Optimization of Mechanism of Boundary Lubrication in Fully Formulated Commercial Engine Oil Using Design of Experiment

Abstract

Design of experiment (DOE) analysis was used to study the desirability factor between contact loads, oil quantity, and surface roughness. The analysis developed a series of interactions between factors to get the best correlations between contact loads and oil quantity that leads to the stabilization of the tribofilm. A closed-loop boundary condition test was developed to examine the behavior of lubricants under boundary conditions. Polished and unpolished testing specimens were established to show the differences in friction and wear profiles under extreme boundary lubrication. The boundary condition test was very reproducible and can be used to study the mechanism of boundary lubrication. The mechanism of antiwear film formation and breakdown was followed carefully by monitoring the friction coefficient over the duration of the test and running scanning electron microscopy (SEM) on selected tests. The thickness of the boundary layer lubricant, which is determined by the concentration of additives in the supplied oil, is optimized for the polished and unpolished test cylinders. The optimized desirability shows the best loading and oil supply condition that leads to greater consistency in the breakdown of the tribofilm for a fixed contact load and fixed amount of fully formulated zinc dialkyldithiophosphate (ZDDP) oil. The number of cycles to breakdown of the protective tribofilm is also consistent with the applied load for a fixed thickness of the boundary lubrication film. It is evident that at lower contact loads a stable tribofilm rich in phosphorous is formed, whereas at higher contact loads the breakdown of the tribofilm results in wear debris and higher sulfur content on the wear surface.

KEY WORDS:

• Boundary Lubrication
• Tribofilm
• Wear
• Surface Roughness
• DOE
• SEM
• EDX

ACKNOWLEDGEMENTS

The authors thank the reviewers for their useful suggestions and the Faculty of Engineering at the University of Balamand. Support provided by the State of Texas as part of a Technology Development and Transfer Grant and Platinum Research Organization LLC is gratefully acknowledged. The authors thank Professor Pranesh B. Aswath, Professor Ronald L. Elsenbaumer, and Dr. Harold Shaub for useful discussions.

Review led by Elaine Yamaguchi