Combustion and Performance Evaluation of Methanol (M15)-Fueled BS-VI Compliant Light-Duty Spark-Ignition Engine

Intergovernmental Panel report (6th) on climate change outlines a rise in global warming by 1.5 °C and heat waves [1]. The transport sector accounts for 23% of total greenhouse gas (GHG) emissions, 16% from road transport and 7% from railroad, marine, and air transport [2]. Spark-ignition (SI) engine powertrains hold a significant market share among passenger vehicles. However, stringent emission norms and continuous rise in fuel prices seek their evolution to become cleaner and more efficient. Alternative fuels such as primary alcohols, compressed natural gas, and hydrogen are emerging alternate fuels for SI engines, which can reduce emissions and hence the dependency on petroleum-based fossil fuels [3–5]. Alcohols have many favorable properties to make them superior alternative fuels. Primary alcohols (ethanol and methanol) can be produced from biodegradable and sustainable resources such as biomass. Using a higher percentage of alcohol in engines requires fuel injection and handling system modifications. However, lower gasoline–alcohol blends can be used in the engine without significant hardware modifications [6]. The fuel-bound oxygen in primary alcohols helps reduce carbonaceous emissions than baseline gasoline. Alcohols have higher octane numbers and flame speeds, which can be used to develop more efficient engines with higher compression ratios.

Currently, Indian vehicles are fueled with a 10% (v/v) ethanol–gasoline blend, and the government plans to increase it to 20% (v/v) by 2025 [7]. The Government of India is also planning to implement methanol utilization in the energy sector. The National Institution for Transforming India (NITI Ayog) has launched the “Methanol Economy” mission to meet future energy demand, reduce oil imports, and GHG emissions. Methanol can be produced from high-ash coal, low-value agricultural biomass residue, municipal solid waste, domestic waste, CO₂ from thermal power plants, and natural gas [8–10]. Methanol has higher autoignition temperature, octane number, flame speed, and flammability limits than baseline gasoline. It increases the knock limit of the engine operating with a high compression ratio and improves the overall thermal efficiency [11]. Higher latent heat of vaporization (LHV) of methanol decreases the intake air temperature, thus improving the engine's volumetric efficiency [12]. NO_x emissions are reduced due to methanol's higher LHV as the peak in-cylinder temperature decreases significantly than baseline gasoline-fueled SI engines. Methanol's higher oxygen content reduces engine-out CO emission [13]. Researchers have studied various aspects of using methanol blends in SI engines. Li et al. [13] compared the combustion, performance, and emission characteristics of different methanol, ethanol, and butanol blends with baseline gasoline for a port-fueled injected (PFI) SI engine. It was reported that adding alcohol led to advanced combustion phasing due to the higher flame speed of alcohol. This resulted in lower brake thermal efficiency (BTE) as ignition timing for alcohol was kept similar to gasoline's maximum brake torque (MBT) timing. Thus, calibration modifications are required to extract the best potential of alcohol-blended gasoline. Balki et al. [14] investigated a small SI engine fueled with M5-M20 and reported improved performance and reduced emissions from a high compression ratio engine. An increasing methanol fraction blended with gasoline increased the BTE, brake-specific fuel consumption (BSFC), and CO₂ while decreased the unburnt hydrocarbons, CO, and NO_x emissions. Prasad et al. [15] reported increased thermal efficiency and NO_x emissions with an increased compression ratio for the methanol blend (M50) at full load conditions. Adding methanol reduced NO_x, and the engine's best performance was observed for 10:1 compression ratio. Chen et al. [16] investigated methanol and ethanol's combustion process and their cyclic variations. They reported improved combustion stability and lower coefficient of variance (COV) of indicated mean effective pressure (COV_imep) for methanol blends. Gheng et al. [17] evaluated three test fuels for low- and high-load conditions, including methanol–gasoline blends (M15 and M45) and 100% gasoline. They reported an earlier rise in the cylinder pressure for methanol blends compared to baseline gasoline. Ignition delay increased upon methanol blending of gasoline, but no influence was observed on the combustion duration. Agarwal et al. [18] assessed 10% and 20% (v/v) methanol-blended gasoline in a medium-duty PFI SI engine and reported identical combustion characteristics but higher BTE for methanol blends. Zhu et al. [19] studied the effect of exhaust gas recirculation (EGR) and spark timings on the performance of a 100% methanol-fueled heavy-duty SI engine. The EGR rate was maintained at 20–30% to improve the BTE. The average BTE of the engine for 100% methanol was ∼40% at engine speeds of 1000–1700 rpm. The BTE was 3% higher for methanol than baseline gasoline. The COV_imep for 100% methanol was below 2%, even for a high EGR rate case. Wu et al. [20] investigated the idle lean-burn combustion, performance, and emission characteristics of a PFI SI engine fueled with 100% methanol and gasoline. They reported improved thermal efficiency for methanol compared to baseline gasoline and achieved a higher indicated thermal efficiency of 24.7% (at λ = 1.4) under idle lean-burn conditions. The combustion was faster, and flame development and propagation duration were shorter for methanol than baseline gasoline. The COV_imep was lower for methanol than gasoline due to its wider flammability limits. Abu-Zaid et al. [21] studied the effect of different methanol–gasoline blends on the engine performance and reported that the best performance was exhibited by M15 and M85. Liu et al. [22] studied the performance, emission characteristics, and cold start behavior of a three-cylinder PFI SI engine for lower methanol–gasoline blends. The addition of methanol improved the cold starting of the engine. The maximum engine power output and torque were lower for methanol, but the BTE was higher than baseline gasoline. HC and CO emissions were also reduced significantly. Most methanol–gasoline blend studies in the literature were performed on lower compression ratio PFI engines, which complied with older emission norms. Therefore, a new-generation Bharat Stage-VI (BS-VI) PFI SI engine with a high compression ratio must be studied to assess M15 implementation on a national level. Therefore, in this study, the combustion and performance characterization of M15 has been done on a BS-VI factory-configured PFI SI engine, and the results are discussed.

Gasoline (G100) with research octane number ∼92–94 was taken as baseline fuel. The blended fuel M15 was prepared by mixing 15% (v/v) methanol, 3% (v/v) isopropyl alcohol, and 82% (v/v) gasoline, where isopropyl alcohol was used to suppress the phase separation of methanol and gasoline. The calorific values of the two test fuels were measured using a bomb calorimeter (6200, Parr, Moline, IL). The densities were measured using a portable density meter (DA-130N, Kyoto Electronics, Kyoto, Japan). These measured properties of test fuels at normal temperature and pressure are given in Table 1.

A three-cylinder, in-line, naturally aspirated, water-cooled BS-VI PFI SI engine was used for this study. The engine was coupled to an eddy current dynamometer (ECB50-200, Dynalec, Pune, India) to conduct steady-state tests. A dynamometer controller was used for varying the engine speed at a constant load or varying the load at a constant engine speed. The schematic of the experimental setup is shown in Fig. 1. The specifications of the test engine are given in Table 2.

The fuel injection system of the test engine had a solenoid-operated port fuel injector, a fuel rail, an electric fuel pump, a fuel filter, and a fuel tank. The fuel injection pressure was maintained at 3.5 bar. A burette installed between the tank and the electric fuel pump measured the volumetric flowrate of the test fuel. A laminar flow element (LFE) (50MC2-2F, Meriam, Cleveland, OH) was installed to measure the test engine's intake air flowrate. A U-tube manometer connected across the orifice of the LFE measured the pressure drop, and the pressure head was used to calculate the mass flowrate of the intake air to the test engine. A spark plug-based piezoelectric pressure transducer (GH13Z-24, AVL, Graz, Austria) was used to measure the in-cylinder pressure. A charge amplifier (AVL, Graz, Austria) converted the charge (in µC) generated by the piezoelectric pressure transducer to a proportional voltage (0–5 V) signal. A precision optical crank angle encoder (2614C11, Kistler, Winterthur, Switzerland) was used to determine crankshaft rotation with 0.1 deg resolution. The lambda values were measured by a lambda sensor (LSU4.9, Bosch, Gerlingen, Germany) installed in the exhaust manifold with a display module. Two current clamps connected to the signal conditioning unit were used to detect the fuel injection and spark timings. All four signals (in-cylinder pressure, crank angle, injection, and ignition signals) were acquired by a high-speed combustion data acquisition system (Indimodule, AVL, Graz, Austria), which also processed the data. Exhaust gas, lubricating oil, and coolant temperatures were measured using thermocouples, and the data were used to ensure the engine's thermally stable operating condition.

The experiments were performed at six engine loads varying from 1 to 6 bar brake mean effective pressure (BMEP) in steps of 1 bar, and the engine speed was varied from 1000 rpm to 4000 rpm in steps of 1000 rpm. The experiments for both test fuels were performed using a factory-configured original equipment manufacturer (OEM) electronic control unit (ECU) tuned for baseline gasoline. However, the fuel injection quantity was slightly adjusted in the ECU mapped for M15 to match the targeted lambda values (Table 3) with respect to baseline gasoline. The air-fuel mixtures were stoichiometric at all the test points, except 5 and 6 bar BMEP at 1000 rpm and 6 bar BMEP at 4000 rpm. The charge has to be enriched to cater to the high load demand at low and high engine speed conditions. The spark timings were identical for M15 and baseline gasoline fueling. The stock ECU controlled the fuel injection and spark timings for different engine operating conditions. The ambient air temperature at the time of experiments was ∼35 °C.

Engine performance parameters were calculated by taking the average of three readings acquired at each engine operating point. Pressure-crank angle (P–θ) data for 250 consecutive engine cycles were recorded by the combustion analyzer, and the averaged data were used for the calculation of combustion parameters, namely heat release rate (HRR), cumulative heat release (CHR), rate of pressure rise, the start of combustion (SOC), end of combustion (EOC), combustion phasing (CP), combustion duration (CD), and COV_imep for the two test fuels at different engine operating conditions.

In this study, the comparative assessment of M15 and baseline gasoline was done for combustion and performance characteristics at a wide range of engine operating conditions. For combustion analysis, in-cylinder pressure, heat release rate, and cumulative heat release trends were assessed. Combustion timings such as the start of combustion, combustion phasing, end of combustion, and combustion duration were also compared. The COV_imep was calculated to assess test fuels’ combustion stability and cyclic variability at different engine operating conditions. BTE, BSFC, BSEC, and exhaust gas temperature (EGT) were compared for the two test fuels at different engine operating conditions to assess engine performance. The results are divided into two sections: (i) combustion characteristics and (ii) performance characteristics.

The variation of in-cylinder pressure and heat release rate with crank angle degrees for gasoline and M15 fuels are shown in Fig. 2. Per the test matrix, these plots are presented for three engine loads at four engine speeds. The spark timings at different operating conditions were identical for both test fuels. These spark timings were programmed in the map in the OEM stock ECU, which was tuned for baseline gasoline operation.

It can be observed that retarded spark timings were used at low engine speeds and high loads. These engine operating conditions have a higher propensity of knocking; hence, knock-limited spark timing is considered in preference to the MBT. At low engine speed, higher absolute time (in ms) is adequate for the autoignition of end-charge in the cylinder once the combustion starts. Hence, spark timing must be retarded to reduce the peak cylinder pressure and temperature. At all engine operating conditions, the peak cylinder pressure occurs at ∼15 deg after top dead center (TDC), except in high-load conditions at 1000 rpm. At 1000 rpm, 6 bar BMEP, a highly retarded spark led to two pressure peaks, one each for compression and combustion. This limited the peak cylinder pressure to 20 bar. At 4000 rpm, 6 bar BMEP, the peak cylinder pressure reached ∼43 bar. Considering the effect of test fuels, both M15 and G100 exhibited almost similar combustion trends at all the engine loads and speeds. It is evident from the open literature that methanol addition to gasoline results in faster combustion since the flame speed of methanol is higher than gasoline [24]. However, this effect was not dominant in this study due to a lower methanol fraction blending. In addition, the higher compression ratio (CR 11:1) of the existing engine might have also affected the thermodynamic conditions of the charge such that the chemical kinetics improved [15]. Slight differences in the peak cylinder pressure can be observed among high engine load conditions at varying engine speeds. The higher engine loads were met by a higher injected fuel mass, which proportionately increased the methanol quantity injected.

The cumulative heat release trends also showed similar behavior from both test fuels, as shown in Fig. 3. The CHR is the total heat released by the combustion of fuel in the engine cylinder. The CHR derived from the instantaneous pressure–volume data is the net heat release, in which combustion inefficiency, heat transfer, crevices, and leakage losses are also explicitly incorporated.

The CHR increased with the increasing engine load at a constant speed due to the increased fuel quantity injection. CHR was mainly influenced by engine load and speed. Increased cylinder pressure and temperature with increased engine load enhanced the combustion kinetics, rapidly releasing the gross heat and increasing the CHR. With increasing engine speed also, the CHR increased. However, with respect to the crank angle position, it did not change.

The SOC, CP, and EOC are shown in Fig. 4 at different engine loads and speeds to reflect different stages of combustion of M15 and G100. CP helps optimize the engine parameters for obtaining the maximum BTE.

The SOC was mainly a few degrees before the TDC, for all engine loads and speeds, except high-load conditions at 1000 rpm, due to retarded spark timings. The SOC for M15 occurred almost concurrently for G100 at all engine operating conditions. It showed that blending 15% methanol had a negligible effect on the ignition delay. Hence, it can be inferred that the overall charge cooling due to methanol's higher latent heat of vaporization would be insignificant due to a small fraction of methanol blending. The combustion phasing and the end of combustion were similar for both test fuels. Although slight variations were present at a few engine operating conditions, it is difficult to interpret since the values lie in the range of error bars. Hence, the engine's combustion kinetics was not affected by the adoption of M15 in place of G100. From Fig. 5, similar observations can be drawn for the trends of combustion duration.

For most engine operating conditions, the CD of M15 was almost equal to or slightly lower than G100. Overall, indifferent results were obtained considering different combustion stages from M15- and G100-fueled engines.

Cyclic variability is an important parameter in assessing combustion stability for transport engine applications. Cycle-to-cycle variations must be minimal for the smooth drivability of the engine. Hence, it was investigated by calculating the COV of combustion parameters from a specific number of engine cycles. COV_imep is usually analyzed to study this aspect in most engine studies. Figure 6 shows the COV_imep values obtained from 250 consecutive engine cycles for M15- and G100-fueled engines.

This study achieved COV_imep lower than 5% at all the engine operating conditions for both fuels. This showed that both test fuels demonstrated combustion stability. However, slight variations in consecutive combustion cycles persist in SI engines due to cyclic variations in the flow field near the spark plug and charge mixing, affecting the initial flame kernel growth. The cyclic variability affects the combustion, sometimes leading to occasional misfires, partially burnt or improper combustion cycles. It can be seen from Fig. 6 that COV_imep was higher at 1000-rpm engine speed as compared to other engine speeds. It decreased with increasing engine load. The mixing rates were slower at low engine speeds, which prolonged the combustion. Thus, it suffered from higher chances of variations in heat release rate. On the other hand, higher engine speed exhibited a higher degree of turbulent mixing and shorter combustion duration, which decreased the chances of cyclic variability during flame propagation.

Further, increasing the engine load led to higher cylinder pressure and temperature, providing suitable conditions for flame growth and superior combustion. Thus, COV_imep decreased at higher engine loads. The variations in COV_imep for both test fuels were not significantly different. Except at 1000 rpm, COV_imep for M15 had similar or lower values than G100 for all engine speeds and loads. It might be due to a lack of homogeneity of the M15 charge at 1000 rpm. Variations in charge homogeneity could be due to methanol's lower volatility and lower evaporation, which dominated at lower engine speeds due to lower cylinder temperature and mixing rates. However, this improved at higher engine speeds due to increased mixing and in-cylinder temperatures. This might have led to stable and improved combustion due to methanol's inherent fuel oxygen. Hence, M15 resulted in nearly similar or more stable combustion at most engine operating conditions than baseline gasoline.

BTE is the most important parameter for defining engine performance. Figure 7 shows the change in BTE percentage for the M15 with respect to the G100-fueled engine at different loads and speeds.

Both test fuels followed similar trends for different engine operating conditions. BTE increased with an increasing engine load at all speeds except high-load conditions at 1000 rpm. Due to lower volumetric efficiency and poor combustion quality, BTE was reduced at lower loads. The lower in-cylinder temperature resulted in unfavorable conditions for efficient combustion. However, at 1000 rpm, 5 and 6 bar BMEP, BTE decreased due to highly retarded combustion, lower peak cylinder pressure, and combustion of a richer mixture. Considering the effect of engine speed, an increased BTE was observed upon increasing the engine speed from 1000 to 2000 rpm. However, further increased engine speed resulted in a similar BTE. At 1000 rpm, the high cycle time prolongs the combustion duration, increasing heat transfer losses. However, beyond 2000 rpm, heat transfer losses might not have changed, possibly due to its smaller timescales. Compared to G100, the M15-fueled engine exhibited higher efficiency at almost the entire engine operating envelope, with an average relative increase in BTE by a factor of 1.04. The increase in BTE was ∼1%. Higher BTE for M15 showed superior combustion quality and higher combustion efficiency due to methanol addition to gasoline. With lower fuel energy input in M15, losses due to combustion inefficiency and heat transfer might be lower than G100 to deliver a similar CHR, as shown in Fig. 3. Further, in the case of M15, slightly higher EGT explains the lower heat transfer losses, despite similar heat release traces with G100, as shown in Fig. 9.

The evaluation of BSFC is essential to study the fuel economy characteristics of the engine and its refueling range. A highly efficient engine is desirable for lower fuel consumption; however, it also depends on the calorific value of the test fuel. Figure 8 shows the BSFC and BSEC variations of G100 and M15 at different engine loads and speeds.

Since the calorific value of M15 (39.3 MJ/kg) was ∼10% lower than G100 (43.9 MJ/kg), BSFC was higher for the M15, although higher BTE with M15 slightly compensated for the increased specific fuel consumption. BSFC for M15 was ∼6% higher on average than G100 for overall engine operating conditions. Figure 8 compares the BSEC of both test fuels on an energy input basis. BSEC for M15 was lower than G100 and can be explained by the higher BTE of M15, as shown in Fig. 7. BSFC and BSEC decreased with increasing engine load at all speeds due to improved engine efficiency. Hence, it can be understood that though the M15-fueled engine operated with slightly higher thermal efficiency, the refueling range slightly decreased by ∼6% for the same fuel tank capacity.

Exhaust gas temperatures for M15 and gasoline at different engine speeds and loads are shown in Fig. 9. EGT helps in the qualitative analysis of the in-cylinder temperature due to combustion.

It was observed that EGT followed a similar trend for varying engine loads and speeds for both test fuels. M15-fueled engine shows slightly higher EGT at most engine operating conditions than baseline G100. The presence of oxygen in methanol molecules improved the combustion and enhanced the oxidation of intermediate combustion species. Lower heat transfer losses during M15 combustion might also be the reason for slightly higher EGT. The effect of higher latent heat of vaporization of methanol was not noticeable in reducing the global peak in-cylinder temperature, which could be due to its lower blending fraction and a higher compression ratio of the test engine. However, studies reported a lower EGT due to methanol blending in gasoline [25,26]. EGT increased with engine load at a similar rate for all engine speeds except at 1000 rpm. The higher variations in EGT at 1000 rpm were observed due to retarded spark timing at higher engine loads. Hence, delayed combustion increased the EGT.

This study investigated the combustion and performance characteristics of an M15-fueled high compression ratio BS-VI PFI engine and its comparison with baseline gasoline fueling. Experiments were performed using the stock ECU having pre-configured fuel injection and spark timing maps tuned for gasoline operation. The lambda values were matched for M15-fueled engine operation with baseline gasoline (G100) fuelled engine operation in the lambda maps to maintain a similar equivalence ratio for both test fuels by changing the fueling manually. The combustion characteristics of the M15-fueled test engine were almost identical to the baseline G100-fueled engine. The deviations in the in-cylinder pressure, heat release rate, and combustion timings were insignificant. This may be because of the lower blending fraction of methanol in gasoline. However, previous studies have reported faster combustion with methanol–gasoline blends due to the higher flame speed of methanol. The higher compression ratio of the test engine might have affected the thermodynamic conditions of the charge such that the combustion kinetics of the gasoline constituents improved. Cyclic variations of the engine were minimized by M15 fueling at most engine operating points due to stable and improved combustion due to methanol's inherent fuel oxygen content. The performance of the engine also improved slightly with M15 fueling. An overall increase in BTE by a factor of 1.04 was observed for M15, possibly due to lower heat transfer losses and reduced cyclic variations. The slightly higher EGT at a few engine operating points indicated that heat losses might be lower during M15-fueled engine operation. However, a lower calorific value of M15 led to ∼6% higher BSFC on average than baseline G100. Overall, it can be concluded that M15 performed similarly to baseline gasoline, with slightly higher BTE and lower fuel economy. Recalibration of ECU and hardware fine-tuning should be assessed to take advantage of test fuel properties to bridge the fuel economy gap. This study highlighted the impact of a 15% methanol blend on engine performance and combustion characteristics after calibration changes for maintaining similar lambda values. For large-scale implementation of methanol in the national economy, other important aspects such as engine durability, in-service emission compliance, and diagnostic functionality must be thoroughly assessed along with required calibration updates.

The authors acknowledge Dr. V. K. Saraswat, Member (S&T) NITI Ayog, Government of India, for directing Engine Research Laboratory (ERL), Indian Institute of Technology (IITK), to take up this study in collaboration with Maruti Suzuki India Limited, Gurugram, India. The authors acknowledge the JC Bose Fellowship by Science and Engineering Research Board, Government of India (Grant EMR/2019/000920) and SBI endowed Chair Professorship from State Bank of India to Professor Avinash Kumar Agarwal, which enabled this work. The authors would like to thank Sh. Roshan Lal, Sh. Hemant Kumar, Sh. Surendra, and other team members of the Engine Research Laboratory, and Mr. Ojase Jain from MSIL for their help in conducting these exhaustive experiments in a time-bound manner. The support of Sh. Anoop Bhat and Sh. P. Panda from MSIL for providing the test engine for this study is also gratefully acknowledged.

There are no conflicts of interest. This article does not include research in which human participants were involved. Informed consent not applicable. This article does not include any research in which animal participants were involved.

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.

BMEP =

brake mean effective pressure

BSEC =

brake-specific energy consumption

BSFC =

brake-specific fuel consumption

BS-VI =

Bharat Stage-VI

BTE =

brake thermal efficiency

CAD =

crank angle degree

CD =

combustion duration

CHR =

cumulative heat release

CO =

carbon monoxide

COV_imep =

coefficient of variance of indicated mean effective pressure

CP =

combustion phasing

DAQ =

data acquisition system

ECU =

electronic control unit

EGR =

exhaust gas recirculation

EGT =

exhaust gas temperature

EOC =

end of combustion

GHG =

greenhouse gas

GtCO₂ =

gigatons of carbon dioxide

HRR =

heat release rate

IPCC =

Intergovernmental Panel on Climate Change

LFE =

laminar flow element

LHV =

latent heat of vaporization

MBT =

maximum brake torque

NITI =

National Institution for Transforming India

NO_x =

nitrogen oxides

OEM =

original equipment manufacturer

PFI =

port fuel injection

SI =

spark ignition

SOC =

start of combustion

TDC =

top dead center