Estimates of the dynamic structure factor for the finite temperature electron liquid via analytic continuation of path integral Monte Carlo data
Thomas Chuna, Nicholas Barnfield, Jan Vorberger, Michael P. Friedlander, Tim Hoheisel, Tobias Dornheim · 2025
Read Estimates of the dynamic structure factor for the finite temperature electron liquid via analytic continuation of path integral Monte Carlo data by Thomas Chuna, Nicholas Barnfield, Jan Vorberger, Michael P. Friedlander, Tim Hoheisel, Tobias Dornheim (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-20433.txt.
Introduction
Thomas Chuna and five co-authors reported a 2025 method for reading how a hot electron liquid moves and vibrates. This uniform system matters for chip physics and for warm dense matter found in planets and fusion work. Direct simulation gives data in imaginary time, so the team had to convert it into a real-time response curve. They used an entropy-based method with simulated data as its starting point.
Building the prior guess from the simulation itself greatly sharpened the result over a simpler classic guess. The team covered a wider range of densities than before, all at the electrons' Fermi temperature. In the strongly coupled range, they found a dip feature and signs of a double peak. Their curves also met known sum rules without being forced to.
Catalog record
| CREATOR | Thomas Chuna, Nicholas Barnfield, Jan Vorberger, Michael P. Friedlander, Tim Hoheisel, Tobias Dornheim |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-20433
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Estimates of the dynamic structure factor for the finite temperature electron liquid via analytic continuation of path integral Monte Carlo data Thomas Chuna, Nicholas Barnfield, Jan Vorberger, Michael P. Friedlander, Tim Hoheisel, Tobias Dornheim Abstract Understanding the dynamic properties of the uniform electron gas (UEG) is important for numerous applications ranging from semiconductor physics to exotic warm dense matter. In this work, we apply the maximum entropy method (MEM), as implemented in Chuna et al. [arXiv:2501.01869], to ab initio path integral Monte Carlo (PIMC) results for the imaginary-time correlation function F ( q , τ ) F(q,\tau) to estimate the dynamic structure factor S ( q , ω ) S(q,\omega) over an unprecedented range of densities at the electronic Fermi temperature. To conduct the MEM, we propose to construct the Bayesian prior μ \mu from the PIMC data. Constructing the static approximation leads to a drastic improvement in S ( q , ω ) S(q,\omega) estimate over using the more simple random phase approximation (RPA) as the Bayesian prior. We find good agreement with existing results by Dornheim et al. [ Phys. Rev. Lett. 121 , 255001 (2018)], where they are available. In addition, we present new results for the strongly coupled electron liquid regime with r s = 50 , … , 200 r_{s}=50,...,200 , which reveal a pronounced roton-type feature and an incipient double peak structure in S ( q , ω ) S(q,\omega) at intermediate wavenumbers. We also find that our dynamic structure factors satisfy known sum rules, even though these sum rules are not enforced explicitly. An advantage of our set-up is that it is not specific to the UEG, thereby opening up new avenues to study the dynamics of real warm dense matter systems based on cutting-edge PIMC simulations in future works. I Introduction The uniform electron gas (UEG) is an important and well-studied physical system [ 1 , 2 , 3 , 4 , 5 ] . Investigations on this topic have continued for decades because there is no closed analytic model which can describe the spatial and temporal structure of the plasma at all experimentally accessible number densities and temperatures. A current frontier of the UEG is the finite temperature electron liquid, where the plasma is both quantum degenerate \Theta=k_{\text{B}}T/E_{F}\approx 1 and strongly coupled \Gamma=\frac{1}{r_{s}}\left((k_{\text{B}}T)^{2}+E_{F}^{2}\right)^{-1/2}\geq 1 . We have adopted Hartree atomic units in this work with r_{s} being the Wigner-Seitz radius that is defined by n_{e}=(4\pi r_{s}^{3}/3)^{-1} (in units of Bohr radius a_{B} ), E_{F} is the Fermi energy in Hartree, and \beta=(k_{\textnormal{B}}T)^{-1} is the inverse temperature. When the plasma is both quantum degenerate and strongly coupled, path integral Monte Carlo (PIMC) simulations are one of the few methods that can describe the structure of the UEG [ 6 ] . In this work, we study PIMC simulations of the electron liquid at fixed \Theta=1 and vary \Gamma between approximately 1 and 150 ; this gives us access to a broad range of physical conditions. In principle, PIMC simulations give one access to the complete equilibrium dynamics of the simulated system in the form of various imaginary-time correlation functions (ITCF) [ 7 , 8 , 9 ] . In particular, the imaginary-time density–density correlation function F(q,\tau)=\braket{\hat{n}(q,\tau)\hat{n}(-q,0)} , which corresponds to the usual intermediate scattering function evaluated at the imaginary time argument t=-i\hbar\tau with \tau\in[0,\beta] and \beta=1/k_{\textnormal{B}}T the inverse temperature in energy units, is directly related to the dynamic structure factor (DSF) S(q,\omega) (in units of inverse energy) via a two-sided Laplace transform \displaystyle F(q,\tau)=\int_{-\infty}^{\infty}d\omega\ e^{-\tau\omega}S(q,\omega)=\int_{0}^{\infty}d\omega\left(e^{-\tau\omega}+e^{-(\beta-\tau)\omega}\right)S(q,\omega)\ , (1) where the second equality follows from the detailed balance condition S(q,-\omega)=S(q,\omega)e^{-\beta\omega} . Matching \beta , \omega is the frequency in energy units. Since the LHS is a function of imaginary-time and the RHS is a function of real-time frequency, estimating the DSF is an analytic continuation problem. The DSF is of key importance e.g. for the interpretation of x-ray Thomson scattering experiments [ 10 , 11 , 12 ] , and directly gives one access to a number of density response and dielectric properties [ 13 , 14 ] . Unfortunately, inverting ( 1 ) to reconstruct S(q,\omega) from F(q,\tau) is a well-known exponentially ill-posed problem [ 15 , 16 ] and plagues many fields of science [ 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ] . For the UEG, Dornheim et al. [ 27 , 28 , 29 ] have partially solved this problem by evoking the fluctuation–dissipation theorem, which has allowed to exactly fulfill a number of exact constrains that are known for the UEG. However, this stochastic sampling approach breaks down with increasing coupling strength, and cannot be easily generalized to real systems beyond the UEG model. Consequently, a more general approach is needed. The recent work of Chuna et al. [ 30 ] has demonstrated that the all-purpose maximum entropy method (MEM) can reliably estimate the DSF from PIMC data in some situations. This method is not specifically tailored for the UEG and can be applied directly to two-component systems that are comprised of both electrons and nuclei. In the present work, we apply the MEM with a dual Newton optimizer to estimate the UEG DSF, at \Theta=1 over a broad range of densities identified by the corresponding Wigner-Seitz radius. In particular, we investigate densities spanning the warm dense matter regime [ 31 ] ( r_{s}=2 ) to the strongly coupled electron liquid [ 32 , 33 , 34 ] ( r_{s}=50,...,200 ). We compare our estimates with the more limited stochastic inversion where it is available, i.e., for r_{s}\leq 10 . As we apply the MEM, we have given special attention to our uncertainty procedure. The MEM requires ITCF data and a best guess of the solution ( i.e. Bayesian prior). In our application, we construct the Bayesian prior directly from the ITCF data based on the static approximation to the DSF that approximates the full dynamic local field correction (LFC) by its exact static limit, i.e., G(q,\omega)\approx G(q,0) . The static approximation has been shown, in many finite temperature electron liquid applications, to yield a good approximation of the DSF [ 27 , 35 , 36 ] . We consider the impact of using such a Bayesian prior that depends on the data, and compare the static approximation to the more simple random phase approximation (RPA) that neglects the LFC entirely. In addition to this, we quantify the uncertainty in our DSF estimates via leave-one-out binning [ 37 , 38 , 39 ] , which estimates the variance arising from uncertainty in the MEM regularization weight and from sampling error in the two-particle correlation function. Thus, this work establishes how to use the general purpose tool, the MEM, to estimate the UEG DSF from PIMC data for the ITCF. In the weak coupling limit \Gamma\rightarrow 0 , our estimates resemble the RPA, as expected. In addition, our estimates match results obtained by stochastically sampling the dynamic local field corrections [ 27 ] for r_{s}\leq 10 , and our set-up works well even at large r_{s} where the latter approach fails. We find a pronounced roton-type feature in our new estimates for S(q,\omega) at