Mathematics and Computers in Simulation 72 (2006) 184–189 The modeling of colloidal fluids by the real-coded lattice gas Y. Sakazaki , S. Masuda, J. Onishi, Y. Chen, H. Ohashi Department of Quantum Engineering and Systems Science, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan Abstract We model colloid-dispersed fluids using the real-coded lattice gas, in which colloidal particles are treated as hard spheres interacting with solvents of the lattice gas particles. We simulate the Brownian motion of a colloidal particle and compare its velocity distribution with that from the theoretical prediction. Both agree well showing that thermal fluctuation is naturally incorporated. We also measured the drag force acting on a colloidal particle and compare the drag coefficient with existing data. It is found that higher space and time resolutions are required in order to model the correct colloidal fluids interactions. © 2006 Published by Elsevier B.V. on behalf of IMACS. Keywords: Real-coded lattice gas; Colloid; Thermal fluctuation; Drag coefficient 1. Introduction The real-coded lattice gas model (RLG) [8], or sometimes called the stochastic rotation dynamics (SRD) method, can be considered as an extension of the lattice gas automata (LGA) [1]. In RLG, space is divided into cells using a regular lattice, and the fluid is represented by fictitious particles moving from cell to cell. The important difference from LGA is that the position and velocity of RLG particles are real numbers, that is, the position and the motion of RLG particles are not restricted by the lattice structure. With this modification, RLG recovers the hydrodynamics without unphysical effects at the macroscopic level, and in the mean time, incorporates thermal fluctuation, naturally. Furthermore, it becomes much easier to handle moving boundaries of spherical solid particles. The above-mentioned features are advantageous particularly in simulations, in which thermal fluctuation and fluid– solid interactions play important roles. These fluid–solid systems with a large amount of interface are hardly treated by the conventional methods such as the finite difference method. LGA has been applied to simulations of colloidal fluid by Ladd and Frenkel[7]. In this paper, we describe a novel simulation model for colloid-dispersed fluids based on the RLG model, and investigate its validity. The rest of this paper is organized as follows. In the next section, we briefly introduce the RLG method, and describe the boundary condition at solid–fluid interfaces. There, we apply a simple algorithm proposed by Hashimoto instead of the bounce-back used in LGA [7] and the lattice Boltzmann method (LBM) [5,6]. In Section 3, we show the validation Corresponding author. E-mail address: sakazaki@crimson.q.t.u-tokyo.ac.jp (Y. Sakazaki). 0378-4754/$32.00 © 2006 Published by Elsevier B.V. on behalf of IMACS. doi:10.1016/j.matcom.2006.05.032