On the rate of quantum ergodicity in Euclidean billiards

A. Bäcker, R. Schubert, P. Stifter

Research output: Contribution to journalArticle (Academic Journal)peer-review


For a large class of quantized ergodic flows the quantum ergodicity theorem due to Shnirelman, Zelditch, Colin de Verdi\`ere and others states that almost all eigenfunctions become equidistributed in the semiclassical limit. In this work we first give a short introduction to the formulation of the quantum ergodicity theorem for general observables in terms of pseudodifferential operators and show that it is equivalent to the semiclassical eigenfunction hypothesis for the Wigner function in the case of ergodic systems. Of great importance is the rate by which the quantum mechanical expectation values of an observable tend to their mean value. This is studied numerically for three Euclidean billiards (stadium, cosine and cardioid billiard) using up to 6000 eigenfunctions. We find that in configuration space the rate of quantum ergodicity is strongly influenced by localized eigenfunctions like bouncing ball modes or scarred eigenfunctions. We give a detailed discussion and explanation of these effects using a simple but powerful model. For the rate of quantum ergodicity in momentum space we observe a slower decay. We also study the suitably normalized fluctuations of the expectation values around their mean, and find good agreement with a Gaussian distribution.
Original languageEnglish
JournalPhysical Review E
Publication statusPublished - 29 Sep 1997

Bibliographical note

40 pages, LaTeX2e. This version does not contain any figures. A version with all figures can be obtained from http://www.physik.uni-ulm.de/theo/qc/ (File: http://www.physik.uni-ulm.de/theo/qc/ulm-tp/tp97-8.ps.gz) In case of any problems contact Arnd B\"acker (e-mail: baec@physik.uni-ulm.de) or Roman Schubert (e-mail: schub@physik.uni-ulm.de)


  • chao-dyn
  • nlin.CD


Dive into the research topics of 'On the rate of quantum ergodicity in Euclidean billiards'. Together they form a unique fingerprint.

Cite this