Effect of di-tertiary-butyl peroxide additive on combustion, performance and emissions of a karanja methyl ester fueled diesel engine

The present work studies the inuence of di-tertiary-butyl peroxide (DTBP) as a cetane-improving additive to karanja methyl ester (KME) on the combustion, performance and emission characteristics of a diesel engine. KME produced by base catalyzed transesterication of non-edible karanja oil was blended with DTBP in different volume proportions to result KMED1 (99% KME + 1% DTBP), KMED2 (98% KME + 2% DTBP), KMED3 (97% KME + 3% DTBP) and KMED5 (95% KME + 5% DTBP) fuel blends. With increase in DTBP content, viscosity was reduced, whereas the cold ow properties, cetane index and caloric value were enhanced. Engine test results exhibited improvement in brake thermal eciency and brake specic energy consumption for all blends compared to neat KME. Combustion analysis showed improved combustion with rise in DTBP content in the blends. The CO, HC and NO x emissions with KME-DTBP blends were less compared to neat KME and the same signicantly reduced with rise in DTBP percentage in the blends. This shows improved combustion due to more oxygen availability and improvement in fuel properties with addition of DTBP to KME. However, the NO x emissions were marginally higher with KME-DTBP blends compared to neat KME and diesel that may be further studied.


Introduction
In the last few decades, increase in global population, growth in industries and automobiles as well as change in human life style have led to increased demand and consumption of energy. Subsequently, fossil-based fuels, being the primary source of energy, are being consumed at a much faster rate for harness of energy. This leads to the risk of rapid depletion of these conventional fuels as well as environmental deterioration. Again, increased use of fossil fuels are responsible for harmful emissions to the atmosphere in the form of greenhouse gases (GHGs) and other pollutants. In recent years, the consequences of environmental degradation in the form of global warming, acid rain, climate change etc. have emerged as possible threat to the survival of humankind. Thus, it is the need of the hour to minimize the dependency on the fossil fuels through development of potential alternate energy sources. In this regard, researchers around the world are putting tremendous effort on development of alternate fuels for internal combustion engines. In recent times, biofuels have emerged as potential substitute fuel for engines. Compression ignition (CI) engines, usually known as diesel engines, are widely used owing to their suitability for medium and large commercial transport purpose. Many published literature have demonstrated the use of straight vegetable oils (SVOs), vegetable oil methyl ester or biodiesel, bioethanol etc. as potential fuel in CI engines (Misra and  Conversely, many researchers also claimed that biodiesels, owing to their poor cold ow characteristics, high viscosity, higher NO x emissions, higher oxygen content, and engine compatibility issues, are not appropriate for long-term usages in diesel engines Yasin et al. 2014;Smith et al. 2010). In order to overcome these issues, efforts were made to improve biodiesel properties. This led to the research on the use of additive-doped biodiesel in CI engines.
Many investigators examined the use of various additives with biodiesel with the intention of improving its fuel properties and combustion characteristics. Nayak and Pattanaik (2014) investigated the use of di-methyl carbonate as an additive with mahua biodiesel and reported signi cant drop in CO, HC, NO x and smoke emissions matched to neat biodiesel and diesel. Musthafa (2017) reported lower CO, HC and NO x emissions for an uncoated diesel engine running on palm oil biodiesel with 1% di-tertiary-butyl peroxide (DTBP) additive. Roy et al. (2014) used different biodiesel blends with Wintro XC 30 additive in a CI engine and reported improved fuel properties, lower CO, HC and NO x emissions with higher BTE compared to diesel fuel. Similarly, many other works on the use of additive blended biodiesel fuel in CI engines have reported better engine performance along with lower exhaust emissions (Lawan et al. 2020;Channappagoudra et al. 2018;Yang et al. 2016). In the current work, efforts were made to use karanja methyl ester (KME) blended with different volume proportions of a cetane enhancing additive (DTBP) in a CI engine for study of various exhaust emission parameters. A comparative analysis of the obtained results were made with diesel fuel to investigate the improvement / deterioration in the engine performance and exhaust emissions. The objective of the current work is to completely replace mineral diesel oil in a CI engine with biodiesel fuel produced from non-edible source. Further, the purpose of using DTBP as an additive to biodiesel is to improve its fuel properties, achieve better combustion and engine performance along with lowering the exhaust emissions.

Materials And Methodology
The methodology involved in the present work includes selection of a suitable SVO, development of biodiesel using base catalyzed transesteri cation, selection of a suitable cetane improving additive, preparation of biodiesel-additive blends in suitable volume proportions, fuel characterization of all the selected test fuels, application of the blended fuels in the selected test engine, carrying out the engine performance and emission tests using the selected fuels, recording the observed data and a comparative analysis of the obtained data. The detailed methodologies are elaborated in the following subsections.

Preparation of Biodiesel and Fuel Blends
In this work, Karanja (Pongamia Pinnata), a non-edible SVO source, was selected for production of Karanja biodiesel or KME. Neat Karanja oil (KO) was procured from M/s Sajjan Agarwal & Co., Karanjia, Baripada, Odisha, India. Neat KO was initially ltered and a selected amount of the same was taken for preparation of biodiesel. A 5 L capacity biodiesel reactor (Make: M/s Gobind Machinery Works, India) was used for preparation of KME from neat KO. The selected quantity of neat KO was fed into the biodiesel reactor followed by preparation of a measured quantity of reagent mixture (Methanol + KOH). The neat KO was initially heated to 40°C followed by mixing of the measured quantity of reagent mixture with it. After proper mixing of the same, transesteri cation was started and continued at 60°C for a period of 1.5 h. After the reaction is over, the products were allowed to settle down overnight. Two distinct layers were visible with the upper layer being KME and the bottom layer being glycerol. Glycerol was collected into a container through the ush bottom valve. This was followed by collection of KME into another container following the same process. The raw KME obtained was then water washed for three times to obtain pure KME. Again, it was dried using a suitable procedure for removal of moisture. DTBP was procured from M/s Manik Chemicals, Kolkata, West Bengal, India. It is available in 100 ml bottle container. DTBP is an organic compound that consists of a peroxide group bounded by two tertiary-butyl groups. It is one of the best stable organic peroxides. It is used as an excellent cetane improver. When the temperature of DTBP is increased beyond 100°C, the peroxide bond undergoes homolysis. Due to this, it is used as a radical initiator in organic synthesis (Musthafa 2017). It has a chemical formula of C 8 H 18 O 2 . 1%, 2%, 3% and 5% v/v of DTBP was suitably mixed with neat KME to result four numbers of KME + DTBP fuel blends named as KMED1 (1% DTBP + 99% KME), KMED2 (2% DTBP + 98% KME), KMED3 (3% DTBP + 97% KME) and KMED5 (5% DTBP + 95% KME).
Fuel characterization of all the prepared fuel blends along with KME and diesel was conducted as per standardized ASTM methods to study various fuel properties. The fuel characterization results are presented in Table 1.

Experimental Engine Setup and Procedure
The prepared fuel blends along with KME and diesel were tested in a computerized single-cylinder four-stroke water-cooled direct injection diesel engine attached to an eddy current dynamometer. The rated power of the engine was 5.2 kW at 1500 rpm. Necessary instrumentation were incorporated with the test engine for measurement of various engine parameters. The eddy current dynamometer is used for applying load on the engine. Two different fuel tanks were used for supply of diesel and biodiesel fuels to the engine. The engine exhaust line was connected to a ve-gas analyzer (AVL Digas 444) and a smoke meter (AVL 437) for measurement of the exhaust emissions. The engine tests for each of the selected fuels was carried out at 0%, 25%, 50%, 75%, 85%, 90%, 95% and 100% of the rated load. During all the engine experiments, diesel was used as the baseline fuel for necessary comparison of obtained data and the acquired data was stored in the attached computer. The main performance parameters studied were brake thermal e ciency (BTE), brake speci c energy consumption (BSEC) and exhaust gas temperature (EGT). On the other hand, the important emission characteristics studied were CO, HC and NO x emissions and smoke opacity. A schematic presentation of the experimental engine setup with all attached instrumentation can be seen in Fig. 1. The technical speci cations of the test engine are presented in Table 2.  analysis of different parameters is carried out. At rst, the engine performance parameters were checked for their respective accuracies which were based on the accuracies of the components (manufacturer declared accuracy) using them. After estimating the total (combined) was carried out for the engine performance parameters using the root mean square method (Panda et al. 2017). The Eq. 1 presented below shows the total uncertainty of a speci ed parameter. Table 3 represents the total uncertainty of brake power and BSEC calculated as per Eq. 1.

Results And Discussion
The obtained engine experimentation results with KME and KME-DTBP blends in the form of engine performance parameters, viz. BTE, BSEC and EGT along with exhaust emissions, viz. CO, HC, NO x and smoke opacity were analyzed critically and compared with those for diesel fuel. Again, the comparative analyses of the in-cylinder peak pressure with the selected blends along with diesel has been presented in this section for better understanding of the combustion pattern of these fuels. These comparative analyses carried out at selected engine loading conditions are presented in the following subsections along with valid discussions in order to ascertain the improvement/deterioration in the said performance and exhaust emissions characteristics.

In-cylinder Peak Pressure
The variations of in-cylinder peak pressure for KME, KMED1, KMED2, KMED3, KMED5, and diesel at 85% load and at rated load with respect to crank movement are presented in Figs. 2 (a) and (b), respectively. It is clearly observed that the highest peak pressure at both 85% and 100% loads is achieved with diesel that shows its better combustion pattern compared to the other fuels due to its higher calori c value and lower viscosity. Similarly, the lowest peak pressure is achieved with KME at both the mentioned loads, which signi es its poor combustion pattern compared to the other blends and diesel owing to its lower cetane index, calori c value and higher viscosity. It is also observed that the peak pressure at both considered loads increases with increment in DTBP percentage in the blends. Among all considered biodiesel blends, KMED5 exhibits highest peak pressure at both conditions. This may be credited to the higher calori c value, higher cetane index and lower viscosity of KMED5 4.2 Brake Thermal E ciency Figure 3 demonstrates the variation of BTE with load. It is detected that BTE increases initially up to 85% load and then somewhat decreases until full load irrespective of the type of fuel used. This shows better combustion at higher loads, which is the common trend of a CI engine. At higher loads, the amount of fuel supplied is more that leads to higher brake power and BTE (Devarajan et al. 2019b

Brake Speci c Energy Consumption
The variation of BSEC with load is presented in Fig. 4. It is noticed that BSEC initially decreases up to 85% load and then marginally increases until full load with all the fuels. Higher BSEC at low loads is owing to incomplete combustion because of lower in-cylinder pressure, low turbulence and poor mixing. On the other hand, higher BSEC at high loads is because of rich mixture formation due to increased fuel supply quantity. BSEC with diesel was found to be lowest at all loads. This may be because of higher calori c value of diesel that leads to better combustion e ciency and lower fuel energy consumption . The BSEC trend also shows that with increase in DTBP percentage in KME, BSEC gradually reduces. This signi es better combustion e ciency with addition of DTBP in KME. The same may be attributed to the increase in cetane index and calori c value as well as reduction in viscosity with addition of DTBP in the blend that leads to improved atomization and superior combustion. These results are well supported by the ndings of the published literature (Yilmaz and Atmanli 2017). Among all selected biodiesel fuels, KMED5 showed lowest BSEC at all loads. At 85% load, BSEC with KMED5 was found to be only 1.09% higher than that with diesel. Again, the same for KMED5 was found to be 2.34% higher matched to that with diesel.

Exhaust Gas Temperature
The variation in EGT with load is depicted in Fig. 5. EGT is found to be increasing with rise in load irrespective of the fuel used. With increase in load, the in-cylinder pressure and temperature tends to increase those result in higher EGT. Diesel produces lowest EGT at all loads. Biodiesel, being an oxygenated fuel, produces higher combustion temperature, which in turn results in higher EGT (Yilmaz and Atmanli 2017; Xue et al. 2011). It is further detected that upsurge in DTBP percentage in KME leads to higher EGT. The probable reason for this is the increase in cetane index and calori c value of KME with increase in DTBP content in the blend. The same results in early start of combustion and a prolonged secondary-phase combustion leading to higher EGT. These ndings and their explanations are in agreement with the published literature (Xue et al. 2011). The highest EGT were observed at full load with all the fuels. At this load, the lowest EGT of 447.6°C was obtained with diesel and the highest of 508°C was obtained for KMED5. Again, EGT for diesel was found to be 7.9%, 8.8%, 9.7%, 10.7%, and 13.5% lower compared to that with KME, KMED1, KMED2, KMED3, and KMED5 respectively, at full load. index as well as decrease in viscosity of the fuel due to addition of the additive, which is re ected in Table 1  load for all the fuels that shows better combustion at this load. The rapid growth in HC emissions for all the fuels after 85% load is attributed to formation of rich fuel-air mixture due to greater injection pressures leading to incomplete combustion (Radhakrishnan et al. 2017). The HC emissions with KME and its blends with DTBP was observed to be lower at all loads matched to diesel. This may be credited to the higher oxygen content in case of biodiesel and the higher cetane index of KME and its blends with DTBP that leads to improved combustion compared to diesel (Radhakrishnan et al. 2017;Doğan 2011). Again, the HC emissions in case of KME was observed to reduce with increase in percentage of DTBP. This is due to the mutual effect of decline in viscosity and enhancement in cetane index and calori c value that leads to improved atomization and superior combustion (Atmanli et al. 2014). In addition, increase in percentage of DTBP in KME lowers the ignition delay and enhances mixing of fuel and air. Consequently, the combustion rate is improved, which lowers the HC  ). This is attributed to the higher oxygen content in biodiesels that leads to increased rate of combustion and higher temperature of combustion gases. Thus, the present trend of NO x emissions are found to be lowest for diesel among all the fuels at all engine loads. It is further witnessed that the NO x emissions slightly increase with increment in DTBP percentage in KME. Addition of DTBP to KME enhances the cetane index and calori c value, whereas it also reduces the viscosity of the blend. This result in a shorter delay period, early start of combustion, increased premixed combustion phase as well as improved atomization and mixing. respectively. Likewise, the same at 100% load was found to be 0.57%, 2.48%, 3.1%, 3.23%, and 9.65% higher compared to those with KMED3, KMED2, KMED1, KME, and diesel, respectively.

Smoke Opacity
Diesel smoke is the combination of aerial particulate matter and gases produced during combustion. The smoke is measured in the smoke meter by means of light absorbed through an exhaust gas column of de ned speci c length. The same is expressed as smoke opacity. The variations of smoke opacity with engine load for diesel, KME, KMED1, KMED2, KMED3, and KMED5 are presented in Fig. 9. It is noticed that the smoke opacity linearly increases with rise in engine load irrespective of the type of fuel used. Further, the same is observed to be increasing signi cantly at higher loads. At higher loads, the amount of fuel supplied is greater in order to maintain a constant power output. This in turn, results in rich fuel-air mixtures that leads to incomplete combustion (Devarajan et al. 2017b;Mahalingam 2018). Hence, smoke opacity are signi cantly higher at high engine loads.
Diesel exhibited higher smoke opacity at all loads compared to the biodiesel fuels. The same can be attributed to the existence of oxygen in biodiesels that ensures better combustion and less smoke formation compared to diesel. Again, addition of DTBP to KME showed reduction in smoke opacity. It is witnessed that smoke opacity tends to reduce with higher percentage of DTBP in KME. This may be attributed to the reduction in viscosity along with enhancement in cetane index due to increase in additive percentage in biodiesel. Reduction in viscosity improves the atomization process and produces better mixing of fuel with air. In addition, improved cetane index lowers the delay period that leads to early commencement of combustion. The combined effect of the same results in improved combustion in the primary combustion phase that produces lower smoke formation (Pandian 2017  The economic factors affecting biodiesel production could be the cost of raw material, capital and chemicals used and capacity of plant and technology used. Out of these, 80% is the raw material cost and the rest includes cost of chemicals (catalyst and methanol) as well as labour (Panda et al. 2018). In this work, the estimated cost of production (US$) of KME from neat karanja oil was found to be 0.7/lit that is inclusive of raw material cost, cost of chemicals and labour expenses. These costs can be lessened if there are some ways to reuse the byproduct glycerol. In addition, if the feedstock is selected appropriately, the overall cost might reduce. These days, as the non-food crops are favourable sources for biofuels production, this can be seen as an opportunity to upsurge the production of biodiesel. Other socio-economic factors, such as regional development, sustainability, agriculture with social structure, supply security etc. may add to the advantage of biodiesel over diesel and other petroleum products. Further, the local economy may be signi cantly promoted through the use of developed biofuels for power production.

Conclusions
The effect of DTBP addition to KME on the engine performance and exhaust emissions characteristics was studied along with a comparative analysis with diesel oil. The key ndings evolved out of the present work are summarized below.
The fuel properties of KME signi cantly improved with addition of DTBP. The cold ow properties, viz.
viscosity, cloud point and pour point reduce with increment in DTBP percentage in the blend. Similarly, DTBP addition with KME enhanced the calori c value and cetane index of the blend.
Improved combustion is achieved with the use of DTBP additive in KME. Further, combustion gets better with rise in DTBP content in the biodiesel blend. KME and DTBP blends produce higher BTE than neat karanja biodiesel. However, BTE with KME-DTBP blends are lower compared to diesel under all working conditions.
Doping DTBP to KME reduces BSEC owing to improved combustion inside the engine cylinder. However, KME-DTBP blends show higher BSEC compared to diesel because of their higher viscosity and low calori c value.
EGT increases with increase in DTBP content in KME-DTBP blends owing to greater availability of oxygen in the blends. Diesel shows lowest EGT at all loads compared to the biodiesel blends.
CO emissions are highest for diesel especially at higher loads owing to lower oxygen availability. Further, the same tend to decline with rise in DTBP percentage in the KME-DTBP blends as a result of improved combustion.
HC emissions are highest in case of diesel under all working conditions due to lower availability of oxygen.
Further, addition of DTBP lowers the HC emissions of KME owing to better combustion as a result of reduction in viscosity along with enhancement in calori c value and cetane index. NO x emissions are greater with all the biodiesel blends matched to diesel at all loads due to their higher oxygen content. Again, the NO x emissions tend to increase with rise in additive percentage in KME-DTBP blends as a consequence of rise in oxygen content, cetane index and calori c value of the blends that gives rise to a longer residence time of high temperature.
Smoke opacity is highest for diesel at all working loads, whereas the same reduces with rise in DTBP content in the KME-DTBP blends. This signi es better combustion with increment in DTBP percentage in the blends.
It may be concluded that addition of DTBP enhances biodiesel performance and lowers the exhaust emissions in a diesel engine. The present work establishes 100% replacement of diesel oil in a diesel engine with KME-DTBP blends, while maintaining satisfactory engine performance along with lower exhaust emissions. However, this work unveils further scope of research to lower the NO x emissions with KME-DTBP blends, which are marginally higher, compared to diesel.  Variation of in-cylinder peak pressure  Variation of HC emissions with load Variation of smoke opacity with load