[Paper Review] What can we still learn from Brownian motion?

Qiuping A. Wang|arXiv (Cornell University)|Jan 3, 2011

Advanced Thermodynamics and Statistical Mechanics11 references3 citations

TL;DR

This paper proposes a generalized mechanics framework for stochastic dynamics by showing that the path probability of conservative or weakly damped Brownian motion under Gaussian random forces follows an exponential dependence on Lagrangian action. This leads to a fundamental principle extending the least action principle, breaking Liouville's theorem and offering a new foundation for the Boltzmann H-theorem, with stochastic dynamics as the general case and regular Hamiltonian mechanics as a special limit when randomness vanishes.

ABSTRACT

Recent result of the numerical simulation of stochastic motion of conservative mechanical or weakly damped Brownian motion subject to conservative forces reveals that, in the case of Gaussian random forces, the path probability depends exponentially on Lagrangian action. This distribution implies a fundamental principle generalizing the least action principle of the Hamiltonian/Lagrangian mechanics and yields an extended formalism of mechanics for random dynamics. Within this theory, Liouville theorem of conservation of phase distribution breaks down. This opens a way to the Boltzmann H theorem. We argue that the randomness is a crucial distinction between two kingdoms of Hamiltonian/Lagrangian mechanics: the stochastic dynamics and the regular one which is a special case of the first one for vanishing randomness. This distinction was missing in the criticisms of this theorem from Loschmidt, Poincaré and Zermelo.

Motivation & Objective

• To investigate the statistical behavior of conservative or weakly damped Brownian motion under Gaussian random forces.
• To determine whether path probabilities in stochastic dynamics follow an exponential dependence on Lagrangian action.
• To establish a generalized mechanics formalism that extends Hamiltonian/Lagrangian mechanics to include randomness.
• To resolve long-standing criticisms of the Boltzmann H-theorem by identifying the missing distinction between stochastic and regular dynamics.
• To show that Liouville's theorem on phase space distribution conservation fails in stochastic systems, implying a new dynamical framework.

Proposed method

• Numerical simulation of stochastic motion under conservative forces and Gaussian white noise.
• Analysis of path probability distributions in phase space for systems with weak damping.
• Derivation of an exponential path probability law proportional to the exponential of the Lagrangian action.
• Formulation of a generalized mechanics principle that unifies deterministic and stochastic dynamics.
• Use of the action-based path probability to derive a new statistical formalism replacing the classical Liouville theorem.
• Application of the formalism to re-express the Boltzmann H-theorem in terms of stochastic path probabilities.

Experimental results

Research questions

• RQ1 Does the path probability of Brownian motion with conservative forces depend exponentially on the Lagrangian action?
• RQ2 How does the generalized action principle differ from the classical least action principle in deterministic mechanics?
• RQ3 Why do traditional criticisms of the H-theorem by Loschmidt, Poincaré, and Zermelo fail to account for the distinction between stochastic and regular dynamics?
• RQ4 In what way does the breakdown of Liouville's theorem emerge in stochastic systems, and what are its implications?
• RQ5 Can the Boltzmann H-theorem be re-derived from a fundamental principle of path probability in stochastic dynamics?

Key findings

• The path probability of weakly damped or conservative Brownian motion under Gaussian noise is exponentially dependent on the Lagrangian action, expressed as P ∝ exp(−S_L / k_B) where S_L is the action and k_B is a constant.
• This exponential path probability implies a fundamental principle that generalizes the classical least action principle, extending it to stochastic dynamics.
• The Liouville theorem, which conserves phase space distribution in deterministic mechanics, breaks down in the stochastic case due to the non-conservative nature of the path probability evolution.
• The framework reveals that regular Hamiltonian mechanics is a special case of stochastic dynamics when the noise amplitude tends to zero.
• The theory provides a new foundation for the Boltzmann H-theorem by deriving its irreversibility from the statistical properties of stochastic paths rather than from assumptions about molecular chaos.