Morgado Welles A M, Duarte Queirós Sílvio M
Department of Physics, PUC-Rio, and National Institute of Science and Technology for Complex Systems, Rua Marquês de São Vicente 225, 22453-900 Rio de Janeiro, RJ, Brazil.
Centro Brasileiro de Pesquisas Físicas and National Institute of Science and Technology for Complex Systems, Rua Dr Xavier Sigaud, 150, 22290-180 Rio de Janeiro, RJ, Brazil.
Phys Rev E Stat Nonlin Soft Matter Phys. 2014 Aug;90(2):022110. doi: 10.1103/PhysRevE.90.022110. Epub 2014 Aug 11.
We discuss the statistical properties of small mechanothermodynamic systems (one- and two-particle cases) subject to nonlinear coupling and in contact with standard Gaussian reservoirs. We use a method that applies averages in the Laplace-Fourier space, which relates to a generalization of the final-value theorem. The key advantage of this method lies in the possibility of eschewing the explicit computation of the propagator, traditionally required in alternative methods like path integral calculations, which is hardly obtainable in the majority of the cases. For one-particle equilibrium systems we are able to compute the instantaneous (equilibrium) probability density functions of injected and dissipated power as well as the respective large deviation functions. Our thorough calculations explicitly show that for such models nonlinearities are irrelevant in the long-term statistics, which preserve the exact same values as computed for linear cases. Actually, we verify that the thermostatistical effect of the nonlinearities is constricted to the transient towards equilibrium, since it affects the average total energy of the system. For the two-particle system we consider each element in contact with a heat reservoir, at different temperatures, and focus on the problem of heat flux between them. Contrarily to the one-particle case, in this steady state nonequilibrium model we prove that the heat flux probability density function reflects the existence of nonlinearities in the system. An important consequence of that it is the temperature dependence of the conductance, which is unobserved in linear(harmonic) models. Our results are complemented by fluctuation relations for the injected power (equilibrium case) and heat flux (nonequilibrium case).
我们讨论了受非线性耦合作用且与标准高斯热库接触的小型机械热力学系统(单粒子和双粒子情形)的统计性质。我们使用一种在拉普拉斯 - 傅里叶空间应用平均值的方法,这与终值定理的推广有关。该方法的关键优势在于有可能避免像路径积分计算等传统替代方法中通常需要的传播子的显式计算,而在大多数情况下这几乎是无法实现的。对于单粒子平衡系统,我们能够计算注入功率和耗散功率的瞬时(平衡)概率密度函数以及各自的大偏差函数。我们详尽的计算明确表明,对于此类模型,非线性在长期统计中无关紧要,其保持与线性情形计算出的完全相同的值。实际上,我们验证了非线性的热统计效应局限于向平衡的瞬态过程,因为它影响系统的平均总能量。对于双粒子系统,我们考虑每个粒子与处于不同温度的热库接触,并关注它们之间的热流问题。与单粒子情形相反,在这个稳态非平衡模型中,我们证明热流概率密度函数反映了系统中非线性的存在。其一个重要结果是电导率的温度依赖性,这在线性(谐波)模型中未被观察到。我们的结果通过注入功率(平衡情形)和热流(非平衡情形)的涨落关系得到补充。