EoS of finite density QCD with Wilson fermions by Multi-Parameter Reweighting and Taylor expansion

TL;DR

This study compares multi-parameter reweighting and Taylor expansion methods in lattice QCD with Wilson fermions to reliably compute the equation of state and related thermodynamic quantities at finite density, demonstrating their consistency within certain chemical potential ranges.

Contribution

It introduces a combined approach using MPR and Taylor expansion with a reduction formula for Wilson fermions, enabling more accurate finite density QCD calculations on small lattices.

Findings

Methods are consistent up to μ/T ≈ 0.8 for EoS and density.

Reweighting and truncation errors are negligible within these ranges.

Derived the reweighting line where fluctuations are minimized.

Abstract

The equation of state (EoS), quark number density and susceptibility at nonzero quark chemical potential $\mu$ are studied in lattice QCD simulations with a clover-improved Wilson fermion of 2-flavors and RG-improved gauge action. To access nonzero $\mu$, we employ two methods : a multi-parameter reweighting (MPR) in $\mu$ and $\beta$ and Taylor expansion in $\mu/T$. The use of a reduction formula for the Wilson fermion determinant enables to study the reweighting factor in MPR explicitly and heigher-order coefficients in Taylor expansion free from errors of noise method, although calculations are limited to small lattice size. As a consequence, we can study the reliability of the thermodynamical quantities through the consistency of the two methods, each of which has different origin of the application limit. The thermodynamical quantities are obtained from simulations on a $8^3\times 4$ lattice with an intermediate quark mass($m_{\rm PS}/m_{\rm V}=0.8)$. The MPR and Taylor expansion are consistent for the EoS and number density up to $\mu/T\sim 0.8$ and for the number susceptibility up to $\mu/T \sim 0.6$. This implies within a given statistics that the overlap problem for the MPR and truncation error for the Taylor expansion method are negligible in these regions. In order to make MPR methods work, the fluctuation of the reweighting factor should be small. We derive the equation of the reweighting line where the fluctuation is small, and show that the equation of the reweighting line is consistent with the fluctuation minimum condition.