generating function for level correlations semiclassical
play

Generating function for level correlations, semiclassical - PowerPoint PPT Presentation

Generating function for level correlations, semiclassical evaluation Alex Altland, Peter Braun, F. H., Stefan Heusler, Sebastian Mller Banff, Feb. 25, 2008 conventional semiclassics E E 2


  1. Generating function for level correlations, semiclassical evaluation Alex Altland, Peter Braun, F. H., Stefan Heusler, Sebastian Müller Banff, Feb. 25, 2008

  2. conventional semiclassics 〈   E    E ′  −  2   ∑ ∗ e i  S a − S b  /  F a F b a , b contributions from b = a (and b =T a for TRI) * * a,b identical up to reconnections in encounters gives non-oscillatory part in full but misses oscillatory part

  3. reminder: RMT − 1  ei 2  unitary R      2 2  4 …  e i 2  orthogonal 1  − 1 …       E − E ′  2  

  4. generating function 〈 det  E   C − H  det  E −  D − H  det  E   A − H  det  E −  B − H    Z   A ,  B ,  C ,  D   exp tr ln det ∂  A ∂  B Z |  A   B   C   D     tr ∂ 2  C     1 1 E    − H tr E −   − H

  5. semiclassical evaluation E  det  E  − H     1 exp   dE ′ tr  1 E ′ − H i S a  E   /  − i  N  E   − ∑ a F a e  exp      exp  − i  N  E   ∑ A F A  − 1  n A e i S A  E   /  det  E − − H    det  E  − H  ∗

  6. Riemann-Siegel lookalike allows for real E , enforcing convergence and reality T A  T H /2 det  E − H   exp  − i  N  E  ∑ F A  − 1  n A e i S A  E  /   c.c. A rigorous for finite matrices, respects unitarity, modelled after Riemann’s ζ not available for inverse determinants

  7. 〈 exp   i  N  E   A  ∑ Z  F A e i S A  E   A  /  A  exp  − i  N  E −  B  ∑ ∗ e − i S B  E −  B  /  F B B T C  T H /2   exp  − i  N  E   C  ∑ F C  − 1  n C e i S C  E   C  /  C T D  T H /2  exp   i  N  E −  D  ∑ ∗  − 1  n D e − i S D  E −  D  /  F D D    Z  1   Z  2  + c.c. Weyl Z  2    A ,  B ,  C ,  D   Z  1    A ,  B , −  D , −  C  symmetry

  8. N  E     N  E    , S  E     S  E    Z  1   e i   A   B −  C −  D  /2  T C , T D  T H /2  ∑ ∗  − 1  n C  n D ∗ F C F D  〈 F A F B A , B , C , D  e i  S A  E  − S B  E   S C  E  − S D  E   e i   A  A   B  B −  C  C −  D  D  /2  contributions only from terms where orbits in A and C are repeated in either B or D, identically (diagonal appr) or at least up to reconnections in encounters (bunches)

  9. diagonal approximation p.o.’s enter as if uncorrelated 〈 i S a  E   /  average ``sees’’ p.o. sum ∑ a F a e as Gaussian random variable, due to central limit theorem Gaussian average most conveniently done in starting expression where four p.o. sums appear in exponent

  10.  1  X  e i   A   B −  C −  D  e Z diag diag X  ∑ a F a e i S a  E  /  i  A  a  ∑ b F b ∗ e − i S b  E  /  i  B  b − ∑ c F c e i S c  E  /  i  C  c − ∑ d F d ∗ e i S d  E  /  i  D  d  exp 〈 X 2  diag  X e diag i   A   B   a − e i   A   D   a ∑ a | F a | 2 e 〈 X 2  diag  i   C   B   c − e i   C   D   c − ∑ c | F c | 2 e

  11. ∑ a | F a | 2 e i  a  − ln  i    const HOdA:  e i   A   B −  C −  D    A   D   C   B   1  Z diag   A   B   C   D   e i   A   B   C   D    A −  C  −  D   B   2  Z diag   A   B  −  D −  C  e 2 i  C diag     2  i   2 − 1 2  i   2

  12. off-diagonal terms from orbit bunches in chaotic dynamics, long p. o.’s do not arise as mutually independent entities but rather in closely packed bunches under weak resolution of configuration space, bunch looks like single orbit orbits in bunch practically identical, apart from reconnections within self-encounters all orbits in bunch generated from single one by reshuffling stretches within self-encounters, so as to differently connect practically unchanged links

  13. ``bunch’’ of 2 orbits: nearly same links, differently connected by the two encounter stretches; action difference can be arbitrarily small respectful bows to Martin & Klaus, and their “disordered precursors”

  14. l- encounter: l orbit stretches mutually close l links encounter stretches can connect links in l ! different ways bunch of l ! (pseudo-)orbits, nearly same ``rigid’’ links; action differences arbitrarily small

  15. bunch of 12

  16. bunch of 72

  17. orbit bunches yield Z  1  asymptotic expansion of in 1/ ε Z  2  Z  1  then from with Riemann Siegel Z  1   Z  2  differentiation of gives correlator R(e), in agreement with RMT; in particular, no addition to diagonal approximation for unitary class n -th order term from bunches of R(e) from bunches with n = L - V +2, where V = # encounters (vertices), L = # links

  18. scene thus set for Sebastian Müller concluding remarks: oscillatory terms through generating function and Riemann-Siegel close correspondence to sigma model of RMT asymptotic series for osc and non-osc terms there arise from perturbative treatment of two saddle points for integral over matrix manifold, Feynman diagrams correspond to orbit bunches, vertices to encounters, links to propagator lines

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend