arXiv:2609.24873 · Physical review. D/Physical review. D.
Measuring correlations using local and nonlocal quenches
Alexey G. Mikhaylenko, Andrew G. Semenov
Abstract
We present a theoretical proposal for measuring correlation functions in a quantum field system based on the use of quantum quenches. Using the examples of massive and massless scalar field theories, we show that a short-term perturbation of the system leads to a dependence of the further evolution of the average field on the initial state's correlation functions. Considering two types of perturbations, namely local and nonlocal quenches, we show which correlation functions can be measured in each of these cases. The proposed procedure can be applied to study non-Gaussian correlations.
AI Summary
The paper asks which initial-state correlations can be inferred from the mean field following an impulsive perturbation of a free scalar quantum field. It considers massive and massless theories in $d$ spatial dimensions, allowing arbitrary initial states, including non-Gaussian ones. A local quench applies $Q=\exp[-i\lambda\varphi_S^n(x_Q)/\hbar]$, where $\varphi_S$ is spatially smeared. Canonical commutators and free propagation give the exact mean-field shift $\Delta\langle\varphi(t,x)\rangle=n\lambda G_S^R(t-t_Q,x-x_Q)\langle\varphi_S^{n-1}(x_Q)\rangle$. Thus, different polynomial quenches encode successive smeared moments in a one-point observable. A nonlocal quench generated by $\lambda\prod_j\varphi_S(x_j)$ instead produces a sum of responses weighted by correlations with one field omitted. Measurements near one quench location, before responses from other locations arrive, isolate the corresponding spatially separated correlation. For Gaussian smearing, the retarded response is evaluated analytically and asymptotically. At the quench point, massive-field responses have an oscillatory envelope proportional to $(t-t_Q)^{-d/2}$. Massless responses approach $-1/2$ in $d=1$, decay as $-1/[2\pi(t-t_Q)]$ in $d=2$, and decay exponentially in time squared in $d=3$. Ground-state examples illustrate both protocols using exactly calculable two-point correlations. The stated procedure requires noninteracting post-quench dynamics; interactions introduce higher-order averages into mean-field evolution. Non-Gaussian-state reconstruction is proposed but not demonstrated in the examples. Spatial smearing removes ultraviolet divergences, while the massless ground-state example in $d=1$ still requires infrared regularization. Rapid response decay in massless $d=3$ makes measurement challenging.
Error found: For the Gaussian smearing used in the paper, nonoverlap of nominal light cones does not exactly isolate one initial-state correlation, because the other smeared sources have spatial tails whose contamination is not quantified. The massive smearing-independent stationary-phase formula requires t−t_Q≫Mε², a condition missing from the stated regime and not implied by its other conditions when Mε≫1.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- DOI
- 10.1103/5nmd-njw7
- Comments
- 10 pages, 5 figures, to be published in Physical Review D