Berthelot proposed the first accurate and widely applicable correlation for the second virial coefficient in 1907. His goal was to develop a method to calculate very accurate departures from the ideal-gas equation of state so that molecular weights could be derived from vapor-density measurements at normal conditions. He developed his model using theoretical considerations as well as judicious approximations based upon the patterns of the experimental data. The Berthelot model was later used, for example, to infer polar effects on the second virial coefficient and to develop the well-known Pitzer-Curl correlation that introduced the acentric factor as the third parameter. This paper analyzes the contribution of Berthelot from a historical perspective to identify and critique good and poor practices in the development of applied thermodynamics. The good practices are that later scientists and engineers built upon the ideas of Berthelot and applied them to problems outside Berthelot's experience. The poor practices include not providing proper credit to Berthelot, incorrectly referencing his papers and reinventing the wheel. The purpose of this paper is use the Berthelot case to strengthen the good practices of applied thermodynamics and to correct the bad ones.

D. Berthelot, Trav. Mem. Bur. Intern. Poids measures, No. 13 (1907)