Clan1
P( N ) correlated
Hadronic Source
Independent production
O.V. Utyuzh et al.
Classical clans:
PPossion ( N Clan )
Plog ( N part )
PNB ( N )
Quantum clans:
PPossion ( N Clan )
PGeom ( N part )
PPA ( N )
Figure 3. Schematic view of our algorithm, which leads to bunches of particles (clans). Whereas in [10] these clans could consist of any particles distributed logarithmically in our case they consist of particles of the same charge and (almost) the same energy and are distributed geometrically to comply with their bosonic character. (1)
above have energy Ei selected from some distribution function peaked at (1) (1) (1) E1 , G E1 − Ei . (4) Repeat points (1) to (2) as long as there is enough energy left. Correct in every event for every energymomentum nonconservation caused by selection procedure and assure that N (+) = N (−) . As result we get a number of EECs with particles of the same charge and (almost) the same energy, which we regard as being equivalent to clans in the [10] (see Fig. 3). These clans are distributed in the same way as the particles forming the seeds for those EEC, i.e., according to Poisson distribution (see Fig. 4, upperleft panel). On the other hand, as was already said, particles in each EEC will be distributed according to geometrical distribution (see Fig. 4, upperright panel). As a result the overall distribution of particles will be of the so called P` olyaAeppli distribution [17]. It ﬁts our examplatory data reasonable well. It is interesting to notice at this point that to get NB distribution resulting from the classical clan model of [10] one should have logarithmic rather than geometrical distribution of particles in EEC, which would then not account for the bosonic character of produced secondaries. In this respect our model diﬀers from this classical clan model and we see that what we have obtained is indeed its quantum version, therefore its proposed name: quantum clan model. The ﬁrst preliminary results presented in Fig. 4 are quite encouraging (especially when one remembers that so far eﬀects of resonances and all kind of ﬁnal state interactions to which C2 is sensitive were neglected here). It remains now to be checked what twobody BEC functions for other components of the momentum diﬀerences and how they depend on the EEC parameters: T , P0 and σ. So far the
252
Numerical modelling of quantum statistics
Figure 4. Upper panels: distribution of cells and particles in a given cell. Lowerleft panel: the corresponding summary P (n) which is convolution of both P (ncell ) and P (np ). Lowerright panel: examples of the corresponding corresponding correlation functions C2 (Q) . Two sets of parameters were used. Data are from [16].
main outcome is that BEC are due to EEC’s only and therefore provide us mainly with their characteristics (it is worth to mention at this point that essentially this type of approach has been also proposed to simulate BoseEinstein condensate phenomenon in [18]). This should clear at least some of many apparently ”strange” results obtained from BEC recently (see Quark Matter 2004 proceedings, especially [19]). The most intriguing is the fact that apparently the ”size” of the hadronizing source deduced from the BEC data does not vary very much with energy and with the size of colliding objects as has been naively expected [2]. In our approach this has simple explanation, see Fig. 5. The point is that BEC are mainly sensitive to the correlation length, which in our case is dimension of the emitting cell, not to dimension of the ”ﬁreball” in which hadronization process takes place. The size of this ﬁreball depends mainly on the number of produced secondaries [19], which in our case is given by the ”partition temperature” pa
253
O.V. Utyuzh et al.
e+e
_ pp(pp)
_ pp(pp) higher energy
PbPb (SPS)
AuAu (RHIC)
PbPb (LHC)
Fireball
cell
energy
Figure 5. Schematic view of how in our approach the size of the EEC compares with the size of hadronizing ﬁreball at diﬀerent energies and for diﬀerent types of projectiles.
rameter T and by the ”chemical potential” parameter µ = T · ln P0 . They are changing with mass M (and therefore with the energy of reaction and the type of projectile). On the other hand dimension of EEC is given entirely by parameter σ describing the spread of energy of particles belonging to this EEC, which is only weakly depending on energy (if at all). We shall close with mentioning that our approach accounts also for multiparticle BEC and, because of this, by intermittence eﬀects seen in data (at least to some extend) [6].
Acknowledgments Partial support of the Polish State Committee for Scientiﬁc Research (KBN) (grant 2P03B04123 (ZW) and grants 621/E78/SPUB/CERN/P03/DZ4/99 and 3P03B05724 (GW)) is acknowledged. References [1] K.J. Eskola, Nucl. Phys. A698 (2002) 78; S.A.Bass et al. Prog. Part. Nucl. 41 (1998) 225. [2] U.A. Wiedemann and U. Heinz, Phys. Rep. 319 (1999) 145; R.M. Weiner, Phys. Rep. 327 (2000) 249; T. Cs¨ orgHo, in Particle Production Spanning MeV and TeV Energies, eds. W.Kittel et al., NATO Science Series C, Vol. 554, Kluwer Acad. Pub. (2000), p. 203 (see also: hepph/0001233); G. Baym, Acta Phys. Polon. B29 (1998) 1839; W. Kittel, Acta Phys. Polon. B32 (2001) 3927; G. Alexander, Rept. Prog. Phys. 66 (2003) 481. [3] T. Osada, M. Maruyama and F. Takagi, Phys. Rev. D59 (1999) 014024. [4] W. Zajc, Phys. Rev. D53 (1987) 3396; R.L. Ray, Phys. Rev. C57 (1998) 2352; J.G.Cramer, Event Simulation of HighOrder BoseEinstein and Coulomb Correlations, Univ. of Washington preprint (1996, unpublished). [5] L. L¨ onnblad and T. Sj¨ ostrand, Eur. Phys. J. C2(1998) 165; K. Fialkowski and R. Wit, Eur. Phys. J. C2(1998) 691; K. Fialkowski, R. Wit and J. Wosiek, Phys. Rev. D58(1998) 094013; T.Wibig, Phys. Rev. D53 (1996) 3586. B. Andersson, Acta Phys. Polon. B29 (1998) 1885; J.P. Sullivan et al., Phys. Rev. Lett. 70, 3000 (1993); K. Geiger, J. Ellis, U. Heinz and U.A. Wiedemann, Phys. Rev. D61(2000) 054002. [6] O.V. Utyuzh, G. Wilk and Z. Polon. B33(2002) 2681.
Wlodarczyk, Phys. Lett. B522 (2001) 273 and Acta Phys.
[7] A. Bialas and A. Krzywicki, Phys. Lett. B354, (1995) 134; H. Merlitz and D. Pelte, Z. Phys. A351(1995) 187 and A357 (1997) 175; U.A. Wiedemann et al., Phys. Rev. C56
254
Numerical modelling of quantum statistics (1997) R614; T. Cs¨ orgHo and J. Zim´ anyi, Phys. Rev. Lett. 80 (1998) 916 and Heavy Ion Phys. 9 (1999) 241.
[8] W.J. Knox, Phys. Rev. D10 (1974) 65; E.H. De Groot and H. Satz, Nucl. Phys. B130 (1977) 257; J. Kripfganz, Acta Phys. Polon. B8(1977) 945; A.M. Cooper, O. Miyamura, A. Suzuki and K. Takahashi, Phys. Lett. B87 (1979) 393; F. Takagi, Prog. Theor. Phys. Suppl. 120 (1995) 201. [9] A. Giovannini and H.B. Nielsen, Stimulated emission model for multiplicity ﬂuctuations, The Niels Bohr Institute preprint, NBIHE7317 (unpublished) and Stimulated emission eﬀect on multiplicity distribution in: Proc. of the IV Int. Symp. on Multip. Hydrodynamics, Pavia 1973, Eds. F.Duimio, A.Giovannini and S.Ratii, p. 538. [10] A. Giovannini and L. Van Hove, Z. Phys. C30 (1986) 381; see also: P. Carruthers and C.S.Shih, Int. J. Mod. Phys. A2 (1986) 1447. [11] M. Biyajima, N. Suzuki, G. Wilk and Z. Wlodarczyk, Phys. Lett. B386 (1996) 297. [12] O.V. Utyuzh, G. Wilk and Z. Wlodarczyk, Multiparticle production processes from the nonextensive point of view, hepph/0410341 and references therein. [13] W.A. Zajc, A pedestrian’s guide to interferometry, in ”Particle Production in Highly Excited Matter”, eds. H.H.Gutbrod and J.Rafelski, Plenum Press, New York 1993, p. 435. [14] G.A. Kozlov, O.V. Utyuzh and G. Wilk, Phys. Rev. C68 (2003) 024901 and Ukr. J. Phys. 48 (2003) 1313. [15] K. Zalewski, Lecture Notes in Physics 539(2000) 291. [16] P.Abreu et al. (DELPHI Collab.), Phys. Lett. B286 (1992) 201 and Phys. Lett. B247 (1990) 137. [17] J. Finkelstein, Phys. Rev. D37 (1988) 2446; Dingwei Huang, Phys. Rev. D58 (1998) 017501. [18] R.Kutner and M.Regulski, Comp. Phys. Com. 121122 (1999) 586; cf. also: R.Kutner, K.W.Kehr, W.Renz and R.Przenioslo , J.Phys. A28 (1995) 923. [19] J.T. Mitchell, J. Phys. G30 (2004) S819; H. Appelh¨ auser, ibidem, S935.
FRONTIERS OF HIGH ENERGY COSMIC RAYS
´ MARIO PIMENTA a b
a,b
LIP, Av. Elias Garcia, 1412, 1000149 Lisboa, Portugal IST, Av. Rovisco Pais,1049001 Lisboa, Portugal
Abstract Ultra high energy cosmic rays are a mystery and a challenge. Its theoretical and experimental status, as well as the expected performance of the future generation of experiments, are brieﬂy reviewed.
Cosmic rays were always a frontier domain. In this short lecture I have no time to give the due credit to the many people involved and to the many breakthroughs of the already 100 years of history. I would nevertheless refer two events: the discovery by Victor Hess in 1912, with pioneer balloon ﬂights, of a strange radiation coming from the space and not correlated with the sun [1]; the discovery of Ultra High Energy Cosmic Rays (UHECR), with an energy greater than 1019 eV, by John Linsley in 1962 at Volcano Ranch [2]. The energy spectrum of cosmic rays extends over more than 11 decades of energy, following an almost perfect power law (see ﬁgure 1) [3]. The ﬂuxes are quite high at low energies (up to 1 particle/m2 /s for energies of the order of the GeV), but extremely small at the highest energies (1 particle/km2 /century at 1020 eV). Under a closer inspection, changes in the slope can be observed around 1015 eV (the knee) and 1018 eV (the ankle). There is a great deal of interest these two regions. This review will however be dedicated to the far end of the spectrum (ﬁgure 2), where just a few events have been observed, mainly in two experiments: AGASA [4] and HiRes [5]. While AGASA seems to be compatible with no slope change in the spectrum at these extreme energies, HiRes seems to suggest a dramatic drop. The statistical signiﬁcance of the diﬀerence between the two experiments is however slightly above 2σ [6]!
255 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 255–261. © 2006 Springer. Printed in the Netherlands.
256
Frontiers of High Energy Cosmic Rays
Figure 1. The cosmic ray spectrum [23].
There is however a consensus: UHECR do exist and have been observed. The most energetic event ever registered was collected by the Fly’s Eye experiment in 1991 and has an estimated energy of 3.2 1020 eV [7]. The geographical distribution of the highest energy data (above 4 1019 eV) collected by AGASA is isotropic at large scales but shows some clustering at smaller scales: one triplet and six doublets with ∆(Θ) = 2.50 [8]. There are many papers discussing possible acceleration mechanisms able to produce particles with such energies. The main idea is to have somewhere a shock wave, and particles undergoing successive crossings of the shock front. It is the so
´rio Pimenta Ma
257
called ﬁrstorder Fermi mechanism [9]. The nonrelativistic case is well understood, but the relativistic case is much more diﬃcult. The regions where such events may occur can be characterised by two variables: the size of the region and the intensity of the existing magnetic ﬁelds. It is possible to have conﬁnement either in small regions, if there is an intense magnetic ﬁeld, or, if the magnetic ﬁeld is weak, in extended regions. Such a feature is usually summarised in the so called Hillas plot (ﬁgure 3) [10]. There are several possible sources for 1020 eV UHECR but at higher energies it is much more diﬃcult to ﬁnd good candidates. Furthermore, energy losses due to Bremsstrahlung have to be taken into account whenever the accelerations are very large [11]. In any case, the observed UHECR would have to propagate through very large distances before reaching the Earth. But there is no empty space. The space is, as it is well known, ﬁlled with the Cosmic Microwave Background (CMB). The interaction of the UHECR with the CMB γs can be quite dramatic if the centreofmass energy is above the ﬁrst inelastic channel threshold. The net result is that UHECR with energies at the source well above 1020 eV would have dropped to about 1020 eV by the time they reach Earth. This is the famous GZK eﬀect [12]. GZK is not however a real cutoﬀ. The threshold energy depends on many parameters (the type of the sources and its distribution in space, the existing magnetic ﬁelds, the particle type, . . . ). But if AGASA is right how to avoid the GZK? One possible solution is to imagine that UHECR are particles with small electromagnetic couplings but able to originate normal air showers. Possible candidates may be neutrinos. In the Z burst model [13], very energetic neutrinos collide with the cosmological neutrino background producing energetic Z 0 bosons near our galaxy. It is an elegant model, but it implies the existence of intense and monochromatic neutrino beams with an energy around 4 1021 eV! There are many other proposed solutions [14] such as: enhanced neutrinonucleon crosssections, the existence of “new” particles (axions, glueballinos, . . . ) or even the violation of Lorentz symmetry which could lead to a change on the relevant thresholds. However in this kind of models (the bottomup scenarios), UHECR would be produced as secondaries of very energetic “ordinary” particles (E > 1022 eV). There is another way out (the topdown scenarios): UHECR may be decay products of superheavy particles, which would be produced continuously by cosmological “objects” left over after the phase transitions that the Universe has suﬀered, or even produced directly in the early Universe. The challenge is to increase the statistics of observed UHECR. If this is achieved, models can be disentangled looking at the energy spectrum power law, the UHECR composition, the geographical distribution, the possible counterparts in other wavelengths  all a program! Another major challenge in cosmic ray frontiers is the observation of high energy cosmic neutrinos. In fact astrophysical neutrino beams should exist [15]! A safe prediction is the existence of the cosmological neutrinos produced in the in
258
Frontiers of High Energy Cosmic Rays
Figure 2. The far end of the cosmic ray sp ectrum [24].
Figure 3. The Hillas plot [10].
´rio Pimenta Ma
259
teraction of the primordial UHECR with the γCM B or with clouds of intergalactic gas. These interactions produce collimated pions and kaons, which will originate energetic neutrino beams. But there are many other models: from the Z burst mechanism, to the possible Zevatron sources (AGNs, GRBs, supernovae, ...) and to the topdown scenarios discussed just above (see ﬁgure 4). These neutrinos can propagate to Earth almost unscattered and so their detection would open an important channel to observe the Universe. However, so far, there is no experimental measurement above 106 GeV. The existing limits, as well as the expected sensitivities of future experiments, are shown in ﬁgure 5. UHECR are the most energetic beams ever accessible. The centreofmass energy of a 1019 −1020 eV UHECR collision in the atmosphere is around 100400 TeV, which is well above the present and future man made accelerators. But UHECR ﬂuxes are small, the accessible kinematic region is very narrow (very forward region) and the detection capability is much poorer than in conventional Particle Physics experiments. The solution for these low ﬂuxes is to have huge detection areas. The surface of the Pierre Auger observatory is around 3000 Km2 [16], while the planned space based experiments like EUSO [17] will cover a surface of around 200000 Km2 . The limit is the Earth surface, which is still a factor 1000 larger. Nevertheless, important particle physics results were already obtained or are expected in the near future. For instances: pp crosssections measurements were extended to higher centreofmass energies (ﬁgure 6) [18]; heavy quark production is very abundant at these energies and may represent an important channel for QCD studies; new particles (Higgs, SUSY, excited fermions, leptoquarks...) with masses above the TeV can be produced; Lorentz symmetry can be tested at very high boost factors; the existence of extradimensions may imply important enhancements in the neutrinoair crosssection [19]. The search for new physics is, therefore, a relevant issue in the next generation of UHECR experiments and an eﬀort for ﬁnding new signatures must be pursued. An example of such possible signatures is the socalled doublebang signature. An example would be the observation, in an experiment like EUSO, of the production and the instantaneous decay of a microscopic black hole (ﬁrst bang), followed, at a measurable distance, by the decay of an energetic tau lepton (second bang) [20]. In the next decade a spectacular increase in the number of observed UHECR is expected (one to two orders of magnitude). Auger is right around the corner. It combines the technique of sampling the shower particles reaching the Earth surface with the technique of detecting the ﬂuorescence light produced by the excitation of Nitrogen molecules by the low energy electrons of the shower. At present, 25% of the 1600 sampling detectors (water tanks) and two of the three sites for ﬂuorescence light detection are already fully working! EUSO, an ESA mission planned for the ﬁrst years of the next decade, is basically a telescope looking downwards to the Earth in order to collect a fraction of the ﬂuorescence light produced in the interaction of the UHECR with the atmosphere. And there are already plans to go further, OWL [21] is a NASA mission under study which will use two satellites, reaching a ﬁeld of view about 5 times larger than EUSO.
260
Frontiers of High Energy Cosmic Rays
Figure 4. Expected neutrino ﬂuxes [15].
Figure 5. Sensitivities for neutrino detection[15].
Figure 6. The PP crosssection [18].
261
´rio Pimenta Ma
Finally, there are several experiments on the way to the detect the coherent radio emission by very energetic showers [22]. New frontiers in Particle Physics, Astrophysics and Cosmology are right ahead of us!
Acknowledgments I would like to thank Catarina Esp´ırito Santo and Pedro Abreu for reading the manuscript and help in preparing these proceedings, and Alessandro De Angelis for the nice atmosphere during all the conference. References [1] //nobelprize.org/physics/laureates/1936/hessbio.html [2] J.Linsley, Phys.Rev.Lett., 10,(1963)146. [3] For a general reference see for example: A.Watson and M.Nagano, Rev.Mod.Phys.,72,689 (2000). [4] M.Takeda et al., Astro.Part.Phys.,19,(2003)499 [5] D.Bird et al., Astrophys.J.,511,(1999),739 [6] D.De Marco,P.Blasi,A.V.Olinto, Astropart.Phys.20(2003)53 [7] http://www.cosmicray.org/reading/ﬂyseye.html [8] N.Hayashida et al., astroph/0008102 (2000) [9] J.Niemiec, astroph/0407531 [10] A.M.Hillas, Ann. Rev. Astron. Astrophys. 22 (1984) 425. [11] M.V.Medvedv, Physical Review E 67, 045401(R) (2003) [12] K.Greisen. Phys.Rev.Lett., 16, (1996),748; G.T.Zatsepin and V.A.Kuzmin, Sov.Phys.JETP Lett. (engl.transl.),4,(1996),78 [13] T.J.Weiler, Phys.Rev.Lett., 40 (1982) 234; D.Fargion et al.,Astrophys.J.517(1999)725 [14] G.Sigl, astroph/0210049 [15] D.V.Semikoz, Theoretical Predictions of UltraHigh Energy Neutrino Fluxes 28th ICRC Tsukuba, 1439 (2003) [16] http://www.auger.org/ [17] http://www.eusomission.org/cgibin/pub.cgi?href=welcome.html [18] M.M.Bock et al., astroph/0004232 (2000) [19] M.Cavagla, Int.J.Mod.Phys. A18,1843 (2003) [20] V.Cardoso,hepph/0405056 [21] http://owl.gsfc.nasa.gov/ [22] H.Falcke, astroph/0409229 [23] from S.Swordy [24] T.AbuZayyad et al., astroph/0208243
OUTLOOKS ON GAMMA RAY ASTROPHYSICS
RICCARDO GIANNITRAPANI a
a
Dipartimento di Fisica dell’Universi` a di Udine and INFN, Italy
Abstract Gamma Ray Astrophysics, extending from MeV to PeV, covers a wide range of phenomena in the domain of both particle physics and astrophysics, providing one of the most active ﬁeld of modern astroparticle physics. Moreover, the study of gamma rays from the universe can unveil some profound implications for the foundation of physics itself (from Quantum Gravity to Cosmology) and provide a challenging domain for technology and computing too, bringing together researchers of diﬀerent domains in an growing interesting crucible of ideas and experiences.
263 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 263–263. © 2006 Springer. Printed in the Netherlands.
NIELSEN IDENTITY, WILSON LINE AND CONSTRAINED EFFECTIVE ACTION: THE HIGH TEMPERATURE STANDARD MODEL
RAFFAELE BUFFA a
a
Dipartimento di Fisica, Universit` a di Udine, via delle Scienze 208, 33100 Udine, Italy
Abstract Using Nielsen Identities we study the high temperature Electroweak Standard Model. We show the gauge ﬁxing parameter independence at the ﬁrst non trivial order of the Fukuda eﬀective action constrained over the phase of the Polyakov loop.
1
Introduction
Nielsen identities originate from enlarged BRST symmetry [1]. They provide a good framework to study problems related to gauge dependence. Moreover they don’t depend on the topology of space time. For this reason they are suitable for studying ﬁnite temperature ﬁeld theory. Thermal gluodynamics break spontaneously the centergroup symmetry developing a A0 condensation, primarily due to the breaking of the Lorentz invariance. The lack of gauge invariance for the eﬀective potential at 2 loops order and for the eﬀective action for a slowly varying background at 1 loop is well known [2],[3]. Moreover the centergroup symmetry is not explicit. This problem has been solved by Belyev [2] by the so called renormalization ! of the Polyakov loop, the trace of the thermal Wilson line: 1/T Ω(x) = tr P exp i 0 dτ A0 (x, τ ) , where P stands for the path ordering. It is clearly a gauge invariant operator. Adding dynamical fermions and scalars in the fundamental representation lifts the centergroup degeneracy, the Z(N) nontrivial
285 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 285–289. © 2006 Springer. Printed in the Netherlands.
286
Nielsen identity, wilson line and action
vacua become metastable with interesting phenomenological consequences [4]. In this presentation we show the link between Nielsen identity and the Fukuda effective action [5] constrained over the phase of the Polyakov loop. As a concrete example we treat the Electroweak Standard Model at temperatures well above the phase transition where we can ignore the bare masses of fermions and the electroweak symmetry is restored. We show for the Standard Model case how the gauge invariance of the oﬀ shell Fukuda eﬀective action comes from the perturbative arrangement of new terms coming from the non linearity of the constraint. Moreover we show a very convenient way to extract the gauge dependence at 1 loop for a slowing varying background without the occurrence of the steepest descendant. We work with the imaginary time formalism and zeta function regularization.
2
Nielsen identity and 1PI eﬀective action
In this section we are interested in the gauge dependence of the perturbative 1PI eﬀective action. Within the context of high temperature SM A0 and B0 condense [4]. Thus we split the ﬁelds into background plus small quantum ﬂuctuations:Aµ = ˜ + gAq , Bµ = δµ0˜b + g Bq where Aµ and Bµ are respectively the SU(2) and δµ0 a U(1) gauge ﬁelds. Other ﬁelds have only the quantum part. For the sake of simplicity we choose a ˜ a time independent diagonal matrix in the Lie algebra of SU(2). We need to ﬁx the gauge, we have chosen to use background gauge ﬁxing: 2 2 2 2 Sg.f. = g2ξ tr Dµa˜ Aµq + gξ ∂µ Bqµ where Dµa˜ = ∂µ + δµ0 [˜ a, .] In this context the Nielsen identity takes the form [1]: δΓO˜b (x) δΓOa˜ (x) δΓ δΓ + = −ξ∂ξ Γ (1) δA0 (x ) δkA0 (x ) δB0 (x ) δkB0 (x ) x x
where Γ is the 1PI eﬀective action, ΓO is the 1PI eﬀective action with a one O ˜ insertion (Oa˜ = 21 η¯A Dµa˜ Aµq , Ob = 21 η¯B ∂µ Bqµ ), kA0 is the source of sAq0 (which is the BRST transformation of Aq0 ) and kB0 is the source of the BRST transformation of Bq0 . We consider ˜b constant in time, so the second term on the left hand side is zero. By the use of identity (1) we see that the gauge dependence at the order of 1 loop is factorized in a classic term plus a 1 loop term. For a constant background ﬁeld the classical action does not depend on a ˜. On the other side < Oa˜ sAq0 > at the classical level is zero. Thus the 1 loop eﬀective action is independent from gauge ﬁxing parameter for constant background. At 1 loop order the eﬀective potential depends on the background and the derivative is no longer zero. Again using eq.(1) we obtain:
ξ∂ξ V 2loop = g 2
δV 1loop < η¯Dµa˜ Aµq D0 η >1loop δA0
(2)
Other terms are zero following the same arguments followed in the 1 loop eﬀective potential gauge ﬁxing independence.
287
R. Buffa
For a slowly varying a ˜ we consider the spatial derivative of the background as a new scale. The eﬀective action already has a non trivial gauge dependence at ˜. For a slowing 1 loop order in the coupling constant and at second order in ∂i a varying background the classical action has a kinetic term diﬀerent from zero of the order ∂i2 a ˜ so that we can use the powerful technique of Nielsen identity to extract ∂i2 a ˜ and treat the rest of the calculation for a constant background. The Nielsen identity takes the form: ξ∂ξ Γ1loop =
3
δScl < η¯Dµa˜ Aµq D0 η >1loop δ˜ a
(3)
Constrained eﬀective action
The constrained eﬀective action ﬁrst introduced by Fukuda et al. [5] has been applied to ﬁnite temperature SU(N) gluodynamics with Nf fermions for a constant background ﬁeld, to a pure gauge Z(N) interface and to a pure gauge Z(N) dimensionally reduced interface [3]. In this section we consider the Polyakov loop to be constant and slowly varying in the z direction and we show how the nonlinearity of the constraint takes care of the gauge dependence of the 1PI eﬀective action. The eﬀective action constrained over the phases of the Polyakov loop is deﬁned in the following way: exp − V Γ(a, b) = DA DB Dψ¯ Dψ Dη¯ Dη DΦ δ (a − p¯(˜ a + gAq0 )) 1 ˜ δ b − p¯(b + g Bq0 ) exp − 2 (Scl + Sg.f. + Sghosts ) (4) g where ψ are left and right fermions, η are the ghosts, Φ is the Higgs, p(.) denotes the phase of the Polyakov loop and ¯. means the spatial average for a constant Polyakov loop and the transverse spatial average for a Polyakov loop varying in the z direction. The second constraint is linear and does not introduce new terms to the eﬀective action. Its treatment is standard [5], in the following we ignore it. ! As in [3] we Fourier transform the δ function: δ(t − t˜) = DΛ exp i gΛ2 (t − t˜) .
We insert this representation of the constraint in eq.(4) and we expand the new ﬁeld into background plus quantum ﬂuctuation: Λ = λ + gΛq . In the constrained eﬀective action formulation the saddle point solution at the thermodynamical limit takes care of the Legendre transformation. We obtain 3 saddle point equations: a−a ˜=0 (1)
iλ¯ p (˜ a)
(1) − gScl (˜ a) (1) Scl (˜ a, ˜b)
=0 =0
where p¯(1) (˜ a) = −g is the development of p¯(.) at ﬁrst order with respect to Aq0 , (1) a) is the linear development of the action with respect to the same quantum Scl (˜
Nielsen identity, wilson line and action
288
(1) degrees of freedom that are linked to p¯(1) (˜ a), Scl (˜ a, ˜b) is what remains of the linear development of the action with respect to quantum ﬂuctuations. The classical action for a constant Polyakov loop does not depend on a ˜ and the saddle point equations impose λ = 0. At one loop order the constraint has no eﬀect other than the Legendre transformation. At 2 loops there is a new vertex coming from the dynamical propagation of Λ1q . It couples the derivative of the 1 loop eﬀective potential to the renormalization of the Polyakov loop (see [3] for more details): 2loops δVconstraint = g2
δV 1loop p¯(2) (˜ a)∗ < Aq0 Aq0 > δ˜ a (p(1) (˜ a))2
(5)
a) ∗ Aq0 Aq0 is the development of p¯(˜ a + gAq0 ) at second order in Aq0 . where p¯(2) (˜ For a background varying in the z direction the classical action is no longer zero g δScl and the saddle point equations give: λ = −i p(1) δ˜ a . At 1 loop order λ couples to
a)∗ < Aq0 Aq0 > and we arrive at the following expression for the constraint p(2) (˜ contribution to the eﬀective action: δΓ1loop constraint = −
δScl p¯(2) (˜ a)∗ < Aq0 Aq0 > δ˜ a g p(1) (˜ a)
(6)
What remains to be done is the explicit calculation of p¯(2) (˜ a)∗ < Aq0 Aq0 > and < η¯Dµa˜ Aµq D0 η >. The ﬁrst gives (3 − ξ)(˜ a/2πT − 1/2) and second ξ(˜ a/2πT − 1/2). We have shown the gauge invariance of the constrained eﬀective action.
4
Conclusions
We have shown the connection between the Nielsen identity and the eﬀective action constrained over the phase of the Polyakov loop. We have shown the gauge invariance of the oﬀshell constrained eﬀective action in the high temperature Standard Model for both constant and slowly varying background ﬁelds. The constrained path integral has also found application in the study of the Casimir eﬀect [6]. We believe that the work undertaken in this presentation can be applied in this context. 4.1 ACKNOWLEDGMENTS I would like to thank C.P.Korthals Altes and S.Bronoﬀ for useful discussions about the constrained eﬀective action. References [1] N.K. Nielsen, Nucl.Phys. B101 (1975), 173; I.J.R. Aitchison, C.M. Fraser, Annals of Phys. 156, (1984), 1; O.M. Del Cima, D.H.T. Franco, O. Piguet, Nucl.Phys. B551 (1999), 813
R. Buffa
289
[2] V.M. Belyaev, Phys.Lett. B254 (1991), 153 [3] C.P. Korthals Altes, Nucl.Phys. 420 (1994), 637; T. Bhattacharya, A. Gocksch, C.P. Korthals Altes, R.D. Pisarski, Nucl.Phys. B383 (1992), 497; S. Bronoﬀ, R. Buﬀa, C.P. Korthals Altes, hepph/9801333 [4] C.P. Korthals Altes, K.M. Lee, R.D. Pisarski, Phys.Rev.Lett. 73 (1994), 1754 [5] R. Fukuda, E. Kyriakopoulos, Nucl.Phys. B85 (1975), 354 [6] M. Bordag, J. Lindig, Phys.Rev. D58 (1998), 045003
WILL ANTIHYDROGEN LIGHT SHINE?
ALESSANDRO VARIOLA a
a
Swansea University of Wales
Abstract In 2002 the ATHENA experiment announced the ﬁrst production of antihydrogen atoms at cryogenic temperature. This achievement is an important step towards antimatter atoms ﬁne spectroscopy. Antihydrogen light is expected to make it possible to carry out very accurate CPT invariance measurements as well as, for the ﬁrst time ever, tests on the gravitational interaction between matter and antimatter with important consequences for cosmology. Over the last two years signiﬁcant results have been obtained in the investigation of antihydrogen production mechanisms. They are all the more important given the strong limitations imposed by the neutral atom trapping techniques required for ﬁne spectroscopy measurements. This talk will focus on the main scientiﬁc motivations in the study of cold antihydrogen, the results obtained so far and the mediumterm plans for the ﬁrst measurements of the interaction between antiatoms and photons.
265 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 265–265. © 2006 Springer. Printed in the Netherlands.
PHYSICS POTENTIAL OF THE ATLAS EXPERIMENT
M. COBAL a , FOR THE ATLAS COLLABORATION a
Dipartimento di Fisica Universit´ a di Udine, via delle Scienze 208, 33100 Udine, Italy
Abstract The ATLAS experiment at the CERN Large Hadron Collider (LHC) will present the most extraordinary challenges that particle physics has ever faced. By studying the collisions of high intensity proton beams at a centerofmass energy of 14 TeV, ATLAS will explore for the ﬁrst time the TeV scale region. This paper reviews the experiment’s potential for discovery of physics beyond the Standard Model (SM) and for the investigation of the nature of electroweak symmetry breaking.
1
Introduction
The SM has been veriﬁed with an impressive accuracy, i.e. to 0.1% or better in most cases [1]. However, there are numerous indications that this is not the ultimate theory of elementary particles. Among them, the evidence for atmospheric [2] and solar [3] neutrino oscillations, and the incapacity of the SM to give satisfactory answers to fundamental questions [4] such as the mass and ﬂavour problems, baryogenesis and the matterantimatter asymmetry in the universe, the origin of dark matter, the size of the cosmological constant, and the uniﬁcation of gravity with the other interactions. One of the most urgent issues is to explain the origin of the particle masses. In the SM particles acquire a mass through their coupling with the Higgs boson. The Higgs mass is not speciﬁed by the theory, which provides only an upper bound of ∼1 TeV. Direct searches performed at LEP have set a lower limit of
267 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 267–272. © 2006 Springer. Printed in the Netherlands.
268
Physics Potential of the ATLAS
mH > 114.4 GeV [5], and a ﬁt of the SM to the data collected by various machines (LEP, Tevatron, SLC) gives a 95% C.L. upper bound on mH of about 250 GeV [1]: experimental data therefore favour a light Higgs boson. The introduction of the Higgs mechanism brings however some problems: the Higgs mass increases with the energy scale Λ up to which the SM is valid and therefore it requires a large amount of “ﬁne tuning” to be stabilized at the electroweak scale. In addition, the generation of fermion masses spoils the simplicity of the SM with a proliferation of unknown parameters. There are several candidate scenarios for physics beyond the Standard Model, including Supersymmetry (SUSY), Technicolour and theories with Extradimensions. All of them predict new particles in the TeV region, as needed to stabilize the Higgs mass. ATLAS [6] is a general purpose detector which will operate at the LHC at centreofmassenergies of 14 TeV from 2007 onwards, in search of the Higgs and of new and even unexpected physics. The design of the detector will allow measurement and identiﬁcation of leptons, photons up to rapidities η < 2.5 and jets up to rapidities η < 5, thus allowing the exploitation of a wide range of physics signatures. Given the highenergy and highluminosity which will characterize the LHC, the rate of events produced in the collisions will be huge and LHC will be a factory of all particles with masses up to a few TeV which have reasonable couplings to SM particles. For example, when the machine luminosity will be about a factor of ten lower than the design luminosity L = 1034 cm−2 s−1 , million of events should be collected for many SM channels over only one year of data taking (i.e.: 107 tt¯, 108 W → eν) and 130 H → γγ with mH = 120 GeV). This will allow not only to explore new territories of discovery, but also to perform many precision measurements (like the determination of the top and W mass, or the study of the Triple Gauge Couplings) with high accuracy. In the following, we will concentrate on the Higgs and SUSY searches.
2
Searches for the SM Higgs boson
The SM predicts that the Higgs couples to fermions and bosons with strength proportional to their masses. Therefore, for mH < 120 GeV the Higgs boson should decay mainly into bb, whereas for larger masses decays into W pairs and Z pairs should dominate. Gluongluon fusion through a topquark loop is the dominant production channel for all masses. Vector boson (W W , ZZ) fusion contributes about 20% of the crosssection for mH ∼ 120 GeV and becomes more and more important with increasing mass. This process leads to the very distinctive topology of a Higgs boson accompanied by two jets emitted in the forward regions of the detector and very little activity in the central region (since no colour lines are exchanged between the two interacting bosons). Higgs production with a tt pair or a W/Z boson has a smaller crosssection; however it allows detection of the purely hadronic H → bb decay mode, because the reconstruction of the particles produced in association with the Higgs provides additional handles against the large QCD backgrounds. A SM Higgs boson can be discovered over
M. Cobal
269
Figure 1. a) The expected SM Higgs signal signiﬁcance in ATLAS in the lowmass region for 30 fb−1 . The total signiﬁcance (full line) and the contributions of the individual channels (symbols) are shown. b) The regions of the (constrained) MSSM plane {mA tanβ} where the various Higgs bosons can be discovered at the LHC through their decays into SM particles. The region below the thick curve has been excluded by LEP.
the full allowed mass range (114.4 GeV to 1 TeV) with only 10 fb−1 of integrated luminosity, which corresponds (in principle) to only one year of LHC operation at low luminosity. The Higgs boson discovery should be easier and faster for masses above 200 GeV, thanks to the “goldplated” H → ZZ → 4 (where = e, µ) channel, which is essentially backgroundfree. In contrast, the most diﬃcult region is the lowmass region close to the LEP limit and at the overlap with the Tevatron reach. The total signiﬁcance of about 4σ per experiment (4+2.2 −1.3 σ including the expected systematic uncertainties) is more or less equally shared among three channels (see Fig. 1 a): H → γγ, ttH production with H → bb, and Higgs production in vectorboson fusion followed by H → τ τ . If a Higgs boson were to be discovered at the LHC, ATLAS should be able to perform several precise measurements of its properties. For example, with the ultimate integrated luminosity of 300 fb−1 the Higgs mass should be measured with the remarkable experimental precision of 0.1% over the mass region up to ∼ 400 GeV. This direct measurement can then be compared to the indirect determination of mH obtained from the measurements of the W and top masses. The expected precisions at the LHC are ∼15 MeV on mW and ∼1 GeV on mtop [8], leading to a 25% (indirect) accuracy on the mass of a light Higgs boson.
3
Searches for Supersymmetry and beyond
SUSY [9], a symmetry relating fermions and bosons, that is matter ﬁelds and force ﬁelds, is one of the relevant scenario today for physics beyond the SM [4]. It does
270
Physics Potential of the ATLAS
not contradict the precise, and therefore very constraining, electroweak data, it predicts a light Higgs boson, as favoured by these data, it allows uniﬁcation of the gauge couplings at the Grand Uniﬁcation scale and a natural incorporation of gravity, it is an essential element of string theories, it provides a candidate particle for the universe cold dark matter. Furthermore, it is able to stabilize the Higgs boson mass, through radiative corrections, provided that the SUSY particles (sparticles) have masses at the TeV scale or below. In SUSY, for each SM particle p there exists a supersymmetric partner p˜ with identical quantum numbers except the spin which diﬀers by half a unit. However, up to now there are no experimental evidences for SUSY. Direct searches for sparticles at LEP and Tevatron have been unsuccessful, and have set mass lower bounds in the range 90300 GeV depending on the sparticle type. Important phenomenological consequences arise from the fact that the theory contains a multiplicative quantum number, called Rparity, which takes opposite values for SM and SUSY particles. The conservation of Rparity, motivated by cosmological arguments, is assumed here. This implies that sparticles are produced in pairs and that the Lightest Superymmetric particle (LSP), to which all sparticles eventually decay, must be stable. In most models the LSP is the lightest neutralino χ01 , which is a stable, massive and weaklyinteracting particle, and therefore an excellent candidate for the universe cold dark matter. At the LHC, the dominant SUSY process is expected to be the production of pairs of squarks or gluinos, because these are stronglyinteracting particles with QCDtype crosssections. For instance, a sample of about 104 q˜q˜, g˜g˜ and q˜g˜ events should be produced over only one year of data taking at L = 1033 cm−2 s−1 if squarks and gluinos have masses of ∼1 TeV. Because these sparticles weigh at least 200300 GeV, given the present Tevatron limits, they are expected to decay through long chains with several intermediate steps, and hence should give rise to very busy ﬁnal states containing in general several jets, leptons and missing transverse energy. Squark and gluino masses of 1 TeV are accessible after only one month of data taking at L = 1033 cm−2 s−1 , once the backgrounds (e.g. tt production, mismeasured QCD multijet events) and the detector performance (in particular tails in the calorimeter response to jets) will have been well understood. The ultimate mass reach is up to ∼3 TeV for squarks and gluinos. As a consequence, SUSY discovery at the LHC could be relatively easy and fast, and if nothing is found at the LHC, TeVscale Supersymmetry will most likely be ruled out, because of the arguments related to stabilizing the Higgs mass mentioned above. On the other hand if SUSY is there, ATLAS should go beyond the mere discovery phase, being able to perform several precise measurements of the sparticle masses, and thus determining the fundamental parameters of the theory with a precision of ∼10% or better in many cases, at least in minimal models like mSUGRA [10]. The correct identiﬁcation of the underlying theory and the measurements of its fundamental parameters will not be easy for general models with many unknown parameters. It is however hoped that, by exploiting the expected richness of the data with a lot of diﬀerent measurements (masses, crosssections, decay modes, etc.) and observations (e.g. excess of events with bquarks or taus), and with some guidance from theory, it will eventually be possible to narrow the a priori large
M. Cobal
271
spectrum of models and pin down the correct framework. A rich phenomenology is also expected from the SUSY Higgs sector, which consists of ﬁve bosons, three neutral (h, H, A) and two charged (H ± ). The mass of the lightest one, h, is predicted to be below 135 GeV, whereas the others are expected to be heavier and essentially massdegenerate over most of the parameter space. The Higgs sector of the Minimal Supersimmetric Standard Model (MSSM) can be described in terms of the mass of the A boson mA and of the parameter tanβ (the ratio of the vacuum expectation values of the two Higgs doublets which give rise to the ﬁve physical states). Figure 1 b shows the regions of this parameter space where the various Higgs bosons can be discovered at the LHC through their decays into SM particles. It can be seen that over a large fraction of the parameter space two or more Higgs bosons should be observed. The only exception is the region at large mA and moderate tanβ, where only h can be discovered at the LHC, unless the heavier Higgs bosons have observable decays into SUSY particles. The LHC may therefore miss part of the SUSY Higgs spectrum. Many other examples of physics beyond the SM have been studied by ATLAS: theories with Extradimensions, Little Higgs models, Technicolour, Compositeness, etc, to exploit all the relevant expected topology (in particular at the trigger level) and to address as many signatures as possible. The LHC discovery potential for some scenarios beyond the SM is illustrated in Table 1. In most cases the direct reach extends well beyond the 1 TeV “threshold”. More details can be found in Refs. [6, 11].
Table 1. LHC discovery potential for some scenarios beyond the Standard Model.
4
Scenario/channel
Reach
Z → W → ν Leptoquarks Compositeness Excited quarks Extradimensions Monopoles
m∼5 TeV m∼6 TeV m∼1.5 TeV Compositeness scale Λ ∼ 40 TeV m∼6.5 TeV Gravity scale M ∼ 9 TeV for 2 extradimensions m∼20 TeV
Conclusions
In about three years from now the LHC will start operation, and CERN and experimental particle physics will enter a new epoch. Given the compelling motivations for new physics at the TeV scale, one can anticipate a profusion of exciting results from a machine able to explore this scale in detail, with a direct discovery potential up to particle masses of ∼56 TeV. As a consequence, the LHC should
272
Physics Potential of the ATLAS
provide deﬁnitive answers about the SM Higgs mechanism, Supersymmetry, and several other TeVscale predictions that have resisted experimental veriﬁcation for decades. References [1] G. Altarelli and M. Grunewald, Precision electroweak studies at LEP, this issue. [2] Y. Fukuda et al., Phys. Rev. Lett. B81 (1998) 1562. [3] Q.R. et al., Phys. Rev. Lett. 87 (2001) 071301. [4] J. Ellis and M. Jacob, Phenomenology of today and tomorrow, this issue. [5] D. Treille, Searches for new particles at LEP, this issue. [6] ATLAS Collaboration, Detector and Physics Performance Technical Design Report, CERN/LHCC/9915, 1999. [7] Large Hadron Collider in the LEP tunnel, Proceedings of the ECFACERN Workshop, ECFA 84/85, CERN 8410. [8] Proceedings of the Workshop on Standard Model Physics (and more) at the LHC, edited by G. Altarelli and M. Mangano, CERN 2000004. [9] For a phenomenological review see for instance P. Fayet and S. Ferrara, Phys. Rep. C32 (1977) 249; H.P. Nilles, Phys. Rep. C110 (1984) 1; H.E. Haber and G.L. Kane, Phys. Rep. C117 (1985) 75. [10] A.H. Chamseddine, R. Arnowitt and P. Nath, Phys. Rev. Lett. 49 (1982) 970. [11] B.C. Allanach et al., Les Houches Physics at TeV Colliders 2003: Beyond the Standard Model Working Group Summary Report, hepph/0402295.
MULTIPHOTON APPROACH ON PAIR PRODUCTION UNDER THE LIGHT OF RECENT EXPERIMENTAL AND THEORETICAL INVESTIGATIONS
a
a
C. KABERIDIS, I. TSOHANTJIS AND S. MOUSTAIZIS a
a
Technical University of Crete, Institute of Matter Structure and Laser Physics, Chania 73100, Crete, Greece
Abstract We explore the range of applicability of the imaginary time approach and the two level on resonance approach of e+ e− pair production from vacuum both theoretically and on the E144 experiment. We show that the resonance approximation is more eﬃcient by many orders of magnitude.
1
Introduction
One of the most intriguing nonlinear phenomenon in QED is particleantiparticle pair production from vacuum in the presence of strong elm ﬁelds[1]. Schwinger [2] implementing the proper time method obtained conditions under which pair production is possible: F = 41 Fµν F µν , G = 41 Fµν F˜ µν , where Fµν and F˜µν are the elm ﬁeld tensor and its dual respectively, must be such that such that neither F =0 , G =0 nor F >0 , G =0 . For the case of a uniform, constant electric ﬁeld m2 πn he obtained a probability density w ∼ n12 e− eE . Brezin and Itzykson [3] examined the case of pair creation in the presence of a pure oscillating electric ﬁeld E by applying a version of WKB approximation and treating the problem in an analogous way as in the ionization of atoms(where the three basic mechanisms multiphoton, tunnelling and over the barrier ionization are present), considering the pairs as bound in vacuum with binding energy 2mc2 . With the critical value 2 18 of electric ﬁeld strength Ec being Ec = mc eλc 1.3 × 10 V /m and under the
279 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 279–283. © 2006 Springer. Printed in the Netherlands.
280
Multiphoton approach on pair production
conditions E Ec , ω mc2 , they obtained a probability per 4Compton 2 e2 E 2 1 + − volume of e e pair creation given by wBI = πc − πm g(γ) , γg (γ) exp eE g(γ)+ 2 21 ! 2 1 1−y where g(γ) = π4 0 1+γ 2 y2 dy . The parameter γ = mcω eE is the equivalent of the Keldysh parameter in the ionization of atoms and the above formula interpolates between the regimes γ 1 (adiabatic nonperturbative tunneling mechanism regime) and γ 1 (multiphoton mechanism regime ) . Popov [4] using the imaginary time method obtained more accurate formulas for pair production taking also in to account interference eﬀects. The condition F n0
γ 1. For γ 1
2n
2 2 π4
Ec
2m4 −5 q (n − n0 ) , and w ≈ 3 (n0 ) 2 π
e 4γ
2n0 ,
(1)
! 2(n−n0 ) t − 1 where q (n − n0 ) = 21 e−2(n−n0 ) 0 e t 2 dt and n0 = 2mc2 /ω is the threshold number of photons. The number of e− e+ pairs created can be obtained by multiplying the above relations by the 4volume λ3 τ = λ4 /c where τ is the pulse duration and λ is the elm wavelength. Specially for the γ 1 regime 3 2
N (τ ) ≈ 2πn0
ω 4γ −2n0 m
e
N ,
(2)
where N = ωτ /2π is the number of oscillations of the electric ﬁeld in a laser pulse. In a recent paper Avetissian et al [5] treated the problem of e+ e− production in a standing wave of oppositely directed laser beams using a two level multiphoton on resonance approximation. Pairs are produced close to antinodes and in spacial dimensions l λ. Transitions can be thought to occur between two energy levels from −E to E by the absorption of n photons and the multiphoton probabilities will have maximum values for resonant transitions 2E = nω. The probability of nphoton e+ e− production process with certain en 2f 2 ergy E, summing over the spin states, obtained as Wn = Ω2n sin2 Ω2n τ , where n 1 mpy p2 cos2 θ 2 E fn = 2py 1 − E 2 nωJn (4ξ Eω ), θ is the angle between the momentum of
produced e− (e+ ) and the amplitude of the electric ﬁeld, the parameter ξ = γ1 , Ωn = ∆2 fn2 + 4n is the ’Rabi frequency’ , τ is the interaction time and ∆n = 2E − nω is the detuning of resonance. With the usual condition Ωn ω, ﬁeld intensities are such that ξ 1 (i.e. γ > 1). The function fn (Rabi frequency on exact resonance) √ and correspondingly the probabilities obtained maximize when n = N = 2 2m/ω. Then the angular distribution of a nphoton probability density on exact reso1 m(n2 ω 2 −4m2 ) 2 cos θ dwn n3 ω 3 n2 ω 2 sin2 θ+4m2 cos2 θ 2 nance is given by do = 64π2 Jn 4ξ where 1 nω 2 2 2 2 2 (n ω −4m ) 2 cos θ
C. Kaberidis et al.
281
+ dwn do = sin θdθdϕ. dw n=n0 do and integrating over the solid angle, the total do = n probability density can be found in [5]. We shall use only dw do at angle θ = 0 and for n = N as this is quite adequate to give an estimate of the power of this approach compared to others.
2
Comparison of the imaginary time and two level resonance approximation
The relation for
dwn do
at angle θ = 0 and for n = N becomes dwN 1 m4 2 θ=0 = √ 2 JN (ξN ) do 2π
(3)
Numerical simulations can be performed for the common Nd laser [5] with ω = 1.7eV , λ = 1.074µm , I = 1.35W/m2 and ξ = 0.9995. As N ∼ 106 , JN (ξN ) ∼ exp(N tanh a−N a) 40 −3 −1 N √ , where a = sech−1 ξ, giving dw s . To do θ=0 = 1.9 × 10 m 2πN ∈tanh a meet the conditions for two level resonance approximations a 4volume V = d2 lτ = (10λ)2 (0.1λ) 10f s can be chosen (i.e maximum laser intensity close to the diﬀraction limit, where the radius of focus is d ∼ 10λ and duration of the pulses τ = 10f s), leading to 108 − 109 number of pairs per laser pulse. On the other hand if we use relation (2) for the same laser characteristics as above we obtain that in the above four volume the number of pairs N (τ ) predicted is approximately 10−216546 per laser shot, which suggests that the multiphoton on resonance approximation is by far more eﬃcient. Moreover requiring the same number of pairs, for ω = 1.7eV , the imaginary time method would require very high E of the order of 1017 V /m while being in the regime γ < 1. The above analysis shows that implementing the resonance approximation, one does not need to use laser such as XFEL [6] or the laser of NIF (National Ignition Facility) [10] in order to observe e+ e− pair creation as present lasers technology is adequate.
3
Application of the imaginary time and the multiphoton resonance approximations on E144 experiment.
The ﬁrst experimental veriﬁcation of e− , e+ pair production took place in SLAC and it is known as E144 experiment [7]. In the ﬁrst stage non linear Compton scattering [9], of highly energetic electrons (maximum 46.6GeV) with n laser photons (527nm, ω = 2.35eV and laser pulses of nominal energy 500mJ ) , e + nω → e + ω , produces backscattered photons ω (2730Gev) which in the second stage interact via the multiphoton Breit Wheeler [8] mechanism with the laser photons, ω + nω → e− e+ ,to produce e− , e+ pairs . The scattering occurs at angle a = 17◦ between the initial electron and the laser beam (and thus for pair creation F < 0 holds). The minimum number of photons required for pair production is given 2 (1+η 2 ) −1 by n0 = ωω2m . The number of positrons measured in (1+β cos a) where η = ξ = γ
282
Multiphoton approach on pair production
21962 laser pulses was 175 ± 13 for a n = 5.1 ± 0.2(statistical)+0.5 −0.8 (systematic) multiphoton order process, in very good agreement with the theory. We shall examine whether the result of E144 experiment can be in agreement with the two methods mentioned above. In E144 experiment ω ∼ 30GeV in the laboratory frame. Changing to the electron’s frame in the beam, both the laser photons and the backscattered ones have the same energy ω = 2.35 × γL eV 46GeV 5 −12 where γL = 0.511M m eV ∼ 10 is the Lorenz factor, λ = 5. 78 × 10 √ and pe−20 riod of the ﬁeld T = 1. 9 × 10 s. Similarly the electric ﬁeld EL = γL 377Ilaser ∼ 3 × 1017 V /m ( with maximum laser intensity at focus Ilaser ∼ 1.35 × 1022 W/m2 ), very close to Ec for pair production. Thus the E144 experiment, in the electrons’ frame, can be viewed as an e+ e− production from vacum in the presence of strong ﬁeld. Applying the imaginary time approach with relativistic invariant γ = 2.7, the probability density wn of (1) (with n0 = 4. 77) has a pick when q(n − n0 ) maximizes and this happens for n 5 , thus verifying that the multiphoton process is a 5th order one . From (1) w5 ∼ 9 × 1048 m−3 s−1 , which for pulse duration τ = 1. 9 × 10−18 s in the 4volume V τ = λ3 τ and for 21962 shots, we obtain ∼ 73 pairs. Moreover from N (τ ) of (2) (where the choice τ ∼ 10−18 corresponds to N ∼ 102 ) we can obtain an estimate for N (τ ) to be 155 to 182 pairs, in very good agreement with E144 experiment . Applying the resonance approximation on the E144 experiment (electron’s √ frame) and using (3) we ﬁrst note that N = 2 ω2m must be taken to be the closest √ integer to the ratio 2 ω2m . For ω = 235keV , N ∼ 6 which suggests that with a ﬁne tuning of ω i.e ﬁne tuning of the electrons’ energy in√the electron beam, we can achieve exact resonance with N = 6. Thus using 2 ω2m = 6, we ﬁnd that the backscattered photons on the electrons frame will have energy 240. 88keV ( 1 λ = 5. 15 × 10−12 and period T = 1. 7 × 10−20 s ). Then with ξ = 2.7 = 0.37, dwn 51 3 1 n = 6 we ﬁnd from (3) do  = 4. 56 × 10 pairs/m s and in the volume 2
θ−0
V τ = (10λ) (0.1λ) 10−18 m3 s1 and for 21962 shots we would have observed on exact resonance, 1. 37 × 105 pairs. Thus had we the opportunity to calibrate the energy of the electrons in the electron beam of SLAC experiment in order to be on exact resonance for a 6th order multiphoton process, we should have observed many orders of magnitude more pairs per laser shot than it was observed. References [1] W. Greiner, B. Muller, J. Rafelski, ‘Quantum Electrodynamics of Strong Fields’, Springer Verlag, Berlin; E. S. Fradkin, D. M. Gitman and Sh. M. Shvartsman, ’Quantum Electrodynamics with unstable vacuum’ SpringerVerlag, Berlin, 1991; [2] J. W. Schwinger, Phys. Rev., 82, 664 (1951); [3] E. Brezin and C. Itzykson, Phys. Rev. D, 2, 1191 (1970); [4] V.S. Popov, JETP Lett. 13 (1971) 185; Sov. Phys. JETP 34 (1972) 709;Sov. Phys. JETP 35 (1972) 659; V.S. Popov and M. S. Marinov, Sov.J. Nucl. Phys.16 (1973) 449; JETP Lett. 18 (1974) 255;Sov. J. Nucl. Phys.,19, (1974) 584;V.S. Popov, Phys. Let. A, 298, 83 (2002); Laser Phys. 10(2000) 1033;
C. Kaberidis et al.
283
[5] H. K. Avetissian et. al., Phys. Rev. E, 66, 016 502 (2002); [6] A. Ringwald, Phys. Let. B, 510, 107 (2001); [7] D.L. Burke et. al., Phys. Rev. Let., 79, 1626 (1997); [8] G. Breit and J.A. Wheeler, Phys. Rev. 46,1087 (1934); [9] A.I. Nikishov and V.I. Ritus, Sov. Phys. JETP, 19, 529 (1964) ; 19, 1191 (1964) ; 20, 757 (1965) ; 25, 1135 (1967);N.B. Narozhny et al., Sov. Phys. JETP 20, 622 (1965) [10] Science of High Energy Lasers : ’From Today to the NIF’, R. Lee, R. Petrasso, R. Falcone
A GLOBAL OPTIMIZATION ALGORITHM FOR FINITE DENSITY QUARK MATTER
RAFFAELE BUFFA a
a
Dipartimento di Fisica, Universit` a di Udine, via delle Scienze 208, 33100 Udine, Italy
Abstract Using a parallel genetic algorithm we study a SU (3)f ×SU (3)c Nambu JonaLasinio model at ﬁnite density and zero temperature within the context of normal and superconducting quark matter. We study the most stable charged conﬁguration in normal quark matter with the use of the Powell constraint techniques and we obtain the phase diagram of superconducting quark matter in the µ, µQ plane for diﬀerent values of the t’Hooft interaction. At the superconducting phase transitions we embed the genetic algorithm in a branch & bound algorithm.
1
Introduction
Stability of compact astrophysical objects and heavy ion collisions are controlled by the behavior of ﬁnite density and low temperature hadronic matter. At these densities QCD must be treated in a nonperturbative way. We choose to carry on our analysis by mean ﬁeld approximation applied to a Nambu JonaLasinio (NJL) model. The mean ingredient in the determination of the ground state of a zero temperature ﬁeld theory is the detection of the global minimum of the eﬀective potential or of the energy. To this aim we adopt a parallel elitist genetic algorithm (for a review about genetic algorithms we send to [1]). We have tried diﬀerent kinds of parallelization and we have found it convenient to adopt the evolution of parallel populations interacting by the crossing of parents of diﬀerent populations. This new crossing is able to give richer chromosome variety to the individuals.
273 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 273–277. © 2006 Springer. Printed in the Netherlands.
274 A global otimization algorithm for finite density quark matter
Figure 1. Normal quark matter; ﬁg.1.a left: the fractions rs = ρs /ρB (solid line) and ru = ρu /ρB = rd = ρd /ρB (dotted line); ﬁg.1.b right: the constituent masses (Mev) Ms (solid line) and Mu = Md (dotted line)
A particular attention must be payed to the proximity of superconducting phase transitions. The genetic algorithm may lose the true minimum of the eﬀective potential. The trivial solution is to increase the number of individuals of the populations and/or the number of generations. We have found it more convenient to embed the genetic algorithm in a simple branch & bound algorithm. The relaxation and the value of the objective function are given by the genetic algorithm applied to the eﬀective potential. Concerning the branching we have divided the variables’ domain in distinct regions corresponding to diﬀerent phases. We have tested the eﬃciency of our approach with respect to results taken from the literature [2], [3] and we have found good agreement except for what concerns ﬁg.5.6 of [2] at zero temperature and for the parameters set I. In the second section we address the problem of ﬁnding the most stable charged solution for SU (3)f × SU (3)c normal quark matter at zero temperature and ﬁnite density. The third section is devoted to the determination of the phase diagram in the µ, µQ plane at diﬀerent values of the t’Hooft interaction for SU (3)f × SU (3)c superconducting quark matter at zero temperature (µ is the chemical potential relative to the total quark number density and µQ is the chemical potential relative to the electric charge density). The two problems are non linear and non convex. In the last section we report the conclusions. The analysis has been carried on on a cluster composed of three nodes.
2
Normal quark matter
In this section we consider a NJL Lagrangian without superconducting interactions: ¯ ∂ − mbare )Ψ + GS (Ψλ ¯ a Ψ)2 + (Ψiγ ¯ 5 λa Ψ)2 L = Ψ(i ¯ + γ5 )Ψ) + Detf (Ψ(1 ¯ − γ5 )Ψ) − K Detf (Ψ(1
(1)
R. Buffa
275
Figure 2. Superconducting quark matter; ﬁg.2.a left: the phase diagram in µ (Mev), µQ (Mev) plane for K = 0 M ev −5 ; ﬁg2.b right: the phase diagram in µ (Mev), µQ (Mev) plane for K = 7.795 10−14 M ev −5 . Region I corresponds to D2 = D5 = D7 = 0, region II corresponds to D2 = 0, D5 = D7 = 0, region III corresponds to D2 D5 D7 = 0
where summation over repeated indices is assumed, λa are the SU(3) ﬂavor matrices and Detf is the determinant in ﬂavor space. Our aim is to determine the zero temperature ground state of the theory in the bulk without leptons and without imposing constraints over the charge and the βequilibrium. The freedom in charge and βequilibrium makes necessary to minimize the energy with respect to the chiral condensates φu , φd , φs and the densities ρu , ρd , ρs of up, down and strange quarks maintaining ﬁxed the total density at the desired value. To ﬁx the total density we use a simple repair algorithm in the generation of the initial populations and a Powell [1] constraint method during the evolution of the populations. We have chosen the following set of parameters: Λ = 602.3, GS = 5.0575 10−06 M ev −2 , K = 1.5590 10−13 M ev −5 , mu = 5.5 M ev, md = 5.5 M ev, ms = 140.7 M ev where Λ is a sharp cutoﬀ in momentum space. We send to [4] for the derivation of the energy of the system in meanﬁeld approximation. We report in ﬁg.1.a and ﬁg.1.b respectively the fractions ρq /ρB of the densities and the constituent masses Mq of up, down and strange quarks with respect to ρB /ρ0 (ρB is the total density and ρ0 is the nuclear density). The most stable solution is positively charged.
3
Superconducting quark matter
In this section we consider a NJL model with superconducting interactions. The total Lagrangian is given by adding Lqq to eq.1: ¯ 5 λa λA C Ψ ¯ T )(ΨT Ciγ5 λa λA Ψ) Lqq = H(Ψiγ where summation over repeated indices is assumed, λA are the SU(3) color matrices and C is the charge conjugation operator. NJL is a low energy eﬀective theory of QCD. There is no reason to assume that the parameters of the NJL at ﬁnite density must reproduce the vacuum quark masses. We would aspect a density dependence of the parameters. Therefore, it is of some interest to study the ﬁnite
276 A global otimization algorithm for finite density quark matter density phase diagram at varying of some parameters. In this section we choose the simple approach of varying the t’Hooft interaction without paying attention to the vacuum properties: we start with a set of parameters that reproduce the vacuum constituent quark masses Λ = 602.3, GS = 6.3926 10−06 M ev −2 , H = GS , K = 0 M ev −5 , mu = 5.5 M ev, md = 5.5 M ev, ms = 112 M ev (taken from [2], parameters set I) and we increase the t’Hooft interaction K maintaining ﬁxed Λ, GS , H, mu , md and ms . We ﬁx the color chemical potentials to zero and we analyze the superconducting zero temperature phase diagram in the bulk at varying of the chemical potentials µ and µQ (µ is relative to the total quark number density and µQ is relative to the electric charge density). We deal with the minimization of the eﬀective potential over the chiral condensates φu , φd , φs and the superconducting condensates D2 , D5 , D7 . The evaluation of the eﬀective potential has been carried out without approximations other than the mean ﬁeld approximation in the same spirit of [5]. We report in ﬁg.2.a and ﬁg.2.b respectively the phase diagrams for K = 0 M ev −5 and for K = 7.795 10−14 M ev −5 . We can see: i) the region of superconductivity is restricted at increasing of the t’Hooft interaction; ii) the line separating the two phases relative to 2SC (D2 = 0, D5 = D7 = 0) and to CFL (D2 D5 D7 = 0) is practically unaﬀected by the change of the t’Hooft interaction.
4
Conclusions
We have applied a parallel genetic algorithm to two non convex problems: i) the determination of the most stable charged solution of zero temperature normal quark matter in the bulk; ii) the determination of the phase diagram in the µ, µQ plane of zero temperature superconducting quark matter in the bulk for diﬀerent values of the t’Hooft interaction. Concerning the ﬁrst application we have obtained a positively charged solution. The result must be taken with care because we have neglected leptons degrees of freedom and βequilibrium. Our result could be useful for mixed phase studies [6]. The second application is much more time consuming and the parallelization becomes really useful. With our choices of the parameters we have found superconducting gaps of the order of 200 Mev. At the superconducting phase transitions we have improved the genetic algorithm by a branch & bound algorithm.
References [1] Z. Michalewicz, ”Genetic Algorithms + Data Structures = Evolution Programs”, SpringerVerlag, 1996; D. Powell, M.M. Skolnick, Proceedings of the Fifth International Conference on Genetic Algorithms, Morgan Kaufmann, San Matteo, CA, (1993), 424 [2] M. Buballa, hepph/0402234 [3] M. Kitazawa, T. Koide, T. Kunihiro, Prog. Theor. Phys. 108, (2002), 929 [4] I.N. Mishustin, L.M. Satarov, H. Stocker, W. Greiner, Phys. Rev. C59, (1999), 3343
R. Buffa
277
[5] M. Buballa, M. Oertel, Nucl. Phys. A703, (2002), 770; A.W. Steiner, S. Reddy, M. Prakash, Phys. Rev. D66,(2002), 094007 [6] N.K. Glendenning, Phys. Rev. D46, (1992), 1274
THE MAGIC EXPERIMENT AND ITS FIRST RESULTS
D. BASTIERI a , R. BAVIKADI b , C. BIGONGIARI a , E. BISESI b , P. BOINEE b , A. DE ANGELIS b , B. DE LOTTO b , A. FORTI b , T. LENISA b , F. LONGO c , O. MANSUTTI b , M. MARIOTTI a , A. MORALEJO a , D. PASCOLI a , L. PERUZZO a , A. SAGGION a , P. SARTORI a , V. SCALZOTTO a AND THE MAGIC COLLABORATION a b c
Dipartimento di Fisica dell’Universi` a di Padova and INFN, Italy Dipartimento di Fisica dell’Universi` a di Udine and INFN, Italy Dipartimento di Fisica dell’Universi` a di Trieste and INFN, Italy
Abstract With its diameter of 17m, the MAGIC telescope is the largest Cherenkov detector for gamma ray astrophysics. It is sensitive to photons above an energy of 30 GeV. MAGIC started operations in October 2003 and is currently taking data. This report summarizes its main characteristics, its ﬁrst results and its potential for physics.
1
Introduction
The MAGIC (Major Atmospheric Gamma Imaging Cherenkov) telescope was designed in 1998 [1] with the main goal of being the Imaging Atmospheric Cherenkov Telescope (IACT) with the lowest gamma energy threshold. It is based on the experience acquired with the ﬁrst generation of Cherenkov telescopes, and includes a large number of technological improvements [2]. With a reﬂector diameter of 17 m, it is the largest Imaging Cherenkov Telescope in the world.
291 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 291–296. © 2006 Springer. Printed in the Netherlands.
292
The Magic Experiment and its First Results
Figure 1. The MAGIC detector.
The study of γ rays is fundamental for our understanding of the universe [3]: γ rays probe the most energetic phenomena occurring in nature, and several signatures of new physics are associated with the emission of γ rays. Photons can travel essentially undeﬂected and unabsorbed in space, and thus they point with excellent approximation to the source of their emission. Together with HESS [4] in Namibia, VERITAS [5] in the USA, and CANGAROO [6] in Australia, MAGIC is one of the “big four” groundbased gamma experiments; among them, it is the only one consisting for the moment of a single telescope. MAGIC comes after the scientiﬁc success of the Energetic Gamma Ray Experiment Telescope (EGRET) instrument on the Compton Gamma Ray Observatory [7]. Launched in 1991, EGRET made the ﬁrst complete survey of the sky above 30 MeV. EGRET increased the number of identiﬁed γ sources producing a catalog which is a reference. Still a large fraction of EGRET sources are unidentiﬁed and, besides other fundamental physics goals, MAGIC aims at their identiﬁcation. MAGIC is located at the Roque de los Muchachos Observatory (ORM) at 2200 m asl (28.8o north, 17.9o west) on the Canary island of La Palma. MAGIC has an eﬀective area of about 4·104 m2 , angular resolution of about 0.2 degrees, relative energy resolution of the order of 20% and can well separate gammas from background (mainly due to hadrons). The MAGIC detector is shown in Figure 1. The sensitivity of MAGIC as calculated from Monte Carlo is shown together with the expected sensitivity from other gammaray detectors in the GeVTeV range in Figure 2. MAGIC has already observed a source, 1ES1959, 10 times fainter than Crab. By observing gamma rays in the energy range from a few tenth of GeV upwards, in overlap with satellite observations and with a substantial improvement in sensitivity, energy and angular resolution, results in several ﬁelds of science can
D. Bastieri et al.
293
Figure 2. Sensitivities for some operating and proposed gamma detectors.
be achieved: • Fundamental physics. – Gamma Ray Horizon, with a larger sensitivity for tests of light propagation eﬀects. – Dark matter, extending the sensitivity to many theoretical models. – Quantum gravity: the search for eﬀects using time delays in the arrival of gammas dependent on the energy, will proﬁt of an improved energy resolution. • Astrophysics. – Gamma Ray Bursts (GRBs): several tests of fundamental physics can be performed based on the characteristics of the transient emission. GRBs are among the most distant and powerful sources in the Universe and key informations on spacetime structure and cosmology can be gained. In addition a more accurate localization of such sources can help in correlating GRBs with known astrophysical objects. – Supernova remnants and plerions: improved energy resolution will help in discriminating between the various acceleration mechanisms assumed to be at the origin of VHE gammas. – Active Galactic Nuclei: the precise gammaray observation of more AGNs at diﬀerent redshifts will much contribute to answer one of the major open questions in extragalactic astronomy, the formation and evolution of galaxies in the early universe. – Pulsars: the known pulsars have cutoﬀ energies of their pulsed emission in the few GeV range, hence their detection will become possible by lowering the energy threshold.
294
The Magic Experiment and its First Results – Unidentiﬁed EGRET sources: this is an enormously rich ﬁeld of activity for detailed studies, possible with modest observation times on nearly half of the observable EGRET sources. – Diﬀuse photon background: the present knowledge of both the extragalactic background radiation and the diﬀuse galactic emission would improve with a more accurate pointing, which could help in separating unidentiﬁed sources from a continuum emission. – Nearby galaxies: their expected steep energy spectrum makes observations at low energy particularly important, as they allow enough ﬂux to be detected in the gamma ray domain. – Studies on the galactic center, where a precise pointing might help in reducing the background on nearby sources.
2
First results
All the results presented here are preliminary since the analysis software is still in its development phase. In the ﬁrst months of data taking, we have mainly concentrated on low zenith angle ( 100 ms ∼ 8 × 10−9 1 × 10−10 1 × 10−11
AGILE 120 kg 30 MeV–50 GeV ∼ 3 sr 4.7◦ (@ 0.1 GeV) 0.6◦ (@ 1 GeV) 0.2◦ (@ 10 GeV) < 100 µs ∼ 6× 10−9 (@ 0.1 GeV) 4× 10−11 (@ 1 GeV) 3 ×10−12 (@ 10 GeV)
Table 1. A comparison between EGRET and AGILE
4
Science Objectives
4.1 GALACTIC DIFFUSE EMISSION The galactic diﬀuse emission is produced in interactions of cosmic rays with gas and ambient photon ﬁelds and thus provides us with an indirect measurement of
F. Longo et al.
307
cosmic rays in various locations in the Galaxy. A signiﬁcant fraction of the diffuse galactic γrays is supposedly produced in decays of neutral pions following inelastic collisions of cosmic ray nucleons. Leptonic emission is particularly important at γray energies below 100 MeV, where bremsstrahlung is presumably the main emission mechanism. Inverse Compton scattering of relativistic electrons on soft ambient photons is expected to provide γrays with a hard spectrum, thus eventually dominating over the π 0 decay γrays at high energies. A new model for Galactic diﬀuse emission suited for the AGILE observations has been developed[4]. 4.2
GALACTIC SOURCES
The capabilities of AGILE with respect of the astrophysics of Galactic γray sources have been recently discussed elsewhere[5]. • Supernova remnants: SNR are considered the most likely sources of galactic cosmic rays. Observational evidence in favor of this scenario has been found only for cosmic ray electrons, not for the nucleons. The signal of π 0 decaying into γγ could indicate the dominant role of the acceleration of nuclei. • Unidentiﬁed EGRET sources: EGRET has left a legacy of about 170 sources not yet identiﬁed ﬁrmly with known sources. Various population studies have been performed to search for correlations with classes of galactic objects. • Pulsars: To date eight pulsars have been identiﬁed in the EGRET data on account of pulsed emission. AGILE will oﬀer the ﬁrst possibility of detecting several young and energetic radio pulsars that have been discovered since the end of the CGRO mission[6]. 4.3
EXTRAGALACTIC SOURCES • Active galactic nuclei These sources show very intense emission, which in many cases is variable. The variability has been observed on all time scales accessible with the available measurement techniques down to about one hour. Simultaneous monitoring of a large number of AGNs per pointing will be possible with the new generation satellites. Several outstanding issues concerning the mechanism of AGN gammaray production and activity can be addressed in the near future (e.g. [7]) including: (1) the study of transient vs. lowlevel gammaray emission and dutycycles; (2) the relationship between the gammaray variability and the radioopticalXrayTeV emission; (3) the correlation between relativistic radio plasmoid ejections and gammaray ﬂares; (4) hard Xray/gammaray correlations. • Gammaray bursts: About ten GRBs were detected by the EGRET spark chamber during ∼ 7 years of operations [8]. This number was limited by
308
Gammaray astrophysics with AGILE the EGRET FOV and sensitivity and, from what we know today, not by the GRB emission mechanism normally producing gammarays above 100 MeV[9]. The small deadtime of the new generation satellites allows a better study of the initial phase of GRB pulses (for which EGRET response was in many cases inadequate). The remarkable discovery of ‘delayed’ gammaray emission up to ∼ 20 GeV from GRB 940217 [10] is of great importance to model prompt and afterglow acceleration processes. Measuring the HE component of GRBs may be critical to the understanding of the charged particle acceleration. AGILE will be able to detect most of the bright long bursts and its two detectors (SuperAGILE and MCAL), with their independent trigger algorithms, compensate each other in detecting soft and high energy GRBs[11].
5
Conclusions
The AGILE scientiﬁc instrument is innovative in many ways, and is designed to obtain an optimal gammaray detection performance despite its relatively small mass and absorbed power. The combination of hard Xray and gammaray imaging capabilities in a single integrated instrument is unique to AGILE. We anticipate a crucial role of SuperAGILE for studies of AGNs, GRBs, and Galactic sources. AGILE will provide an important step forward in γ astronomy. We are conﬁdent that the partnership between High Energy Physics and High Energy Astrophysics will be the source of new discoveries over a wide range of subjects. References [1] Hartman R.C. et al., ApJS, 123, 79, 1999 [2] Hunter S.D. et al., ApJ, 481, 205, 1997 [3] Tavani M. et al., Proceedings of Gamma2001 Conference, AIP Conf. Proceedings, eds. Ritz S., Gehrels N. and Shrader C.R., 2001, Vol. 587, p. 729 [4] Giuliani A., PhD thesis, University of Milano [5] Tavani M., Pellizzoni A. and Vercellone S. (eds.), X and GammaRay Astrophysics of Galactic Sources, Rome 2004 [6] Pellizzoni A. et al., Adv.Sp.Res, 33, 625, 2004 [7] Vercellone S., Soldi S., Chen A.W., Tavani M., MNRAS, 353, 890, 2004 [8] Schneid E.J. et al., in AIP Conf. Proc., Vol. 384, p. 253, 1996 [9] Tavani M., PRL, 76, 3478, 1996 [10] Hurley K. et al., Nature, 372, 652, 1994 [11] Ghirlanda G. et al., AIP Conf. Proc., 727, 704, 2004
SIMULATING THE HIGH ENERGY GAMMARAY SKY SEEN BY THE GLAST LARGE AREA TELESCOPE
F. LONGO a , P. AZZI b , D. BASTIERI b , G. BUSETTO b , Y. LEI b , R. RANDO b , O. TIBOLLA b , L. BALDINI c , M. KUSS c , L. LATRONICO c , N. OMODEI c , M. RAZZANO c , G. SPANDRE c , P. BOINEE d , A. DE ANGELIS d , M. FRAILIS d , M. BRIGIDA e , F. GARGANO e , N. GIGLIETTO e , F. LOPARCO e , M.N. MAZZIOTTA e , C. CECCHI f , P. LUBRANO f , F. MARCUCCI f , M. PEPE f , G. TOSTI f , A. LIONETTO g , A. MORSELLI g , C. PITTORI g a
University of Trieste and INFN, Sezione di Trieste University of Padova and INFN, Sezione di Padova c University of Pisa and INFN, Sezione di Pisa e University of Bari and INFN, Sezione di Bari d University of Udine and INFN, Sezione di Trieste f University of Perugia and INFN, Sezione di Perugia g University of Roma “Tor Vergata” and INFN, Sezione di Roma2 b
Abstract This paper presents the simulation of the GLAST high energy gammaray telescope. The simulation package, written in C++, is based on the Geant4 toolkit, and it is integrated into a general framework used to process events. A detailed simulation of the electronic signals inside Silicon detectors has been provided and it is used for the particle tracking, which is handled by a dedicated software. A unique repository for the geometrical description of the detector has been realized using the XML language and a C++ library to access this information has been designed and implemented. A new event display based on the HepRep protocol was implemented. The full simulation
309 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 309–314. © 2006 Springer. Printed in the Netherlands.
310
Simulating the High Energy GammaRay Sky was used to simulate a full week of GLAST high energy gammaray observations. This paper outlines the contribution developed by the Italian GLAST software group.
1
Introduction
The Gammaray Large Area Space Telescope (GLAST) is an international mission that will study the highenergy phenomena in gammarays universe [1]. GLAST is scheduled for launch in mid 2007. GLAST is instrumented with a hodoscope of Silicon planes with slabs of converter, followed by a calorimeter; the hodoscope is surrounded by an anticoincidence (ACD). This instrument, called the Large Area Telescope LAT, is sensitive to gamma rays in the energy range between 30 MeV and 300 GeV. The energy range, the ﬁeld of view and the angular resolution of the GLAST LAT are vastly improved in comparison with those of its predecessor EGRET (operating in 19912000), so that the LAT will provide a factor of 30 or more advance in sensitivity. This improvement should enable the detection of several thousands of new highenergy sources and allow the study of gammaray bursts and other transients, the resolution of the gammaray sky and diﬀuse emission, the search for evidence of dark matter and the detection of AGNs, pulsars and SNRs. A detailed description of the scientiﬁc goals of GLAST mission and an introduction to the experiment can be found in [2]. GLAST is a complex system, and detailed computer simulations are required to design the instrument, to construct the response function and to predict the background in the orbit. To accomplish these tasks an objectoriented C++ application called Gleam (GLAST LAT Event Analysis Machine) was adopted and implemented by the GLAST LAT collaboration. A brief description of Gleam could be found in [3]. Its structure is described in ﬁgure 1. An entire week of the gammaray observations by GLASTLAT was simulated using Gleam in order to develop and test GLAST LAT scientiﬁc analysis softaware.
2
Simulation and Reconstruction Software
The GLAST oﬀline software is based manly on Gaudi, a C++ framework, originally developed at CERN [4]. Gaudi manages the loop of particles to be simulated, then a series of algorithms are applied to each of them to get the result of the complete simulation and reconstruction chain. The Source Generation is the ﬁrst algorithm called within the particle loop. Its task is to generate particles according to certain characteristcs. This algorithm must store the information on the temporal and spectral behaviour of the source, as well as on the orbital characteristics of GLAST. It provides a user interface to produce additional incoming particles and is responsible for setting the current time, the particle energy, direction, and type. Within this package a series of default sources are implemented.
F. Longo et al.
311
Figure 1. General scheme for simulation and reconstruction within the GLAST oﬀline software framework
They include source for testing purposes as well as the description of astrophysical spectra and the expected particle and albedo gamma backgrounds. An extension of this framework has been implemented for simulating transient sources such as GammaRay Bursts (GRB). It can be used for studying the capability of GLAST for the observation of rapid transient ﬂuxes in general[5]. The algorithm which is responsible for generating the interactions of particles with the detector is based on the Geant4 MonteCarlo toolkit [6] which is an Object Oriented (OO) simulator of the passage of particles through matter. Its application areas include high energy physics and nuclear experiments, medical science, accelerator and space physics. Within the Gleam application the simulation is managed by the Gaudi algorithm G4Generator. Figure 2 shows an event generated using Geant4 within the GLAST LAT experiment. The next algorithm to be applied is the digitization which transforms the hits generated by the Monte Carlo simulations into the signal as read by the electronics. To implement a detailed digitization of the Tracker system a full simulation code has been developed[7]. Starting from Monte Carlo hits in the Silicon detectors, the current signals induced on each strip are evaluated and are converted into voltage signals using the transfer function associated to the detector electronics, taking
312
Simulating the High Energy GammaRay Sky
Figure 2. High energy gammaray interacting with the GLAST LAT detector
into account the detector noise as well as the noise associated to the electronics. The ﬁred strips and the time over threshold are then determined. The signals in the Tracker are then analysed by the reconstruction package. It generates a series of clusters, that are used to ﬁnd and ﬁt the best track candidates. This last procedure is done using alternative pattern recognition algorithms and a Kalman Filter based algorithm. Finally, using the best track found, another algorithm ﬁnds the best vertex candidate for gamma events. Although it is not part of the simulation, the visualization package is essential for the use of the simulation itself. A new version [8, 9] of the event display based on the HepRep protocol was developed and integrated in the oﬄine software. The ﬁgure 3 shows a recent FREDbased event display of GLAST.
3
Simulating the GLAST LAT High Energy Sky
Using Gleam it’s possible to compute and store time, direction and energy of each incoming simulated and reconstructed gamma ray. The Data Challenge One (DC1), organized by the LAT Collaboration from December 2003 to February 2004, represented the ﬁrst opportunity to test the complete simulation chain, and the ﬁrst attempt to perform scientiﬁc analyses on simulated data. For DC1, only
F. Longo et al.
313
Figure 3. GLAST LAT event display based on FRED
the gammaray sky was simulated, while the cosmicray ﬂux (about 104 times greater) was modeled separately for development of the background rejection algorithms. These algorithms were then applied to the simulated gammaray data. The gammaray sky simulated for DC1 was quite rich: a variety of sources was included. The simulation software takes into account the relative ﬂuxes and computes from which source the next photon arrives. A map of the simulated sky is reported in ﬁg. 4.
Figure 4. Full simulated gammaray sky seen by GLAST LAT
Based on the simulated data, several groups tried to develop the best algorithms to detect unknown gammaray sources. GLAST observations contain photons from astrophysical sources, convoluted with the spatial and spectral instrument
314
Simulating the High Energy GammaRay Sky
response. Moreover in most astronomical gammaray images a large fraction of sources is near the detection limit. Thus a careful statistical treatment is needed to determine their existence and properties. Many tools (parametric methods) need a priori model to ﬁt the data and estimate their parameters. No model or hypotesis on the data are requested by the wavelet method[10]. Through an iterative procedure the wavelet method allows a blind search of gammaray point sources. Another goal for the DC1 was to test the possibility to trigger GammaRay Bursts using only LAT information. Several groups within the LAT collaboration are prototyping trigger and alert algorithms for detecting transient signals in DC1 data, and diﬀerent algorithms were studied[11]. Diﬀerent algorithms were successfully applied for searching for transient signals in DC1 data. Bright bursts (with ﬂuence greater than 10−5 erg/cm2 between 20 keV and 1 MeV) can be detected with simple algorithms. Further studies will include the particle background, and the possibility to implement an onboard LAT alert algorithm. All of these items will be addressed for the next Data Challenge (DC2), in which one month of simulated data will be produced.
4
Conclusions
The Gleam simulation program has been developed in the last few years and now it’s ready for simulating the full GLAST satellite and is being used for deriving the ﬁnal instrumental parameters and for generating a full set of events for the developing of scientiﬁc analysis software. References [1] http://glast.gsfc.nasa.gov/ [2] GLAST Science Brochure (March 2001), http://glast.gsfc.nasa.gov/science/resources/ [3] P.Boinee et al. (2003), “Gleam: the GLAST Large Area Telescope Simulation Framework”, in “Science with the New Generation of High Energy Gammaray Experiments. Between Astrophysics and Astroparticle Physics”, eds. S. Ciprini, A. De Angelis, P. Lubrano, O. Mansutti, Forum Editrice Universitaria, Udine, arXiv:astroph/0308120 [4] Gaudi project, http://projgaudi.web.cern.ch/projgaudi/ [5] N. Omodei (2005), PhD thesis, University of Siena [6] S. Agostinelli et al., (2003), NIMA, 506, 250 [7] M. Brigida et al. (2002), LAT internal note, LATTD1058 [8] M. Frailis and R. Giannitrapani, “The FRED Event Display: an Extensible HepRep Client for GLAST,” arXiv:cs.gr/0306031. Talk given at 2003 Conference for Computing in HighEnergy and Nuclear Physics (CHEP 03), La Jolla, California, 2428 Mar 2003. Published in eConf C0303241:MOLT010,2003 [9] http://www.fisica.uniud.it/∼riccardo/research/fred [10] F. Marcucci (2004), PhD thesis, Uinversity of Perugia [11] F. Longo et al., in press on the proceedings of the 4th workshop on “Gamma Ray Bursts in the Afterglow Era”, (2005)
DARK MATTER DETECTION IN GAMMA ASTROPARTICLE EXPERIMENTS
` MARIOTTI b , ERICA BISESI a , MOSE b VILLI SCALZOTTO a b
Department of Physics, Udine University and INFN, Italy Department of Physics, Padua University and INFN, Italy
Abstract The content of matter in the Universe is estimated to be the 27% of its critical density. It is almost universally accepted that most ot this matter is nonbaryonic. Constraints from primordial nucleosynthesis and cosmic background radiation measurements impose that the baryonic content of the Universe cannot exceed the 4% of the critical density, so the nature of the remaining 23% has yet to be identiﬁed. In this sense, one of the most promising candidates is represented by supersymmetric neutralinos. If they exist, they give rise to relic densities in the required range, and are very well motivated in the framework of theoretical extensions of the Standard Model of particle physics. In addition to direct neutralino searches and collider experiments, neutralino annihilation into gamma rays, neutrinos and synchrotron emission from the charged products represents a reliable way of detecting these intriguing particles. The strongest signals are expected to come from the Galactic Center and from the nearest dwarf spheroidals. Clumps of dark matter in galactic haloes are well predicted by high resolution cold dark matter numerical simulations. In this work we present our studies on the gammaray emission from the Galactic Center and from the Draco dwarf spheroidal. We investigate the eﬀect of clumpiness on the detection of signals from neutalinos for diﬀerent mass density proﬁles. One of the scientiﬁc goals of the MAGIC telescope are just searches for the stable lightest supersymmetric particle in the diﬀerent physical scenarios in which they are produed. Assuming MAGIC speciﬁcations, we draw some conclusions about the potentialities of this telescope in such a kind of investigation.
315 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 315–320. © 2006 Springer. Printed in the Netherlands.
316 Dark matter detection in gamma astroparticle experiments 1
Introduction
One of the most promising candidates for halo dark matter are weakly interacting massive particles (WIMPs), although it is not excluded that particles of other kinds, not yet predicted by particle physics models, might represent the solution of this issue. These particles give a relic density which is of the right order of magnitude to explain the dark matter on all scales. Neutralino is the lightest stable supersymmetric particle in most models, so we focus on detection prospects of such a candidate, working either in the MSSM or in the mSUGRA frameworks. High resolution numerical simulations of dark haloes formation suggest that the strongest signals are likely to come from the Galactic Center and from the nearest dwarf spheroidals. The persistence of substructures in these simulations induces to argue that at least a fraction of the dark matter in haloes is clustered in clumps. Taking a phenomenological approach, we here discuss the implications of clumpiness on neutralino dark matter searches.
2
Particle physics models
Minimal supersymmetric model has many free parameters, but with some assumptions we are left with seven parameters in the MSSM model and with ﬁve parameters in the mSUGRA setup, namely the supersymmetric extension of the Standard Model deﬁned in a supergravity inspired framework. For details about the parameters which fully deﬁne the action of MSSM and mSUGRA see Ref. [6, 7]. Table 1 shows the range of parameter values used in our scan of the MSSM space. Choosing the cosmologically interesting relic density range 0.094 < Ωχ h2 < 0.129, we generate in this framework 500000 models and impose that they are not excluded by accelerators constraints. We sample the 5dimensional mSUGRA parameter space choosing a few values of tg β and A0 , and slices along the m1/2 , m0 plane for both sign µ. We consider both the slepton and the stop coannihilation regions and calculate the relic density with all coannihilations. Visited benchmark points are indicates at the upper left of Fig. (3). Table 1. Scans of the MSSm space
Parameter Unit
µ GeV
M2 GeV
tgβ 1
mA GeV
m0 GeV
Ab /m0 1
At /m0 1
Min Max
10 10000
10 10000
1 60
10 1000
50 10000
3 3
3 3
317
E. Bisesi et al. 3
Dark matter distribution models
We focus on indirect searches of neutralinos in the Milky Way and in the Draco dwarf spheroidal. We model haloes of these galaxies using the following dark matter proﬁles: • Navarro−Frenk−White cuspy model [4]: ρcusp (r) = γ = 1;
ρ0 (r/rs )γ (1+(r/rs ))3−γ ,
• Moore & al. cuspy model [3]: the same as above with γ = 1.5; • a milder cuspy proﬁle [2]: the same as above with γ = 0.5; • Burkert & al. proﬁle [5]: ρBurk (r) =
4
ρ0 (1+(r/rs )) (1+(r/rs )2 ) .
Gamma ray ﬂux
Neutralino annihilation in the galactic halo produces both a gammaray ﬂux with a continuum energy spectrum and monochromatic gammaray lines. Considering a detector with an angular acceptance ∆Ω pointing in a direction of galactic longitude and latitude (l, b), the gammaray ﬂux from neutralino annihilation at a given energy E is: & < σ vF > dNγF < J(l, b) > (∆Ω) cm−2 s−1 sr−1 m2χ dE F (1) for the continuum gamma. If we assume a spherical dark matter halo in the form of a perfectly smooth distribution of neutralinos, < J(l, b) > (∆Ω) is equal to: 1 < J(l, b) > (∆Ω) = const dΩ ρ(L, ψ )2 dL, (2) ∆Ω ∆Ω lineof sight Φγ (E, ∆Ω, l, b) = const.
where L is the distance from the detector along the line of sight, ψ is the angle between the direction of observation and that of the center of the galaxy and the integration over dΩ is performed over the solid angle ∆Ω centered on ψ. 4.1
EFFECT OF CLUMPINESS
To discuss the implications of clumpiness on neutralino dark matter searches, we follow the phenomenological approach of Bergstr¨ om & al. (1999) [1]. This model is mainly focused on a many small clumps scenario, where substructures are light, with Mcl less than 104 − 106 M . Postulating that a fraction f of the dark matter is concentrated in clumps, they ﬁnd that the increase of the signal compared to a smooth halo is determined by the enhancement factor f δ, where δ is the eﬀective contrast between the dark matter density in clumps and the local halo density ρ0 .
318 Dark matter detection in gamma astroparticle experiments
Figure 1. Values of < J(l, b) > (∆Ω) for diﬀerent halo proﬁles for the Milky Way and Draco.
Assuming that the clumps can be regarded as pointlike sources, authors derive the anoalogous of Eq. (2) in the clumpy scenario: fδ < J(l, b) > (∆Ω) = const dΩ ρ(L, ψ )dL. (3) ρ0 ∆Ω ∆Ω lineof sight
Fig. (1) illustrates the values of < J(l, b) > (∆Ω) for three of the halo proﬁles introduced above, giving the smooth and clumpy components in the cases of the Milky Way and Draco respectively. The clumpiness enhancement factor is taken reasonably equal to 20. Values of the scale lenght and local halo densities follow prescriptions of Ref. [1, 2].
5
Results and discussion
As the background follows a poissonian statistics, the minimum detectable ﬂux of gamma rays from an ACT telescope is determined by the condition: Φγ Aef f t ∆Ω √ ≥ 5, Nb
(4)
for a 5σ detection level. Nb is the number of background counts, hadrons and electrons, which is obtained on the ground of Ref. [1]. We assume the following
E. Bisesi et al.
319
Figure 2. The minimum detectable < σ v >cont versus mχ for the NFW, Moore & al. and Burkert & al. proﬁles. Dots are points of the parameter space of MSSM, lines represent the 5σ detection level for the MAGIC telescope. Only models corresponding to SUSY points above the curves yield a detectable signal.
MAGIC speciﬁcations: Eth = 60 GeV, Aef f = 109 cm2 , t = 250 h, ∆Ω = 10−5 sr and an energy resolution of 25%. Plotting inequality (4) with the equality sign onto the SUSY parameter space, we divide it into the detectable (above the line) and undetectable (below the line) regions. Results for the Galactic Center are shown in Fig. (2) and Fig. (3), for the MSSM and mSUGRA scenarios respectively and for a clumpiness enhancement factor of 20. This factor is anyway uninﬂuential in this case. As we can see from Fig. (2)−(3) plots, detectability of SUSY particles is very sensitive to the choice of the dark matter proﬁle. We ﬁnd that the scenario which gives the best opportunities for the MAGIC telescope is the NavarroFrenkWhite. Anyway, our model doesn’t take into account distribution functions for substructures; an extension of our investigations at higher galactic latitudes does need this is to be modeled in detail, so we will address our future interests in this direction. References [1] L. Bergstr¨ om, J. Edsj¨ o, P. Gondolo & P. Ullio, Phys Rev D 59, 043506 (1999). [2] N.W. Evans, F. Ferrer & S. Sarkar, Phys Rev D 69, 123501 (2004). [3] B. Moore, F. Governato, T. Quinn, J. Stadel & G. Lake, ApJ 499, L5 (1998). [4] J.F. Navarro, C.S. Frenk & S.D. White, ApJ 490, 493 (1997). [5] A. Burkert, ApJ 447, L25 (1995). [6] J. Edsj¨ o, M. Schelke, P. Ullio & P. Gondolo, arXiv:hepph/0301106 v2 (2003). [7] J. Edsj¨ o, M. Schelke & P. Ullio, arXiv:astroph/0405414 v1 (2004).
320 Dark matter detection in gamma astroparticle experiments
Figure 3. The minimum detectable < σ v >cont versus mχ for the NFW, Moore & al. and Burkert & al. proﬁles. Dots are points of the parameter space of mSUGRA, lines represent the 5σ detection level for the MAGIC telescope. Only models corresponding to SUSY points above the curves yield a detectable signal.
[8] A. Tasitsiomi & A.V. Olinto, Phys Rev D 66, 083006 (2002). [9] L. Bergstr¨ om, P. Ullio & J.H. Buckley, arXiv:astroph/9712318v1 (1997). [10] http://hegra1.mppmu.mpg.de/MAGICWeb/
CONTRIBUTION OF PULSARS TO THE GAMMARAY BACKGROUND AND THEIR OBSERVATION WITH THE SPACE TELESCOPES GLAST AND AGILE
ERICA BISESI a
a
Department of Physics, Udine University, via delle Scienze 208, 33100 Udine, Italy
Abstract Luminosities and ﬂuxes of the expected population of galactic gammaray pulsars become foreseeable if physical distributions at birth and evolutive history are assigned. In this work we estimate the contribution of pulsar ﬂuxes to the gammaray background, which has been measured by the EGRET experiment on board of the CGRO. For pulsar luminosities we select some of the most important gammaray emission models, taking into account both polar cap and outer gap scenarios. We ﬁnd that this contribution strongly depends upon controversial neutron star birth properties. A comparison between our simulation results and EGRET data is presented for each model, ﬁnding an average contribution of about 10%. In addition, we perform the calculation of the number of new gammaray pulsars detectable by GLAST and AGILE, showing a remarkable diﬀerence between the two classes of models. Finally, we suggest some improvements in the numerical code, including more sophisticated galactic m odels and diﬀerent populations of pulsars like binaries, milliseconds, anomalous pulsars and magnetars.
1
Introduction
The estimation of the contribution of pulsars to the gammaray background is based on the comparison between theoretical predictions on the ﬂux emitted by these sources and experimental observations. Each model for the emission of radiation by a stellar population is speciﬁed by the galactic distribution of sources and their individual luminosities.
321 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 321–326. © 2006 Springer. Printed in the Netherlands.
322
Contribution of pulsars to the gammaray background
Figure 1. Distributions of pulsars ages (t), distances (d), magnetic ﬁelds (B0 ), periods (P0 ), period derivatives (P˙ ). Dot lines represent a situation with constant magnetic ﬁeld, solid lines are referred to the ﬁeld decay case (E. Bisesi, 2002).
2
Galactic model
2.1 SPATIAL DISTRIBUTIONS We here present a numerical simulation of the galactic population of radio pulsars giving all initial spatial and physical parameters in the framework of the galactic models of Paczy´ nski (1990) and of Gonthier & al. (2002) [4]; we then make them evolve accordingly with the velocity model of Sturner & Dermer (1994) [10]. We show histograms of distributions of various pulsar properties in Fig. (1), assuming both constant magnetic ﬁeld and ﬁeld decay. 2.2 PULSAR LUMINOSITIES In the commonly accepted hypotesis that every pulsar irradiates on the whole electromagnetic spectrum, we assign a gamma luminosity to each source. We consider a suitable choise of emission models. POLAR CAP MODELS: 0.95 −1.7 • Harding (1981) [5]: Lγ (>100 MeV) = 1.2 × 1035 B12 P ph s−1 ;
• Zhang & Harding (2000) [12]: Lγ (I) = 5.87×1035 B12 6/7 P −1/7 ph s−1 Lγ (II) = 1.0 × 1035 B12 P −9/4 ph s−1
for B12 1/7 P −11/28 > 6.0, for B12 1/7 P −11/28 < 6.0;
E. Bisesi
323
Figure 2. Map of the gammaray sky as seen by EGRET (E. Bisesi, 2002).
Figure 3. Comparison between the EGRET gammaray background (in blue) and predictions for the polar cap model of Harding (1981) (in violet). We also give the simulated number of pulsars above the EGRET detection threshold for this model. Plots on the right side are enlargements of the simulated proﬁles (E. Bisesi, 2002).
• Sturner & Dermer (1994) [10]: Lγ = 6.25 × 1035 B12 3/2 P −2 ph s−1 .
OUTER GAP MODELS with the death line 5 log B − 12 log P ≤ 72: • Romani & Yadigaroglu (1995) [9]: Lγ = 1.56×1036 B12 0.48 P −2.48 ph s−1 ; • Cheng & Zhang (1996) [2]: Lγ = 3.93 × 1037 B12 0.3 P −0.3 ph s−1 . We calculate the integrated ﬂux on the galactic latitude −10◦ < b < 10◦ as a function of the longitude l; results are plotted together with the experimental points as they have been measured by the EGRET experiment. Fig. (2) and (3) respectively show the map of the gammaray sky as seen by EGRET and our results for the polar cap model of Harding (1981). The contribution of pulsars to the gammaray background is less than 10% for this model; the foreseen number of sources above the EGRET detection threshold of 10−7 ph cm−2 s−1 in not far from the observed value.
324
Contribution of pulsars to the gammaray background
Figure 4. Diagram period−magnetic ﬁeld for diﬀerent pulsar families. Points indicate young isolated pulsars, circle−dots pulsars in binary systems. The population of millisecond pulsars is visible at the bottom left (from Phinney & Kulkarni, 1994 [8]).
3
Comparison among models
In the previous discussion we have forced the physical parameters involved in the calculation of gamma ray luminositied and ﬂuxes to assume some deﬁnite values. In order to give a meaningful comparison among diﬀerent emission models an extension to a more general parameter space is required. So we deﬁne a parameter space whose extremes are corresponding to the limit conditions for pulsating gamma neutron stars; we then follow the evolution of a representative point inside this space. The most relevant physical parameters are the rotation period, the magnetic ﬁeld and the velocity fron the Galactic Centre. 3.1 PULSAR POPULATIONS Diﬀerent families of pulsars follow distinct evolutive histories, and this has a direct impact on their physical properties. Furthemore, physical mechanisms ruling over the gammaray emission are peculiar of each class. Fig. (4) shows a diagram period−magnetic ﬁeld for young isolated, binaries and millisecond pulsars. Magnetars are characterized by very high magnetic ﬁelds, typically 1014 − 1016 G. We bind the parametric space in this way: • P0min = 0.01 s (when the centrifugal force equals molecular bonds of the star, yielding its survival); • P0max = 1 s (ﬁxed from observations); • B0min = 1011 G (the lowermost value for young isolated pulsars); • B0max = 4.413 × 1013 G (condition for the magnetar regime);
E. Bisesi
325
Figure 5. Comparison among models (E. Bisesi, 2002).
• v0min = 120 km/s (Gonthier & al., 2002 [4]); • v0max = 350 km/s (Lyne & Lorimer, 1994). We show a tridimensional representation of the parametric space log P0 , log B0 , V in the left side of Fig. (5). The bidimensional portion on the right shows the positions of two models of interest [4, 3]. We compare predictions for the ﬁve models introduced above for a set of points of the parametric space. As physical mechanisms for the gammaray emission diﬀer from one pulsar family to another, we would give luminosity function expressions for each of them. We purpose to improve our work in such a way in future; in this context we restrict our analysis to young isolated pulsars. An example relevant to the point X3 is shown in Fig. (6).
Figure 6. Comparison between EGRET observations and simulation results for the point X3 of the parametric space (E. Bisesi, 2002).
326 4
Contribution of pulsars to the gammaray background Discussion
For each point considered, we evaluate the percentual excess between the foreseen number of gammaray pulsars for each model and their actual number of 7. The total percentual excess for each model gives us the possibility of select the most reliable model, which we ﬁnd being that of Harding (1981) [5]. We ﬁnally estimate the number of gammaray pulsars detectable by the next space telescopes GLAST and AGILE for the model selected. GLAST and AGILE detection thresholds are 6 × 10−9 ph cm−2 s−1 and 10−7 ph cm−2 s−1 respectively. Our results for the ﬁve points of the right side of Fig. (5) are shown in Table 1. Predictions for GLAST are very optimistic, remarkably we expect to detect a very large number of new gammaray pulsars, opening very promising frontiers in understanding these mysterious and fascinating objects. Table 1. Total percentual excess for each model on the whole parametric space for young pulsars. Model
Total percentual excess =
H ZH SD RY CZ
P8
i=1
27 220 174 30506 2277
Percentual excess(Xi )
References [1] Bailes, M. & Kniﬀen, D.A., ApJ 391, 659666 (1992). [2] Cheng, K.S. & Zhang, J.L., ApJ 463, 271283 (1996). [3] Emmering, R.T. & Chevalier, R.A., ApJ 345, 931938 (1989). [4] Gonthier, P.L. & Harding, A.K., ApJ 425, 767775 (1994). [5] Harding, A.K., ApJ 245, 267273 (1981). [6] Harding, A.K., Gamma Ray Pulsars: Models and Predictions, arXiv:astroph/0012268 (2000). [7] Lyne, A.G., Manchester, R.N. & Taylor, J.H., MNRAS 213, 613639 (1985). [8] Phinney, E.S. & Kulkarni, S.R., ARA&A 32, 591639 (1994). [9] Romani, R.W. & Yadigaroglu, I.A., ApJ 438, 314321 (1995). [10] Sturner, J.S. & Dermer, C.D., ApJ 461, 872883 (1996). [11] Thompson, D.J., Gamma Ray Pulsars: Observations, arXiv:astroph/101039 (2001). [12] Zhang, B. & Harding, A.K., ApJ 532, 11501171 (2000). [13] http://glast.gsfc.nasa.gov/ [14] http://agile.ifctr.mi.cnr.it/Homepage/collaboration.shtml
DATA MINING IN GAMMA ASTROPHYSICS EXPERIMENTS
MARCO FRAILIS a , ALESSANDRO DE ANGELIS a , VITO ROBERTO b a b
INFN and Dipartimento di Fisica, Universit` a degli Studi di Udine, Italy Dipartimento di Matematica e Informatica, Universit` a degli Studi di Udine, Italy
Abstract Data mining techniques, including clustering and classiﬁcation tasks, for the automatic information extraction from large datasets are increasingly demanded in several scientiﬁc ﬁelds. In particular, in the astrophysical ﬁeld, large archives and digital sky surveys with dimensions of 1012 bytes currently exist, while in the near future they will reach sizes of the order of 1015 . In this work we propose a multidimensional indexing method to eﬃciently query and mine large astrophysical datasets. A novelty detection algorithm, based on the Support Vector Clustering and using density and neighborhood information stored in the index structure, is proposed to ﬁnd regions of interest in data characterized by isotropic noise. We show an application of this method for the detection of point sources from a gammaray photon list.
1
Characterization of the astrophysical datasets
At present, several projects for the multiwavelength observation of the Universe are underway, for example SDSS, GALEX, POSS2, DENIS, etc. In the next years, new spatial missions will be launched (e.g. GLAST, Swift), surveying the wall sky at diﬀerent wavelengths (gammaray, Xray, optical). In the Astroparticle and Astrophysical ﬁelds, data is mostly characterized by multidimensional arrays. For instance, in Xray and Gammaray astronomy, the data gathered by detectors
327 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 327–331. © 2006 Springer. Printed in the Netherlands.
328
Data mining in gamma astrophysics experiments
are lists of detected photons whose properties include position (RA, DEC), arrival time, energy, error measures both for the position and the energy estimates, quality measures of the events . Source catalogs, produced by the analysis of the raw data, are lists of point and extended sources characterized by coordinates, magnitude, spectral indexes, ﬂux, etc. 1.1 MINING MULTIDIMENSIONAL DATA Data mining applied to multidimensional data analyzes the relationships between the attributes of a multidimensional object stored into the database and the attributes of the neighboring ones. Several queries are required by this kind of analysis: • point queries, to ﬁnd all objects overlapping the query point; • range queries, to ﬁnd all objects having at least one common point with a query window; • nearestneighbor queries, to ﬁnd all objects that have a minimum distance from the query object. Another important operation is the spatial join, which in the astrophysical ﬁeld is needed to search multiple source catalogs and crossidentify sources from diﬀerent wavebands. This multidimensional (spatial) data tend to be large (sky maps can reach sizes of Terabytes) requiring the integration of the secondary storage, and there is no total ordering on spatial objects preserving spatial proximity. This characteristic makes it di cult to use traditional indexing methods, like B+trees or linear hashing.
2
An optimized Rtree structure
The Rtree is a data structure meant to eﬃciently index multidimensional point data or objects with a spatial extent. The structure of an Rtree is the following: • an inner node of the Rtree has entries of the form (cp, MBB), where cp is the address of a child node and MBB is the ndimensional Minimum Bounding Box of all entries in that child node; • a leaf node has entries of the form (cp, MBB ), where cp refers to a record describing a particular object and MBB is its minimum bounding box, or (Point, Attributes), where Point is a coordinate in the ndimensional space and Attributes are data associated to that point. An optimized index, in terms of construction time, memory occupied and query performances, can be built using a priori information on the dataset by means of bulk loading algorithms. We have followed a topdown construction method called VAMSplit algorithm to build and optimized RTree. This method preserves
M. Frailis et al.
329
Figure 1. The structure of the optimized Rtree built a photon dataset
the spatial proximity between sibling nodes, resulting in a partition of the dataset with no overlapping between MBBs. Moreover, the volume of the data space covered by each node (at a particular level) is variable and dependent on data density. The main idea of this method is to recursively split the dataset on a near median element along the dimension with maximum variance. In particular, at each recursive step, child subtrees capacity is calculated by: 3 3 44 N cscap = B · F logF B −1
where B is the page capacity, F is the internal fanout and N is the number of elements to index in the current step. The near median element is computed by: 5 6 78 1 N med = cscap · · 2 cscap Our implementation uses a sampling strategy to ﬁnd a good pivot value in the partition step and reduce the number of I/O operations; a caching strategy has been adopted to partition the data into the secondary storage. The total construction N time is O N M log B B . B
3
A scalable novelty detection algorithm
The structure of the optimized Rtree can help exploring the data and ﬁnding regions of interest. For this purpose, other information can be added to each node: the total number of data points covered by the node, their mean and variance, other statistical moments. In this work we propose a novelty detection algorithm based on the Support Vector Clustering (SVC).
330
Data mining in gamma astrophysics experiments
3.1 THE SVC ALGORITHM The SVC algorithm estimates the support of a high dimensional distribution. It computes the hypersphere with minimal radius which encloses the data points when mapped to a high dimensional feature space. Given a set of points x1 , . . . , xN , with xi ∈ X ⊆ Rd , it ﬁnds the hypersphere (c, r) that solves the optimization problem: N & min r2 + C ρi c,r,ρ
i=1
s.t. φ(xi ) − c ≤ r2 + ρi ρi ≥ 0,
i = 1, . . . , N
where ρi are slack variables and C is a positive constant. Deﬁning the Lagrangian, the solution is obtained solving the dual problem: max α
N &
αi k(xi , xi ) −
i=1
s.t. 0 ≤ αi ≤ C,
N &
αi αj k(xi , xj )
i,j=1
i = 1, . . . , N
where αi are the lagrange multipliers and k(xi , xj ) = φ(xi ), φ(xj ). 3.2 FEATURES FROM THE RTREE The partition generated at a given level of the optimized Rtree is used as the input space of the novelty detection algorithm. For each node of the partition, the input parameters include the center c of its bounding box and the density (the ratio between the number n of elements covered by the node and the volume V of its bounding box). These features are not orthogonal. Therefore, we ﬁrst apply the PCA method to ﬁnd the directions along which the variance is higher and project the features on the corresponding eigenvectors. The projected features are passed to the SVC algorithm. 3.3 GAMMA SOURCE DETECTION Point sources are mostly characterized by a stronger ﬂux, with respect to the surrounding, focused on a small angular region. The area covered by a point source depends also on the instrument point spread function. An optimized Rtree index can be built on a dataset including photons gathered in a certain range of time (we are using, for the analysis, a minimum interval of 6 days). To ﬁnd static or strong variable sources (e.g. gammaray bursts) we use only a bidimensional indexing on the RA and DEC values. The algorithm estimates the support of the diﬀuse background. The output of the SVC algorithm is ﬁltered to single out the nodes with lowest density. Point sources are considered as the remaining outliers.
M. Frailis et al.
331
Figure 2. The novelty detection algorithm applied to the anticenter
Figure 2 shows the application of our method to the anticenter region. Green boxes represent the background (support) while yellow boxes are support vectors and the red ones are the outliers. In particular,the three major sources of the anticenter are highlighted as novelty.
4
Conclusions
In this work we have realized a multidimensional indexing method to eﬃciently access and mine large multidimensional astrophysical data. The index adapts the VAMSRtree to large datasets. The partition generated by the optimized Rtree is used to scale the SVC algorithm and ﬁnd regions of interest where a more accurate analysis can be performed. References [1] D. A. White, R. Jain, Similarity Indexing: Algorithms and Performance, Proc. SPIE, 2670:6273, San Diego, USA (1996) [2] C. B¨ ohm, H.P. Kriegel, Eﬃcient Bulk Loading of Large HighDimensional Indexes, DaWaK 99, 31:251260 (1999) [3] A. BenHur et al., Support Vector Clustering, Journal of Machine Learning Research 2(Dec), 125137 (2001) [4] J. ShaweTaylor, N. Cristianini, Kernel Methods for Pattern Analysis, Cambridge (2004)
GRID SERVICES FOR THE MAGIC EXPERIMENT
A. FORTI a , S.R. BAVIKADI a , C. BIGONGIARI b , G. CABRAS a , A. DE ANGELIS a , B. DE LOTTO a , M. FRAILIS a , M. HARDT c , H. KORNMAYER c , M. KUNZE c , M. PIRACCINI a AND THE MAGIC COLLABORATION a b c
INFN and Dipartimento di Fisica, Universit´ a degli Studi di Udine, Italy INFN and Dipartimento di Fisica Galileo Galilei, Padova, Italy Institut f¨ ur wissenschaftliches Rechnen, Forschungszentrum Karlsruhe, Germany
Abstract Exploring signals from the outer space has become an observational science under fast expansion. On the basis of its advanced technology the MAGIC telescope is the natural building block for the ﬁrst large scale ground based high energy γray observatory. The low energy threshold for γrays together with diﬀerent background sources leads to a considerable amount of data. The analysis will be done in diﬀerent institutes spread over Europe. Therefore MAGIC oﬀers the opportunity to use the Grid technology to setup a distributed computational and data intensive analysis system with the nowadays available technology. Beneﬁts of Grid computing for the MAGIC telescope are presented.
1
MAGIC
The MAGIC telescope has been designed to search the sky to discover or observe high energy γrays sources and address a large number of physics questions [1]. Located at the Instituto Astrophysico de Canarias on the island La Palma, Spain,
333 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 333–337. © 2006 Springer. Printed in the Netherlands.
334
Grid services for the MAGIC experiment
at 28◦ N and 18◦ W, at altitude 2300m asl, it is the largest γray telescope in the world. MAGIC is operating since October 2003, data are taken regularly since February 2004 and signals from Crab and Markarian 421 was seen. The main characteristics of the telescope are summarized below: • A 17m diameter (f/d=1) tessellated mirror mounted on an extremely light carbonﬁber frame (< 10 tons), with active mirror control. The reﬂecting surface of mirrors is 240 m2 ; reﬂectivity is larger than 85% (300  650nm). • Elaborate computerdriven control mechanism. • Fast slewing capability (the telescope moves 180◦ in both axes in 22s). • A higheﬃciency, highresolution camera composed by an array of 577 fast photomultipliers (PMTs), with a 3.9◦ ﬁeld of view. • Digitalization of the analogue signals performed by 300 MHz FlashADCs and a high data acquisition rate of up to 1 KHz. • MAGIC is the lowest threshold (≈ 30 GeV) IACT operating in the world. Gammarays observation in the energy range from a few tenth of GeV upwards, in overlap with satellite observations and with substantial improvement in sensitivity, energy and angular resolution, leads to search behind the physics that has been predicted and new avenues will open. AGNs, GRBs, SNRs, Pulsars, diﬀuse photon background, unidentiﬁed EGRET sources, particle physics, darkmatter, quantum gravity and cosmological γray horizon are some of the physics goals that can be addressed with the MAGIC telescope.
2
Grid
The idea of computational and data grids dates back to the ﬁrst half of the 90’s. The vision behind them is often explained using the electric power grid metaphor. The electric power grid delivers electric power in a pervasive and standardized way. You can use any device that requires standard voltage and has a standard plug if you are able to connect it to the electric power grid through a standard socket. When you use electricity you don’t worry were it is produced and how it is delivered, you just plug your device into the wall socket and use it. Currently we have millions of computing and storage systems all over the planet connected through the Internet. What we need is an infrastructure and standard interfaces capable of providing transparent access to all this computing power and storage space in a uniform way. This is the concept behind Grid. More precisely, Grid is a kind of parallel and distributed system that enables the sharing, selection and aggregation of services of heterogeneous resources distributed across multiple administrative domains based on their availability, capability, performance, cost, and users qualityofservice requirements [2]. As network performance has outpaced computational power and storage capacity, this new paradigm has evolved to enable the sharing and coordinated use of geographically distributed resources.
A. Forti et al. 2.1
335
VIRTUAL ORGANIZATION
The Virtual organization is an important Grid concept. Grid allows a pool of heterogeneous resources both within and outside of an organization to be virtualized and form a large, virtual computer. This virtual computer can be used by a collection of users and/or organizations in collaboration to solve their problems. The rules governing the participants providing the resources and the consumers using the resources, as well as the conditions for sharing, dictate the nature of the virtual organization. Hence, a virtual organization groups people and resources without worry about their physical location or institute boundaries. For security reasons, the use of the resources is constrained by an authentication process.
3
Beneﬁts of Grid computing for Magic
The collaborators of the MAGIC telescope are mainly spread over Europe, 18 institutions from 9 countries, with the main contributors (90% of the total) located in Germany (MaxPlanckInstitute for Physics, Munich and University of Wuerzburg), Spain (Barcelona and Madrid), Italy (INFN and Universities of Padova, Udine and Siena). The geographical distribution of the resources makes the management of the experiment harder. This is a typical situation for which Grid computing can be of great help, because it allows researchers to access all the resources in a uniform, transparent and easy way. The telescope is in operation during moonless nights. The average amount of raw FADC data recorded is about 500600 GB/night. Additional data from the telescope control system or information from a weather station are also recorded. All these information have to be taken into account in the data analysis. The MAGIC community can leverage from Grid facilities in areas like ﬁle sharing, Monte Carlo data production and analysis. In a Grid scenario the system can be accessed through a web browser based interface with single signon authentication method. We can brieﬂy summarize the main beneﬁts given by the adoption of Grid technology for the MAGIC experiment: • Presently, users analyzing data must know where to ﬁnd the required ﬁles and explicitly download them. In a Grid perspective, instead, users don’t care about data location and ﬁles replication policies improve access time and fault tolerance. • Grid workﬂow tools can manage the MAGIC Monte Carlo simulation. The resources from all the members of the MAGIC community can be put together and exploited by the Grid. Easy access to data production for every user, or accordingly to the VO policies. • Analysis tools can be installed on the Grid. They are thus shared and available for all the users (no need for single installations), moreover, they can
336
Grid services for the MAGIC experiment
Figure 1. MAGIC Monte Carlo Simulation workﬂow
exploit the facilities of a distributed system. 3.1 MMCS The MAGIC Monte Carlo simulation workﬂow is a series of programs which simulate the properties of diﬀerent physics processes and detector parts (ﬁgure 1): • CORSIKA: air shower and hadronic background simulation. The output contains information about the particles and the Cherenkov photons reaching the ground around the telescope. • Reﬂector : simulates the propagation of Cherenkov photons through the atmosphere and their reﬂection in the mirror up to the camera plane. The input for the Reﬂector program is the output of CORSIKA. • StarﬁeldAdder : simulation of the ﬁeld of view. It adds light from the nondiﬀuse part of the night sky background, or the eﬀect of light from stars, to images taken by the telescope. • StarResponse: simulation of NSB (night sky background) response. • Camera: simulate the behavior of the photomultipliers and of the electronic of the MAGIC camera. It also allows to introduce the NSB (optionally), from the stars and/or the diﬀuse NSB.
A. Forti et al. 3.2
337
THE PRESENT
The EGEE (Enabling Grids for Escience in Europe) project brings together experts from 70 organizations and 27 countries with the common aim of building on recent advances in Grid technology and developing a service grid infrastructure in Europe which is available to scientists 24 hoursaday. Recently MAGIC has became part of the EGEE project. This is the ﬁrst step for enabling MAGIC on the Grid. This process migration is a big eﬀort and must be divided in smaller steps. The ﬁrst step was chosen to be the migration of the MAGIC Monte Carlo simulation workﬂow and it is now working [3] thanks to the eﬀort of the Udine, Padova and CNAF (Bologna) groups.
4
Conclusions
Grid technologies promise to change the way organizations tackle their complex computational and dataintensive problems. The vision of large scale resource sharing is nowadays becoming a reality in many areas. However, it must be realized that Grid is an evolving ﬁeld in computer science, where standards and technology are still being under development to enable this new paradigm. Presently, many eﬀorts are being made to attract a wide range of new users to the grid. MAGIC has caught this big opportunity and now another step is made towards the realization of a wide project that wants to connect the compute and storage resources of the astroparticle institutions in order to collaborate across the institute borders as well as across the collaboration borders. This new challenge will be the ASTROPA Grid project [4]. References [1] Lorenz, E., New Astron. Rev. vol. 48, 339344 (2004). [2] Foster, I., Kesselman, C. (eds.). The Grid: Blueprint for a New Computing Infrastructure. Morgan Kaufmann, 1999. [3] M. Piraccini, tesi di Laurea in Fisica Computazionale, Udine 2004. [4] ASTROPA Grid, proposal by the MAGIC Collaboration to EGEE, 2004.
Part IV
Complex Systems
TOWARDS A PHYSICAL MODEL FOR MEMORY
LEON COOPER a
a
Brown University, Providence, USA
Abstract The Interaction of theory with experiment?normal in physics since Galileo, but novel in neuroscience? has proven fruitful in neuroscience. Theory has suggested experiments that have uncovered new phenomena: a. Long Term Depression (LTD) and bidirectional synaptic modiﬁcation dependent on the depolarization of the postsynaptic cell in agreement with the BCM synaptic modiﬁcation function. b. The sliding modiﬁcation threshold. This has provided experimental veriﬁcation of the postulates of the BCM theory of synaptic plasticity. Theory has been able to relate diverse observations in diﬀerent brain regions such as LTD/LTP in hippocampus to monocular deprivation and reverse suture results in visual cortex. Theory has also shown that one underlying calcium dependent mechanism can account for the various methods of inducing synaptic plasticity. Cellular and molecular mechanisms that underlie this synaptic plasticity have been proposed, and some of these have been conﬁrmed experimentally. The interaction of local and global modulatory signals, the probable basis for memory consolidation, as well as the implication of LTD mechanisms in such pathologies as the Fragile X Syndrome are being explored.
341 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 341–341. © 2006 Springer. Printed in the Netherlands.
PATTERNS AND DISSIPATIVE WAVES, INCLUDING SOLITONS, IN LATTICES AND AT INTERFACES
MANUEL G. VELARDE a
b
a,b
Instituto Pluridisciplinar, Universidad Complutense de Madrid (UCM), Paseo Juan XXIII 1, 28040 Madrid, Spain International Centre for Mechanical Sciences (CISM), Palazzo del Torso, Piazza Garibaldi 18, 33100 Udine, Italy
Abstract A succint account is given here of recently discovered salient features of selforganizing processes (disorderorder transitions) leading to cellular convective patterns and dissipative (and solitonic) waves in ﬂuid layers with an interface and in active lattices. Emphasis is given to the recently predicted transition from (Drude) Ohm’s linear electric conduction to a form of (hightemperature) superconduction mediated by the formation of electronsoliton (solectron) dynamic bound states in a nonlinear lattice, a phenomenon reminiscent of surﬁng.
1
Introduction
The apparently unrelated phenomena of cellular convective patterns and waves occurring at ﬂuid interfaces and in discrete lattices can be explained using a common ideology, if not a theory. I refer to the onset of cooperative phenomena driven by a constraint that, on the one hand, leads to a (nonequilibrium) phase transition to a more ordered state and, on the other hand, provides a steady (dynamic) balance between energy input and dissipation. Beyond threshold the form and evolution of the newly formed “dissipative structures” [1] depend on the level of that energy balance sustaining the opportunistic mode (pattern or wave) dominating the others. This enslaving mode or “order parameter” in the supercritical regime is
343 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 343–354. © 2006 Springer. Printed in the Netherlands.
344
Patterns and dissipative waves
the “master/conductor” in the selforganizing, synergetic process occurring in the system [2]. Lack of space permits me to just itemize a few salient features with emphasis, in particular, on a recently predicted curious phenomenon triggered by dissipative solitons in a lattice operating as an electric circuit: the transition from (Drude) Ohm’s linear electric conduction to an apparently purely classical form of electric (hightemperature) superconduction (the more ordered phase), an intriguing result indeed [3, 4].
2
Cellular patterns at interfaces
The onset and development beyond threshold of (steady) cellular convection driven by interfacial agents, e.g. surface tension gradients, is a typical case of selforganization and cooperative processes in a ﬂuid layer subjected to a thermal constraint, like heating or ad/absorption and (mass) diﬀusion of a (surface active agent) surfactant. The engine providing the (steady) energy balance that allows the disorderorder transition is the Marangoni eﬀect [5][8]. Here a variation of temperature and/or (excess) surfactant concentration at the interface creates shear stress and ﬂow through the coupling between the “thermal” ﬁeld and the velocity. The essential character of this process is that the engine operates not in the bulk of the ﬂuid but rather at an open surface or interface between two ﬂuids (e.g. an airliquid interface). The enslaving of the velocity ﬁeld by the “thermal” one feeds forward on the bulk equations where the “energy” (Fourier or, alternatively, Fick) equation couples to the (NavierStokes) equations of motion. Literally, a strict enslaving of the velocity by e.g. the temperature occurs with heated layers of high Prandtl number ﬂuids like silicone oils. Fig. 1 illustrates the onset of cellular convection in one such liquid. It shows a disorderorder transition from a homogeneous, diffusion, motionless state to a ﬁnitewavenumber, cellular, and hence more ordered convective state. This is not the only possibility oﬀered by the dynamics or observed in experiment. Indeed, a long wavelength (zerowavenumber) convective mode is also possible, and may even coexist with the above mentioned case. The control, or bifurcation, parameter of the dynamics is the temperature diﬀerence or the surfactant concentration diﬀerence or gradient (e.g. across the layer) creating the surface tension gradient that leads to and sustains the “ordered” ﬂow. To study the (nonlinear) evolution of the cellular convection illustrated in Fig. 1 both the (Newton/NavierStokes) equations of motion and the energy equation are needed. However, this poses a formidable task. A drastic albeit signiﬁcant and universal approximation to the dynamics valid near and slightly above threshold is possible. Generic arguments permit one to reduce the problem to a set of LandauGinzburg equations for the amplitudes of the velocity and the thermal ﬁeld [9, 10]. This is indeed the natural extension of Landau’s approach to disorderorder equilibrium phase transitions where now the amplitudes of the cellular ﬂow are the enslaving modes or order parameters in the evolution. A notable phenomenon is the (unavoidable) appearance of defects in such cellular structures. The defect not only contains signiﬁcant information in itself but, moreover, contains the whole
M. G. Velarde
345
Figure 1. Benard cells. A shallow (mm depth) layer of silicone oil open to the ambient air is heated from the bottom side. As the temperature diﬀerence across the layer increases there is the transition from the motionless, heat diﬀusion phase to the cellular convective one. Once the critical temperature diﬀerence for the onset of convection is achieved the sequence proceeds with a constant value of this constraint. Velocities are about a hundredth of a centimeter per second. [Details can be found in Refs. 6, 8 and 11].
dynamics as it is known in crystal growth processes. The evolution of a defect, e.g. the screw dislocation is enough to grow the full crystal, perfect or otherwise. The reader is referred to several recent publications where indepth analyses are provided and due comparison with experimental observations given [8][11].
3
Patterns in (active, nonlinear) lattices
Cellular patterns arising from a disorderorder transition with a supporting (steady) inputoutput energy balance also appear in (active) lattices. Here (think about action potential transmission and pattern dynamics in brain activity or about arrays of lasers or collections of linked mechanical or electronic oscillators or the discrete version of reactiondiﬀusion systems) the equations of motion may be Newton’s equations (in a generalized sense) for the units considered as (nearestneighbor, n.n. in short) particles interacting with a given force (again in a generalized sense). The “activity” of the units and the role played by the bonds (springs, resistors, etc.) coupling the units necessitates the energy equation. Diﬀusion, noise, etc. provide the disorder when we consider not a purely deterministic case but a stochastic one. The latter is easy to visualize if we take the units to be immersed in a “heat” bath where the noise level (via Einstein’s ﬂuctuationdissipation theorem) also provides a temperature. Then the disorderorder transition could depend on the temperature as well as on the control or bifurcation parameter of the dynamics that drives the change from one mode of operation to another in the time and/or space evolution of the system. Such a bifurcation parameter may aﬀect only the “activity” of the units or the “interaction” between units or both. By “activity” of a unit I mean that if left alone it has an intrinsic evolution independent of the others. These units may all be the same, e.g. electronic oscillators with a rich panoply of dynamic states, from the simplest motionless one to bistability of steady states or oscillations, excitability or even timechaotic states. Through cou
346
Patterns and dissipative waves
Figure 2. Typical replication of a pattern (a) using raw material (b) following the mutual interaction of two “active” lattices (130×130 units). The two (identical) oﬀsprings, (c) and (d), almost perfectly reproduce the pattern but this need not be the case in general; combination/competition of patterns is also possible. The outcome of “evolution” depends on parameter values controling the “activity” of units and bonds/interactions, here between n.n. [Details can be found in Refs. 12 and 13].
pling and suitable synchronization patterns may appear in the lattice. Note that in a 2D, N×N lattice, if a unit possesses two accessible states, diﬀusive coupling, 2 a resistor in an electric circuit/lattice, there may be as many as 2N patterns, an extraordinary wealth indeed. Furthermore we can replicate a given pattern in a (father) lattice by dynamically “imprinting” it on a disordered (mother) lattice coupled to the former; the disordered state is also a possible steady pattern of the system [12, 13]. Fig. 2 illustrates two steps, initial and ﬁnal, of one such replication process with two identical oﬀsprings, almost cloning the father but this need not to be so. It is a dynamic process and not the mere action of a photocopying machine. We see again how signiﬁcant the role of defects may be in this replication process. Replicating a given pattern is in general a diﬃcult problem in the transfer process. Hence for purposes of information only it may be enough to replicate a defect in the original and nothing more. As a lattice and its dynamics with active units may be considered as a discrete version of a (continuous) reactiondiﬀusion system, the evolution of the “order parameters” (patterns) in many such problems can again be described, in approximate way, by a set of generic LandauGinzburg equations, much as in the case of cellular ﬂuid ﬂows described above.
4
Dissipative waves and solitons in ﬂuid layers
Earlier we illustrated the onset of (steady) cellular convective patterns occurring at the open surface of a heated liquid layer. As indicated in the caption of Fig. 1 the heating is done from the liquid side, i.e., from the solid support below the liquid, while the open atmosphere is (mechanically passive) air. [N.B. The dynamic or shear viscosity of air is much lower than that of most liquids, hence air exerts negligible traction at the open surface]. In fact, if the ambient atmosphere is another liquid, cellular patterns may appear for either way of heating albeit depending on the values of the ratios of the corresponding (heat or mass)
M. G. Velarde
347
diﬀusivities and kinematic viscosities [14]. [N.B. This is a signiﬁcant statement because the latter quantities control the diﬀusion of vorticity while the dynamic or shear viscosity comes in the stress balance at the interface and accounts for the relative traction of one liquid over the other (hence controlling momentum transfer)]. Restricting consideration to the case of a liquid layer open to air, what is the expected outcome of heating from the ambient atmosphere or cooling from the solid support? Theory predicts and experiments have conﬁrmed that various types of waves, with and without signiﬁcant surface deformation, transverse or longitudinal waves (and even internal waves in the otherwise stably stratiﬁed layer) can be excited and, eventually, sustained past threshold, again by the Marangoni “engine” operating at the open surface. The argument applies verbatim to the case of ad/absorption of a “light” surfactant that lowers the surface tension of the liquid in the layer. A surprising result is that some of the waves appearing at the open surface have solitonic features [11, 15]. For instance, as solitary waves or periodic wave trains (series of periodic wave peaks) they travel undeformed for long times, they cross/collide with each other with no appreciable change of form, they reﬂect at solid surfaces exhibiting MachRussell’s stems, or not, according to the incident angle (a phenomenon also occurring in wide/obtuse angle collisions). These solitonic waves are very much like surf waves or hydraulic jumps (bores in English and mascarets in French) moving (upstream) in certain rivers and shallow channels and straits. [N.B. Solitary waves were found in canals. They were beautifully described by Russell, around the midXIXth Century. He also proposed, in 1840, that solitary waves could be used as a clever way to traverse canals (like surﬁng) in boats whose keel lengths might be greater than the depth of the canal. However, solitons were ﬁrst described in the evolution of nonlinear lattices and their (quasi) continuum approximations around midXXth Century. Pioneering work by Fermi, Pasta and Ulam was followed by that of Zabusky and Kruskal (who coined the word soliton) and others]. Fig. 3 illustrates dissipative solitonic wave trains [16]. Here, the threshold for instability occurs in the long wavelength limit (zero wavenumber). In the energy balance the input of energy comes in this long wave range while dissipation (due to viscosity) occurs in the short wave range. In the zerowavenumber case the approximation to the equations of motion and the energy equation cannot be a set of LandauGinzburg equations because there is a Goldstone mode (the conservation of liquid does not permit lifting the surface all at once in a homogenous way). The natural approach is that opened by Boussinesq and, subsequently, taken by others leading to a wave equation for the corresponding order parameters (e.g., for a onesided propagation the BoussinesqKortewegde Vries equation), augmented by the energy balance [11]. Mathematically, the original hyperbolic equations are then modiﬁed by (super) parabolic terms (e.g., the dissipation modiﬁed BKdV equation). Hence their solutions have been called dissipative solitons [17] as they are the generalization of the soliton concept introduced in conservative (hyperbolic) systems (for ﬂuids those obeying the Euler equations and not the NavierStokes equations). As in the earlier cases discussed here there is a (nonequilibrium) transition from a disordered state, the motionless case or even a ﬂow regime, to
348
Patterns and dissipative waves
Figure 3. Typical dissipative periodic wave trains (like cnoidal waves) whose wave peaks/crests have been shown to exhibit solitonic features. The experiment corresponds to heating the liquid layer from the ambient air side or ad/absorbing a “light” surfactant, e.g., pentane vapor ad/absorbed by liquid toluene or hexane. Velocities are about a centimeter per second, hence two orders of magnitude higher than the velocities in Fig. 1 corresponding to steady convection. [Details can be found in Refs. 11 and 16]
a solitonic wave regime, the more ordered state, sustained by the steady inputoutput energy balance.
5
Dissipative solitons in (active) lattices and superconducting circuits
In the ﬁnal section of this short text let me describe what can occur in a lattice of “active” units if we take ions as the units, electrons as “boats”, and an electric ﬁeld as the “wind” and/or powerful horses, all immersed in a heat bath. Can an electron surf on a nonlinear lattice of strongly interacting (positive) ions? What follows is a rough account of work done with W. Ebeling and A. P. Chetverikov [3, 4]. Toda showed that a 1D lattice with a suitable exponential (hence nonlinear) interaction between n.n. units of the form [18] U T (r) =
a br e¯ − 1 + ar; a, b > 0, b
(1)
has an exact solution in the form of solitonic (cnoidal or solitary) waves. The force
M. G. Velarde
349
F (r) = a e¯br − 1 or else Fn = a e¯brn − 1 .
(2)
corresponding to (1) is
Thus for positive r the force is attractive (and constant) while for negative r it is repulsive (and very much like hard rods when b is large and a → 0; this is the case of strong compression of the spring deﬁned by (2)). [N.B. A more realistic interaction is the Morse potential, a combination of two exponentials that treats adequately the attractive component and ﬁts well the LennardJones interaction. However, we have no exact solutions of the corresponding equations of motion. Both the Toda and Morse potentials are easy to implement electronically]. The Taylor expansion of (1) yields the (cubic and quartic) anharmonic interactions used by Fermi, Pasta and Ulam in their pioneering computational physics work (with the earliest computer available at the Los Alamos Labratory) and of current use in solid state theory to explain equipartition, thermal expansion and heat transfer in a lattice (for b small the compressionexpansion obeys Hooke’s law and the motions are harmonic vibrations or phonons in the lattice). [N.B. The pure Toda lattice is transparent to heat but here we use more than the pure Toda interaction]. The solution found (constructed) by Toda is (using the force) e¯brn − 1 = sinh2 κ sec h2 (κn − βt + δ) ,
(3)
hence deﬁning a solitary wave propagating supersonicaly with velocity c=
β y=y κ
ab m
1/2
sinh κ > κ
ab m
1/2 ,
(4)
where y accounts for the mean interparticle distance between the units (of mass, m). For a 1D lattice of length L, with periodic b.c. xi+N = xi + L we have U=
N &
UiT (ri ),
(5)
i=1
with UiT given by (1) with r → ri − σ, ri = xi+1 − xi , σ > 0. Then the (Newton) equations of motion of the units are m
dvk ∂U + = 0. dt ∂xk
(6)
To these deterministic and Hamiltonian equations we shall add an energy equation after having also added the following: (i) the consideration of the units as Brownian particles in a heat bath (white Gaussian noise, delta correlated in space discrete units and time), (ii) as Brownian particles they experience dissipation/friction and ﬂuctuations and, eventually, they are able to pump energy, ﬁrst internal and, subsequently, kinetic energy/motion from the (disordered) noise, (iii) the units are charged particles, e.g., ions with positive charge (+e), and (iv) the lattice/system
350
Patterns and dissipative waves
is subjected to an external electric ﬁeld, E. Thus the equations of motion (6) become stochastic (Langevin) equations, m
√ dvk ∂U + = ek E + F (vk ) + 2Dξk (t), dt ∂xk
(7)
where the force F accounts for two terms. We set F (vk ) = −mγ(vk )vk , with γ = γ0 + γ1 . Here γ0 is the standard (ﬂuctuating) friction coeﬃcient obeying Einstein’s ﬂuctuationdissipation theorem. The other term, γ1 , is taken to be nonﬂuctuating and with values at our “will” in view of the energy balance to be satisﬁed in the evolution of the system. Let us introduce now a particular form of the energy balance [19]. Each Brownian particle is considered as “active”, capable of pumping internal energy from the bath (noise), dissipating internally and transforming internal energy into motion, hence feed forwarding the equations of motion. [N.B. We are now exploring the possibility of direct energy pumping by “passive” (γ1 = 0) particles from the electric ﬁeld alone taken indeed as “wind”]. Let (t) be the energy, then we may write d = q(t) − c(t) − d(v)(t), (8) dt which is reminiscent of metabolic processes in biosystems. The quantity q(t) accounts for the energy ﬂux from the bath to the unit’s “internal” depot. The internal dissipation is assumed proportional to the actual energy level with a constant rate of energy loss, c. The third term describes how internal energy is converted into motion, with a rate depending on the velocity value, d(v), and indeed proportional to the actual energy level. For simplicity, we assume a steady state, q = q0 , and d(v) = dv 2 , with d constant and positive. Hence if at the initial time (0) = 0 we have 0 = q0 / c + dv 2 . (9) Then we should write γ1 = −q0 d/ c + dv 2 . However, to further simplify we shall use γ1 ≈ γ1R ≈ γ0 v 2 , which is a function introduced by Lord Rayleigh [20] that approximates γ1 well for negative values, when γ1 acts opposite to γ0 , as an energy pumping (negative friction) mechanism. The two quantities γ1 and γ1R diﬀer drastically for positive values where γ1 saturates while γ1R diverges when v is large. Hence we write v2 F (v) = mγ0 v µ − 2 , (10) v1
where for convenience we have rewritten γ1 in terms of two new parameters µ and v1 . The former, µ, is now the key parameter of the problem. Indeed for µ = 0 the montionless state is the only stable phase of the system that for µ positive yields a √ new state with motion (expected to be the more ordered phase), v = ±v0 = v1 µ. To the 1D lattice of ions we now couple a 1D lattice of electrons. For simplicity we take only one electron (e) whose mass is a thousand times smaller than that of the ions (hence me γ0e kc (the critical wall fugacity) the situation can be resumed as follows: i) a unique free energy minimum at z = 0 (for k = 2.6100 > kc ) which delocalizes into a wide one at k = 2.6075 kc (i.e. a continuous unbinding transition in the pure component wedges); ii) double minima proﬁles for k = 2.6050, 2.6025 kc with the minima placed at z = d1 −d and z = d1 (i.e. coexistence between the state localized at the SW central ridge height and the bulk unbound state); iii) dominance of bulk unbound state (k = 2.6000 > kc ). This scenario supports an interface unbinding in two steps, with a ﬁrst–order unbinding from the SW central ridge.
Figure 3. Average interface distances (black dots) from 3D SW central ridge (panel (a)) and bottoms (panel (b). The data come from Metropolis Monte Carlo simulations at constant t = 2.0 and system sizes L = 100, W = 50, wall unit slope and d1 /d = 2 (i.e. d = 16). In the insets is the detail of interface distance distributions at the u values marked by the dashed lines
G. Giugliarelli
363
Surprisingly, such a mechanism, as shown in Fig. 3, is fully conﬁrmed by our Monte Carlo simulations for 3D SWs. In fact also in this case we have a discontinuous unbinding of the interface from the central ridge. The only diﬀerence is between 2D and 3D problems seems to be related to the fact that in 3D continuous and discontinuous unbindings seem to be separated by a ﬁnite gap. Our investigation demonstrate that interface unbinding in SWs involves also a discontinuous transition. In this respect, this result opens new possible extensions to the wetting problem of random rough surfaces. In particular, our arguments give some more eﬀorts on the key role of surface roughening exponent ζS in determining the nature of the wetting transition in self–aﬃne rough substrates.
References [1] S. Dietrich, in Phase Transition and Critical Phenomena, edited by C. Domb and J. L. Lebowitz (Academic Press, London, 1988), vol. 12, p. 1; G. Forgacs, R. Lipowsky, and T. M. Nieuwenhuizen, in Phase Transition and Critical Phenomena, edited by C. Domb and J. L. Lebowitz (Academic Press, London, 1991), vol. 14, p. 135. [2] P. Pfeifer, Y. J. Wu, M. W. Cole, and J. Krim, Phys. Rev. Lett. 62, 1997 (1989); M. Kardar and J. O.Indekeu, Europhys. Lett. 12, 161 (1990); A. O. Parry, A. J. Wood, and C. Rasc` on, J. Phys.: Condens. Matter 12, 7671 (2000); C. Rasc` on and A. O. Parry, Nature 207, 986 (2000); L. Bruschi, E. Carlin, and G. Mistura, J. Phys.: Condens. Matter 15, S315 (2003). [3] A. O. Parry, C. Rascon, and A. J. Wood, Phys. Rev. Lett. 83, 5535 (1999); [4] G. Giugliarelli and A. L. Stella, Phys. Rev. E 53, 5035 (1996); G. Giugliarelli and A. L. Stella, Physica A 239, 467 (1997); A. L. Stella and G. Sartoni, Phys. Rev. E 58, 2979 (1998); G. Giugliarelli and A. L. Stella, J. Phys.: Math. Gen. 32, 5409 (1999). [5] G. Giugliarelli, submitted to Phys. Rev. Lett. (2004); G. Giugliarelli, to be published; [6] D. B. Abraham, Phys. Rev. Lett. 44, 1165 (1980).
Part V
New Approaches to Physics Teaching
LEARNING PROBLEMS RELATED TO THE CONCEPT OF FIELD
a
a
FRANCESCA BRADAMANTE, MARISA MICHELINI, a ALBERTO STEFANEL
Research Unit in Physics Education University of Udine, Italy
Abstract The concept of ﬁeld has formed itself through a process of elaboration of concepts both on the historical level, and on that of mathematic formalism (integral or diﬀerential approach, Lagrangian and Hamiltonian formalism), and in the relationship between mathematics and physics. In didactic tradition it is used as an example of the formalised analogical process. Its phenomenological aspects and its characteristics of time and space are only partially described for the various types of ﬁeld, mixing the descriptive and interpretative levels. In this review we select some of the main teachinglearning problems and students’ diﬃculties with learning the concept of ﬁeld, that many researches have stressed, such as: the concept of ﬁeld and its representation by lines of force, the diﬀerence between ﬁeld and force, the connection between ﬁeld and its sources, the concept of the ﬁeld’s superposition and the interpretation of mathematical formulas, the application of the third principle of dynamics, the motion of the ﬁeld’s characteristic particles in the ﬁeld itself. We consider the ﬁeld in static situations, because these learning problems are fundamental and basic for when we will pass to dynamic situations (electromagnetic ﬁeld).
1
Introduction
The concept of ﬁeld has formed itself through a process of elaboration of concepts both on the historical level and on that of mathematics formalism, and in the relationship between mathematics and physics. This concept is central in the
367 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 367–379. © 2006 Springer. Printed in the Netherlands.
368
Learning problems related to the concept of Field
description of the real world that physics gives nowadays. It is of fundamental importance when dealing with interaction processes, both in classic physics and in quantum physics, even if with diﬀerent meanings. The extension of the Lagrangian and Hamiltonian formalisms in systems with a limited or deﬁnite number of levels of freedoms in dealing with continuous systems provides the formal tools for a uniﬁed treatment of classical ﬁelds and for their following quantization. The concept of ﬁeld is no longer used only when dealing with interaction mediators, even in classical physics. It is more often used to describe physical quantities in relation to space and time within phenomenologies where local variations of physical dimensions are followed by global variations of the same quantities or of connected ones. The uniﬁed treatment of ﬁelds gives a uniﬁed view of the two diﬀerent formal approaches that have been historically developed: the diﬀerential one, starting from the mathematic apparatus of ﬂuiddynamics and of physics of continuous systems based on the use of curl and divergence, and that of diﬀerential geometry, which uses integraldiﬀerential formalism. The ﬁrst approach refers to many instruments which are typical of plane and geometry (curves and vectors), which initially facilitate the understanding for a broader public, but that have later a complex formal use in the forward developments and cause wrong local visions, inhibiting or strongly contrasting the formation of a global vision of the matter. The second approach is developed with a less common formalism and is therefore thought to be more complex, but it is actually more complete, general and elegant than the ﬁrst. The concept of ﬁeld can also be built without having to use other notions such as “force”. Both approaches, in order to be fully developed, require mathematical instruments that are usually acquired during the ﬁrst two years of a degree course in a scientiﬁc subject, but in all cases remain references for dydactic approaches of ﬁeld concept also in secondary school. Historically [1, 2, 3] ﬁeld lines and ﬁeld concept, as “ﬁeld of force”, were introduced in 1845 by Faraday in the debate of action at a distance [4]. According to Faraday the propagation of forces takes place through a medium and not at a distance. Also, lines of force have a physical reality and represent the structure of material substances and their interactions. According to Maxwell “Faraday . . . saw lines of force traversing all space where the mathematicians saw centres of force, . . . saw a medium where they saw nothing but distance . . . ” [5]. Afterwards, Faraday’s concept of ﬁeld was adopted and changed by Thomson and Maxwell. The ﬁrst gave a mathematical representation of Faraday’s theory of magnetic ﬁeld and developed a theory on the “ethereal continuum” as a physical expression of Faraday’s “plenum of forces”. Maxwell, as well as completing the electromagnetic theory with his equations, formulated the ﬁrst clear deﬁnition of a ﬁeld, giving it the role of mediator of the interactions between systems. The ﬁeld of forces was seen as a continuous or discrete system formed by interacting particles (the ether) or as a spatial distribution of forces. Maxwell also elaborated a “geometrical model” of ﬁeld, in which Faraday’s lines of force constituted a purely geometric representation of the ﬁeld’s structure. A conﬁrmation of Maxwell’s theory came from Hertz’s experiments, which allowed to directly highlight the propagation of electromagnetic waves. Lorentz described
F. Bradamante, M. Michelini, A. Stefanel
369
the interaction between ﬁelds and matter, separating the concepts of ether and matter, ﬁnally setting up a theory that developed a universal physics based on electromagnetism. With the ridding of the ether and the discovery of the quantum nature of the electromagnetic ﬁeld (Einstein, 1905) it was possible to pass from classical physics to modern physics, based on the quantum theory, and reach a modern concept of ﬁeld. The theory of ﬁeld brought a redeﬁnition not only of the physical model, but also of the meanings and the role given to the ﬁeld, and it also drew attention on the need to acquire a unitary vision of the ﬁelds. The study on learning problems, on mental models and on the students’ reasoning patterns is carried out considering these two problems: the historical point of view and the mathematical formalism. As shown in literature, among the pupils’ learning knots [6], there are often the ones1 that scientists encountered in the process of building physics theories: they have a privileged position in oﬀering diﬀerent and partial representations, which are presented in an alternative and separate way to the physical meaning of the entity. The concept of ﬁeld is one of these. In mathematics it has an autonomous nature. In physics, it is reinterpreted with a role of synthesis and generalization, referring to diﬀerent types of ﬁeld and a complex nature in relation to various phenomenologies. It is a typical physical entity2 , object of interest also from the point of view of disciplinary foundations. Considered by physicists themselves only a formal (mathematical) object, an object/physics entity (for example the electromagnetic ﬁeld), a property of a physical system (for example the ﬁeld of the velocity in a ﬂuid), a property of space or a structure or a tool for describing interaction [7, 8], it occupies a privileged position among the basic concepts for the building of physics culture. In didactic tradition it is used as an example of the formalised analogical process. Its phenomenological aspects and its characteristics of time and space are only partially described for the various types of ﬁeld, mixing the descriptive and interpretative levels3 . The local characters are never discussed with the general ones and properties are presented depending on the type of ﬁeld, without giving a role to the compared physical meaning of formal entities, which have the unifying power of the ﬁeld concept (ﬁeld lines, potential, ﬂux, circuitation, etc.). As happens for other abstract and/or general concepts4 , which recall the human toil needed to acquire familiarity with their representative power, many ignore it
1 The diﬃculties children meet obviously are not only the historical interpretative knots or those unifying concepts which single out new physical entities: there are also various cognitive aspects, which complicate the problem. 2 The nature of the concept of ﬁeld summarises physics’ modalities in the interpretative processes. 3 The confusion between the descriptive and interpretative levels is a typical problem of formal didactics and a lot of proof may be found in research on learning processes [22] 4 The case of fractions in maths is famous: though handled with ease by primary school children, they are a complex learning problem at the ages of 1416 [7].
370
Learning problems related to the concept of Field
or consider it in a reductive way in school practics, while even younger children have intuitions that allow to deal with the physical meaning at a high level of abstraction and symbolic representation [9]. Our research aims at building didactic proposals that precociously introduce the concept of ﬁeld, giving a modern vision of physics, able to overcome learning problems or avoid the obstacles in learning that many researches have shown and that are encountered also by university students [10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21]; for example those relating to the concept of ﬁeld and its representation by lines of force, the diﬀerence between ﬁeld and force, the connection between ﬁeld and its sources, the concept of the ﬁeld’s superposition and the interpretation of mathematical formulas, the application of the third principle of dynamics, the motion of the ﬁeld’s characteristic particles in the ﬁeld itself. In this paper we present a critical overview of the learning problems related to the concept of ﬁeld, which have been pointed out in literature. These problems, together with the analysis of the fundamental principles of the concept of ﬁeld, both on the historical and on the formal level, allow us to consider the relationship between science and mathematics and to discuss the meaning of the concept of ﬁeld, putting in correlation the knowledge problems with the conceptual aspects. Here we take into consideration only static ﬁelds; however, the learning diﬃculties and problems students encounter and that we analyse concerning the ﬁeld concept foundations, are therefore important also when considering dynamic situations (electromagnetic ﬁeld).
2
Learning problems related to the concept of ﬁeld
In this work we select some of the main problems and diﬃculties students encounter when learning the concept of ﬁeld, which have already been pointed out in literature. This review, concerning only static ﬁelds, is the base for future researches on the construction of learning proposals for teaching and learning the ﬁeld as a unifying concept in static and dynamic contexts, that can be proposed from primary school to secondary and university level. 2.1 THE ROLE OF ACTION AT A DISTANCE IN OPPOSITION TO THE CONCEPT OF FIELD AS A MODIFICATION OF SPACE POINTS The explanation of interaction as action at a distance has always been problematic, either on historical level [1, 2, 3] or common sense one [16, 23]. The action at a distance concept implies some presuppositions that causes learning diﬃculties in the comprehension of ﬁeld concept. These presuppositions, as some researchers pointed out [23] in their studies with pupils of 918 age years old, are: the idea that the interaction is instantaneous, the need for a connection between the objects interacting at a distance (many pupils expressed a need for air as a conducting medium for magnetic and gravitational attraction [16, 23]), the need for a medium for the transmission of interaction, how various forces support and focus each
F. Bradamante, M. Michelini, A. Stefanel
371
other, and the idea of the ﬁeld’s action as a limited region of space. Bar [23, 24] analyzed children’s conceptions about “gravity” and found that they considered the atmosphere as the limit of the Earth’s gravitational attraction; while Borges’ and Gilbert’s studies [15] reveal one of the models of magnetism as a “cloud” or a “ﬁnite region of inﬂuence”. On the other hand a completely diﬀerent layout is to consider the diﬀerent phenomenologies, interpreted up to now in terms of action at a distance, in a unifying framework where the concept of ﬁeld acts as a cognitive organizer. The analysis of diﬃculties of students in learning the concept of ﬁeld, have been the bases for some proposals [17, 18, 25, 26] for the introduction of this concept in Secondary Education and High School. The purpose of these proposals is to overcome some learning problems, as: 1) the students’ idea of ﬁeld as a limited space region or a volume; 2) the students’ identiﬁcation of the ﬁeld with the force and not with a physical entity and a modiﬁcation of space which allow interactions; 3) the students’ diﬃculty in understanding the diﬀerences between Newtonian conception of interactions and the conception of interaction described by ﬁelds. 2.2
GRAPHIC REPRESENTATIONS OF A FIELD AND THE ROLE OF FIELD LINES
Here are some learning problems related to ﬁeld representation by ﬁeld lines [13, 14, 27]: • Interpretation of ﬁeld representation by ﬁeld lines [13, 15, 27] • Objectualization of ﬁeld lines [13, 15, 27] • Diﬃculties to recognize diﬀerences and to integrate alternative representations of ﬁelds (ﬁeld lines, equipotential surfaces. . . ) [28, 29] • Interpretation of the representation of objects in a ﬁeld and situations involving ﬁelds. The concept of ﬁeld and ﬁeld lines are sources of confusion among students even at university level. Tornkvist, Pettersson and Transtromer in a research on the electrostatic ﬁeld [13] have shown that university students dont treat ﬁeld lines using arguments in terms of mathematical concepts (uniqueness and continuity, proportionality and isomorphic mapping). In fact, students often consider ﬁeld lines as isolated entities in the Euclidean space, not as a set of curves representing a vectorial property of that space; ﬁnally they don’t fully understand the hierarchical sequence between the concepts, for instance: charge geometry, ﬁeld line, force vector, velocity vector, trajectory (ﬁg.1). The authors suggest that this confusion by representation is, at least partly, the cause of the wellknown conceptions students have on the concept of force, reported by Johansson and Viennot [10].
372
Learning problems related to the concept of Field
Figure 1. From [13]
Some other studies [29] on the graphical representation of electric and magnetic ﬁelds, suggest to not limit to ﬁeld lines but also to draw the orthogonal equipotential surfaces. The authors believe that in such a ﬁeld diagram it is possible not only to clearly distinguish the distribution of the “ﬂux sources”, but also that of the “circulation sources”. Maxwell already used lines of force and “level surfaces” in all his graphical representations of ﬁelds, with and without circulation [5] (ﬁgure 2). In the case of the gravitational ﬁeld there are some studies that investigate the way pupils envisage the Earth, its shape and its relationship with the direction of the gravitational ﬁeld [9, 23, 24, 30, 31, 32, 33, 34]. In the 80’s Nussbaum [30, 31] made a ﬁrst classiﬁcation of children’s ideas about Earth, ﬁnding three essential elements: the Earth’s shape, the feature of the sky and the space, and the direction of falling objects from diﬀerent locations of the Earth. We found this type of classiﬁcation in Vosniadou’s studies about conceptual change and teaching of science [32, 33], where she classiﬁed Mental models of the Earth. For the author (Vosniadou) the representation of Earth as ﬂat is based on the assumption that the Earth is a physical body, and children must understand that Earth is an “astronomical” rather than a “physical object”. In particular she observes that in the area of astronomy students understand the spherical shape of the Earth only after they have acquired an elementary notion of gravity; and that it seems particularly important to teach children something about gravity in order to understand how people can live on a spherical, rotating Earth. On the other hand, Arnold [34] investigates the development of children’s concept of the Earth and the direction of the gravitational ﬁeld, and the relationship between the development sequence of children’s drawings of the Earth’s shape and
F. Bradamante, M. Michelini, A. Stefanel
373
Figure 2. Maxwell’s diagram of the superposition of the magnetic ﬁeld of a linear current and a homogeneous magnetic ﬁeld, in [5] quoted by [29]
the quality of their drawings of human ﬁgures. 2.3
THE FIELD AND ITS SOURCES: THE NATURE OF FIELD SOURCES; FIELD SUPERPOSITION
Even at university level, students don’t identify correctly magnetic ﬁeld sources, and most of them have diﬃculties in the distinction between electrostatic and magnetic interaction, i.e. they consider an electrostatic charge as a source of magnetic ﬁeld [20, 35]. Students from 13 to 20 years old identify the magnet as a source of magnetic ﬁeld, and motivate it considering its magnetic nature; but they think that the objects interact diﬀerently with it because of the diﬀerence of their electrostatic charge [35]. Borges Gilbert has investigated the ideas of students from 15 to 18 years old about the sources of the magnetic ﬁeld and its action on objects [15]; they classify the answers in ﬁve categories: the models of magnetism. Viennot and Rainson, while investigating university students’ reasoning on electrostatics identify two main problems: “ﬁeld only if mobility” and “cause in the formula” [11, 12]. The ﬁrst one means that students have diﬃculty in accepting the existence of an electric ﬁeld in a medium where charges are motionless. Students are not able to understand the diﬀerent behaviours of conductors and insulators [11, 36] and to recognize the possible presence of an electric ﬁeld in insulators. The second one (“cause in the formula”) is a result of a questionnaire on Gauss’s Theorem [12] and consists in ignoring the sources of the ﬁeld that are not represented by their symbol in the formula which expresses the ﬁeld. In these researches the authors describe a teaching sequence for the superposition of electric ﬁelds, which
374
Learning problems related to the concept of Field
Figure 3. From reference [19, 20]
is a prerequisite for a basic understanding of electrostatics and for a uniﬁed view of electrostatics and electric circuits that can be fostered by teaching strategies, which emphasize causal aspects and transient phases. The authors suggest that the principle of ﬁeld superposition is not obvious to students, and that it is useful to work on it in static situations before analysing electric circuits. 2.4 THE DYNAMICS OF INTERACTIONS MEDIATED BY A FIELD Researches show that university students do not have a complete comprehension of the following: • the objects on which a magnetic ﬁeld can act, for example they think that a magnetic ﬁeld can interact with electrostatic charges and can move them [19, 20] ( Fig.3) • the application of the third Dynamics Law and its applicability in contexts other than the mechanics one. At every level from primary school [23, 24] to university [14, 19] students are not able to recognize that interaction is reciprocal in the diﬀerent ﬁelds: the gravitational one [16, 23, 24, 37, 38], the electric one [18, 19, 25], the magnetic and electromagnetic one [14, 19, 20]. Most of them consider the force only in one direction: this shows the importance to teach force concept not as a cause that produces an eﬀect but as an interaction (reciprocal). The motion of the ﬁeld’s characteristic particles in the ﬁeld itself implies learning problems on: • the recognition of the role of initial conditions when the particle is moving in a ﬁeld, • the distinction between trajectories of the particles and ﬁeld lines [13, 14, 18, 20] (Fig.4) Other studies [12, 20] reveal that at university level students have some learning problems related to the application of the magnetic force law: F = qv × B, due mostly, for the authors, to the incomprehension of the vectorial product.
F. Bradamante, M. Michelini, A. Stefanel
375
Figure 4. From reference [13]
2.5
CORRELATION BETWEEN STUDENTS’ CONCEPTIONS
GRAVITY
AND
MAGNETISM
IN
Among the diﬃculties pointed out in research literature concerning these two types of interaction, in the last twenty years several studies [16, 23, 24, 34, 37, 38] have investigated children’s commonsense ideas about gravity and its relationship with magnetism, and reveal that students (918 years old) identify gravitational eﬀects with magnetic ones, considering the gravitational attraction as a magnetic one. They also connect gravity and magnetism on the Earth, considering gravity as necessary for a magnetic ﬁeld on the Earth, a uniqueness of the Earth’s system; in fact some children think that “the magnet cannot function without gravity”, while others aﬃrm that “electrostatic force cannot function without gravity”, founding in that case a link between gravity and electrostatics.
376
Learning problems related to the concept of Field
2.6 THE STUDENTS’ IDEAS OF MAGNETISM DESCRIBED BY ELECTROSTATIC EFFECTS Many studies [15, 20, 21] reveal some diﬃculties in the comprehension of the magnetic ﬁeld at university or during the last years of secondary school and the confusion of magnetism with electrostatics: • the description of the magnetic ﬁeld using electric ﬁeld representations [20] • models of magnetism in terms of electrostatic eﬀects [15] Some authors [23] consider that the “link between electricity and magnetism should be emphasized in education, enabling pupils to understand Ampere’s explanation of magnetism. This relates magnetism to microcurrents within particles of matter and electromagnetic waves.”
3
Conclusions
The analysis of ﬁeld concept foundations, on the historical and formal level, and also the analysis, on the didactical level, of teachinglearning problems and schemes of reasoning, allow us to consider on one hand the relationship between science and mathematics, together with the role of formalism in giving interpretative models of physics, and on the other hand it lays down the problem of the meaning of this concept, both from the scientiﬁc point of view and from that of the students’ conceptions. The learning problems are related to knowledge problems of the concept of ﬁeld (from the mathematical and historical point of view), thus they have to be considered together. For example, the problem of identiﬁcation of the ﬁeld’s lines with something “real” (their objectualisation) can be found both in students and in children, who give a physical meaning to the ﬁeld lines, thus getting very near to Faraday’s concept. Can the ﬁeld be considered as a property of space? Or should it be considered as a ﬁeld of force momentum, or as a system? Or does it allow us to deﬁne the structured properties of a system? And ﬁnally, is it correct to observe all ﬁelds in only one way? From the point of view of mathematics, one could answer aﬃrmatively, since scalar ﬁelds are a subclass of vectorial ﬁelds; however, from the physical point of view the answer is more diﬃcult and questionable. In light of the above, it is necessary to interpret the ways in which students and scientists consider this concept, and how they are associated to ideas and to the capacity to represent the concept dealt with. In view of a cognitive research for the development of a didactical proposal a question that could therefore be asked is: which are the representations that are triggered when the concept of ﬁeld is explained by using an approach based on ﬂuids’ mathematics? Which are instead the ones when the approach adopted is based on diﬀerential geometry? What are the children’s analogical representations according to the diﬀerent approaches? The problem here is that of knowledge and of the building of knowledge itself, together with the didactic problem. If we limited ourselves to the present requests
F. Bradamante, M. Michelini, A. Stefanel
377
of secondary school (applicative science conception), we could adopt a conceptual model of the nineteenth century. But if we want to give a unitary description of physical phenomena and a cognitive organisation of the diﬀerent parts of physics, such as mechanics and electromagnetism, then it is necessary to adopt a unitary model (scientiﬁc competence conception) ﬁnding out new modalities to introduce in a simple way otherwise complex formalism. Whatever choice is made from a didactical point of view, it is necessary to carry it on rigorous way, both with regard to the choice of the model and with regard to the formalization of the model itself, highlighting its qualities and limits. This work has only taken into consideration the fundamental problems related to static ﬁelds. It is a preliminary work for future researches concerning the possible teaching proposals on ﬁeld as a unitary and unifying concept, starting from elementary school and carrying on until secondary school and university.
References [1] M.B. HESSE, Forces and Fields, the Concept of Action at a Distance in the History of Physics, Thomas Nelson and Sons Ltd., Edinburgh (1961). [2] M. JAMMER, Concepts of Force: A Study in the Foundations of Dynamics, Harvard University Press edition (1957). [3] E.J. DIJKTERHUIS, Il meccanicismo e l’immagine del mondo, pp 640643, Feltrinelli (1980). [4] M. FARADAY, Experimental Researches in Electricity, Richard Taylor and William Francis, Printers and Publishers to the University of London, Vol I, II, III. (1855). [5] J.C. MAXWELL, A Treatise on Electricity and Magnetism, Dover Publications (1954). [6] R. DUIT, Constraint on knowledge acquisition and conceptual change: the case of physics, Paper presented at the Symposium on Constraints on Knowledge Construction and Conceptual Change. Sixth European Conference for Learning and Instruction, Nijmegen (1995). [7] P. GUIDONI, private communication (2003). [8] R. GIANNITRAPANI, private communication (2004). [9] F. BRADAMANTE, M. MICHELINI, Children’s ideas about gravitation, investigating a model of gravitational ﬁeld, Proceedings of Girep, Ostrava (2004). [10] L. VIENNOT, Spontaneous reasoning in elementary dynamics, Eur. J. Sci. Educ. 1, pp 205222 (1979). [11] R. RAINSON, L. VIENNOT, Design and evaluation of a researchbased teaching sequence: the superposition of electric ﬁeld, Int. J. Sci. Educ. 21(1), pp 116 (1999). [12] R. RAINSON, L. VIENNOT, Students’ reasoning about superposition of electric ﬁelds, Int. J. Sci. Educ. 14(4), pp 475487 (1992). [13] S. TORNKWIST, K.A. PETTERSSON, G. TRANSTROMER, Confusion by representation: On student’s comprehension of electric ﬁeld concept, Am. J. Phys. 61(4), pp 335338 (1993). [14] I. GALILI, Mechanics backgroud inﬂuences students’ conceptions in electromagnetism, Int. J. Sc. Educ. 17(3), pp 371387 (1995). [15] A.T. BORGES, J.K. GILBERT, Models of Magnetism, Int. J. Sc. Educ. 20(3), pp 361378 (1998).
378
Learning problems related to the concept of Field
[16] I. GALILI, Weight versus gravitational force: historical and educational perspectives, Int. J. Sc. Educ. 22(10), pp 10731093 (2001). [17] J. MARTIN, J. SOLBES, Diseno y evaluacion de una propuesta para la ensenanza del concepto de campo en ﬁsica, Ensenanza de las Ciencias 19(3), pp 393403 (2001). [18] C. FURIO, J. GUISASOLA, La ensenanza del cocepto de campo electrico basada en un modelo de aprendizaje como investigacion orientada, Ensenanza de las Ciencias 19(2), pp 319334 (2001). [19] D.P. MALONEY, T.L. O’KUMA, C.J. HIEGGELKE, A. Van HEUVELEN, Surveying students’ conceptual knowledge of electricity and magnetism, Phys. Educ. Res.,Am. J. Phys. Suppl. 69(7), pp S12S23 (2001). [20] J. GUISASOLA, J.M. ALMUDI, M. CEBERIO, Concepciones alternatives sobre el campo magnetico estacionario. Seleccion de cuestiones elaboradas para su deteccion y tratamiento, Ensenanza de las Ciencias 21(2), pp 281293 (2003). [21] J. GUISASOLA, J.M. ALMUDI, J. ZUBIMENDI, Diﬃculties in Learning the Introductory Magnetic Field Theory in the First Years of University, Wiley Periodicals, Inc. pp 443464 (2004). [22] S. BOSIO, V. CAPOCCHIANI, M. MICHELINI, J.S. PUGLIESE, C. SARTORI, M.L. SCILLIA, A. STEFANEL, Playing, Experimenting, Thinking: Exploring Informal Learning Within An Exhibit Of Simple Experiments, In New Way For Teaching, Girep Book, Ljubljana University Press (1997). [23] V. BAR, B. ZINN, E. RUBIN, Children’s ideas about action at a distance, Int. J. Sci. Educ. 19(10), pp 11371157 (1997). [24] V. BAR, B. ZINN, GOLDMUNTZ, C. SNEIDER, Children’s concept about weight and free fall, Science Education 78(2), pp 14969 (1994). [25] C. FURIO, J. GUISASOLA, Concepciones alternatives y diﬁcultades de aprendizaje en electrostatica. Seleccion de cuestiones elaboradas para su deteccion y tratamiento, Ensenanza de las Ciencias 17(3), pp 441452 (1999). [26] C. FURIO, J. GUISASOLA, The teaching of the concept of electric ﬁeld: mountain or hill, Proc. Esera Conf., Kiel (2001). [27] E.R. SAVERLSBERG, T. DE JONG, M. FERGUSONHASSLER, Situational Knowledge in Physisc: the case of Electrodinamics, J. Res. Sci. Teach. 39(10), pp 928951 (2002). [28] I.M. GRECA, M.A. MOREIRA, The kinds of mental rapresentationsmodels, propositions and imagesused by college physics students regarding the concept of ﬁeld, Int. J. Sci. Educ. 19(6), pp 711724 (1997). [29] F. HERRMANN, H. HAUPTMANN, M. SULEDER, Representations of electric and magnetic ﬁelds, Am. J. Phys. 68, pp 171174 (2000). [30] J. NUSSBAUM, J.D. NOVAK, An assessment of children’s concepts of earth utilizing structured interviews, Science Education 60, pp 535550 (1976). [31] J. NUSSBAUM, The Earth as a cosmic body Children’s Ideas in Science, Driver Gnesnes Tibershien (Eds) Open Univ Press Miltan Keynes, chapter 9 (1985). [32] S. VOSNIADOU, W.F. BREWER,, Mental Models of the Earth: A study of conceptual change in childhood, Cognitive Psychology 24, pp 535585 (1992). [33] S. VOSNIADOU, Conceptual change Research and Teaching of Science, h. Behndt et al. (eds) Research in Science Education Past, Present, and Future, Kluwer Academic Publischers Printed in the Netherlands, pp 177188 (2001). [34] P. ARNOLD, A. SARGE, L. WORRALL, Children’s knowledge of the earth’s shape and its gravitational ﬁeld, Int. J. Sci. Educ. 17(5), pp 635641 (1995). [35] A. MAAROUF, S. BENYAMNA, La construction des sciences physiques par les representations et les erreurs: cas des phenomenes magnetiques, Didaskalia, 11, pp 103120 (1997).
F. Bradamante, M. Michelini, A. Stefanel
379
[36] J. PARK, I. KIM, M. KIM, M. LEE, Analysis of students’ process of conﬁrmation and falsiﬁcation of their prior ideas about electrostatics, Int. J. Sci. Educ. 23(12), pp 12191236 (2001). [37] S. RUGGIERO, A. CARTELLI, F. DUPRE, M. VICENTINIMISSONI, Weight, gravity and air pressure: Mental representations by Italian Middle School pupils, European Journal of Science Education 7(2), pp 181194 (1985). [38] D. PALMER, Student’s alternative conceptions and scientiﬁcally acceptable conceptions about gravity, Int. Scie. Educ. 23(7), pp 691706 (2001).
ELASTIC WAVES: MENTAL MODELS AND TEACHING/ LEARNING SEQUENCES
GIOVANNI TARANTINO a
a
Dipartimento di Fisica e Tecnologie Relative, Universit` a di Palermo, viale delle Scienze Ed. 18, 90128 Palermo, Italy
Abstract In last years many research studies have pointed out relevant student diﬃculties in understanding the physics of mechanical waves. Moreover, it has been reported that these diﬃculties deal with some fundamental concepts as the role of the medium in wave propagation, the superposition principle and the mathematical description of waves involving the use of functions of two variables. In the context of preservice courses for teacher preparation a teaching/learning (T/L) sequence based on using simple RTL experiments and interactive simulation environments aimed to show the eﬀect of medium properties on the propagation speed of a wave pulse, has been experimented. Here, preliminary results of investigations carried out with a 120 traineeteacher (TT) group are reported and discussed.
1
Introduction
According to experiences reported by many physics teachers, students are in trouble whenever they have to face up to elastic wave phenomena. Investigations [1] have shown that the diﬃculties students have with wave physics mainly involve the mathematical description of wave phenomena, requesting the use of twovariables functions, the concept and the application of the superposition principle, often mistaken by a mere overlap of waveforms, and the role that has to be assigned to the medium through which wave is propagating . This role is sometimes considered passive in the sense that the propagation is described as not directly aﬀected by the elastic properties of the medium.
381 B.G. Sidharth, F. Honsell and A. de Angelis. (eds.), Frontiers of Fundamental Physics, 381–384. © 2006 Springer. Printed in the Netherlands.
382
Elastic waves: Mental Models and T/L Sequences
Nevertheless, in real life, it is possible to ﬁnd some examples which can be considered a clear proof that probably people perceive what a wave is. To produce the ”Ola” (the typical wave in a stadium), for example, the fans know well when they have to stand up or sit down. On the other hand, if we change the context, what happens is that students seem to apply inappropriate cognitive resources to describe the event propagation [2]. Consider, for example, a pulse wave travelling along a slinky. Many researches [1] have shown that a large group of students, even at university level, describe the pulse as if it were a moving body. Consequently, the medium is thought as playing a passive role and the propagation properties are wrongly considered as depending on the way the pulse has been generated. When we think of a propagating pulse as a whole (for example a pulse travelling along an elastic rope), we usually attribute to the pulse some objectlike features (shape, width, speed etc.). But, while in the ”Ola” case, the two levels of representation (on the whole and in terms of individual agents) are well distinguishable and people recognize that what is propagating is an event (standing up or sitting down), in the ”slinky” case the switching between the two levels of representation is harder. Then, they are unable to recognize that they’re dealing with the propagation of an event and use cognitive and perceptual resources which are adequate for describing the propagation of an object but inadequate to interpret the propagation of events. Things get worse when we deal with sound waves because the diﬃculty of discerning between event and particle propagation is hampered by the imperceptibility of the medium: the perception of phenomenon is auditory and not visual and what is moving must be indirectly inferred. In the next section, we discuss the results obtained in the preliminary phase of the research study carried out with a group of 120 traineeteachers attending the teachers preparation school (SSIS) at University of Palermo concerning the understanding of the role of medium in elastic waves propagation.
2
The PreTest
By means of a questionnaire, students have been requested to describe three different situations dealing with propagation of pulses through elastic media. The situations were: 1. The free end of an horizontal elastic rope ﬁxed on the wall is moved quickly up and down. A travelling pulse is generated. What happens if the hand generating the pulse is moved more rapidly? 2. The end of a two meters long metal bar is hit by an hammer. Two microphones are positioned at the same distance from the hit point but one is near the other end of the bar, while the second probe is far away from it. Which probe does the sound pulse reach before?
G. Tarantino
383
3. Consider a dust particle in front of a loudspeaker emitting a sound of a given frequency. Describe the motion of the particle. The analysis of pretest results has shown that the students answered to the questionnaire by using two diﬀerent mental models. The ﬁrst model may be deﬁned as the Active Medium Model (AMM). In this model the medium is thought as playing an active and essential role in the propagation mechanisms. Those who used this model gave correlated interpretations of the three problematic situations of pretest according to which the propagation speed is not aﬀected by the way the pulse is generated (question 1), the sound pulse propagates faster along the metal bar because of the diﬀerent elastic properties of metal and air (question 2) and the dust particle (question 3) moves back and forth. The second mental model has been called the Passive Medium Model (PMM). Predictions made according this model have been based on the representation of the medium as a passive element in the propagation. While the pulse is travelling, it is subjected to the resistance exerted by the medium and depending on the medium density, just like a body moving in a viscous medium. Those who applied this model provided some interpretations like: • The pulse travels faster if the hand is moved faster • Concerning question 2, the sound pulse reaches before the microphone placed far from the bar end because of resistance oﬀered by the metal bar • The dust particle in front of the loudspeaker moves forward or it keeps still. All these wrong representations show that when the eventlike nature of waves is not recognized, objectlike properties are attributed to the wave pulse.
3
The Teaching/Learning Sequences
The teaching/learning sequences have been aimed to stimulating the use of the appropriate cognitive resources for description of wave phenomena and have been based on two diﬀerent kinds of tools: • RTL experiments • Simulation environments The experiments concerned essentially the measurement of the sound speed in diﬀerent media (air and metals) with the aid of real time data logger. Regarding the second kind of tools, we used the Interactive Physics environment to build up several simulations of linear chains of identical masses interacting through a linear nearestneighborhood coupling. Students used such kind of tools in a constructivistic context stimulating the learning by means of worksheets and adoption of a PEC (PredictionExperimentComparison) cycle.
384
Elastic waves: Mental Models and T/L Sequences
The analysis of materials produced by the students (worksheets, interviews etc.) has shown the following two main results: • The understanding of elastic wave phenomena is improved by choosing an approach based onto the representation of waves in terms of individual agents, rather than based ”on the whole” • The ”pulse” approach is more eﬀective than the ”sinusoidal” one. First result may be explained by considering that, when we observe a wave phenomenon, its wavelike nature is recognizable not until we think of it in terms of the behavior of the individual agents. In terms of conceptual resources, we ﬁnd that the appropriate resources for description of wave phenomena are mostly stimulated by an approach based on individual agents representation. On the other hand, the approach to waves based on the properties of sinusoidal and/or other periodic waves focus on the periodicity of a wave and not on its propagation. For this reason we think that, in order to explain the role of medium in the propagation process, a pulse approach be more eﬀective. We found that students often associate the wavelike properties of a system to its shape. The term wave evokes the waves propagating through the surface of liquids (the sea waves, for example) and the sinusoidal approach fosters this kind of misunderstanding.
4
Conclusions
The starting point of the research study here described is based on the measured diﬃculties the students show to have in interpreting elastic wave phenomena. By now, we are still analyzing the students worksheets together with other materials and improving them on the basis of the feedback. However, the tutorials have met with measurable success and preliminary results obtained by examination of questionnaires, tests and interviews have shown an improvement in understanding and in the ability to describe the correct physics. References [1] M.C. Wittmann, R.N. Steinberg, E.F. Redish, Making Sense of how Students Make Sense of Waves, The Physics Teacher, 37 1521 (1999) [2] D. Hammer, Student Resources for Learning Introductory Physics, Phys Educ. Res., Am. J. Phys., Suppl. 68 (7) (2000)
INDEX OF AUTHORS
AlfonsoFaus, A., 31 Altaisky, M., 155 Ansoldi, S., 45, 69 Argan, A., 303 ATLAS Collaboration Azzi, P., 309
De Angelis, A., 291, 297, 309, 327, 333 De Lotto, B., 291, 333 Del Monte, E., 303 Di Cocco, G., 303 Di Persio, G., 303 Donnarumma, I., 303 , 267
Baiotti, L., 75 Baldini, L., 309 Barbarino, F., 297 Barbiellini, G., 303 Basset, M., 303 Bastieri, D., 291, 309 Bavikadi, S.R., 291, 333 Beck, C., 137 Bigongiari, C., 291, 333 Bisesi, E., 291, 315, 321 Boffelli, F., 303 Boinee, P., 291, 297, 309 Bradamante, F.a, 367 Bradamante, F.o, 221 Brigida, M., 309 Buffa, R., 273, 285 Bulgarelli, A., 303 Busetto, G., 309 Cabras, G., 333 Caraveo, P., 303 Carmeli, M., 3, 13 Cattaneo, P.W., 303 Cecchi, C., 309 Celesti, E., 303 Chen, A., 303 Citossi, M., 355 Cobal, M., 267 Cocco, V., 303 Cooper, L., 341 Costa, E., 303
Enders, P., 239 Feroci, M., 303 Ferrari, V., 83 Fiorini, M., 303 Font, J.A., 75 Forti, A., 291, 333 Frailis, M., 309, 327, 333 Froysland, T., 303 Frutti, M., 303 Fullana i Alfonso, M.J., 115 Galli, M., 303 Gargano, F., 309 Giannitrapani, R., 263 Gianotti, F., 303 Giglietto, N., 309 Giordano, P., 183 Giugliarelli, G., 355, 359 Giuliani, A., 303 Goradia, S., 37 Greiner, W., 211 Hardt, M., 333 Hartnett, J.G., 21 Hawke, I., 75 ’t Hooft, G., 29 Iovane, G., 183 Ivanov, M.A., 51 Kaberidis, C., 279
385
386
Index of Authors
Keller, J., 163 Kirilova, D.P., 55 Klauder, J.R., 61 Kornmayer, H., 333 Koschmieder, E.L., 231 ¨ ger, H., 123 Kro Kunze, M., 333 Kuss, M., 309 Labanti, C., 303 Lapshov, I., 303 Latronico, L., 309 Lazzarotto, F., 303 Lei, Y., 309 Lenisa, T., 291 Liello, F., 303 Lionetto, A., 309 Lipari, P., 303 ¨ ffler, F., 75 Lo Longo, F., 291, 303, 309 Loparco, F., 309 Lubrano, P., 309 MAGIC Collaboration, 291, 333 Mansutti, O., 291 Marcucci, F., 309 Mariotti, M., 291, 315 Marisaldi, M., 303 Mastropietro, M., 303 Mauri, A., 303 Mauri, F., 303 Mayburov, S.N., 191 Mazziotta, M.N., 309 Melkonyan, G., 123 Mereghetti, S., 303 Michelini, M., 367 Mizushima, M., 103 Montero, P.J., 75 Moralejo, A., 291 Morelli, E., 303 Morselli, A., 303, 309 Moustaizis, S., 279 Nottale, L., 107 Omodei, N., 309
Ord, G.N., 175 Pacciani, L., 303 Paradis, F., 123 Pascoli, D., 291 Pellizzoni, A., 303 Pepe, M., 309 Perotti, F., 303 Peruzzo, L., 291 Picozza, P., 303 Pimenta, M., 255 Piraccini, M., 333 Pittori, C., 303, 309 Pontoni, C., 303 Porrovecchio, G., 303 Prest, M., 303 Rando, R., 309 Rapisarda, M., 303 Razzano, M., 309 Rezzolla, L., 75 Roberto, V., 327 Rossi, E., 303 Rubin, S.G., 123 Rubini, A., 303 ´ez Mila ´n, D.P., 115 Sa Saggion, A., 291, 297 Salerno, S., 183 Sartori, P., 291 Scalzotto, V., 291, 315 Seidel, E., 75 Selleri, F., 195 Seriu, M., 93 ShalytMargolin, A.E., 131 Sidharth, B.G., 147 Sindoni, L., 45, 69 Soffitta, P., 303 Spandre, G., 309 Stefanel, Al., 367 Stergioulas, N., 75 Strazhev, V.I., 131 Tarantino, G., 381 Tavani, M., 303 Tibolla, O., 309
Index of Authors Tosti, G., 309 Traci, A., 303 Trifoglio, M., 303 Tsohantjis, I., 279 Utyuzh, O.V., 247 Vallazza, E., 303 Variola, A., 265 Velarde, M.G., 343 Vercellone, S., 303 Wlodarczyk, Z., 247 Wilk, G., 247 Zacchello, M., 297 Zanello, D., 303
387