March 28, 2026 Dataset Open Access

Temperature dependence of the dynamic structure factor of the electron liquid via analytic continuation

Chuna, Thomas; Boehme, Maximilian; Dornheim, Tobias

All results have been computed for the unpolarized UEG, i.e., with an equal number of spin-up and spin-down electrons $N^\uparrow=N^\downarrow=N/2$ for $N=34$. For $r_s = 20$ and at $\Theta=0.75$, there are 1000 independent MCMC seeds, at $\Theta=1, \, 2$ there are $280$ seeds, at $\Theta=4, \,8$ there are $277$ seeds. To compute the data, we conduct leave-one-out binning across the seeds and for all the data, the variance of the mean is $\delta F \approx 10^{-3}-10^{-4}$. This error estimate is computed for each leave-one-out-bin via Hatano's error formula~\cite{hatano1994data} and verified using leave-one-out binning~\cite{berg_book_2004}. This is the online repository with the PIMC results for $F(\mathbf{q},\tau)$ and analytic continuation results for $S(q,\omega)$  seen "Temperature dependence of the dynamic structure factor of the electron liquid via analytic continuation" article. 

The data contained here is (1) leave-one-binned imaginary time correlation functions F(tau) (units dimensionless) over tau (units 1/Hartree) and there error (2) the dynamic structure factors (units 1/Hartree) over omega (units Hartree) obtained using Bryan's MEM with the static approximation as the Bayesian prior (3) the dynamic structure factors (units 1/Hartree) over omega (units Hartree) obtained using  PyLIT with the static approximation as the Bayesian prior (4) The omega->0 limit of the ideal gas susceptibility chi(q,0)/n/beta (units dimensionless) over q (units 1/Bohr).