arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.05455cond-mat.stat-mech
收藏

有限热储与正则分布

Finite thermal reservoirs and the canonical distribution

William Griffin, Michael Matty, Robert H. Swendsen

摘要:

微正则系综一直是热力学从统计力学发展的起点。然而,这种方法存在两个问题。首先,它预言熵仅定义在有限量子系统的离散能量集上,而热力学要求熵是能量的连续函数。其次,它不满足经典和量子系统的一级跃迁的稳定性条件($ΔS / ΔU < 0$)。Swendsen最近指出,这些问题的根源在于微正则系综本身,它只包含能量本征态,不包括它们的线性组合。相反,如果感兴趣的系统曾经与另一个系统热接触,它将被描述为在许多本征态上的概率分布,该概率分布等价于足够大的系统的正则系综。Novotny等人最近通过量子力学模型的动力学数值计算支持了这一观点,他们展示了多达40个量子自旋的正则分布的方法。通过将问题简化为只计算平衡性质,我们能够将演示扩展到超过一百万个粒子。

英文摘要:

The microcanonical ensemble has long been a starting point for the development of thermodynamics from statistical mechanics. However, this approach presents two problems. First, it predicts that the entropy is only defined on a discrete set of energies for finite, quantum systems, while thermodynamics requires the entropy to be a continuous function of the energy. Second, it fails to satisfy the stability condition ($ΔS / ΔU < 0$) for first-order transitions with both classical and quantum systems. Swendsen has recently shown that the source of these problems lies in the microcanonical ensemble itself, which contains only energy eigenstates and excludes their linear combinations. To the contrary, if the system of interest has ever been in thermal contact with another system, it will be described by a probability distribution over many eigenstates that is equivalent to the canonical ensemble for sufficiently large systems. Novotny et al. have recently supported this picture by dynamical numerical calculations for a quantum mechanical model, in which they showed the approach to a canonical distribution for up to 40 quantum spins. By simplifying the problem to calculate only the equilibrium properties, we are able to extend the demonstration to more than a million particles.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑