(Editor's note: The following is taken from the Web site of the Air-Conditioning and Refrigeration Institute, www.ari.org.) The Air-Conditioning and Refrigeration Technology Institute continues to release interim and final results from its Materials Compatibility and Lubricants Research program. Funded by a U.S. Department of Energy grant and the air conditioning and refrigeration industry, the program is evaluating the properties and compatibilities of lubricants and non-chlorofluorocarbon refrigerants. This article summarizes the findings of Selda Gunsel and Michael Pozebanchuk of Pennzoil Products Co., in their investigation of elastohydrodynamic lubrication properties of R-134A and R-410A with various lubricants, using R-22 and mineral oil as the baseline.


Properties of refrigeration lubricants were investigated under high-pressure nonconforming contacts at different conditions of temperature, rolling speed and refrigerant concentration. The test program was based upon the recognition that the lubrication regime in refrigeration compressors is generally elastohydrodynamic or hydrodynamic, as determined by the operating conditions of the compressor and the properties of the lubricant. Depending on the design, elastohydrodynamic lubrication conditions exist in many rolling and sliding elements of refrigeration compressors such as roller-element bearings, gears and rotors.

The formation of an elastohydrodynamic film separating rubbing surfaces is important in preventing the wear and failure of compressor elements. It is, therefore, important to predict the elastohydrodynamic or "EHD" performance of lubricants under realistic conditions. This is, however, difficult as the lubricant properties that control film formation are critically dependent on pressure and shear, and cannot be evaluated using conventional laboratory instruments.

In this study, the elastohydrodynamic behavior of refrigeration lubricants, with and without the presence of refrigerants was investigated using the ultra-thin film, EHD "interferometry" technique, modified by the researchers for the study of refrigerant-lubricant mixtures. This technique enables very thin films, down to less than five nanometers, to be measured accurately within an EHD contact under realistic conditions of temperature, shear and pressure. Film thickness measurements were obtained on refrigeration lubricants as a function of speed, temperature, and refrigerant concentration in the lubricant. The effects of lubricant viscosity, temperature, rolling speed, and refrigerant concentration on EHD film formation were investigated. From the film thickness measurements, effective pressure-viscosity coefficients were calculated.

The lubricants studied in this project included two naphthenic mineral oils or NMOs, four polyolesters or POEs, and two polyvinyl ether, called PVE, fluids, representing viscosity grades of ISO 32 and ISO 68. Refrigerants studied included R-22, R-134A, and R-410A. Film thickness measurements were conducted at 23?C, 45?C, and 65?C with refrigerant concentrations ranging from zero percent to 60 percent by weight.

Results as expected

All of the lubricants studied behaved as expected from the EHD theory under air. EHD film thickness increased with speed and dynamic viscosity and decreased with temperature. Effective pressure-viscosity coefficients (alpha) calculated from the film thickness data showed the effect of chemical structure on the pressure-viscosity characteristics of the fluids. The fluids were ranked with respect to their pressure-viscosity coefficients in the following order: NMO, PVE, POE.

Differences were observed in the pressure-viscosity characteristics of the polyolesters studied. This was related to the degree of branching in the ester structure. Esters with branching have higher alpha values than those with linear structure. Effective pressure-viscosity coefficients also change with temperature, decreasing as the temperature increases.

Refrigerants have a significant effect on reducing the EHD film formation ability of lubricants. EHD film thickness decreases drastically in the contact as the refrigerant concentration in the lubricant increases. Even at the low refrigerant concentration of 10 percent, the reduction in film thickness ranges from 30 percent to 65 percent depending on the test temperature. Refrigerants reduce dynamic viscosity as well as the pressure-viscosity coefficients of lubricants; however, this effect decreases as the temperature increases. The thickness of the EHD film formed by the refrigerant-lubricant mixtures shows a similar dependence on speed as that formed by the lubricant itself. Under some conditions (high refrigerant concentrations, temperatures and pressures), some deviations from the EHD theoretical slope for the film thickness/speed relationship are observed.


Effective pressure-viscosity coefficients for the refrigerant-lubricant mixtures were calculated from the film thickness data using the theoretical relationship of Dowson-Hamrock and dynamic viscosity data available for the mixtures in the literature. The accuracy of the calculated pressure-viscosity coefficients depends strongly on the accuracy of the dynamic viscosity data used in the calculations.

For a given refrigerant and lubricant mixture, pressure-viscosity coefficient increases linearly in proportion to the logarithm of dynamic viscosity. This finding suggests that the same fundamental molecular properties govern changes in both dynamic viscosity and pressure-viscosity properties of fluids.

The ranking obtained with respect to the pressure-viscosity characteristics of the lubricants under air was also observed under refrigerant environments. However, the differences became smaller as the refrigerant concentration and the temperature increased. In general, refrigerant (R-134A or R-410A) mixtures with PVEs have higher alpha values than those with POEs. Mixtures of NMO and R-22 have higher alpha-values than those of both POEs and PVEs with R-134A or R-410A.

The presence of R-410A in the POE and PVE lubricants results in thinner EHD films than those produced by the same lubricants in the presence of R-134A.

A graphical ranking of all the tested lubricant/refrigerant combinations is available in the final report. The film thickness data reported in this chart were obtained under the same conditions of temperature (22.5¿C) and speed for ISO 32 and lubricants and their mixtures with refrigerants.

(For information on ordering the report, write to ARTI Database, c/o James M. Calm, Engineering Consultant, 10887 Woodleaf Lane, Great Falls, VA 22066-3003.)