Accuracy, Repeatability, and Limitations for Determination of Chemical Activities from Vapor Pressure Osmometry.

Abstract

Vapor pressure osmometry presents a convenient method to measure chemical activity. The work presented here was carried out to provide confidence in using this technique for a VPO model that does not utilize the "hanging-drop" method. While validation studies are available for certain models of vapor pressure osmometers, none were located for the UIC Jupiter 833 osmometer. This study addresses that need by providing a comparison between original experimental data on sodium chloride, calcium chloride, and sodium sulfate solutions to values calculated using the Pitzer equations. A comparison is also made for experimental data on sucrose with a literature correlation. This study briefly reviews the assumptions going into the equation used to relate the osmometer signal to the diluent activity in order to identify potential sources of the error and the noise in the data. The experimental data shows that the UIC 833 osmometer yields diluent activity values accurate to an average of 0.02%, allowing calculation of osmotic coefficients and solute activity coefficients. Further studies need to be conducted on the accuracy at concentrations above 1.4 m. Qualitatively, however, comparison suggests the UIC Jupiter 833 osmometer yields more scatter in the experimental data than the Knauer instruments. Using more uniform mesh caps on the thermistors could possibly reduce that scatter. Finally, we show that replacing the glass thermistors with in-house made thermistors with Teflon incorporated in the structure give reproducible results and that certain modifications to the design are possible without losing accuracy in the measurements.