Journal of Chromatography A, 1265 (2012) 144–148 Contents lists available at SciVerse ScienceDirect Journal of Chromatography A j our na l ho me p ag e: www.elsevier.com/locate/chroma Determination of partition coefficients of refrigerants by gas liquid chromatographic headspace analysis Michael H. Abraham a, , Javier Gil-Lostes a , Stuart Corr b , William E. Acree Jr. c a Department of Chemistry, University College London, 20 Gordon St, London WC1H 0AJ, UK b 31 Foxhills Close, Appleton, WA4 5DH Cheshire, UK c Department of Chemistry, 1155 Union Circle Drive #305070, University of North Texas, Denton, TX 76203-5017, USA a r t i c l e i n f o Article history: Received 1 June 2012 Received in revised form 23 September 2012 Accepted 24 September 2012 Available online 1 October 2012 Keywords: Partition coefficient Headspace analysis Refrigerants a b s t r a c t Gas–water partition coefficients, K w , and gas–solvent partition coefficients, K s , have been determined for chlorodifluoromethane and for 1,1,1,3,3,3-hexafluoropropane by headspace analysis, using a very simple experimental procedure. These partition coefficients then yield water–solvent partition coefficients, P s . Where comparisons can be made there is excellent agreement with literature values for K w and P s . The obtained values of K s and P s can be used to obtain physicochemical properties, or descriptors, for the refrigerants. Combination of these descriptors with previous equations we have developed enables partition coefficients to be obtained for a host of systems. © 2012 Elsevier B.V. All rights reserved. 1. Introduction A standard procedure for the determination of water–solvent partition coefficients is the ‘shake-flask’ method. A compound is partitioned between water and an immiscible, or a partly immiscible solvent, and after equilibration the concentration of the solute in each phase is determined, often through UV/vis analy- sis [1,2]. Another well-known method involves determination of the capacity value of a compound on a suitable reversed-phase HPLC system that has been calibrated using a number of reference solutes [3]. Analysis is again often carried out using UV/vis spec- trometry. There are a number of compounds, however, for which determination of partition coefficients by these methods is diffi- cult. These include compounds that are not UV/vis active and very volatile compounds for which specific manipulation techniques have to be used. We wished to obtain water–solvent partition coefficients for 1,1,1,3,3,3-hexafluoropropane, R-236fa, bp -1 C, and chlorodifluoromethane, R-22, bp -41 C, and decided to use GLC headspace analysis to obtain gas–water partition coefficients, K w , and gas–solvent partition coefficients, K s , from which the corre- sponding water–solvent partition coefficients, P s , can be obtained. The use of headspace analysis to obtain gas–solvent partition coefficients or activity coefficients in a given solvent is well- known [4–6], but we wished to introduce a modification that leads Corresponding author. E-mail address: m.h.abraham@ucl.ac.uk (M.H. Abraham). to gas–solvent partition coefficients with almost no calibration required. 2. Materials and methods Gas–water, K w , or gas–solvent, K s , partition coefficients can be defined in terms of equilibrium concentrations of the solute, through Eq. (1). Then K w and K s are dimensionless. K s = Conc. of solute in solvent, in mol dm -3 Conc. of solute in the gas phase, in mol dm -3 (1) Water–solvent partition coefficients, P s , can be obtained through Eq. (2). If K w and K s refer to the pure water and sol- vent phases, then P s will refer to transfer of a solute from water to the pure solvent phase. However, P s values determined by the ‘shake flask’ method refer to transfer from water saturated with the organic solvent to the organic solvent saturated with water, so care has to be taken over possible differences between the two types of partition coefficient. P s = K s K w (2) The usual method for the determination of K s by head space analysis is to equilibrate the solute between a solvent and the headspace above the solvent and then to determine the concen- tration of the solute in each phase by GLC analysis. Zenkevich and Makarov [6] have suggested the use of two reference solutes, but we found this to be unnecessary. 0021-9673/$ see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2012.09.085