3.2. Friction Behavior Analysis
Figure 3 shows the friction behavior of DF-CN48, which was influenced by the POB percentages in the blends. It can be seen that the friction coefficient fluctuated during the test time. Based on the results, the friction coefficient value decreased with an increasing POB percentage in DF-CN48 due to the lubricating properties of the methyl ester compounds. Specific confirmation of the positive impact of POB on the lubricity of diesel fuel is found in the research conducted by Lapuerta et al. [
25]. Adding 10% biodiesel to e-diesel (7.7% ethanol
v/
v in diesel fuel) reduces the corrected WSD from 327 μm to 275 μm compared with pristine e-diesel (without % biodiesel
v/
v).
The highest average friction coefficient value is 0.150, which is the average friction coefficient for DF-CN48, and the lowest is for POB, with an average friction coefficient value of 0.126. The friction coefficient value with an increasing POB percentage looks tenuous, even though there are some points at specific minutes whose values are close together. From
Figure 3, it can be analyzed that the film percentage value is not directly proportional or inversely proportional to the POB volume in DF-CN48, in contrast with the friction coefficient value, which decreases with the addition of POB to DF-CN48. According to Mei et al. and Sundus et al. [
26,
27], the decrease in the friction coefficient with the increasing POB percentage could be attributed to the methyl ester compound in the DF-CN48 blend. The ester group in POB has lubrication properties whereby they are adsorbed on the metal surface to soften the friction [
26]. Interestingly, increasing levels of POB in DF-CN48 resulted in smoother WSD values and smaller diameters. This phenomenon is due to the adsorption of ester molecules on the metallic surfaces, which acted as a protective layer during the rubbing process [
13,
28].
Figure 4 shows the values of the friction coefficients, which decrease with an increasing volume of POB. It is clear that methyl esters have high lubricating properties, resulting in a reduced friction coefficient and increased film percentage with the effect of the POB volume in DF-CN51. Based on the results, DF-CN51 has the highest friction coefficient value with an average of 0.293 compared with the POB friction coefficient of 0.126. The experimental results show that adding biodiesel decreases the friction coefficient from 0.30 to 0.14–0.12 for B10–B90. It is worth noting that having no lubricating film between the contacting surfaces during the test led to a very high friction coefficient for DF-CN51. After adding POB, steady-state conditions were achieved due to it forming a thin lubricating protective layer between the metallic surfaces, leading to a lower friction coefficient. The pure POB sample showed a minimum friction coefficient due to the adsorption of ester molecules on the metallic surfaces, which acted as a protective layer during the rubbing process [
29,
30]. It is clear that the ester molecules in POB in the DF-CN51 fuel samples assisted in forming a protective lubricating film between the contacting surfaces. From
Figure 3 and
Figure 4, it is evident that POB acts as a lubricant additive and a friction modifier for diesel fuel (DF-CN48 and DF-CN51) based on the principle of film formation on metal surfaces. On these protected engine components, the lubricant additive has a mechanism of being adsorbed on the surfaces to reduce the friction coefficient of the engine components in contact with each other [
17,
31].
The effect of biodiesel in HVO on the fuel’s lubrication properties is shown in
Figure 5. An interesting trend was found in the friction coefficient value with the increasing POB content in the HVO. However, it is worth noting that a non-equal contribution towards the friction coefficient was also observed when the POB concentration was beyond 60–90%-
v/
v. As a result, pristine HVO has a friction coefficient of 0.125, and the value decreases gradually to the lowest at B30 with a friction coefficient of 0.121. At B40, the friction coefficient value increases gradually until the highest value at B80 of 0.128. The film percentage produced with B0 is 93%, the same as that with B100, and increases for B10, B20, B30, B40, and B50 with a film percentage of 95%. Then, the film percentage decreases to 91% for B60, B70, B80, and B90. Based on the results, 10–50% POB blends in HVO showed noticeable lubricity improvements with friction coefficients lower than that for pristine HVO. According to the literature, mixing biodiesel and HVO must produce effective lubricity properties to prevent the removal of the adsorbed film formed under low-stress conditions at the test temperature and load. Nevertheless, the higher ester content in POB-HVO blends did not contribute to the lubricity capacities of the fuel blends (for 60–90% POB). It is clear that a change in the saturation degree in higher-POB blends did not cause a strong improvement in the HVO blends’ lubricity capacities [
12,
28].
3.3. Wear Characteristics
The wear characteristics of each fuel blend sample in the HFRR test occurred on the surface of the test material on a friction node determined by the physical processes. The appearance of the wear scar on a ball can vary with the fuel type, particularly when lubricity additives are present. The wear scar appears to be a series of scratches in the ball’s direction of motion, somewhat larger in the x direction than in the y direction. In some cases, for example, when low-lubricity reference fluids are tested, the boundary between the scar and the discolored (but unworn) area of the ball is distinct, and it is easy to measure the scar size. In other cases, the central scratched part of the scar is surrounded by less distinct wear, and there is no sharp boundary between the worn and unworn areas of the ball, and it shows cracks in different directions.
An increase in the methyl ester (POB) in the diesel fuels (DF-CN48 and DF-CN51) increased the oxygen molecules in the samples. The presence of oxygen is conducive to tribochemical processes, which, in turn, leads to a change in the chemical and phase composition of the material [
13,
32]. It is clear that the evolution of the chemicals and phases led to metal oxides forming and a reduction in the friction coefficient of the friction node. The lubricating properties of generated metal oxides and their mixtures are determined by their ionic potentials, as pointed out in [
15,
33,
34]. A higher ionic potential carries out a stronger polarization effect leading to a lower friction coefficient. Low ionic potential in fuel can increase the friction coefficient and intensify wear, with a higher wear scar diameter (WSD) value [
8,
35]. According to Kuszewski et al. [
17], at low oxygen concentrations in a sample, the wear is decreased by the oxygen due to the formation of anti-adhesive oxide layers. Higher concentrations lead to excessive oxidation of the friction surface, leading to more intensification of wear. However, the wear characteristics phenomenon in the POB blends in DF-CN48 and DF-CN51 were associated with the anti-adhesive oxide layer formation and wear resistance increase at the beginning of the test. Clearly, such a phenomenon is dominant in generating wear scars [
36,
37]. As shown in
Figure 6,
Figure 7 and
Figure 8, confirmation of the tribochemical reactions within the test friction pair is given in the photographs showing wear scars on the test balls. The effect of POB addition into DF-CN48, DF-CN51, and HVO was analyzed for its degree of conformity correlation with the WSD formation and its significance model to evaluate the results further.
Figure 6 shows the wear scars produced with each additional volume of POB in DF-CN48. The results show decreased WSD values when increasing the POB volume in DF-CN48. Successively, rising levels of POB in DF-CN48 resulted in smoother WSD values and smaller diameters. DF-CN48 has an immense WSD value of 288.5 μm, which decreases with the POB volume in DF-CN48. Meanwhile, the smallest value of the wear scar diameter with POB is 206.5 μm, with a smooth and precise WSD character. Clearly, every 10%-
v/
v addition of POB to DF-CN48 results in a 3% decrease in WSD formation. As shown in
Figure 6, the addition of biodiesel improved the lubricity of DF-CN48, with a polynomial fitting chosen for the generated model. Furthermore, the calculated
R2 value (0.9951) of the polynomial fitting was also considerably high and considered to be in good agreement. It is clear that all the DF-CN48 and biodiesel blends following the model satisfy the limit of ±5%.
The wear scar characteristics resulting from the variation in POB in DF-CN51 are shown in
Figure 6. The wear scar looks more precise and smoother with the increase in POB in DF-CN51. Methyl ester is an active compound that can become concentrated on a metal surface, forming a thin adsorbed layer. The coating can naturally change the molecular structure (via intermolecular forces) and surface characteristics of the metal. It causes a change in the process kinetics involved in transferring substances across the surfaces that rub against each other [
13,
15]. As a result, an increase in POB in DF-CN51, results in a change in the conditions of the molecular interactions between the two contacting surfaces and their effect on minimizing WSD formation [
13,
17]. It is worth noting that adding the first 10%-
v/
v biodiesel to DF-CN51 resulted in the optimal WSD improvement, with a decrease of 22%. Adding 10% POB further reduced WSD formation by 2–7% (B20–B90). These results indicate that the effect of biodiesel as a lubricant improver additive is very influential on the type of fuel used. In this study, DF-CN48 and DF-CN51 have differences in their compositional hydrocarbon compositions, which are determined via the differences in their distillation ranges, as shown in
Figure 1. For further analysis, the polynomial fitting expresses the effect of biodiesel addition on DF-CN51, as demonstrated in
Figure 7. According to the results, the model’s calculated
R2 value (0.8828) is considerably high, implying that the model accurately navigates the correlations of all the samples following it satisfying the limit of ±10%.
Figure 8 shows that increasing the WSD increased the percentage of POB in the HVO mixture. The wear scar character in the HVO mixture tends to be longer and smoother as the percentage of the POB volume increases. Many dominant vertical strokes are found on B70, B80, and B90. The test results show that biodiesel in HVO only affects the WSD by 1–3% (B10–B50). However, it is worth noting that a non-equal contribution to the WSD was also observed when the POB concentration was beyond 60–90%-
v/
v. These results suggest optimizing biodiesel as an additive improver to increase the effectiveness of the material’s lubricity is imperative. According to [
17,
29], compounds with good lubrication properties are adsorbed as thin layers on metal surfaces. These layers can be easily cut off and easily distracted by rough moving surfaces. The most exciting thing is that all these compounds (methyl esters) function via adsorption on the metal surface, and the adsorbed layer can be easily removed due to friction on the rough surface. After removing the stress, the adsorbed layer can return to its original position [
7]. This phenomenon can be described in the experimental results of the POB-HVO blends. In such a non-ideal mixture, the methyl ester in POB and the paraffinic compounds in HVO show effects of lubricity that compete determinatively and attractively with each other, resulting in a non-ideal contribution to the friction coefficient and WSD value.
According to the requirements for fuels determined in the Worldwide Fuel Charter for category 1, 2, and 3 fuels for compression-ignition engines, the WSD (determined using an HFRR) cannot exceed 460 μm, whereas for category 4 and 5 fuels, this value cannot be greater than 400 μm. Thus, all the blends of DF-CN48 and HVO with POB in
Figure 8 meet the requirements for all fuel categories for compression-ignition engines. Interestingly, with POB blends, DF-CN51 can fulfill the requirement in the WWFC, whereby pristine DF-CN51 has the highest WSD (467 μm).
Table 5 summarizes the results of a generated model using ANOVA analysis. The ANOVA suggested that the polynomial model was the best based on the results. It represents the relationship between the lubrication properties of the fuel test samples (
Y) and the concentration of POB (
x). As shown in
Table 3, the calculated F-value for the generated model of each blend of DF-CN48, DF-CN-51, and HVO with POB was 402.85, 57.51, and 19.10, respectively. The results indicate that the model is significant, with a 0.01% chance of occurring due to noise. Furthermore, the considerable values of the error probability (
P) were also found to be less than 0.05 (<0.0001 for DF-CN48 and DF-CN51 with POB blends and <0.005 for HVO with POB blends). This suggests that the addition of POB blends significantly affects the values of the DF and HVO blends’ wear scars with a probability level of 95%. Therefore, the generated polynomial model is sufficient for describing the fuels’ lubrication properties.
The specific wear rate for each fuel blend was determined in this study according to the literature [
11]. The specific wear rates (
Wball) were calculated as per Equations (2)–(4):
where
d is the average wear scar diameter on rubbed balls,
r is the tested ball radius,
S is the sliding distance, and
N is the normal load.
Figure 9 shows the impact of frictional species growth kinetics on the surface material wear. According to the results, DF-CN48, DF-CN51, and HVO have specific wear rates of 126, 311, and 66 μm
3/Nm, respectively (where pristine POB has the lowest specific wear rate of 64 μm
3/Nm). The addition of POB decreased the specific wear rate. As a result, the POB used in this work contains methyl ester compounds that act as lubricity improvers, leading to the lower typical wear rates measured in DF-CN48, DF-CN51, and HVO. It is clear that the specific wear rates correlate linearly with the wear scar diameters and friction coefficients. The lower the friction coefficient of the fuel blends, the lower the wear diameter and specific wear rate they have, signifying the higher lubricity of the fuels.