Teaching
At the GHZ Lab, we are thermodynamics enthusiasts. We teach thermodynamics in every shape and form: macroscopic and molecular, classical and statistical, theoretical and computational. Whether we are predicting how components distribute between phases or generating molecular trajectories one time step at a time, the underlying question is the same: how do molecular interactions and thermodynamic constraints determine observable behavior?
Our teaching pedagogy emphasizes physical intuition, rigorous problem formulation, and the ability to connect mathematical models with real molecular and engineering systems. We want students to understand not only how to perform a calculation, but also what the calculation means, what assumptions it contains, and when its answer can be trusted.
CHEE 3333 develops the thermodynamic framework needed to understand multicomponent, multiphase, and reactive systems. The course begins with equilibrium, spontaneity, free energy, equations of state, and fugacity before moving to the thermodynamics of mixtures and the behavior of ideal and nonideal solutions.
Students learn to predict how components distribute among vapor, liquid, and solid phases using partial molar properties, activity coefficients, excess properties, and phase-equilibrium models. Applications include vapor–liquid and liquid–liquid equilibrium, gas solubility, flash calculations, ternary phase diagrams, solid–liquid equilibrium, and reaction equilibrium. Throughout the course, these concepts are connected to engineering applications such as distillation, fractionation, and other separation processes.
The broader goal is to help students see thermodynamics as a coherent physical framework rather than a collection of unrelated equations. Fugacity, activity, and free energy become tools for answering concrete questions: Will a process occur spontaneously? Which phases will form? Where will each component go? How will temperature, pressure, and composition change the outcome?
CHEE 6385 provides a foundation in modern molecular modeling and simulation. The course connects statistical mechanics with the algorithms used to generate molecular trajectories, calculate thermodynamic properties, and explore free-energy landscapes. Programming experience is helpful but not required; students develop the necessary computational skills through progressively more advanced assignments.
Rather than beginning with a prepackaged simulation program, students first learn how the underlying methods work. They develop their own molecular dynamics and Monte Carlo codes for a Lennard-Jones liquid and use computational assignments to connect microscopic configurations and motions with macroscopic properties. This foundation prepares them to use established simulation packages more critically and effectively.
Topics include quantum and classical statistical mechanics, model resolution, empirical force fields, potential-energy landscapes, molecular-dynamics integrators, periodic boundary conditions, simulation ensembles, statistical uncertainty, Metropolis Monte Carlo, free-energy calculations, and potentials of mean force. The course also introduces enhanced and rare-event sampling methods—including replica exchange, umbrella sampling, multiple-histogram analysis, metadynamics, and on-the-fly biasing—as well as transport properties, multiscale modeling, and advanced Monte Carlo techniques.
Weekly problem sets and programming assignments are complemented by a final project that allows students to investigate a molecular-simulation question connected to their own research interests. The course emphasizes good coding practices, thoughtful selection of model resolution, proper matching of computational methods to the quantities of interest, and critical evaluation of simulation results.