Cui Lei, Zhang Hou-Dao, Zheng Xiao, Xu Rui-Xue, Yan YiJing
Hefei National Laboratory for Physical Sciences at the Microscale and Department of Chemical Physics and Synergetic Innovation Center of Quantum Information and Quantum Physics and Collaborative Innovation Center of Chemistry for Energy Materials (iChEM), University of Science and Technology of China, Hefei, Anhui 230026, China.
J Chem Phys. 2019 Jul 14;151(2):024110. doi: 10.1063/1.5096945.
The Fano spectrum decomposition (FSD) scheme is proposed as an efficient and accurate sum-over-poles expansion of Fermi and Bose functions at cryogenic temperatures. The new method practically overcomes the discontinuity of Fermi and Bose functions near zero temperature, which causes slow convergence in conventional schemes such as the state-of-the-art Padé spectrum decomposition (PSD). The FSD scheme fragments Fermi or Bose function into a high-temperature reference and a low-temperature correction. While the former is efficiently decomposed via the standard PSD, the latter can be accurately described by several modified Fano functions. The resulting FSD scheme is found to converge overwhelmingly faster than the standard PSD method. Remarkably, the low-temperature correction supports further a recursive and scalable extension to access the near-zero temperature regime. Thus, the proposed FSD scheme, which obeys rather simple recursive relations, has a great value in efficient numerical evaluations of Fermi or Bose function-involved integrals for various low-temperature condensed physics formulations and problems. For numerical demonstrations, we exemplify FSD for the efficient unraveling of fermionic reservoir correlation functions and the exact hierarchical equations of motion simulations of spin-boson dynamics, both at extremely low temperatures.
提出了费诺谱分解(FSD)方案,作为低温下费米函数和玻色函数的一种高效且精确的极点求和展开。新方法实际上克服了费米函数和玻色函数在接近零温度时的不连续性,这种不连续性在诸如最新的帕德谱分解(PSD)等传统方案中会导致收敛缓慢。FSD方案将费米函数或玻色函数分解为一个高温参考和一个低温修正。前者通过标准PSD有效分解,而后者可以用几个修正的费诺函数精确描述。结果发现,所得的FSD方案比标准PSD方法收敛速度快得多。值得注意的是,低温修正进一步支持递归和可扩展扩展,以进入接近零温度区域。因此,所提出的FSD方案遵循相当简单的递归关系,在对各种低温凝聚态物理公式和问题中涉及费米函数或玻色函数的积分进行高效数值评估方面具有很大价值。为了进行数值演示,我们以FSD为例,展示了在极低温下对费米子库关联函数的高效解析以及自旋玻色子动力学的精确层次运动方程模拟。