Interactions of Hydrofluorocarbons and Polyolester Lubricants

A variety of hydrofluorocarbon (HFC) compounds, both as pure substances and mixtures, are being developed as alternative refrigerants to replace the exsting chlorofluorocarbons and hydrofluorocarbons which have been linked to stratospheric ozone depletion. The use ofHFC's in refrigeration systems will require new lubricants, and polyolesters are being widely promoted as the lubricant of choice for HFC refrigerants. Understanding the chemical and physical properties of mixtures of the new lubricants and proposed HFCs is of utmost importance for applying the alternatives to existing or new equipment. Recently, we have developed a new theoretical model for oiUrefrigerant interactions (solubility and viscosity). In the present report, experimental data are compared with the model. In addition, a new method to determine the miscibility gap of refrigerant/oil mixtures has been developed and tested. INTRODUCTION Various physical and chemical properties of HFC-lubricant mixtures are needed to develop useful products as alternative refrigerants for new and existing equipment. The chemical stability of HFC's in the presence of oil is one concern, as are the thermodynamic/thermophysical properties such as mutual solubility (miscibility and immiscibility) and viscosity of mixtures. Since HFCs are soluble (or at least partially miscible) with polyolester oils, the oil will act as another comi>onent of the refrigerant. This means the refrigerant (pure or mixture) contains an additional compound, and all thermodynamic properties such as vapor pressure, compositions of vapor/liquid phases, etc. will be affected due to the extra component. In this report, we are particularly interested in the thermodynamic aspects of 1-WC/polyolester oil mixtures. A recent development of our theoretical modeling on the oiUrefrigerant interactions<1> will be further examined. The model is constructed using the vapor-liquid equlibrium (VLE) data of binary systems (oiUrefrigerant). Since it is based on thermodynamic theory, it should be able to predict all thermodynamic properties, including the vapor-liquid-liquid equilibrium (VLLE) of mixtures. The VLLE prediction is known to be a severe test for the validity of such a model, while the experimental method used commonly to obtain such data is tedious and often contains large uncertainties. Therefore, we have developed a new experimental method to obtain such data to test the model prediction. Another test for the validity of the model is to compare predicted vapor compositions of oiUrefigerant mixtures with data obtained from Gas Chromatography (GC) experiments. EXPERIMENTS 1) VLLE measurments When two partially miscible fluids are mixed and allowed to reach equilibrium, one can visually observe the formation of two separate liquid phases. In the case where the oil density is heavier than that of the refrigerant at the experimental temperature (T), an oil-rich layer forms on the bottom and a refrigerant-rich layer on the top. Traditional