Thin Solid Films, 150 (1987) 105-114 GENERAL FILM BEHAVIOUR 105 PHASE TRANSITIONS IN ADSORBED LAYERS VIII: EVALUATION OF PHASE DIAGRAMS FOR MONOLAYERS OF ARGON, KRYPTON AND XENON ADSORBED ON GRAPHITE A. PATRYKIEJEW AND S. SOKOLOWSKI Institute of Chemistry, Maria Curie-Sklodowska University, 20031 Lublin (Poland) (Received November 7, 1986; accepted January 14, 1987) The theory of monolayer adsorption of simple atomic gases on crystalline solids is presented and is used to calculate the phase diagrams for noble gases adsorbed on graphite. It is shown that the third-order correction terms to the interaction energy between adsorbed atoms must be included in order to obtain results that are in agreement with experimental data. It is also demonstrated that our theory predicts the formation of a commensurate two-dimensional solid phase for krypton monolayer films and an incommensurate two-dimensional solid phase for both argon and xenon monolayer films on graphite. 1. INTRODUCTION During the last few years interest in the theory of first-order phase transitions in monolayer films on solid surfaces has increased. Particular attention has been focused on the formation of solid two-dimensional phases in adsorbed monolayersl-3. Recent theories formulated to describe this type of phenomena have been based on the assumption that the beginning of the formation of a two- dimensional solid phase is indicated by the bifurcation of solutions to the equations describing the equilibrium distribution of particles in the system4-8. So far, the relevant equations for the density distribution have been developed using functional perturbation expansion methods and treating the two-dimensional liquid phase as a reference system4-a. However, as recent investigations of bulk phases have indicated9, the expansion of the properties of a solid phase around those of a liquid phase is often quite a poor approximation and is suspected to be one of the two main sources of discrepancies between theoretical predictions and computer-simulated data. The second source of possible errors is connected with the use of a truncated Fourier series representation of the crystal density. Because of strong non- uniformity in the system, the Fourier expansions are slowly convergent and their early truncation may cause significant errors. The theory of freezing of bulk systems, developed recently by Baus and Colot 9, utilizes the so-called density functional theory of Saam and Ebnerl° to calculate the thermodynamic properties of the system and assumes the superposition of gaussian 0040-6090/87/$3.50 © Elsevier Sequoia/Printed in The Netherlands