Performance, combustion and emission characteristics of a direct injection VCR CI engine using a Jatropha curcas oil microemulsion: a comparative assessment with JCO B100, JCO B20 and petrodiesel

Abstract

A Jatropha curcas oil (JCO) based hybrid microemulsion fuel (MHBF) comprising pretreated JCO–1-butanol–ethanol was prepared. The fuel characteristics of the hybrid microemulsion fuel (MHBF) were determined as per standard tests methods recommended by ASTM and compared with JCO B20, JCO B100 and petroleum diesel. The performance, combustion and emission characteristics of MHBF were studied for the first time in a variable compression ratio (VCR) direct injection CI engine. The fuel properties of the MHBF formulation were comparable to that of JCO B100 (viscosity of 5.9 mm² s⁻¹, density 0.872 g cm⁻³ and higher heating value of 38.25 MJ kg⁻¹), while the cold flow properties were found to be superior than JCO B100 and JCO B20. In terms of engine performance, the brake specific fuel consumption (BSFC) of MHBF was comparable to B100 and B20 but higher than petrodiesel; while brake thermal efficiency (η_bte) was close to that of JCO B100 but slightly lower than petrodiesel at almost all load conditions. In terms of emission quality, MHBF was found to be superior to all other tested fuels under all conditions of loading. Most significantly it was observed that up to 80% NO_x reduction as compared to petrodiesel was achieved when operated at full load with the MHBF at the maximum compression ratio of 17.5. Obviously, this is a significant finding and a factor of motivation and investigation for future biofuel applications.

---

1. Introduction

Steep escalation of oil costs and detrimental environmental issues over fossil fuel utilization enhanced the application of vegetable oil derived biofuels in Compression Ignition (CI) engines in recent years. Biofuels such as biodiesel and hydrotreated vegetable oil are considered as possible petrodiesel substitutes due to their similarities with petrodiesel and low life cycle GHG emission. In fact, the use of vegetable oils as fuel is not new and the idea of using vegetable oils as a fuel arose a century back during the early diesel locomotives invented by Dr Rudolph Diesel (1893). Despite that, the use of vegetable oils in CI engines is generally thought to be unsatisfactory and unfeasible because of their high viscosities and low volatilities specifically for fuel injection systems. The high viscosity of vegetable oil causes poor fuel atomization and because of inefficient mixing of fuel with air results incomplete combustion. Lower volatility leads to oxidative polymerization during combustion results more gum deposit, injector coking, carbon, ring sticking, lubrication oil dilution and degradation etc.^1–4

Regardless of the above technical difficulties, the major impediments in large scale commercialization and application of vegetable oil based fuels like biodiesel originates due to the complicated and expensive conversion technology needed to reduce viscosity for CI engine applications. In addition, the role of biodiesel as renewable fuel has been criticised by many experts due to the increased NO_x emissions resulting from higher combustion temperatures and with rigid environmental regulations aiming to virtually eliminate NO_x emissions from CI engines; the intensity of the positive arguments for biodiesel as a diesel fuel substitute has become debatable.^2,4–6

In sight of these difficulties, microemulsification technique has been regaining importance as an economically viable and ‘green’ approach for liquid biofuel production from vegetable oils.^5,7–11 Especially, in the recent 2–3 years several research articles dealing with production, characterization and application of hybrid diesel fuels by microemulsification method were reported.^{7,8,10,12–14} The advantages of microemulsification approach over the chemical routes (transesterification, hydrocracking, hydroprocessing etc.) include lower production costs, short production time, simple and easy implementation (on the farm blending) and no by-product formation due to noninvolvement of chemical reactions. In addition, due to the presence of more volatile oxygenates the combustion temperature of the emulsion and microemulsion fuels are much lower, which results in a drastic reduction in emissions of NO_x, CO, black smoke and particulate matter. Some other added benefit of such oxygenate fuels is to reduce ignition delay whilst engine efficiency also remain comparable to that of petrodiesel.^7,8

Bora, Konwar and co-workers⁷ recently reported a novel method for the preparation of microemulsion based hybrid biofuels (MHBF) directly from low grade acidic oils (non-edible oil and waste cooking oils). Their method makes use of the naturally occurring monoglyceride, diglyceride and free fatty acids (or esters in case of pretreated oils) components present in these low grade oils as the surfactants to produce thermodynamically stable, isotropically clear microemulsion systems (Windsor Type IV) with anhydrous ethanol as dispersed phase and 1-butanol or 2-butanol as the co-surfactant.^8–10

Keeping in view the importance of microemulsion systems like MHBF as renewable and low cost fuel alternative to reduce CI engine exhaust and maintain the engine efficiency, an investigation of the recently introduced MHBF was found necessary to confirm its potential utility as a fuel. Accordingly, the CI engine performance, combustion and emission analysis of Jatropha curcas oil (JCO) based MHBF were studied in a single cylinder direct injection CI engine and reported herewith. A comparative study was conducted with JCO B100, JCO B20 and petrodiesel with respect to engine efficiency, exhaust emission and combustion profile parameters such as brake specific fuel consumption (BSFC), brake thermal efficiency (η_bte) etc. All the attributes of the investigation have been compared with established biofuel systems.

2. Experimental

2.1. Materials

Ethanol (<99.9%), methanol (99.9%), n-heptane (99.9%), pyridine (HPLC grade), Na₂SO₄ (99.5%), H₂SO₄ (98%), NaHCO₃ (99%), n-hexane (LR) were purchased from Merck India Ltd. 1-Butanol (Reagent PlusR, ≥99%), methyl heptadecanoate (99%), N-methyl-N-(trimethylsilyl)trifluoroacetamide (98%), FAME mix (37 component), ASTM® D6584 Individual Standard Solution and Internal Standards Kit (Supelco) were purchased from Sigma-Aldrich (India). All chemicals and reagents were used without further purifications. Jatropha curcas oil (JCO) was purchased from Bhushan oil mill (Rajasthan, India). Petrodiesel (LDO, Indian Oil Corporation) was obtained from a local petrol pump in Kapurthala district (Punjab).

2.2. Preparation of MHBF

MHBF was prepared using the standard procedures reported earlier.^3,4,6 Briefly, in a typical MHBF synthesis, moisture free and pretreated JCO, 1-butanol and ethanol were mixed by gently stirring at 150 rpm to a mechanical stirrer for 5 min at room temperature at predetermined proportions (see below) (Scheme 1). In this process, 1-butanol was employed as a co-surfactant while the FAME, monoglycerides (MG) and diglycerides (DG) present in the pretreated JCO functioned as surfactants. In our previous studies, typically the formulations with oil, 1-butanol and ethanol at proportions between 60 : 30 : 10 and 50 : 40 : 10 were found to be best suited for CI engine applications as they represent systems with higher heat value (HHV) and diesel range viscosity (see below).^7,8,10 Subsequently, several MHBF samples with varying volume fractions of each of the reagents (in the above specified range, Table 2) were formulated. The density, viscosity and HHV of the formulations were compared and only the optimum formulation was selected for engine testing. For comparison, JCO B100 and JCO B20 were also prepared according to the standard procedures described in literature.¹² The detailed descriptions of the individual preparation methods have been included in the ESI file.†

Scheme 1 Preparation of MHBF from crude JCO.

Crude JCO (∼9.05 wt% FFA, acid value 18.12 mg KOH per g) was subjected to pretreatment with methanol to convert the FFA into FAME, which is an essential step to reduce the acid value (acidity) of oil to 1.1 mg KOH per g and improve fuel quality of MHBF.^3–6 In a typical pretreatment reaction, 800 g preheated oil was added to 800 mL 2% (wt%) methanolic H₂SO₄, stirred and heated at 60 °C for 30 min in 2 L capacity jacketed reactor (Radleys, United Kingdom) equipped with mechanical stirring and reflux condenser. After completion of the reaction, excess methanol was removed with a rotary evaporator (Heidolph Hei-VAP Advantage, Germany). The resulting product was neutralized with 1 M NaHCO₃ solution, followed by washing with deionised water. The oil layer was recovered in a separatory funnel and dried over anhydrous Na₂SO₄ prior to use. The composition (wt%) and physiochemical properties of the pretreated JCO (pretreated-JCO) are summarized in Table 1.

Table 1 Properties of esterified JCO

Fatty acid composition, name of the acid (wt%)
Palmitic	16.168
Stearic	6.915
Palmitoleic	0.856
Oleic	39.103
Erucic	0.261
Nervonic	0.187
Linoleic	35.305
Gamma-linolenic	0.200
Alpha-linolenic	0.279
cis-13,16-Decosadienoic	0.202
cis-4,7,10,13,16,19-docosahexanoic	0.228
Others	0.296
Physicochemical properties
Density @ 15 °C (g cm^−3)	0.896
Viscosity @ 40 °C (mm^2 s^−1)	19.230
Calorific value (MJ kg^−1)	38.680
Different components of MHBF (wt%)
FAME	10.2
Monoglyceride	9.9
Diglyceride	6.7
Triglyceride	73.1

---

2.3. Characterization of feedstock, MHBF and biodiesel

Fatty acid composition of the JCO was determined on an Agilent 7890A gas chromatograph (GC) equipped with FID detector and HP-INNOWax GC column (30 m length; 0.25 mm internal diameter, 0.25 μm film thickness). The methyl esters were prepared according to AOCS Official Method 1998; Ce 1e62 and Ce 2e66. The individual components were identified by comparison of retention times with the standard FAME mix and the wt% of the individual fatty acids were calculated based on methyl heptadecanoate internal standard. The column temperature was held at 50 °C for 2 min, and then heated to 200 °C @ 10 °C min⁻¹. The temperature program was subsequent to raised from 200 to 240 °C at 5 °C min⁻¹ and finally held for 10 min. Helium was used as the carrier gas at a flow rate of 1 mL min⁻¹.

The glyceride content in the oil samples, as well the conversion of fatty acids (FFA) to FAME (pretreatment), was also determined by GC according to ASTM D6584 method on an Agilent 7890A gas chromatograph equipped with FID detector and Agilent (Select Biodiesel for glycerides, 30 m length; 0.30 mm internal diameter; 0.10 μm film thickness) GC column.^7,8 The column temperature was held at 50 °C for 1 min, 15 °C min⁻¹ to 180 °C, 7 °C min⁻¹ to 230 °C, 30 °C min⁻¹ to 380 °C and finally held for 5 min. Helium was used as the carrier gas at a flow rate of 3 mL min⁻¹.

The physicochemical fuel properties (density, viscosity, flash point, calorific value and pour point) of JCO, JCO B100, JCO B20, petrodiesel and MHBF were determined using standard test methods recommended by ASTM (list of test methods in Table S3, ESI†). The fatty acid composition, weight percentages of different components and major physicochemical properties of JCO are presented in Table 1.

2.4. Experimental set-up (CI engine test rig)

A single cylinder, 4 stroke direct injection VCR CI engine (Kirloskar India Ltd.) with rated power 3.5 kW @ 1500 rpm was used to conduct the experiments (Fig. 1). The test rig was connected to a water cooled type eddy current dynamometer for loading on crankshaft with the help of electromagnetic force. Six numbers of allen bolts are used for varying the CR without stopping the engine and without altering the combustion chamber geometry. A load cell was used for load variation on CI engine through dynamometer. The load sensor, fixed inside the dynamometer was used to send the load signals in terms of kilogram load on crank shaft. Two rotameters were used in the experimental set-up to provide the cooling water in the jackets of engine block, cylinder head and calorimeter water flow measurement. The technical specifications of the engine are given in Table S1.† Two piezometric sensors were used to measure the combustion pressure and fuel line pressure fitted on cylinder head and fuel injector respectively. An optical crank-angle sensor was used to deliver the signals for each degree rotation of crank shaft. These signals were used to count the rpm of the crank shaft through interfacing with computer. Six thermocouples were used at various locations in the setup for measurement the inlet and outlet temperature of water and exhaust gases. The setup had a control box consisting with air box, fuel tank, manometer, fuel measuring burette and central processing unit. The fuel measurements were performed by fuel flow transmitter made by Yokogawa, Model no. EJA110A-DMS5A-92NN. It was mounted on a fuel line and the signal of flow rate was transferred through data acquisition device (DAD) made by National Instrument. The DAD was connected to the computer with USB port. All the analogue signals were recorded from different positions of the test rig and transmitted to ‘Enginesoft LV Version 9.0’ (Apex Innovation Pvt. Ltd.) software for performance and combustion analysis.

Fig. 1 The schematic diagram of CI engine test rig used in the work.

The exhaust gas analysis was conducted through an AVL DI Gas 444 Analyzer (India). The resolution, accuracy, measuring range and technical specifications of the gas analyzer are presented in Table S2.† The exhaust gases namely CO, CO₂, HC, O₂, and NO_x were analyzed under the test procedure approved by the Ministry of Road Transport and Highways, Government of India, and specified in MoRTH/CMVR/TAP-115/116, Issue no. 3, Part-VIII for 4-gas analyzer. In a typical analysis, initially the analyzer was allowed to suck in air, which in turn cleansed the passage joining the sensors through various filters and condensation trap. Then the gases from the exhaust were allowed to enter the analyzer through a probe during steady engine operation. The analyzer display showed the concentration of different components in exhaust gas and the results were manually recorded or printed for further analysis.

2.5. Experimental procedure

To validate the fact that CR has impact upon the combustion and efficiency, CR 15 was considered initially as baseline experiment for MHBF. However, as it was already established in our earlier work that other test fuels (JCO B20 and petrodiesel) showed significantly poor performance, improper combustion and higher emission characteristics at CR 15, experiments with these fuels were not repeated in this study at CR 15.³ Thereafter, the comparative analysis with VCR CI engine was conducted at standard specifications; CR 17.5 and fuel injection timing (IT) of 23BTDC for different test fuels. The initial run was performed with petrodiesel followed by JCO B20, JCO B100 and MHBF. The order for analysis was petrodiesel, JCO B20, JCO B100 and MHBF. Prior to each test, the engine was initially run at no load for 30 min (warm up time) enabling proper combustion of fuel. During the measurement engine was tested with 0% (no load), 20%, 40%, 60%, 80%, and 100% load conditions. For measurement at specified load condition, the engine was allowed to run on that condition for 30 min and the temperatures at the outlet of cooling water and exhaust gas were monitored until the engine reached at a steady state (this showed that the combustion inside the cylinder becomes steady and the engine is ready for analysis). The readings of temperatures, air flow rate, fuel flow rate, engine speed, in-cylinder pressure and fuel line pressure were automatically recorded by the engine software (Enginesoft LV Version 9.0, Apex Innovations Pvt. Ltd.) and these values were used by the engine software to generate the performance and combustion parameters. The calculations were based on the equations as given in ES1.†^15,16

3. Results and discussion

3.1. Fuel properties for MHBF, JCO B100, JCO B20 and petrodiesel

From the prospective of CI engine application viscosity, density and HHV are the most important parameters which determine the suitability of a certain MHBF formulation as fuel. It was established in our previous studies that these attributes were primarily determined by the concentration of surface active agents (FFA, FAME, MG and DG) present in the oil and the ratio of oil, 1-butanol and ethanol in MHBF formulations. As already outlined in Section 2.2 several MHBF formulations with oil, 1-butanol and ethanol at proportions ranging between 60 : 30 : 10 to 50 : 40 : 10 were prepared. Typical MHBF formulations within this range were recommended (best suited) for engine applications that represent thermodynamically stable systems having the highest calorific value and viscosity in petrodiesel range.^7,8,10 The results summarized in Table 2 showed that the pretreated JCO feedstock used for MHBF formulation with 1-butanol and ethanol in (v/v/v) ratio 55 : 33 : 12 exhibited optimum fuel properties for CI engine application (highest calorific value with petrodiesel range viscosity and density) and accordingly this particular formulation were selected for all engine testing experiments. Overall, these results are in good agreement with literature and confirm the effect of surfactant components on MHBFs properties.^7,8

Table 2 Effect of oil : butanol : ethanol (v/v/v) ratio on density, viscosity and HHV of JCO derived MHBF

Oil (v)	1-Butanol (v)	Ethanol (v)	Density @ 15 °C g cm^−3	Viscosity @ 40 °C mm^2 s^−1	GCV MJ kg^−1
52	33	15	0.869	6.72	37.81
54	35	11	0.871	5.95	38.34
55	33	12	0.870	5.91	38.25
60	30	10	0.891	7.32	n.d

---

Table 3 presents a comparative summary of the main fuel characteristics of the optimum MHBF formulation with petrodiesel and JCO B100 and JCO B20. The data in Table 3 clearly indicated the superiority of MHBF over JCO B100 in terms of viscosity, density and cold flow properties. Nevertheless, the calorific value of MHBF was found to be somewhat lower than that of JCO B100. Further, to get an idea of the combustion characteristics of MHBF the activation energy of the optimum JCO MHBF was determined by thermogravimetric analysis under air flow (Coats and Redfern method) and compared with the activation energies reported for JCO B100, JCO B20 and petrodiesel.¹⁵ The values summarized in Table 3 indicated that the energy barrier for thermal combustion of MHBF were comparable to petrodiesel, but lower than that of JCO B100 and JCO B20. This suggested that the combustion characteristics of MHBF was similar to petrodiesel and validated its suitability as an alternative fuel for CI engine.¹⁵ The observation was also co-related with the relatively low flash point (37 °C) observed for MHBF, which resulted from the presence of 45% volatile light alcohols (ethanol and butanol) in the fuel system. In contrast, JCO B100 (the pure esters) exhibited the highest flash point (112 °C) and activation energy of combustion (Table 3).¹⁷

Table 3 Fuel properties of JCO derived MHBF, B100 and B20 in comparison with petrodiesel

	Density @ 15 °C g cm^−3	Viscosity @ 40 °C mm^2 s^−1	GCV MJ kg^−1	Cetane no.	Flash point (°C)	Pour point (°C)	Activation energy kJ mol^−1
a Values are reported for the optimum MHBF formation exhibiting highest HHV and viscosity within the ASTM D6751 limit: JCO oil (v) : butanol (v) : ethanol (v) = 55 : 33 : 12.b Values taken from ref. 11.c Stored in diesel tank for 60 days under normal environmental conditions.
MHBF^a	0.870	5.91	38.25	—	37	−7	13.72
MHBF^a^,^c	0.869	5.90	38.26	—	37	−6	—
B100	0.894	6.09	39.45	57.35	112	1	30.05^b
B20	0.850	3.26	40.98	n.d	79	n.d	17.66^b
Petrodiesel	0.813	2.55	42.83	45	65	12	16.47^b

---

3.2. Engine performance analysis

The trend of BSFC for CI engine under different load conditions with petrodiesel, JCO B100, JCO B20 and MHBF are illustrated in Fig. 2(a) at CR 17.5 and 15. The results showed that irrespective of the fuel type, the BSFC gradually decreased with increasing load on crank shaft (at both CR) which is in agreement with the fact that fuel burns more efficiently at high load conditions because of proper combustion inside the engine cylinder. Similar findings were also reported by Qi and co-workers who found a comparable trend of BSFC for microemulsion based fuel and petrodiesel.¹² It is important to note that, the BSFC of CI engine depends on (i) the amount of fuel injected, (ii) density, (iii) kinematic viscosity (iv) CR and (v) calorific value of the fuel and the dissimilar trends obtained for the test fuels were a consequence of these effects.^12–14 As compared to petrodiesel and JCO B20, higher amount of MHBF was needed to produce the same amount of energy as MHBF exhibited considerably lower calorific value (Table 3). However, MHBF was found to exhibit almost similar consumption trend to JCO B100 at all load conditions particularly at CR 17.5 (attributed to comparable calorific values of the two fuels, at CR 15 as the combustion was incomplete/partial, also refer to Section 3.3). In other words, at CR 15 the BSFC increased considerably at all load conditions in order to compensate the lower combustion pressure (i.e. more MHBF was required to give the same power output at CR 15). This revealed the fact that MHBF was not suitable for operation at CR 15. Moreover, due to the presence of 45% alcohol, the viscosity of MHBF was significantly lower than that of JCO B100, which could facilitate the formation of apposite droplets during the spray. This resulted in better evaporation and atomization for proper mixing with air and eventually reflected by the superior combustion behavior of MHBF as compared to JCO B100.

Fig. 2 Variation in (a) BFSC (b) brake power (c) mechanical efficiency and (d) indicated thermal efficiency with respect to load (the engine was operating at constant speed of 1500 rpm).

Fig. 2(b) shows the variations of η_bte of CI engine for petrodiesel, JCO B100, JCO B20 and MHBF under variable load conditions at CR 17.5 and 15. Here, the different trends observed for the different test fuel originated from the difference in their higher heating (calorific) values (refer to eqn (4), ESI†). Irrespective of the fuel type η_bte showed an increasing trend up to the full load beyond which it gradually decreased because of overload condition (which is the typical for constant speed CI engines).¹⁸ At no load condition, the time available for heat dissipation to the cylinder walls was relatively long as compared to part load and full load, and hence significant amount of heat loss occurred. When the load was increased the neat heat release rate also increased which resulted high combustion pressure that led to an equivalent increase in η_bte (Fig. 2). Here, the η_bte for CI engine fueled with MHBF was found to be higher than JCO B100 (for all load conditions at CR 17.5), which could be attributed to the presence of 45% light alcohol in MHBF system. Overall, under full load conditions at CR 17.5 the order of η_bte was found to be petrodiesel (31.75%) > JCO B20 (30.71%) > MHBF (29.8%) > JCO B100 (28.01%). The low η_bte was observed for MHBF tested at CR 15 (21.55%) which could be attributed to the incomplete combustion. Furthermore, at a low compression ratio (CR 15), combustion pressure and net heat release rate were also lowered which resulted low η_bte up to full engine load conditions. The slightly increased in η_bte of MHBF as compared to JCO B100 could also be attributed to the higher reaction activity of MHBF in the fuel-rich zone.^5,12–14 Besides, due to the lower flame temperature of light alcohols (ethanol and 1-butanol) in comparison to the other test fuels (petrodiesel and biodiesel consisting long chain hydrocarbons and fatty esters) the heat losses in the engine cylinder also decreased when using MHBF (this is correlating with the lowest flash point for MHBF, Table 3).^14,19,20

The trend of η_mech obtained with different test fuels under variable load conditions clearly showed that the mechanical efficiency of CI engine fueled with MHBF was comparable to that of petrodiesel at CR 17.5 and outperformed both petrodiesel and JCO B100 at CR 15 (Fig. 2(c)). The order of η_mech at full load at CR 17.5 is in the order JCO B20 ≥ MHBF = petrodiesel > JCO B100. This was because the 20% oxygenated biodiesel in JCO B20 (blend) facilitated improved combustion with high rate of brake power which in turn resulted higher mechanical efficiency at full load.^20–22 Not surprisingly at CR 15, the η_mech value reduced further for all test fuels possibly due to incomplete combustion.¹⁴

η_ite of CI engine is related to the indicated power developed in the combustion chamber during power stroke as defined by the ratio of indicated power to the total energy released from fuel air mixture. From the Fig. 2(d) it could be clearly seen that among all the tested biofuels (JCO B100, JCO B20 and MHBF) MHBF exhibited the highest η_ite at 20% of load and in fact the value was found to be comparable with petrodiesel and higher than JCO B100 under certain load conditions at CR 17.5. This observation can be attributed to the high combustion pressure generated in CI engine cylinder with MHBF due to its superior atomization and combustion behavior.^23–27 However, due to improper combustion at CR 15 the η_ite value of MHBF lowest (also refer to Section 3.3).

Considering the calorific values of the test fuels, as expected petrodiesel produced highest BP while MHBF and JCO B20 were found to exhibit similar BP at all load conditions (particularly at CR 17.5, ESI†). However, the BP of JCO B100 (4.28 kW) was lowest to JCO B20 (4.33 kW), petrodiesel (4.33 kW) and MHBF (4.3 kW) on full load condition. It was also likely that in addition to the lower calorific value, higher density and viscosity of JCO B100 caused poor spray behavior which contributed to the reduce BP. On the other hand, due to the presence of light alcohols the BP of MHBF was significantly brought down under no load condition (1-butanol and ethanol causes a higher latent heat of evaporation which causes cooling effect in the combustion chamber and this negatively influences the thermal efficiency on no burden and at the commencement of the CI engine causing BP to decrease); the result was particularly prominent at CR 15 where upon MHBF combustion was incomplete (ESI data, Fig. S1†). However, this effect was overcome with increased engine load and at a high CR 17.5.^28–30

3.3. Combustion characteristics

The combustion characteristics of different test fuels were studied with respect to variation in cylinder pressure, because cylinder pressure variation in the combustion chamber is directly related to the combustion characteristics of test fuel. Cylinder pressure data of the CI engine for different test fuels such as petrodiesel, MHBF, JCO B100, JCO B20 and MHBF at CR 17.5 and 15 were acquired through data acquisition system and the results represented in terms of (i) cylinder pressure vs. crank angle and (ii) rate of heat release vs. crank angle plots, for the analysis of combustion characteristics (Fig. 3). It was evident from Fig. 3(a) that irrespective of the fuel type, the engine cylinder pressure was increased for CI engine from compression stroke to power stroke of the engine. Further, it could be determined from the Fig. 3(a) that MHBF followed the similar pressure path to that of petrodiesel at CR 17.5, but the pressure path was significantly different from petrodiesel when MHBF was tested at CR 15 and in agreement with incomplete combustion or combustion in stages under low in-cylinder pressure at this CR. The peak cylinder pressure value (CP_max) for MHBF (60.36 bar) was found to be slightly higher than that of petrodiesel (57.78 bar) and also the respective crank angle on that peak pressure value occurred at nearby positions to petrodiesel. The CP_max were 60.36, 58.73, 58.70, 57.78, 38.73 bars for MHBF, JCO B100, JCO B20, petrodiesel and MHBF at CR 15, respectively. Besides, the crank angle corresponding to CP_max was almost same (370°) for MHBF, JCO B100, B20 and petrodiesel. But for MHBF tested at CR 15 the value (381°) was considerably higher which suggested late or incomplete combustion. The CP_max for MHBF was higher because of the low flash point of anhydrous 1-butanol and ethanol.^17,31–36

Fig. 3 Variation of (a) net heat release rate and (b) cylinder pressure at different crank angle (the values are average of all readings obtained for the test fuels under different load conditions, the engine was operating at constant speed of 1500 rpm).

Further, from Fig. 3(b) it was seen that for all test fuels, the net heat release (NHR) were increasing from compression stroke to power stroke and there was no observable difference in the trends of NHR at CR 17.5. However, as expected for MHBF tested at CR 15, the NHR pattern was significantly different and unveiled late combustion on crank angle 376° (also refer to Fig. 3(a)). Nonetheless, the maximum heat release rate (NHR_max) was found to be higher (45.48 J at CA 363°) for MHBF and lowest (33.24 J at CA 359°) for JCO B100. In fact, the ignition delay is dependent on latent heat of vaporization and auto ignition temperature of the fuel/air mixture; since, ethanol and 1-butanol have higher latent heat of vaporization as well as lower auto ignition temperature, this leads to premixed combustion and ignition delay in case of MHBF (the effect is more easily recognizable at CR 15, Fig. 3). Further, due to proper air fuel mixing, during this fraction of time, the net heat release as well as combustion pressure gets increased (Fig. 3).¹⁷ The nonlinear trend of combustion observed for MHBF could be ascribed to its multi component nature. Further, as the ignition delay increased, more fuel was required at a lower CR and the NHR was increased in combustion phase. This trend was clearly seen in the results where petrodiesel and biodiesel showed longer controlled combustion period and MHBF showed a longer premixed combustion period (Fig. 3(b)).^22,35–37 Therefore, crank angle of 359–363° was optimum for combustion in CI engines at CR 17.5 with respect to no ignition delay.

3.4. Emission analysis

The emission parameters of the test fuels were compared at CR 17.5 as the performance and combustion parameters of MHBF was comparable to the other test fuels at this CR. CO emission from CI engine is a consequence of incomplete combustion, slow burning of the soot during final combustion phase and also due to poor air to fuel ratio. In our experiments, MHBF and petrodiesel exhibited almost similar CO emission trends at CR 17.5; but at full load condition emission from MHBF was found to be considerably lower than that of petrodiesel (Fig. 4(a)). Further, from Fig. 4(a) it is clear that at high engine loads the CO emission for the oxygenated fuels JCO B20 and JCO B100 was almost similar with MHBF. Whereas, when engine CR was reduced, CO emission with MHBF was also increased, doubled at all loads resulted from incomplete combustion (also refer to Section 3.3). As stated earlier, the presence of 45% alcohol in MHBF increases the oxygen content of fuel which improves its atomization and combustion and eventually contributed to lower CO emission levels. At the same time, alcohols also generate a cooling effect (particularly at lower engine loads) which can reduce in-cylinder gas temperature and contributed to poor combustion and increased in CO emissions, particularly at low engine loads. As the combustion temperature was enhanced for partial and full load conditions that counter weighted the cooling effect of ethanol. Thus, our results suggested that alcohol content of MHBF and related oxygenated fuels influenced their combustion and emission characteristics.^27,29,30

Fig. 4 Variation of (a) CO (b) HC (c) CO₂ and (d) NO_x emission profiles with respect to load (at CR 15, the engine was operating at constant speed of 1500 rpm).

Unburnt HC is another important parameter for determining the emission characteristics of a fuel, resulted from incomplete combustion in a CI engine due to inadequate combustion temperature and insufficient air supply. The solutions obtained in this study evidenced that, at CR 17.5, compared to the other test fuels, HC emission from MHBF was higher (at all load conditions). This observation could be accredited to the incomplete combustion of the heavier triglyceride component of MHBF at the end of combustion phase. Further, it was also observed from Fig. 4(b) that MHBF tested at CR 15 gave highest HC emission as combustion was incomplete (also refer to Section 3.3). The reason for the relatively higher HC of MHBF could be mainly attributed to the lower volatility of the high molecular weight triglyceride component of JCO as compared to the other carbon fuels as stated (petrodiesel, biodiesel and blends).^27–30

CO₂ is the byproduct of complete combustion and an increase of CO₂ in the exhaust is an indication of proper combustion inside the CI engine combustion chamber. In case of incomplete combustion CO and HC emission always increases in the reciprocal rate of CO₂.^26,27 The variation of CO₂ emissions with engine load have been illustrated in Fig. 4(c). For all the test fuels, the increasing trend of CO₂ emission with increased engine load showed that the combustion was more effective at higher loads (also refer to Section 3.2). However, the CO₂ emission for MHBF was fairly lower than all other test fuels (at all load conditions) as it was an oxygenated fuel, thus lesser amount of CO₂ was produced as combustion byproduct. While the highest CO₂ emission of petrodiesel at full load condition was at par with its complete combustion and non-oxygenated nature. Similarly, the other oxygenated fuels (JCO B100 and JCO B20) also exhibited reasonably reduced CO₂ emissions. However, these values increased significantly and even outperformed petrodiesel at 60–80% load. Further, MHBF tested at CR 15 showed the similar CO₂ emission trend, although it exhibited very high value of CO emission which may be again due to the incomplete combustion of this fuel at a lower compression ratio.^27,29,30

The main sources of NO_x emissions from IC engine are thermal NO_x and fuel NO_x. In fact, increased NO_x emissions from traditional biofuels such as biodiesel have been one of the major bottlenecks in their large scale realization.⁴ As N content of all the tests fuels were insignificant, during our experiments thermal NO_x (resulting from very high combustion temperature ≤ 1261 °C) was likely to be the most relevant source of emission.^4,37 The rate of NO_x formation is mainly dependent upon combustion (reaction) temperature, nitrogen residue in the air and the amount of oxygen content in the fuel.^4,34 Here, similar trends of NO_x emission were obtained for all the oxygenated fuels (JCO B100, JCO B20 and MHBF) that showed an increasing trend up to 80% load, after which it decreased slightly (at full load) (Fig. 4(d)). Conversely, the non-oxygenated petrodiesel was exceptional in this regard and showed an almost exponential increase of NO_x emission at full engine load. The order of NO_x emission can be related to the activation energy of the fuel, higher the activation energy higher is the NO_x emission.^17,33,34 The activation energy presented in Table 3 for the tested fuels and the NO_x emission pattern as shown in Fig. 3 were in agreement with literature.¹² The results obtained can be explained in terms of the reduced combustion temperatures for the oxygenated fuels and accordingly for MHBF (containing the highest amount oxygenate: ∼45% alcohol by volume) NO_x emission was low (∼5 times lower than petrodiesel at full engine load).^17,33,34 Further reduction of NO_x was observed at CR 15, but this reduction was mainly due to incomplete combustion at low combustion pressure (also refer to Fig. 3 and Section 3.3).^27–30

4. Conclusion

The experimental investigation of performance, combustion and emission characteristic of JCO based MHBF unveils the following facts:

• The BSFC of MHBF was comparable with the other oxygenate fuels (JCO B100 and JCO B20) but slightly higher than petrodiesel under all engine load conditions.

• MHBF showed delayed and prolonged combustion at CR 15, but at CR 17.5 combustion of MHBF was sufficiently smooth and comparable to that of petrodiesel.

• The emissions levels (CO and CO₂) of MHBF were comparable to that of petrodiesel for 0–80% load, while at full engine load the emissions reduced appreciably.

• HC emission for MHBF was higher than that for petrodiesel, JCO B100 and JCO B20, respectively, and was observed to increase with engine load at CR 17.5 and 15. The HC emission from MHBF could be attributed to the unburned triglyceride part of oil at the end of combustion. Nevertheless, HC emissions from MHBF could be reduced by operating the engine at a higher CR.

• A salient finding was that the NO_x emissions from MHBF were considerably lower than all other test fuels (JCO B100, JCO B20 and petrodiesel) at all engine loads and CR. There was almost 5 fold reduction in NO_x emissions observed for MHBF at full engine load and at CR 17.5 as compared to petrodiesel.

In summation, these results confirm the uniqueness and stability of MHBF as a fuel for CI engine application and can be recommended for application in engines with CR 17.5 without any modification of engine components.

Nomenclature

MHBF	Microemulsion based hybrid fuel
FFA	Free fatty acid
JCO	Jatropha curcas oil
FAME	Fatty acid methyl ester
BP	Brake power
MG	Monoglyceride
CI	Compression ignition
DG	Diglyceride
BSFC	Brake specific fuel consumption
TG	Triglyceride
BTDC	Before top dead center
CPmax	Maximum cylinder pressure
FID	Flame ionization detector
NHRmax	Maximum net heat release rate
VCR	Variable compression ratio
ηbte	Brake thermal efficiency
CR	Compression ratio
ηite	Indicated thermal efficiency
CA	Crank angle
ηmech	Mechanical efficiency
DAD	Data acquisition device
Cd	Coefficient of discharge of orifice
rpm	Revolutions per minute
Wden	Water density
IP	Indicated power
Aden	Air density
HHV	High heating value

Acknowledgements

The authors Himansh Kumar, Lakhya Jyoti Konwar and Mohammad Aslam acknowledge the financial support received from SSS-NIBE (MNRE, Govt. of India) in the form of SSS NIBE-Bio-Energy Promotion Fellowship.

References

1. M. I. Arbab, H. H. Masjuki, M. Varman, M. A. Kalam, H. Sajjad and S. Imtenan, RSC Adv., 2014, 4, 37122–37129.
2. N. Kumar and U. Khare, SAE Tech. Pap. Ser., 2004 DOI:10.4271/2004-28-0040.
3. A. Sanjid, M. A. Kalam, H. H. Masjuki, S. M. Ashrafur Rahman and M. J. Abedin, RSC Adv., 2014, 4, 36973–36982.
4. N. Singh, H. Kumar, M. K. Jha and A. K. Sarma, J. Therm. Anal. Calorim., 2015, 122(2), 907–916.
5. P. J. Singh, J. Khurma and A. Singh, Renewable Energy, 2010, 35(9), 2065–2070.
6. A. Sanjid, H. H. Masjuki, M. A. Kalam, S. M. Ashrafur Rahman, M. J. Abedin and I. M. Rizwanul Fattah, RSC Adv., 2015, 5, 13246–13255.
7. P. Bora, L. J. Konwar, M. M. Phukan, D. Deka and B. K. Konwar, Fuel, 2015, 157(1), 208–218.
8. P. Bora, L. J. Konwar, J. Boro, M. M. Phukan, D. Deka and B. Konwar, Appl. Energy, 2014, 135(15), 450–460.
9. P. Bora, J. Boro, L. J. Konwar and D. Deka, Energy Sources, Part A, 2012 DOI:10.1080/15567036.2012.747036.
10. P. Bora, J. Boro, L. J. Konwar and D. Deka, J. Energy Inst., 2015 DOI:10.1016/j.joei.2015.07.001.
11. G. R. Kannan and R. Anand, Energy, 2011, 36, 1680–1687.
12. D. H. Qi, H. Chen, R. D. Matthews and Y. Z. H. Bian, Fuel, 2010, 89(5), 958–964.
13. P. Singh, J. Khurma and A. Singh, Am. J. Environ. Sci., 2010, 6(1), 69–75.
14. P. Singh, J. Khurma and A. Singh, Renewable Energy, 2010, 35(9), 2065–2070.
15. A. P. S. Chouhan, N. Singh and A. K. Sarma, Fuel, 2013, 109, 217–224.
16. http://apexinnovations.co.in/pdf/ICEngineSoft-8.5VTraining.pdf, accessed 05.2.2016.
17. M. Assawadithalerd, N. Chollacoop, J. Mungkalasiri and C. Tongcumpou, Fuel, 2016, 134–140.
18. J. B. Heywood, Internal combustion engines fundamentals, McGraw Hill, Newyork, 2nd edn, 1998.
19. D. L. Do, A. Witthayapanyanon, H. J. Harwell and A. D. Sabatini, J. Surfactants Deterg., 2009, 12(2), 91–99.
20. S. M. Kumar, J. Bellettre and M. Tazerout, Proc. Inst. Mech. Eng., Part A, 2009, 223, 729–742.
21. S. Fernando and M. Hanna, Energy Fuels, 2004, 18(6), 1695–1703.
22. C. Attaphong, L. Do and A. D. Sabatini, Fuel, 2012, 94(4), 606–613.
23. M. Jindal, P. Rosha, S. K. Mahla and A. Dhir, RSC Adv., 2015, 5, 33863–33868.
24. C. J. Huang, D. Y. Wang, D. S. Li, P. A. Roskilly, D. H. Yu and F. H. Huifen Li, Appl. Therm. Eng., 2009, 29(11–12), 2484–2490.
25. D. C. Rakopoulos, K. A. Antonopoulos, C. D. Rakopoulos, T. D. Hountalas and G. E. Giakoumis, Energy Convers. Manage., 2006, 47, 3272–3287.
26. H. E. Pryde, J. Am. Oil Chem. Soc., 1984, 61, 1609–1610.
27. H. Aydin and C. Ilkiliç, Appl. Therm. Eng., 2010, 30, 1199–1204.
28. J. Campos-Fernández, M. J. Arnal, J. Gómez and P. Dorado, Appl. Energy, 2012, 95, 267–275.
29. A. Srivastava and R. Prasad, Renewable Sustainable Energy Rev., 2000, 4, 111–133.
30. C. Sayin, Fuel, 2010, 89, 3410–3415.
31. S. Altun, H. Bulut and C. ONer, Renewable Energy, 2008, 33, 1791–1805.
32. H. Chen, J. S. Shuai and X. J. Wang, Proc. Combust. Inst., 2007, 31, 2981–2989.
33. H. Huang, Q. Liu, R. Yang, T. Zhu, R. Zhao and Y. Wang, Energy Convers. Manage., 2015, 106, 748–758.
34. H. Huang, C. Zhou, Q. Liu, Q. Wang and X. Wang, Appl. Energy, 2016, 219–231.
35. S. Imtenan, H. H. Masjuki, M. Varman and I. M. Rizwanul Fattah, RSC Adv., 2015, 5, 17160–17170.
36. Y. H. Teoh, H. H. Masjuki, M. A. Kalam, M. A. Amalinaa and H. G. Howa, RSC Adv., 2014, 4, 50739–50751.
37. G. P. Boulter, J. Borken-Kleefeld and L. Ntziachristos, Handb. Environ. Chem., 2013, 26, 31–54.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04795e

---