Keith Gubbins received his B.S. in Chemistry from Queen Mary College in the University of London. He went on to receive his Ph.D. degree in Chemical Engineering from King's College at the University of London in 1962. The topic of his doctoral research, which was largely experimental, was the kinetics of reactions in fluidized beds. He then joined the University of Florida in Gainesville as a postdoctoral fellow to work on the mass transport of gases in fuel cells with Robert D. Walker. In 1964 he became an assistant professor in the Department of Chemical Engineering at the University of Florida and then moved rapidly through the ranks to become a full professor in 1972. While at the University of Florida he co-authored, with the late Tim Reed, Applied Statistical Mechanics (McGraw-Hill, 1973), the first such text aimed at practical applications to fluids, with a particular orientation towards chemical engineering. In 1973 very few chemical engineers were aware of the potential of statistical mechanics in their field. In 1976 he moved to Cornell University as the Thomas R. Briggs Professor of Engineering, where he spent the next 22 years, serving 7 years as Director of the School of Chemical Engineering there (1983–1990). At Cornell, he was a member of the Graduate Fields of Chemistry, Applied Physics, Applied Mathematics and History of Science, in addition to Engineering. During his period as Director, he led a successful effort to raise funds for a renovation of all of the teaching and research laboratories and was responsible for hiring seven new faculty members. While at Cornell, Keith established a style of conducting research that has become synonymous with him. Using the resources of Cornell and those of various funding agencies he built a vibrant international collaborative laboratory in statistical mechanics and molecular modeling, to which he welcomed many international visiting scientists. In 1998 he moved to North Carolina State University as the W.H. Clark Distinguished University Professor.
Following his largely experimental studies in his doctoral and postdoctoral research, his interests turned increasingly to the application of statistical mechanics to problems in chemical engineering and applied physics. Much of chemical engineering involves the handling and processing of dense fluids and liquids so that a knowledge of their thermodynamic and transport behaviour is essential to the design of process equipment. It was to these areas that he addressed himself, and he has been largely responsible for the introduction of statistical mechanics and atomistic simulation methods (Monte Carlo and molecular dynamics) into chemical engineering. His research has focused on the development of reliable predictive methods based on statistical mechanics for phase and chemical equilibria for mixtures; equations of state for complex fluid mixtures, particularly those involving associating liquids and polymers; thermodynamics, transport in interfaces, including small droplets, gas-liquid and liquid-liquid interfaces and nano-porous media; prediction of viscosity, diffusion coefficients, and thermal conductivities; natural gas storage in porous media.