Effects of compression ratio, swirl augmentation techniques and ethanol addition on the combustion of CNG–biodiesel in a dual-fuel engine

Abstract

The diminishing resources and continuously increasing cost of petroleum in association with their alarming pollution levels from diesel engines has led to an interest in finding alternative fuels to diesel. Emission control and engine efficiency are two of the most important parameters in current engine design. The impending introduction of emission standards such as Euro IV and Euro V has forced research towards developing new technologies for combating engine emissions. This paper examines the effects of compression ratio, swirl augmentation techniques and ethanol addition on the combustion of compressed natural gas (CNG) blended with Honge oil methyl esters (HOME) in a dual fuel engine. The present results show that the combustion of HOME and 15% ethanol blend with CNG induction in a dual-fuel engine operated in optimized parameters at an injection timing of 27° Before Top Dead Centre and a compression ratio of 17.5 resulted in acceptable combustion emissions and improved brake thermal efficiencies. The implementation of swirl augmentation techniques increased brake thermal efficiencies (BTEs) and considerably reduced combustion emissions such as smoke, HC, CO and NO_x. The addition of ethanol also increased BTEs. However, at more than 15% of ethanol in HOME, NO_x emissions increased dramatically.

Keywords::

• dual fuel engine
• compression ratio
• swirl
• groove
• bridge
• ethanol

Introduction

The demand for petroleum fuel will rise steeply with every passing decade and is expected to reach an amount of nearly 250 million metric tonnes by the year 2024 (Agarwal 2006; Lei et al. 2010; Barnwal and Sharma 2005). A large amount of energy is required to sustain industrial growth and agricultural production. The diminishing and continuously increasing cost of petroleum resources associated with their alarming pollution levels from diesel engines have led to an interest in finding renewable and sustainable fuels alternative to diesel. The use of such alternative fuels in diesel engines will ensure energy security of many developing countries such as India and simultaneously address environmental concerns. Biodiesel is one of the best substitutes for petroleum fuels because it is environmentally benign, sustainable and economically suitable for agricultural countries (Raheman and Phadatare 2004). In particular, biodiesel derived from non-edible of Honge oil has been found to be a suitable alternative by virtue of its abundant and local availability (Murugesan et al. 2009; Banapurmath, Tewari, and Hosmath 2008, 2009). Ethanol is a renewable energy and can be derived from many raw materials including sugar cane, molasses, waste biomass materials, corn, sugar beets, etc. by using already improved and demonstrated technologies. In diesel engines, ethanol not only reduces oil usage but also significantly reduces PM and dry soot emissions. Ethanol and biodiesel blends form a high oxygen content fuel and, with both being renewable resources, together can significantly provide better combustion efficiency. Moreover, they can facilitate the total elimination of petroleum products (Banapurmath and Tewari 2010).

The clean burning characteristic of compressed natural gas (CNG) has gained more interest in its application in gas engines. CNG has become the first choice among other gaseous fuels as a clean fuel for implementation in certain metropolitan cities, where pollution emitted from conventional fuels has become intolerable and has been proved to create serious health hazards (Banapurmath et al. 2011; Carlucci et al. 2008; Karim and Burn 1980; Karim and Amoozegar 1983; Karim 1987; Karim et al. 1989; Papagiannakis et al. 2010). Natural gas comprises methane (CH₄), the shortest and lightest hydrocarbon molecule. It is lighter than air, and so it tends to dissipate. Biodiesel and CNG are thus popular contenders for use as alternatives to diesel in dual-fuel engines, and many investigations have been carried out to check their feasibility in the literature (López et al. 2009; Nwafor 2007). CNG can be utilised in diesel engines by applying gas manifold induction (Karim 1991; Karim et al. 1991; Papagiannakis and Hountalas 2003; Nwafor 2007; Papagiannakis, Hountalas, and Rigopoulos 2007), port injection (Selim 2005; Selim, Radwan, and Saleh 2008) or direct in-cylinder gas injection (Ishida, Tagai, and Ueki 2003; McTaggart-Cowan et al. 2004).

Several engine parameters have been found to affect the performance of combustion in gas engines (Mohamed Selim 2005; Sahoo, Sahoo, and Saha 2009). These parameters include injection timing (Nwafor 2007; Sayin, Uslu, and Canakci 2008), compression ratio (Mohamed Selim 2005; Sahoo, Sahoo, and Saha 2009; Budzianowski and Miller 2009b), exhaust gas recirculation (Ishida, Tagai, and Ueki 2003; Hountalas, Mavropoulos, and Binder 2008; Pierpont, Montogomery, and Reitz 1995; McTaggart-Cowan et al. 2004; Budzianowski 2010a, 2010b; Budzianowski and Miller 2009c) and swirl augmentation techniques (Budzianowski and Miller 2009a).

The increase in compression ratio results in an increase in the cycle-to-cycle variation in indicated mean effective pressure. Knock starts earlier when a high compression ratio is used and an increased compression ratio generally increases the combustion noise.

Changes in the swirl rate alter fuel evaporation and fuel–air mixing processes. They also affect wall heat transfer during compression and hence the charge temperature at injection. Swirl augmentation can result in almost complete combustion and thus increased thermal efficiency and reduced emissions. In addition, swirl augmentation leads to decreased carbon deposits, fuel consumption, exhaust gas temperature and the tendency to knock (Heywood 1988; Benajes et al. 2004; Van Gerpen, Huang, and Borman 1985; Rao, Winterbone, and Clough 1993; Timotey 1985). Swirling can be created by providing grooves on a bowl-shaped cavity as well as with manifold designs on the piston (Prasad et al. 2011). The effect of an initial swirl can be the solution for overcoming the power loss in a dual-fuel engine and for the prevention of knock occurrence. Reducing the NO_x emission is another advantage of the effect of increasing the swirl. This decrease is due to increased heat loss to the wall and reduced temperature. Swirl also influences the ignition delay by exchanging the portion between physical and chemical ignition delays. This is important because the introduction of natural gas delays the autoignition of diesel.

Ethanol is a renewable energy carrier and can be derived from many raw materials including sugar cane, molasses, waste biomass materials, corn, sugar beets, etc. by using already improved and demonstrated technologies (Banapurmath and Tewari 2010; Lei, Bi, and Shen 2011; Xingcai et al. 2004). In general, in biodiesel–ethanol blends, when the blending ratio is increased, cold flow characteristics will improve and specific gravity and viscosity of the blend will decrease. The addition of ethanol to Honge oil methyl esters (HOME) improves the efficiency of dual-fuel engines. This is because the addition of ethanol to HOME lowers the viscosity of the blend and increases the volatility of the mixture (Banapurmath and Tewari 2010).

Gas engines have recently become increasingly important in both mobile and stationary energy applications. The increased interest in gas engines has been associated with the observed rapidly growing production of the following factors: (i) transportation biofuels (Banapurmath, Tewari, and Hosmath 2008; Karim and Burn 1980; Karim and Amoozegar 1983; Karim 1987; Karim et al. 1989), (ii) biosyngases derived from biomass gasification systems (Budzianowski 2011b) and (iii) biogases produced by the anaerobic digestion of wet biomass (Budzianowski 2011c). Combustion reactions and heat transfer are two major factors that cause exergy destruction in combustion-based power systems (Dunbar and Lior 1994; Dincer and Rosen 2004; Ahmadi, Dincer, and Rosen 2011; Szargut 2010; Budzianowski 2010b, 2011a). In addition, fuel combustion is a source of unwanted emissions to the environment. These emissions are particularly large for internal combustion engines utilising petroleum fuel. HOME and CNG are popular contenders for use as alternatives to diesel, and this emerging area requires rigorous research to ascertain the feasibility of myriad fuels available in nature. The substitution of petroleum fuels by biofuels is a necessary shift towards sustainable economy (Budzianowski 2011b, 2011c). The goal of increased combustion efficiency with a simultaneous reduction of pollutant emissions can be achieved through optimising key constructional and operating parameters of combustion systems in gas engines (Banapurmath, Tewari, and Hosmath 2008; Budzianowski and Miller 2009c; Karim and Burn 1980; Karim and Amoozegar 1983; Karim 1987; Karim et al. 1989). Therefore, the current experimental study seeks such optimised conditions for the combustion of renewable CNG–biodiesel fuels in a dual-fuel engine. The focus is on both combustion efficiency and emission reduction through analysing the effects of compression ratio, swirl augmentation techniques and ethanol addition.

Characterisation of the investigated fuels

Vegetable oil is converted to simple esters of methyl and ethyl alcohols to form methyl or ethyl esters using a transesterification process called biodiesel. Biodiesel fuels can be derived from vegetable oils and animal fats. Biodiesel derived from Honge oil called HOME was prepared in the laboratory using a simple transesterification process.

Chemical composition of Honge oil and its structure

Triglycerides are the main constituents of vegetable oils. The fatty acid contribution in vegetable oils varies from oil to oil. Many samples of Honge oil were taken from different places and they vary in their chemical structure. The reasons for the variation in the fatty acid composition of vegetable oils obtained from different locations could be due to prevailing weather and soil conditions. Three such samples of fatty acids from Honge oil were considered and analysed (Mahanta, Mishra, and Kushwash 2006; Meher, Naik, and Das 2004; Yaliwal et al. 2010), and the one measured at Bangalore Test House Laboratory is presented in Table 1.

Table 1 Fatty acid composition of Honge oil (Mahanta, Mishra, and Kushwash 2006; Meher, Naik, and Das 2004; Yaliwal et al. 2010).

Sl no.	Composition	Chemical name	No. of bonds	Structure	Saturated/unsaturated	Chemical formula	Molecular weight
1	Palmitic	Hoicexadecan	–	16:0	Saturated	C18H32O2	284
2	Stearic	Octadecanoic	–	18:0	Saturated	C18H36O2	282
3	Oleic	Cis-9–octadecanoic	Single	18:1	Unsaturated	C18H34O2	280
4	Linoleic	Cis-9,cis-12-octadecanoic	Double	18:2	Unsaturated	C18H32O2	280
5	Linoleic	Cis-9,cis-12,cis-15-octadecanoic	Triple	18:3	Unsaturated	C18H30O2	278
6	Arachidic	Etcosanoic	–	20:0	Saturated	C20H40O2	312
7	Lignoceric	Tetracosanoic	–	24:0	Unsaturated	C24H48O2	368
8	Behenic	Docosanoic	–	22:1	Unsaturated	C22H44O2	340

Methane is the principal component of natural gas. Typical natural gas comprises more than 90% methane. Natural gas can be compressed, so it can be stored and used as CNG. Natural gas is safer than gasoline in many respects, and the ignition temperature of natural gas is higher than that of gasoline and diesel fuel (Banapurmath et al. 2011; López et al. 2009; Nwafor 2007).

Renewable ethanol is derived from agricultural feedstocks through already improved and demonstrated technologies. The properties of Honge oil, methyl esters and natural gas fuels are presented in Tables 2 and 3. Biodiesel is blended with ethanol to obtain different fuel blends. A stirrer is mounted inside a fuel tank in order to prevent phase separation. Figure 1 shows a mechanical stirrer used to mix the biodiesel–ethanol blend.

Table 2 Properties of diesel, Honge oil and their respective esters and biodiesel–ethanol blends.

Sl no.	Properties	Diesel	Honge oil	HOME	Ethanol	HOME+15% ethanol	ASTM standards
1	Viscosity (cSt at 40°C)	4.59	56	5.6	1.2	3.7
2	Flash point (°C)	56	270	163	13.5	32	ASTM D93D3278 – 96
3	Calorific value (kJ/kg)	45,000	35,800	36,010	27,300	33,060	ASTM D5865
4	Density (kg/m^3 at 15°C)	830	930	890	780	–	ASTM D4052
5	Cetane number	45–55	40	40–42	–	–	–
6	Cloud point (°C)	15	13	–	–	–	–
7	Pour point (°C)	1	–2 to − 5	–	–	–	–
8	Carbon residue (%)	0.1	0.66	–	–	–	ASTM D4530
9	Type of oil	Fossil fuel	Non-edible	Non-edible	–	–	–

Table 3 Properties of natural gas.

Properties	Natural gas
Boiling range (K at 1 atm)	147
Density (kg/m^3, at 1 atm and 15°C)	0.77
Flash point (K)	124
Octane number	130
Flammability limit range
Rich	0.5873
Lean	1.9695
Flame speed (cm/s)	33.80
Net energy content (MJ/kg)	49.5
Autoignition temperature (K)	923 (650°C)
Combustion energy (kJ/m^3)	24.6
Vaporisation energy (MJ/m^3)	215–276
Stoichiometric A/F (kg air/kg fuel)	17

Figure 1 Fuel tank with a mechanical stirrer and its position in the experimental set-up.

Experimental set-up

The experiments were carried out on a single-cylinder four-stroke CI engine test rig to operate in a dual-fuel mode. Figure 2 shows the overall view of the test rig that was modified to operate in a dual-fuel mode. The specifications of the engine are given in Table 4. The engine was always operated at a rated speed of 1500 rpm. A piezoelectric pressure transducer was mounted within the surface of the cylinder head to measure cylinder pressure for the combustion study. The engine tests were conducted with a dual fuel using the CNG and HOME–ethanol blends at both 80% and 100% load conditions. The engine supplier has prescribed an IOP of 200–225 bar for diesel engine operation. However, the IOP of HOME being more viscous than that of diesel (nearly twice) was varied from 200 to 240 bar. The IOP was varied directly in the injector by changing the spring stiffness. This was then checked with a standard calibrated nozzle tester. An optimum injection pressure of 230 bar resulted in overall better performance. The injection timing was kept fixed at 27° Before Top Dead Centre (BTDC) and the compression ratio was varied from 15 to 17.5. Masking was provided on the inlet valve with the angle of mask varying from 50°, 70° and 90°, respectively, as shown in Figure 3. In addition to the inlet valve masking, the cylinder heads were also suitably provided with different grooves and bridges to study the effect of swirl augmentation on dual-fuel engine performance (Hountalas, Mavropoulos, and Binder 2008). In cylinder head 1 (CH1), one bridge and two grooves were provided while in differently configured cylinder head 2 (CH2), two bridges and three grooves were provided, as shown in Figure 4. Figure 4 clearly shows the number of bridges and grooves used in both cylinder heads. In order to intensify the air swirl, the number of elliptical grooves was made on the cylinder heads with additional bridges being provided to compensate for the change in compression ratio. Hence the compression ratio of 17.5 was maintained during all the engine-operating conditions in order to study the effect of swirl on the performance of the dual-fuel engine.

Figure 2 Overall view of the engine test rig with a dual-fuel arrangement.

Table 4 Specifications of the CI engine.

Sl no.	Parameters	Specification
2	Type	TV1 (Kirlosker make)
3	Software used	Engine soft
4	Nozzle opening pressure	200–225 bar
5	Governor type	Mechanical centrifugal type
6	No. of cylinders	Single cylinder
7	No. of strokes	Four stroke
8	Fuel	H.S. Diesel
9	Rated power	5.2 kW (7 HP at 1500 rpm)
10	Cylinder diameter (bore)	0.0875 m
11	Stroke length	0.11 m
12	Compression ratio	17.5:1
Air measurement manometer
13	Made	MX 201
14	Type	U-Type
15	Range	100–0–100 mm
Eddy current dynamometer
16	Model	AG – 10
17	Type	Eddy current
18	Maximum	7.5 (kW at 1500–3000 rpm)
19	Flow	Water must flow through a dynamometer during the use
20	Dynamometer arm length	0.180 m
21	Fuel measuring unit – range	0–50 ml

Figure 3 Inlet valve masking with the angle of mask varying from 50°, 70° and 90°.

Figure 4 Modified cylinder heads with different numbers of grooves and bridges.

Tables 5 and 6, respectively, present the specifications of the exhaust gas analyser and the smoke meter used for the emission measurements with measurement accuracies and uncertainties. The emissions of HC and NO_x were measured in ppm, while those of CO and CO₂ were measured in percentage (volume based). Smoke was measured in Hartridge smoke units (HSU). In order to reduce the error in the measurement of emissions, five readings were recorded and only their averages are presented in the graphs. The uncertainty of the measured parameters can be estimated with confidence limits of ± 2σ (95.45% of the measured data lie within the limits of ± 2σ around the mean). The percentage uncertainty of the measured parameters can be estimated using the following relation:

Table 5 Specifications of the exhaust gas analyser.

Type	DELTA 1600S
Object of measurement	Carbon monoxide (CO), carbon dioxide (CO2) and hydrocarbons (HC)
Range of measurement	HC = 0–20,000 ppm as C3H8 (propane)CO = 0–10%CO2 = 0–16%O2 = 0–21%NOx = 0–5000 ppm (as nitric oxide)
Accuracy	HC = ± 30 ppm HCCO = ± 0.2% COCO2 = ± 1% CO2O2 = ± 0.2% O2NOx = ± 10 ppm NO
Resolution	HC = 1 ppmCO = 0.01 vol.%CO2 = 0.1 vol.%O2 = 0.01 vol.%NOx = 1 ppm
Warm-up time	10 min (self-controlled) at 20°C
Speed of the response time	Within 15 s for a 90% response
Sampling	Directly sampled from the tail pipe
Power source	100–240 V AC/50 Hz
Weight	800 g
Size	100 mm × 210 mm × 50 mm

Table 6 Specifications of the smoke meter.

Type	Hartridge Smoke Meter-4
Object of measurement	Smoke
Measuring range opacity	0–100%
Accuracy	± 2% relative
Resolution	0.1%
Smoke length	0.43 m
Ambient temperature range	− 5°C to +45°C
Warm-up time	10 min (self-controlled at 20°C)
Speed of the response time	Within 15 s for a 90% response
Sampling	Directly sampled from the tail pipe
Power supply	100–240 V AC/50 Hz10–16 V DC at 15 amps
Size	100 mm × 210 mm × 50 mm

Results and discussion

The engine tests were conducted with a dual fuel using the CNG+HOME and CNG+HOME+ethanol blends at both 80% and 100% load conditions.

Dual-fuel mode of operation on the HOME+CNG blend with a variable compression ratio

Effect of the compression ratio

The experiments conducted enabled to analyse the effect of the compression ratio on the performance of the CNG+HOME dual-fuel engine. The engine was operated at a constant injection timing of 27° BTDC with a CNG flow rate of 0.5 kg/h. The nozzle opening pressure of the injector was maintained at 230 bar for the injected liquid fuel blends.

Figure 5 shows the cylinder gas pressure plotted for three compression ratios of 15, 16 and 17.5. The cylinder pressure showed a slight decrease in the peak pressure value with the reduction in compression ratio. This low cylinder pressure was due to the low compression ratio. The effect of the compression ratio on the rate of heat release was much mild. The engine operation at a higher compression ratio beyond 17.5 resulted in an unstable engine operation.

Figure 5 In-cylinder pressure versus crank angle for compression ratios at the 100% load condition.

Figure 6 shows the rate of heat release with different compression ratios. Similar to the behaviour of cylinder pressure versus compression ratio, the rate of heat release decreased as the compression ratio decreased.

Figure 6 Rate of heat release versus crank angle for compression ratios at the 100% load condition.

Effect of the compression ratio on brake thermal efficiency (BTE)

Figure 7 shows variation in BTE versus variation in compression ratio for the HOME+CNG fuel combinations, respectively. As the compression ratio increased, the BTE also increased. This is due to the increased pressure and temperature within the combustion chamber and facilitated complete combustion.

Figure 7 Brake thermal efficiency versus compression ratio at the 80% load condition.

Effect of swirl augmentation techniques

The effect of swirl augmentation techniques in the cylinder CH1 on the performance of the CNG+HOME dual-fuel engine is examined in this section. The engine was operated at a constant injection timing of 27° BTDC with a CNG flow rate of 0.5 kg/h. The nozzle opening pressure of the injector was maintained at 230 bar for the injected liquid fuel blends. The angle of mask was varied from 50° to 90° in steps of 20^o angle of mask.

Effect of inlet valve masking: results obtained with modified CH1 (one bridge and two grooves)

Brake thermal efficiency

Figure 8 shows variation in BTE versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the BTE increased considerably with the dual-fuel combinations until the mask angle of 90°. The reason for this increase in BTE is due to the better mixing of the fuel combinations inside the engine cylinder. However, there was a decrease in BTE at the 120° mask angle due to a drop in volumetric efficiency and incomplete combustion.

Figure 8 Effect of the angle of mask on brake thermal efficiency.

NO_x emission

Figure 9 shows variation in NO_x emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As shown in this figure, the angle of mask increased as the NO_x emission decreased considerably with the dual-fuel combinations. Increasing the initial swirl reduced the in-cylinder temperature, which in turn reduced thermal NO_x emission. However, there was a slight increase in NO_x emission being observed at the 120° mask angle due to a drop in volumetric efficiency.

Figure 9 Effect of the angle of mask on NO_x emission.

CO emission

Figure 10 shows variation in CO emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the CO emission decreased considerably with the dual-fuel combinations until the 90° mask angle. From this figure, it follows that the reduction in CO emissions is due to the enhanced mixing rate and turbulence in the combustion chamber. However, with 120° angle of masking, there was a slight increase in HC emissions observed and may be due to decreased volumetric efficiency which in turn resulted in incomplete combustion.

Figure 10 Effect of the angle of mask on CO emission.

Smoke emission

Figure 11 shows variation in smoke emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the smoke emission decreased considerably with the dual-fuel combinations until the 90° mask angle. As shown in the figure, swirl improved fuel combustion by reducing the smoke emissions considerably. However, at the 120° mask angle, there was a slight increase in smoke emissions being observed because as the angle of mask increased, the volumetric efficiency decreased, which in turn resulted in incomplete combustion.

Figure 11 Effect of the angle of mask on smoke emission.

HC emission

Figure 12 shows variation in HC emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the HC emission decreased considerably with the dual-fuel combinations considered. From this figure, it follows that the reduction in smoke opacity with earlier combustion initiation leads to a higher rate of heat release and is affected by swirl augmentation. However, at the 120° mask angle, there was a slight increase in HC emissions being observed because as the angle of mask increased, the volumetric efficiency decreased which in turn resulted in incomplete combustion.

Figure 12 Effect of the angle of mask on HC emission.

Effect of inlet valve masking: results obtained with modified CH2 (two bridges and three grooves)

Brake thermal efficiency

Figure 13 shows variation in BTE versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the BTE increased considerably with the dual-fuel combinations. The reason for this increase in BTE is due to the better mixing of the fuel combinations inside the engine cylinder. However, there was a decrease in BTE at the 120° mask angle due to a drop in volumetric efficiency.

Figure 13 Effect of the angle of mask on brake thermal efficiency.

Smoke emission

Figure 14 shows variation in smoke emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the smoke emissions increased considerably with the dual-fuel combination. As shown in this figure, swirl improved fuel combustion by reducing the smoke emissions considerably. The smoke emission reduced with swirl intensity. However, at the 120° mask angle, there was a slight increase in smoke emissions being observed because as the angle of mask increased, the volumetric efficiency decreased which in turn resulted in incomplete combustion.

Figure 14 Effect of the angle of mask on smoke emission.

CO emission

Figure 15 shows variation in CO emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. As the angle of mask increased, the CO emission decreased considerably with the dual-fuel combinations. From this figure, it follows that the reduction in CO emissions is due to the enhanced mixing rate and turbulence in the combustion chamber. However, at the 120° mask angle, there was a slight increase in CO emissions being observed because as the angle of mask increased, the volumetric efficiency decreased which in turn resulted in incomplete combustion.

Figure 15 Effect of the angle of mask on CO emission.

HC emission

Figure 16 shows variation in HC emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. HC emissions decreased considerably as the angle of mask increased with the dual-fuel combinations. As shown in this figure, the reduction in HC emissions due to earlier combustion initiation that leads to a higher rate of heat release is affected by swirl augmentation. However, at the 120° mask angle, there was a slight increase in smoke emissions being observed because as the angle of mask increased, the volumetric efficiency decreased which in turn resulted in incomplete combustion.

Figure 16 Effect of the angle of mask on HC emission.

NO_x emission

Figure 17 shows variation in NO_x emissions versus angle of mask for the HOME+CNG fuel combinations, respectively. NO_x emissions decreased as the angle of mask increased considerably with the dual-fuel combinations. Increasing the initial swirl ratio reduces the in-cylinder temperature which in turn reduces thermal NO_x emission. However, there was a slight increase in NO_x emission being observed at the 120° angle of mask due to a drop in volumetric efficiency.

Figure 17 Effect of the angle of mask on NO_x emission.

Comparison made between the modified cylinder head configurations CH1 and CH2

Brake thermal efficiency

Figure 18 shows variation in BTE versus two modified cylinder heads for the proposed dual-fuel engine operation. As shown in this figure, the increase in the number of bridges and grooves increased the BTE at the 80% load condition. The reason for this increase in BTE is due to the better mixing of the fuel combinations inside the engine cylinder. Moreover, as shown in this figure, BTE was significantly affected by various methods of swirl augmentation. This may also be due to a higher rate of heat release during the early stage of combustion. BTE increased from the CH2 configuration towards the CH1 configuration. This might be due to the enhanced mixing rate in the case of CH2 induced by turbulence in the combustion chamber.

Figure 18 Variation in brake thermal efficiency versus modified cylinder head configurations.

CO emission

Figure 19 shows variation in CO emission versus two modified cylinder heads for the proposed dual-fuel engine operation. From this figure, it follows that the reduction in CO emissions is due to the enhanced mixing rate and turbulence in the combustion chamber.

Figure 19 Variation in CO emissions versus modified cylinder head configurations.

NO_x emission

Figure 20 shows variation in NO_x emission versus two modified cylinder heads for the proposed dual-fuel engine operation. From this figure, it can be observed that the effect of swirl decreased NO_x emission when compared with CH1 with a three-groove 90° mask cylinder head. Generally, increasing the initial swirl ratio reduces the in-cylinder temperature which in turn reduces thermal NO_x emission.

Figure 20 Variation in NO_x emission versus modified cylinder head configurations.

HC emission

Figure 21 shows variation in HC emission versus two modified cylinder heads for the proposed dual-fuel engine operation. As shown in this figure, swirl improved combustion efficiency by reducing the HC emission considerably. The further decrease in HC emission with the CH2 configuration when compared with the CH1 with earlier combustion initiation leads to a higher rate of heat release and is affected by swirl augmentation.

Figure 21 Variation in HC emission versus modified cylinder head configurations.

Smoke opacity

Figure 22 shows variation in smoke opacity versus two modified cylinder heads for the proposed dual-fuel engine operation. As shown in this figure, swirl improved fuel combustion by reducing the smoke emissions considerably. The smoke emission reduced with swirl intensity.

Figure 22 Variation in smoke opacity with modified cylinder head configurations.

When compared with all the above results obtained, modified CH2 with the 90^o angle mask is best suited when compared with modified CH1 and with the other angles of mask (50° and 70°).

Effects of ethanol addition

Brake thermal efficiency

Figure 23 shows the variation in BTE at the 80% and 100% load conditions. The addition of ethanol to HOME improved the efficiency of a dual-fuel engine. This effect arises from the fact that ethanol addition lowers the viscosity of the blend and thus increases the volatility of the mixture. BTE for the CNG+HOME, CNG+HOME+7.7% ethanol and CNG+HOME+15% ethanol blends were found to be 24.3%, 25.8%, 26.75% and 26.25%, respectively, at the 80% load condition and 25.8%, 26.25%, 28% and 27.7%, respectively, at the 100% load condition. However, the addition of ethanol to HOME increased the NO_x emission. This negative effect was alleviated by the induction of 10% exhaust gas recirculation (EGR) (Figure 20). EGR induction beyond 10% would dilute the charge and hence lower the BTE.

Figure 23 Variation in BTE versus different CNG+HOME+ethanol blends.

HC emission

Figure 24 shows the variation in HC during the 80% and 100% load conditions. HC emissions in diesel engines are caused by lean mixture during the delay period and under mixing of the fuel that exits in the fuel injector nozzle at a lower velocity. The addition of ethanol to HOME reduced the HC emission. Also, the addition of ethanol to HOME improved overall combustion characteristics in terms of increased premixed combustion resulting in improved combustion. This is because with CNG being common, the addition of ethanol to HOME lowers the viscosity of the blend as well as increases its volatility. HC emissions for the CNG+HOME, CNG+HOME+7.7% ethanol and CNG+HOME+15% ethanol blends were found to be 74, 55, 52 and 57 ppm, respectively, at the 80% load condition and 106, 64, 61 and 64, respectively, at the 100% load condition.

Figure 24 Variation in HC emissions versus different CNG+HOME+ethanol blends.

Smoke opacity

Figure 25 shows the variation in smoke opacity at the 80% and 100% load conditions. The addition of ethanol to HOME reduced the smoke emissions. This is because the presence of the bonded oxygen reduces the probability of the formation of soot nuclei in locally rich zones (Banapurmath and Tewari 2010; Lei, Bi, and Shen 2011; Xingcai et al. 2004). The enrichment of the oxygen content in the fuel due to the addition of oxygenates by both HOME and ethanol resulted in a more complete combustion. The main reason for this reduced smoke emission is due to the improved spray characteristics of the injected liquid fuels. Also, with CNG being common, the addition of ethanol to HOME lowers the viscosity of the blend as well as increases its volatility. Smoke opacity values for the CNG+HOME, CNG+HOME+7.7% ethanol and CNG+HOME+15% ethanol blends were found to be 60, 22, 20 and 24 HSU, respectively, at the 80% load condition and 65, 25, 24 and 26 HSU, respectively, at the 100% load condition.

Figure 25 Variation in smoke opacity versus different CNG+HOME+ethanol blends.

CO emission

Figure 26 shows the variation in CO at the 80% and 100% load conditions. CO is a toxic by-product and is a clear indication of incomplete combustion of the premixed mixture. CO emission is mainly associated with incomplete combustion prevailing inside the combustion chamber. The addition of ethanol to HOME improved combustion and hence reduced the CO emissions of a dual-fuel engine (Banapurmath and Tewari 2010; Lei, Bi, and Shen 2011; Xingcai et al. 2004). This is because with CNG being common, the addition of ethanol to HOME lowers the viscosity of the blend as well as increases its volatility. CO emissions for the CNG+HOME, CNG+HOME+7.7% ethanol and CNG+HOME+15% ethanol blends were found to be 0.18%, 0.105%, 0.098% and 0.11%, respectively, at the 80% load condition and 0.18%, 0.126%, 0.112% and 0.124%, respectively, at the 100% load condition.

Figure 26 Variation in CO emissions versus different CNG+HOME+ethanol blends.

Figure 27 shows the variation in NO_x at the 80% and 100% load conditions. The addition of ethanol to HOME reduced NO_x emissions of a dual-fuel engine. The NO_x emission behaviour for all the fuels was found to be similar at a lower load condition with a slight increase in its magnitude at a higher load condition. This is because at a lower load condition, the adiabatic flame temperature of stoichiometric air/ethanol flame is slightly lower. However, the low cetane number of ethanol led to an increase in the ignition delay and a greater increase in the rates of pressure, resulting in higher peak pressure and combustion temperatures. The exhaust gas temperature increased with an increasing ethanol ratio in the fuel mixture. This high peak temperature increased the NO_x emissions. Hence as the ethanol concentration increased, the NO_x emission also increased. Also the oxygenates of ethanol–ester fuel combinations resulted in higher NO_x emissions due to a more complete combustion than the ester alone (Banapurmath and Tewari 2010; Lei, Bi, and Shen 2011; Xingcai et al. 2004). This is because with CNG being common, the addition of ethanol to HOME lowers the viscosity of the blend as well as increases its volatility. NO_x emissions for the CNG+HOME, CNG+HOME+7.7% ethanol and CNG+HOME+15% ethanol blends were found to be 1160, 1000, 1025 and 940 ppm, respectively, at the 80% load condition and 1200, 1050, 1075 and 1000 ppm, respectively, at the 100% load condition. However, the addition of ethanol to HOME unfortunately increased the NO_x emission. This negative effect was alleviated by the induction of 10% EGR.

Figure 27 Variation in NO_x emissions versus different CNG+HOME+ethanol blends.

Conclusions

The combustion of HOME+15% ethanol blended with CNG in the dual-fuel engine operated with optimised parameters at an injection timing of 27° BTDC and a compression ratio of 17.5 resulted in reduced pollutant emissions and improved brake thermal efficiency. Based on the experimental findings presented here, the following detailed conclusions were drawn with reference to three parameters investigated in the current study:

• Compression ratio. The experimental results showed that with the increasing compression ratio from 15 to 17.5, the BTE increased at both 80% and 100% load conditions. The smoke opacity, HC and CO emissions decreased significantly, while the NO_x emissions increased slightly. The optimal compression ratio was found to be 17.5.

• Swirl augmentation techniques. The effect of swirl provided with a varied angle of masking showed that the BTE increased by 1% for CH1 and by 2% for CH2. Emissions such as smoke, HC, CO and NO_x reduced considerably. The results obtained with an angle of masking beyond 90° are not presented here as the results were not encouraging. It was demonstrated that the BTE increased from the CH1 towards CH2 cylinder head configuration, i.e. with the increased number of grooves and bridges. Emissions such as smoke, HC, CO and NO_x reduced considerably. Consequently, swirl augmentation was found to be the main factor that could ensure complete combustion in biofuel-blended dual-fuel engines, including problematic NO_x emissions.

• Ethanol addition. The experimental findings showed that the addition of ethanol (in HOME–ethanol blends) increased the BTE. However, at more than 15% of ethanol in HOME, NO_x emission increased dramatically. To mitigate this NO_x increase, the exhaust gas recirculation (10%) was proposed and experimentally proved to be the technique for controlling the NO_x level. Furthermore, it was found that for HOME–ethanol blends enriched with ethanol (>15%), it was difficult to achieve a homogeneous mixture and two distinct layers were formed that adversely affected biofuel combustion.