Application of Response Surface Methodology in Characterization of Biolubricant with Titanium Oxide Nanoparticles / Nor Syahirah Mohamad and Salmiah Kasolang

An optimized model is often deployed to reduce trial and error in experiment approach and to obtain the multi-variant correlation. In this study, Response Surface Methodology (RSM) namely Box-Behnken design (BBD) approach has been used to investigate the characteristic of lubricant. In BBD, this approach is based on multivariate analysis whereby the effect of different parameters is considered simultaneously. The effect of three parameters namely speed, load and concentration of TiO2 on the coefficient of friction (CoF) was investigated in this study. By using this approach, the number of experiment has reduced to 15 from 100 experiments using optimization method. The result obtained from BBD has shown that the most influential parameters were speed and load. Analysis of variance (ANOVA) indicated that the proposed experiments from quadratic model successfully interpreted the experimental data with a coefficient of determination R2 = 0.9931. From the contour plot of BBD, the optimization zone for interacting variables can be determined. The zone that indicates two regions of lower friction values (<0.04) were: (i) at a speed 1300 to 2000 rpm for a normal load at 10 to 16 kg and (ii) at a speed 700 to 1500 rpm for a normal load in the range of 19 to 20kg. The optimized condition shows that the minimum value of CoF (0.0159) is at the speed of 2000 rpm, load of 10 kg and TiO2 concentration of 1.0 wt%.