Towards nationwide implementation of 40% biodiesel blend fuel in Indonesia: a comprehensive road test and laboratory evaluation

This research focused on evaluating the technical viability of using biodiesel with a blending ratio of 40% v/v, which is expected to be implemented soon in Indonesia. Two kinds of biodiesel blends were prepared, a blend of 60% diesel fuel and 40% biodiesel (B40) and a blend of 60% diesel fuel, 30% biodiesel, and 10% hydrogenated vegetable oil (HVO) (B30D10). The fuels were tested on EuroII vehicles without any engine modifications through a 50,000 km endurance road test. Laboratory tests were also performed at certain traveled distances to evaluate various engine parameters, including power, fuel economy, exhaust emissions, and used engine oil properties. Engine components were inspected upon the completion of the road test. Cold-start ability was also examined to confirm the suitability of the investigated biofuels at low-temperature operating conditions in Indonesia. The road test results showed that vehicles fuelled with B40 and B30D10 could reach a distance of 50,000 km without encountering any technical issues. The laboratory evaluation during the road test indicated that B30D10 had a higher power and fuel economy than B40, with a maximum difference of 2%. Furthermore, B30D10 emitted lower CO, HC, and PM emissions than B40 throughout the distance traveled, with maximum differences of 11.4%, 14.7%, and 22.6%, respectively, but led to 15% higher NOx. Engine component inspection and used engine oil analysis confirmed the fulfillment of the manufacturer's recommendations for both B40 and B30D10. Finally, B40 and B30D10 were suitable for operating at low ambient temperatures in Indonesia, confirming them as practical options to be implemented in the nationwide 40% biodiesel blend fuel.


Introduction
Indonesia has shifted to a biodiesel blend fuel since 2013 to reduce its significant dependence on fossil diesel fuel. Biodiesel is currently implemented with a volume ratio of 35% (B35). Shortly, Indonesia intends to increase the volumetric percentage of biodiesel in fossil diesel fuel to 40%. Utilization of biodiesel is considered a strategic effort to reduce the amount of oil imported and increase the share of renewable energy to achieve a 23% renewable energy mix by 2025. Moreover, implementing biodiesel is also one strategy to help Indonesia reach net zero CO2 emissions in the transportation sector by 2060 (IEA, 2022).
Utilizing a high blend ratio of biodiesel has some advantages, including drop-in fuel characteristics, low emission, and good lubricity, but some concerns should also be considered. For instance, Jaroonjitsathian et al. (2016) raised issues of filter clogging, degradation in fuel injection systems, and fuel tank corrosion with 50% of biodiesel blended fuel. They reported that vehicle maintenance schedules should be corrected for biodiesel blending ratios above 20%. Furthermore, biodiesel could also increase the harmful unregulated emission of formaldehyde, acetaldehyde, benzene, toluene, and xylene (Jaroonjitsathian et al., 2016). Metal contaminants, including Cu, Fe, Mn, Na, Mg, Al, and Ca, can be found due to friction wear or biodiesel contaminants, which lead to filter clogging (Patel et al., 2022). The properties of biodiesel, including sources and process, play a significant role in performance, emission, and engine components` lifetime; therefore, it is important to define specific biodiesel characteristics for high blending ratio (Markov et al., 2021) as well as the optimization of engine parameters to achieve better performance and emissions (Vijay Kumar et al., 2018). Moreover, improving biodiesel blend fuel properties could improve engine component lifetime comparable to neat diesel fuel (Reksowardojo et al., 2020). Cardeño et al. (2020) recommended controlling the saturated monoglycerides (SMG) content of neat biodiesel to below 0.4% to minimize fuel clogging issues for efficiently utilizing a high blend ratio of biodiesel fuel. Paryanto et al. (2022) also reported similar results that high saturation at lowtemperature conditions, below the biodiesel cloud point, could increase precipitate formation in fuel filters. The formation of precipitate in biodiesel blend fuel could be suppressed with urea inclusion (Liu and Tao, 2022) and the utilization of hydrotreated vegetable oil (HVO) as a blended fuel (Mansur et al., 2019). HVO is a paraffinic biofuel which results from a hydrotreated process. It does not contain monoglyceride, so that it can be used as an alternative for high biofuel blends. Further, HVO has a higher cetane number and calorific value, better oxidation stability, and lower filter-blocking tendency than fatty acid methyl esters (FAME). HVO has been subject to extensive investigation as a blended fuel with neat diesel fuel, FAME (Mansur et  Overall, enhancing FAME characteristics or the utilization of HVO might be used to implement high-ratio biodiesel blend fuel in real operating conditions. However, few studies have examined high-ratio biodiesel blend fuels without engine modification in real-world road operating conditions. Hence, this study was conducted to fill the gap in existing research by investigating the feasibility of using a 40% biodiesel blend fuel as a substitute for 30% biodiesel blend fuel (B30) in Indonesia's current vehicle technology. To achieve the aims of this research, two types of biodiesel blend fuel were prepared. The first blend consisted of 60% diesel fuel and 40% FAME, designated as B40, and the second blend was a mixture of 60% diesel fuel, 30% FAME, and 10% HVO, designated as B30D10. This study used the improved properties of FAME, including SMG, water content, and oxidation stability, to ensure the proper utilization of 40% biodiesel blends in current vehicles available in Indonesia, which typically operate on B30. Both candidate fuels were evaluated on currently high-population vehicles in Indonesia under 50,000 km road test conditions, as well as in a laboratory at certain traveled distances. Both B40 and B30D10 were thoroughly investigated to ensure they would not negatively impact vehicle performance, emissions, fuel economy (FE), engine components, engine lubricant, and cold start-ability.

Investigated fuels
In this study, B40 and B30D10 were used as test fuels, and the properties of B40 and B30D10 were evaluated based on the requirements of the Decree of the Director General of Oil and Gas No. 146.K/10/DJM/2020 (KESDM, 2020). The testing of biodiesel used for blending into B40 and B30D10 was performed based on the requirements of the Decree of the Director General of Renewable Energy and Energy Conservation No. 189 K/10/DJE/2019 (DJEBTKE, 2019). Additional parameters proposed for developing fuel specifications included the water content parameter from a maximum of 350 to 320 mg/kg, the minimum oxidation stability from 600 to 720 min, and the maximum monoglyceride content from 0.55 wt% to 0.50 wt%. Meanwhile, the characteristics of HVO were investigated in compliance with the requirements stated in the Decree of the Director General of Renewable Energy and Energy Conservation No. 95.K/EK.05/DJE/2022 (DJEBTKE, 2022).

Performance, emission, and fuel economy evaluation
B40 and B30D10 fuels were evaluated on two common rail vehicles that complied with Euro II emissions regulations. The tested vehicle was selected based on its high population in the Indonesian market. Table 1 shows the vehicle specifications. A road test was conducted on Java Island roads for 50,000 km. The route was selected to represent road characteristics in Indonesia, consisting of 40.6% general road and 59.4% toll road. The traveled distance for each vehicle was around 650 km/d.
The emissions, FE, and vehicle performance tests were conducted in a vehicle laboratory at the National Research and Innovation Agency of Indonesia and according to the ECE-04 R83 (Euro II standard), ECE R101, and ECE R85 standards, respectively. In this study, we modified the carbon balance formulas for B40 and B30D10 to obtain better accuracy in FE   Eq. 2 where FE stands for fuel economy in km/L; D is fuel density in g/cm 3 ; mHC denotes HC emission in g/km; mCO is CO emission in g/km; and mCO2 stands for CO2 emission in g/km.
The emissions test and fuel consumption were run by using the New European Driving Cycle (NEDC), which consisted of the Urban Driving Cycle (UDC) and Extra Urban Driving Cycle (EUDC), as shown in Figure 1.
A chassis dynamometer was used for road load simulation in the laboratory, and emissions were sampled by constant volume sampling systems. An analysis of carbon monoxide (CO), carbon dioxide (CO2), hydrocarbons (HC), nitrogen oxides (NOx), and particulate matter (PM) were performed using nondispersive infrared, heated flame ionization, chemiluminescence, and particulate sampling systems, respectively. The specification of the testing facility for this study is provided in Table 2. In this work, uncertainty was calculated based on a statistical analysis of measurement data and the calibration certificate of measurement equipment. The uncertainty is expressed at a 95% confidence level and a coverage factor 2.

Engine components evaluation
The analysis of engine components is an essential part of ensuring the safety and reliability of the engine when using alternative fuel. In this study, inspection of engine components was done before (0 km) and after the road test (50,000 km) for both B40 and B30D10.

Engine oil analysis
The physicochemical characteristics of used engine oil were evaluated at the Lubricant Laboratory of the Product Application Testing Group (BBPMGB LEMIGAS). Engine oil analysis refers to the quality limits of the International Council on Combustion Engines (CIMAC) Recommendation No. 30 (CIMAC, 2011). Sampling of used oil was carried out when replacing (draining) lubricating oil, following the vehicle manufacturer's recommendations. The scheme for collecting used vehicle lubricants was carried out at a distance of 10,000, 20,000, 30,000, 40,000, and 50,000 km.

Cold start ability test
The cold start ability test was conducted to determine the ability of vehicles to start in low-temperature environments when using B40 and B30D10 fuels. Before the test, eight vehicles were serviced, including fuel filter replacement and fuel tank drainage, followed by the filling of B40 or B30D10 fuels. After filling the tank, the vehicles were driven 100 km to ensure the fuel was evenly distributed. Then, vehicles were grouped to be soaked for 7, 14, 21, and 28 d. After the soaking phase was completed, the engine starting phase commenced with an inspection to ensure the battery was in excellent condition. During this stage, the duration of the engine start was measured using a stopwatch with a resolution of 0.001 s, calibrated with an uncertainty value of 0.3 s.

Parameter Unit Specifications
Year  Table 3 shows the characteristics of the tested fuels used in this study. Generally, it was found that all the tested fuels used in the road test activities complied with the limit determined by the Indonesian government.

Fuel characteristics
B30D10 had a lower density compared with B40, which could be due to the paraffinic chain and lower-end boiling point of HVO (Dimitriadis et al., 2018;Setiapraja et al., 2019). The viscosity of B30D10 was lower than B40; however, both fuels were still in the limit value of 4.5 mm 2 /s. The HVO contained in B30D10, owing to its long straight chain molecule structure (Dimitriadis et al., 2018;d'Ambrosio et al., 2022), could lead to a higher cetane number and calorific value than B40, as shown in Table 3. Furthermore, B30D10 also offers other advantages, including better filterblocking tendency value, lower viscosity, monoglycerides, and water content, as well as higher oxidation stability than B40. However, B30D10 has lower lubricity characteristics compared to B40, which could be Table 3.
The characteristics of the tested fuels used in this study. attributed to the very low sulphur content of HVO (Mikkonen et al., 2013). In addition to regulated parameters, fuel cleanliness value was also measured. The result showed that B30D10 had a better fuel cleanliness value than B40 ( Table  3).

Performance and fuel economy
Performance test results of the light-duty vehicles over 50,000 km are shown in Figure 2. B40 achieved maximum power of 78.8, 78.1, and 78.3 kW, while B30D10 exhibited maximum power of 80.5, 80.9, and 83.1 kW at traveled distances of 0, 30,000, and 50,000 km, respectively. No deterioration in maximum power was observed in B30D10 while covering a distance of 50,000 km. Overall, the performance of the vehicles fuelled with B30D10 was approximately 2.14% higher than those powered by B40. Moreover, the B30D10-powered vehicles showed an increase in power of up to 3.26% during the travel distance of 50,000 km. In contrast, the power produced by vehicles operated on B40 fuel did not show a significant change. The presence of HVO in B30D10 fuel could be highlighted as the main reason for higher power compared to using B40, which could be explained by HVO's higher cetane number and higher mass (calorific value multiplied with density) and volumetric calorific value compared to FAMEs (Bortel et al., 2019;Serrano et al., 2021). On the other hand, B30D10 also possessed a lower density and viscosity than B40, which could also potentially result in a higher power through enhancing fuel spray characteristics and the consequent improved atomization (Mujtaba et al., 2020;Ropandi et al., 2022). Figure 3 shows the FE in km/L based on the R101 test cycle. FE test indicated that the B40 exhibited FE of 9.57, 9.72, and 10.01 km/L, whereas the B30D10 showed FE values of 9.67, 9.91, and 9.95 km/L across the distances of 0, 30,000, and 50,000 km, respectively. Despite the increasing trends observed for both fuels, these differences were insignificant.
The results revealed that the difference in the calorific value between B40 and B30D10 did not influence FE due to characteristics of the EUDC cycle in R101 rarely requiring maximum power for tested vehicles. Karavalakis et al. (2016) also reported similar results where the HVO and FAME blended fuels did not significantly differ in heavy-duty diesel truck transient cycle (HHDDT).

Exhaust emissions
Exhaust gas emissions of the vehicles fuelled with B40 and B30D10 were evaluated by using Euro II standard test method. Vehicles were soaked at room temperature of 23 O C before emission testing for one day. Figure 4 shows the result of the exhaust gas emissions test for B40 and B30D10 at travel distances of 0, 30,000, and 50,000 km.
CO and NOx emissions of B40 and B30D10 tended to decrease as travel distance increased (Figures 4a and c), while HC and PM showed a reverse trend (Figures 4b and d). Emission results revealed that both vehicles fuelled with B40 and B30D10 could fulfill the emission limit of UN-ECE R83-04 (Euro II) for the whole evaluation distance of the road test. The fuel blends containing HVO or FAMEs, have the advantages of a high cetane number and ultra-low sulphur content, as well as the oxygen content for FAME. These characteristics could lead to better combustion efficiency and improved air-fuel mixing, reducing emissions (Bortel et al., 2019; d'Ambrosio et al., 2022; Szeto and Leung, 2022). B30D10 had lower CO, HC, and PM emissions compared with B40, while NOx emission was higher with B30D10 (Fig. 4). This trend was different from a previous comparative study where B30 and a blend of 70% diesel, 20% FAME, and 10% HVO resulted in comparable exhaust gas emissions (Setiapraja et al., 2019).
Overall, despite B30D10's several superior properties compared to B40, including higher cetane number, lower viscosity, and lower distillation temperature ( Table 4), these enhanced properties did not significantly improve exhaust gas emissions. It can also be deduced that the better FE (Fig. 3), resulting in less fuel burned, might have contributed to the lower emission levels of HC, CO, and PM observed in B30D10 compared to B40. Another reason for different trends can be attributed to engine setting, including injection strategy, spray characteristics, and other related parameters to combustion in the ECU of tested vehicles which might have been more suitable for the characteristics of B30D10 (  The viscosity value of the used lubricating oil was lower than the new oil (Fig. 5b), which could be attributed to the presence of fuel entering into the lubrication system, known as the fuel dilution phenomenon (Neale, 1996 viscosity compared to B30D10, as shown in Figure 5b; this can be attributed to the higher oxygen contents of B40 than B30D10, leading to a higher degree of oxidation in the B40 and fuel-engine oil mixture. Figure 5c presents the percentage of fuel dilution in engine oil, measured before the engine oil change at an interval of 10,000 km. Overall, the fuel dilution percentage trends of both fuel types across the five distance millages are quite constant, which could be due to the regular maintenance of the engine and oil changes keeping the engine in good condition and reducing the risk of fuel dilution. Nevertheless, fuel dilution only slightly affects engine oil properties (Kaminski, 2022), and the presence of a fuel diluent in used lubricating oil could be considered in normal conditions if it does not exceed the specified maximum threshold value.

Engine oil analysis
The insoluble content in engine oil indicates the formation of an insoluble phase containing contaminants. Pentane insolubles can encompass substances not soluble in oil, including materials originating from the degradation of oil or additives, as well as other non-oil soluble resinous matter (Agarwal, 2003;ASTM D893-14, 2018). On the other hand, toluene insolubles can arise from various sources, such as external contamination, carbon deposits from fuel, oil, additive degradation, or materials resulting from engine wear and corrosion.
Any noticeable change in the concentrations of pentane insolubles, toluene insolubles (with or without coagulant), and insoluble resins indicates oil changes that may cause problems with the lubrication system (ASTM D893-14, 2018). According to Figure 6, it is clear that the total insoluble values increased noticeably with increasing travel distance using both fuels, especially B30D10. However, they did not exceed the 3 wt% upper limit mandated in CIMAC recommendations No. 30 (CIMAC, 2011).
Metal wear in engine oil is commonly derived from components such as bearings, piston rings, piston cylinders, and other moving metal parts within the engine. Additionally, certain metal compounds like Calcium (Ca) and Zinc (Zn) are commonly added to lubricants as additives to enhance their performance. The analysis of metal content provides valuable insights into the condition of the lubricating oil, revealing the amount of additive content during engine operation and identifying potential wear points. Table 4     tabulates the metal elements found generally in the lubricants and their possible sources. Figures 7 and 8 present the metal analysis results in engine oil during the road test. Based on the data presented in Figure 7, the wear metal contents, including Fe, Cr, Al, and Cu, were still below the wear threshold or warning level, i.e., 100 ppm for Fe, 20 ppm for Al, 10 ppm for Cr and 30 ppm for Cu.   Figure 8 illustrates the contents of commonly found metallic additives in lubricants. The metallic additive Ca is typically employed as a detergent or lubricant dispersant (Lakshminarayanan and Nayak, 2011). The relatively stable Ca content during road tests indicates that the amount of additive remained within the initial specifications, even after lubricant replacement intervals. On the other hand, the Zn content in lubricants can originate from metallic additives, serving purposes such as anti-wear corrosion and anti-oxidation. Additionally, Zn may also originate from alloying elements in components like bearings, thrust washers, and galvanized cases (Lakshminarayanan and Nayak, 2011). The increase in Zn content can be attributed to component wear. Notably, the variation in fuel types B40 and B30D10 did not significantly affect the Ca and Zn metal contents in the lubricants.
Overall, the engine oil analysis results of the samples obtained from the vehicles powered by B40 and B30D10 fulfilled the minimum quality standard specified in CIMAC Recommendation No. 30 (CIMAC, 2011). In other words, the currently employed lubricant changing intervals were still adequate to achieve optimal lubricity for vehicles fuelled with B40 and B30D10. This result can contribute to stable vehicle performance and FE over a 50,000 km road test, as described in Section 3.2.

Engine components evaluation after 50,000 km travel distance
The effect of the tested fuels on engine component deterioration following the 50,000 km road test was assessed by comparing engine component conditions before and after the road test. A general visual inspection of the pistons and cylinder liners of the vehicles powered by B40 and B30D10 showed no significant wear of these components (Fig. 9). More specifically, none of the piston rings became stuck, indicating that the ring would still work perfectly. The piston skirt and cylinder wall experienced normal wear in the trust position. Superior lubricity was observed for both B40 and B30D10, as indicated by their considerably smaller wear scar diameter of 235 and 242.5 µm, respectively, compared to the 460 µm limit shown in Table 3. It should be noted that the high oxidation stability of the tested fuels, B40 and B30D10, would also effectively contribute to preventing excessive wear. Furthermore, the engine oil experienced normal degradation during the 50,000 km endurance tests, as explained earlier (Section 3.4), further enhancing its ability to provide optimal protection for engine components.
Carbon deposits in the cylinder head of the vehicles powered by B40 and B30D10 are shown in Figure 10. The cylinder head top for B40 had a slightly larger deposit formation than B30D10. However, the deposit thickness of each cylinder head was measured less than 0.2 mm, which is still within the manufacturer's recommended limit for the tested vehicle. The low-carbon residues (0.02 wt%) of B40 and B30D10 compared to the quality limit requirement (0.1 wt%) ( Table 3) could explain the normal deposit formation during the 50,000 km endurance test. Figure 11 presents the results of carbon deposits measured on the intake valve and combustion chamber after brushing with a soft brush, collecting deposits on paper, and weighing the paper with a microbalance. Combustion chamber deposits for B40 and B30D10 were weighted at 0.423 and 0.417 g, respectively. Further, the average amount of deposits formed on the intake valve using B40 was 0.641 g, compared to 0.231 g when using B30D10. Carbon deposits on the intake and exhaust valves of cylinder 1 of the vehicle powered by B40 and B30D10 are shown in Figure 12. Carbon deposits on compression ignition (CI) engine components generally originate from thermal and oxidative degradation of lubricating oil and incomplete combustion (Patel et al., 2022).
The normal wear and less deposit formation in the intake valve and combustion chamber during the 50,000 km endurance test can also contribute to improved FE (Section 3.2). More specifically, less deposit formation on the intake valve would maintain the air charge mixing pattern and level, resulting in a smaller decrease in engine volumetric efficiency (Ziejewski et al.,1986;Stepien, 2014). On the other hand, less deposit formation in the combustion chamber would prevent reductions in thermal efficiency, maintaining an effective heat transfer during compression stroke (Stepien, 2014; Dong and Truong, 2019). The lower deposit formation of B30D10 compared to B40 (Fig. 11) must have also contributed to the observed increased FE of B30D10 over B40. Similar results were also reported by Hidayat and Sugiarto (2020), arguing that the deposit mass was less when the engine was operated on HVO compared to B30.
The deposit formation in an engine is a complex process, and it appears to depend on a combination of different parameters such as fuel, surface material, temperature, pressure, combustion chamber environment, etc. However, wall temperature is one of the most important parameters that affect deposit formation (Arifin et al., 2009). The lower calorific value of B40 compared to B30D10 would result in lower combustion chamber temperatures and component surfaces, resulting in higher deposit formation. The other reason for the higher deposit formation using B40 could be its lower volatility than B30D10, increasing evaporation time. The faster the evaporation time, the smaller the deposit area will be (Suryantoro et al., 2016).

Cold start ability evaluation
Eight vehicles were prepared for the cold start ability to evaluate cranking time at soaking periods of 7, 14, 21, and 28 d at Dieng of Central Java with an environment temperature range of 13.6-18.1 o C. B30D10 and B40 did not significantly affect the time required for the engine to start for the different soaking periods ( Table 5). The time required from engine start until the stable idle condition for the eight tested vehicles was in the range of 0.69 to 2.15 s. These results aligned with the finding of Sugiyama et al.    11. The results of deposit analysis at the intake valve and combustion chamber for the vehicles powered by B40 and B30D10 after the 50,000 km endurance test.
(2011), who reported that the low-temperature start ability of neat HVO till reaching the stable ide ranged between 1.3 to 2.3 s. At temperatures between 13.6 and 18.1 o C, the cloud point might have met, but there was still a smooth flow of fuel entering the combustion chamber from the fuel system, causing the engine to start easily. Since B40 and B30D10 did not lead to a significant difference in starting time, it can be concluded that they also did not affect the engine's ignition timing, confirming the safe use of these alternative fuels under Indonesia's operating conditions.

Conclusions and Prospects
Based on the evaluation results of the investigated fuels, namely B40 and B30D10, on light-duty vehicles using road and laboratory tests, the following conclusions can be drawn: Vehicles fuelled with B40 and B30D10 encountered no technical problems during the 50,000 km endurance test. II.
Engine components inspection and used engine oil analysis of the vehicles powered by B40 and B30D10 showed normal conditions with all the investigated parameters aligning with the manufacturer's  Overall, based on the present study's findings, both B40 and B30D10 are viable options for implementing 40% biodiesel blend fuel in Indonesia. However, it is still crucial to consider the sustainability aspects of Indonesia's biodiesel program by considering various factors, including supply-demand chain analysis as well as techno-economic and environmental impact assessments (Aghbashlo et al., 2022), which were not included in this study.
Biodiesel availability should also be investigated by securing sustainable resources. For instance, in a techno-economic study, Hor et al. (2023) highlighted palm oil waste sludge as a promising resource for biofuel production in Indonesia. In addition, the effect of 40% biodiesel on unregulated emissions such as aldehyde should also be assessed in more detail. The various aspects of HVO utilization at a large scale should also be taken into consideration. precursor. Additionally, further research should be conducted to explore the implementation of higher biodiesel ratios (>40%), aiming to determine the maximum limit of the biodiesel blending ratio in alignment with advanced vehicle technologies.
Please cite this article as: Mokhtar