Effect of Fuel Injection Timing on the Emissions of a Direct-Injection (DI) Diesel Engine Fueled with Canola Oil Methyl Ester−Diesel Fuel Blends

2675r 2010 American Chemical Society pubs.acs.org/EF

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451nPublished on Web 03/11/2010

Effect of Fuel Injection Timing on the Emissions of a Direct-Injection (DI)Diesel Engine

Fueled with Canola Oil Methyl Ester-Diesel Fuel Blends

Cenk Sayin,† Metin Gumus,† and Mustafa Canakci*,‡,§

†Department of of Automotive Engineering Technology, Marmara University, Istanbul 34722, Turkey,‡Department of Automotive Engineering Technology, Kocaeli University, Izmit 41380, Turkey, and

§Alternative Fuels Research and Development Center, Kocaeli University, Izmit 41275, Turkey

Received December 1, 2009. Revised Manuscript Received March 1, 2010

Biodiesel is the name of a clean burning monoalkyl-ester-based oxygenated fuel made from natural,renewable sources, such as new/used vegetable oils and animal fats. The injection timing plays an importantrole in determining engine performance, especially pollutant emissions. In this study, the effects of fuelinjection timing on the exhaust emission characteristics of a single-cylinder, direct-injection diesel enginewere investigated when it was fueled with canola oil methyl ester-diesel fuel blends. The results showed thatthe brake-specific fuel consumption and carbon dioxide and nitrogen oxide emissions increased and smokeopacity, hydrocarbon, and carbon monoxide emissions decreased because of the fuel properties andcombustion characteristics of canola oil methyl ester. The effect of injection timing on the exhaust emissionsof the engine exhibited the similar trends for diesel fuel and canola oil methyl ester-diesel blends. When theresults are compared to those of original (ORG) injection timing, at the retarded injection timings, theemissions of nitrogen oxide and carbon dioxide increased and the smoke opacity and the emissions ofhydrocarbon and carbon monoxide decreased for all test conditions. On the other hand, with the advancedinjection timings, the smokeopacity and the emissions of hydrocarbon and carbonmonoxidediminished andthe emissions of nitrogen oxide and carbon dioxide boosted for all test conditions. In terms of brake-specificfuel consumption, the best results were obtained from ORG injection timing in all fuel blends.

Introduction

Increasing air pollution is one of the most importantproblems of countries. Exhaust emissions from diesel engineshave amain role on this pollution.TheEuropeanCommissionpublished some directives (2005/55/EC for Euro 4/5, etc.) toreduce exhaust emissions from light- and heavy-duty dieselengines. Hence, the vehicle manufacturers and researchershave been directed to produce diesel engines with high per-formance and low emissions. The vehicle manufacturers canmeet the diesel engine emissions within the accepted levelusing the highest injection pressure, multipoint injection,different catalyst types (oxidation catalyst, nitrogen oxideabsorber, etc.), exhaust gas recirculation, particulate traps,and controlling the start of injection timing. Nevertheless, allexhaust emissions from diesel engines are difficult to reducesimultaneously. One approach to solve this problem is to useoxygenated fuels. Therefore, the development in alternativefuel sources is important for diesel engine applications.1,2

Numerous vegetable oil esters have been tried as an alter-native to diesel fuel. Biodiesel is made from renewable biolo-gical sources, such as vegetable oils and animal fats. Biodiesel,being renewable and widely available from a variety of

sources, is sustainable as a result of applicability to dieselengines without anymodification.3,4 A lot of researchers havereported that using biodiesel as a fuel in diesel engines causes adiminution in harmful exhaust emissions as well as equivalentengine performance with diesel fuel.5-11 Several studies havefound that biodiesel seems to emit far less of the mostregulated pollutants than standard diesel fuel. Decreasing

*To whom correspondence should be addressed. Telephone: þ90-262-3032285. Fax: þ90-262-3032203. E-mail: [email protected].(1) Ozsezen, A. N.; Canakci, M.; Sayin, C. Effects of biodiesel from

used frying palm oil on the exhaust emissions of an indirect injection(IDI) diesel engine. Energy Fuels 2008, 22 (4), 2796–2804.(2) Ozsezen, A. N. Investigation of the effects of biodiesel produced

from waste palm oil on the engine performance and emission character-istics. Ph.D. Dissertation, Kocaeli University, Izmit, Turkey, 2007;pp 4-7 (in Turkish).

(3) Puhan, S.; Vedaraman, N.; Ram, V. B.; Sankarnarayanan, G.;Jeychandran, K. Mahua oil (Madhuca indica seed oil) methyl ester asbiodiesel preparation and emission characteristics. Biomass Bioenergy2005, 28 (1), 87–93.

(4) Ramadhas, A. S.; Jayaraj, S.; Muraleedharan, C. Use of vegetableoils as IC engine fuels;A review. Renewable Energy 2004, 29 (5), 727–742.

(5) Graboski, M. S.; McCormick, R. L. Combustion of fat andvegetable oil derived fuels in diesel engines. Prog. Energy Combust.Sci. 1998, 24 (2), 125–164.

(6) Altun, S.; Bulut, H.; Oner, C. The comparison of engine perfor-mance and exhaust emission characteristics of sesame oil-diesel fuelmixture with diesel fuel in a direct injection diesel engine. RenewableEnergy 2008, 33 (8), 1791–1795.

(7) Ramadhas, A. S.; Muraleedharan, C.; Jayaraj, S. Performanceand emission evaluation of a diesel engine fueled with methyl esters ofrubber seed oil. Renewable Energy 2005, 30 (12), 1789–1800.

(8) Canakci,M.; Ozsezen, A.N.; Turkcan,A. Combustion analysis ofpreheated crude sunflower oil in an IDI diesel engine.Biomass Bioenergy2009, 33 (5), 760–767.

(9) Lee, S. W.; Herage, T.; Young, B. Emission reduction potentialfrom the combustion of soy methyl ester fuel blended with petroleumdistillate fuel. Fuel 2004, 83 (11-12), 1607–1613.

(10) Alptekin, E.; Canakci, M. Determination of the density and theviscosities of biodiesel-diesel fuel blends. Renewable Energy 2008, 33(12), 2623–2630.

(11) Huzayyin, A. S.; Bawady, A. H.; Rady, M. A.; Dawood, A.Experimental evaluation of diesel engine performance and emissionusing blends of jojoba oil and diesel fuel.EnergyConvers.Manage. 2004,45 (13-14), 2093–2112.

2676

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

carbon dioxide (CO2) using biodiesel contributes to a reducegreenhouse effect. In another sense, diminishing carbonmon-oxide (CO), hydrocarbons (HCs), and smoke opacity im-proves air quality.12-17

In efforts to achieve the reduction of engine emissions andfuel consumption while keeping other engine performances atan acceptable level, the fuel injection plays an important role.The most important injection characteristics are injectiontiming, injection duration, and injection pressure. Fuel injec-tion timing affects the combustion and exhaust emissions. Thestate of air into which the fuel is injected changes as theinjection timing is varied, and thus, ignition delay varies. If theinjection starts earlier, the initial air temperature and pres-sure in the cylinder are low; therefore, the ignition delay willincrease. If the injection starts later, which means the piston iscloser to top dead center (TDC), the temperature and pressurewill be high and a decrease in ignition delay will occur. Thus,variation in the start of injection timing has effects on theengine performance and exhaust emissions, especially on thebrake-specific fuel consumption (BSFC), brake thermal effi-ciency (BTE), and nitrogen oxide (NOx) emissions, because ofthe changing of themaximumpressure and temperature in theengine cylinder.18,19

Several studies have shown that the injection timing affectsthe level of exhaust emissions of diesel engines when usingbiodiesel. Bari et al.20 examined the influence of advancedinjection timing on the CO and NOx emissions of a direct-injection (DI) diesel engine using waste cooking oil (WCO)and diesel fuel. The original (ORG) start of injection timing ofthe test engine was 15� crank angle (CA) before top deadcenter (BTDC). The tests were carried out at the engine speedof 3600 revolutions per minute (rpm) and repeated for theinjection timings of 16.3 and 19� CA BTDC. With theinjection timing advanced by 4�, the CO emission reducedby 9.9% forWCO and 44.9% for diesel but theNOx emissionwas increased by 77.6% for WCO and 91.4% for diesel.

Aktas and Sekmen21 investigated the effect of advancedinjection timing on theCO,NOx, andHCemissions of a dieselengine fueled with biodiesel. The ORG injection timing of theenginewas 24.9�CABTDC.The testswere conducted at three

different injection timings (24.9, 26.6, and 28.5� CA BTDC)for 20 N m constant load and six different engine speeds(1400-3400 rpm) with 400 rpm intervals for each injectiontiming. When the injection timing was advanced to 26.6� CABTDC, CO and HC emissions decreased, while NOx emis-sions increased by 4-11%, when the engine was run withbiodiesel.

Nwafor et al.22 researched the effects of advanced injectiontiming on the HC emissions of a diesel engine using rapeseedoil. The ORG injection timing of the engine was 30� CABTDC. When the injection timing was advanced by 5� CABTDC (35� CA BTDC), HC emissions decreased by 12.2%.

From the literature review, it seems that the effects ofretarded and advanced injection timings on the exhaust emis-sionsof a diesel enginehavenot been clearly studiedwhenusingcanola oil methyl ester (COME) blended diesel fuel. Therefore,in the present study, the effects of both injection timing andCOME-blended diesel fuel on the exhaust emissions of a DIdiesel engine were experimentally investigated.

Experimental Section

The experiments were performed on a Lombardini 6 LD 400,single-cylinder, naturally aspirated, air-cooled DI diesel engine.The basic specifications of the engine are shown in Table 1. Aschematic layout of the experimental setup is depicted in Figure 1.A Cussons-P8160-type single-cylinder test bed, which is equippedwith an instrument cabinet (column mounted), fitted the torquegauge, electrical tachometer, and switches for load remote control,measurement instruments was used in the experiments. The dy-namometer is a DC machine rated at 380 V and 10 kW. Aninductive pickup speed sensor was also used tomeasure the enginespeed. Air consumption was measured using a sharp-edged orificeplate [ISO 5167(1980)] and inclined manometer (error (3%).Different digital thermocouples (error (1%) monitored the tem-peratures of intake air, engine oil, coolant inlet and outlet, andexhaust. Fuel consumption was determined using a calibratedburet with an accuracy of 0.1% and a stopwatch with an accuracyof 0.5%. A Bilsa 2100 model exhaust gas analyzer was used tomeasure the concentrationof emissions, suchasHC,CO,andCO2.ASun1500-type smokemeterwas employed tomeasure the smokeopacity of the exhaust gas emitted from the diesel engine. AKane-May Quintox exhaust gas analyzer was used to measure the NOx

emissions. These analyzers were calibrated before the experiments.Table 2 shows the accuracies of the measurements and theuncertainties in the calculated results.

The ordinary diesel fuel was obtained from the T€upras- Petro-leumCorporation.COMEwas purchased fromEKOBiodiesel, acommercial supplier. Some properties of the both fuels are shownin Table 3. The nomenclature BX represents a blend includingX%COME; i.e., B5 indicates a blend including 5%COME, andB100 represent pure COME.

Table 1. Technical Specifications of the Test Engine44

engine type Lombardini6 LD 400

cylinder number 1bore (mm) 86stroke (mm) 68total cylinder volume (cm3) 395injector opening pressure (MPa) 20number of nozzle holes 4start of injection timing (deg CA BTDC) 20compression ratio 18:1maximum torque (N m at 2200 rpm) 21maximum power (kW at 3600 rpm) 8

(12) Zhang, Y.; Van Gerpen, J. H. Combustion analysis of esters ofsoybean oil in a diesel engine. SAE Tech. Pap. 960765, 1996.(13) Raheman, H.; Phadatare, A. G. Diesel engine emissions and

performance from blends of karanja methyl ester and diesel. Fuel 2004,27 (4), 393–397.(14) Canakci, M. Combustion characteristics of a turbocharged DI

compression ignition engine fueled with petroleum diesel fuels andbiodiesel. Bioresour. Technol. 2007, 98 (6), 1167–1175.(15) Canakci, M. The potential of restaurant waste lipids as biodiesel

feedstocks. Bioresour. Technol. 2007, 98 (1), 183–190.(16) Ozsezen, A. N.; Canakci, M.; Turkcan, A.; Sayin, C. Perfor-

mance and combustion characteristics of a DI engine fueled with wastepalme oil and canola oil methyl esters. Fuel 2009, 88 (4), 629–636.(17) Baldassarri, L. T.; Battistelli, C. L.; Conti, L.; Crebelli, R.;

Berardis, B. D.; Iamiceli, A. L. Emission comparison of urban busengine fueledwith diesel oil and biodiesel blend.Sci. Total Environ. 2004,327 (1-3), 147–162.(18) Sayin, C.; Uslu, K. Influence of advanced injection timing on the

performance and emissions of CI engine fueled with ethanol-blendeddiesel fuel. Int. J. Energy Res. 2008, 32 (11), 1006–1015.(19) Sayin, C.; Uslu, K.; Canakci, M. Influence of injection timing on

the exhaust emissions of a dual-fuel CI engine. Renewable Energy 2008,33 (6), 1314–1323.(20) Bari, S.; Yu, C.W.; Lim, T. H. Effect of fuel injection timingwith

waste cooking oil as a fuel in a direct injection diesel engine. J. Automob.Eng. 2003, 218 (1), 160–172.(21) Aktas, A.; Sekmen, Y. The effects of advance fuel injection on

engine performance and exhaust emissions of a diesel engine fuelledwithbiodiesel. J. Fac. Eng. Archit. Gazi Univ. 2008, 23 (1), 199–207 (inTurkish).

(22) Nwafor, O. M. I.; Rice, G.; Ogbonna, A. I. Effect of advancedinjection timing on the performance of rapeseed oil in diesel engines.Renewable Energy 2000, 21 (3-4), 433–444.

2677

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

Dynamic fuel injection timing is adjusted on the basis of theopening time of the needle. Fuel properties, such as bulk modulusanddensity, affect thedynamic fuel injection timing. Static injectiontiming, in other words fuel delivery advance timing, is the timingwhen fuel starts to pressurize in the injection pump. In this study,static injection timing was taken into account. The ORG injectiontiming of the test engine is 20� CA BTDC. Injection timingadjustment is accomplished by adding or removing the shims inthe injection pump. The addition or removal of 0.25 mm of shimchanges the injection timingapproximately 5�CA.The experimentswere carried out at three different injection timings (15, 20, and 25�CA BTDC). For each injection timing, the engine was run atconstant engine speed (2200 rpm) and four different loads (5, 10,15, and 20 Nm). The values of the engine oil temperature, exhaustgas temperature, and emissions, such as smoke opacity, CO, HC,NOx, and CO2, were recorded during the experiments. Each testwas repeated 3 times. The values given in this study are the averageof these three results. All data were collected after the enginestabilized. The engine was sufficiently warmed up at each test,and the engine oil temperature was maintained around 85-90 �C.During the experimental study, all fuels had no trace of knock atdifferent injection timing and it was detected that the engine noisewas qualitatively less than that of petroleum-based diesel fuel(PBDF) when the engine was running with B100.

Results and Discussion

COEmissions.TheCOemissions in theexhaust represent lostchemical energy that is not fully used in the engine.Generally, theCO emission is affected by the equivalence ratio, fuel type,

combustion chamber design, atomization rate, start of injectiontiming, engine load, and speed. Themost important among theseparameters is the equivalence ratio.1,23 The variation of COwithengine load for the fuels is presented in Figure 2. While the fuelsproduceda lowamount ofCOemissionat high-load levels, thoseresulted in more emissions at light loading conditions. The COemissionswere found todecreasewith the increasing load.This isa typical result for internal combustion engines because thecombustion temperature increaseswith the engine load, as shownin Figure 8. Therefore, CO emissions start to decrease.7 Theresults obtained in this study confirmed this statement. At ORGinjection timing, while the CO emission was measured to be0.22% with B20 at 20 N m load, it was 0.43% at 5 N m.

Figure 2 shows that the CO emission level decreased with theincreasingCOMEpercentage in the fuel blend.COMEcontainsabout 11%oxygen. This helps for the complete combustion.24,25

For all engine loads, CO emissions for B100, B50, B20, and B5decreased by 40.11, 25.82, 15.18, and 3.73% compared to thoseof B0, respectively. The effect of fuel injection timing on theCOemissions is shown in Figure 3. CO emissions increased withretarded fuel injection timing. Retarding the fuel injectiontiming decreases the amount of fuel burned in the premixedcombustion phase and increases the amount of fuel burned inthe subsequent diffusive combustion phase. The latter phasealways takes place in a rich mixture environment and easilyproduces the incomplete burning product CO.26 Retarding theinjection timing by 5� (from 20� to 15� CA BTDC) caused theCO emission, which increased by 31.25% for B50 at 20 N m

Figure 1. Experimental layout.

Table 2. Accuracies of theMeasurements and the Uncertainties in the

Calculated Results

measurements accuracy

load (N m) (2speed (rpm) (10time (%) (0.5temperatures (�C) (1

calculated results uncertainty

power (%) (2.55BSFC (%) (2.60

(23) Heywood, J. B. Internal Combustion Engines; McGraw-Hill:New York, 1988, pp 491-499, 297, 571.

(24) Gumus, M. Evaluation of hazelnut kernel oil of Turkish origin asalternative fuel in diesel engines.Renewable Energy 2008, 33 (11), 2448–2457.

(25) Huang, Z. H.; Lu, H. B.; Jiang, D. M.; Zeng, K.; Liu, B.; Zhang,J. Q.; Wang., X. B. Combustion behaviors of a compression ignitionengine fuelled with diesel/methanol blends under various fuel deliveryadvance angles. Bioresour. Technol. 2004, 95 (3), 331–341.

(26) Ma, Z.; Huang, Z. H.; Li, C.; Wang, X. B.; Miao, H. Effects offuel injection timing on combustion and emission characteristics of adiesel engine fueled with diesel-propane blends. Energy Fuels 2007,21 (3), 1504–1510.

2678

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

load. However, the CO emission decreased with the advancedinjection timing. The advanced injection timing produces ahigher cylinder temperature, as seen in Figure 9, and increasesthe oxidation process between carbon and oxygen molecules.This causes a reduction inCOemission.Advancing the injectiontiming by 5� (from 20� to 25� CA BTDC) caused a lower COemission, which decreased by 18.75% for B50 at 20 N m load.

HC Emissions. HC emissions from DI diesel engines aremainly caused by the fuel injected andmixed beyond the leancombustion limit during the ignition delay and fuel effusingfrom the nozzle sac at low pressure.27 The variation of HCwith engine load for different fuels tested is shown in

Figure 4. Typically, HC emissions are a serious problem atlight loads for diesel engines. At light loads, the fuel is less toimpinge on surfaces, but because of poor fuel distribution,large amounts of excess air, and low cylinder temperature,lean fuel-air mixture regions may survive to escape intoexhaust.28 The results acquired in this study supported thisexplanation. At the ORG injection timing, while the HCemission was measured to be 23 particulates per million(ppm) with B0 at 15 N m load, it was 28 ppm at 5 N m.

As shown in the Figure 4, the HC emission was steadilydecreasedwhen the amount ofCOME increased in the blend.This is also due to the oxygenated nature of COME, wheremore oxygen is available for burning and reducing HCemissions in the exhaust.29 On average, HC emissions forall engine loads for B100, B50, B20, and B5 decreased by65.25, 48.1, 39.82, and 12.47% compared to those of B0,respectively.

Figure 5 shows the effect of fuel injection timing on HCemissions. When combustion was retarded, which happenswith the retarded start of injection, the maximum gastemperatures were lower, as shown in Figure 9. While thevolume increases during the expansion stroke, HC emissionsincreased drastically.30 Retarding the injection timing by

Table 3. Some Properties of the Fuels Used in the Experiments

units EU limits (EN 14214) U.S.A. limits (ASTM D 6751) COME45 diesel46

typical formula C18.08H34.86O2 C14.16H25.21

molecular weight g/mol 284.17 195.50sulfated ash content mass % maximum of 0.02 maximum of 0.02 0.0004 0.0015density kg/m3 (at 15 �C) 860-900 885 840.3carbon/hydrogen ratio 1:1.93 1:1.78flash point �C minimum of 120 minimum of 130 74.1 61.5carbon residue mass % maximum of 0.30 0.0004 0.067heating value kJ/kg 38730 42930cetane number minimum of 51 minimum of 41 60.4 56.5kinematic viscosity mm2/s (at 40 �C) 3.5-5.0 1.9-6.0 4.39 3.18acid value mg of KOH/g maximum of 0.50 maximum of 0.80 0.15oxidation stability 1 h (at 110 �C) minimum of 6.0 10.1distillation

initial boiling point �C 331 164.790% recovered �C maximum of 360 348 351.1

Figure 2. CO emissions versus engine load for the fuels at ORGinjection timing.

Figure 3. CO emissions versus injection timing at 20 N m engineload.

Figure 4. HC emissions versus engine load for the fuels at ORGinjection timing.

(27) Lakshminarayanan, P. A.; Nayak, N.; Dingare, S. V.; Dani,A. D. Predicting hydrocarbon emissions from direct injection dieselengines. J. Eng. Gas Turbines Power 2002, 124 (3), 708–807.

(28) Sayin, C.; Ilhan,M.; Canakci,M.; Gumus,M. Effect of injectiontiming on the exhaust emissions of a diesel engine using diesel-methanolblends. Renewable Energy 2009, 34 (5), 1261–1269.

(29) Agarwal, A. K. Biofuels (alcohols and biodiesel) applications asfuels for internal combustion engines. Prog. Energy Combust. Sci. 2007,33 (3), 233–271.

(30) Payri, F.; Benajes, J.; Arregle, J.; Riesco, J. M. Combustion andexhaust emissions in a heavy-duty diesel engine with increased premixedcombustion phase by means of injection retarding.Oil Gas Sci. Technol.2006, 61 (2), 247–258.

2679

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

5� (from 20� to 15� CA BTDC) caused the emission toaugment by 9.1% for B5 at 20 Nm load. On the other hand,the advanced injection timing caused an earlier start ofcombustion relative to the TDC. Because of this, the cylindercharge, being compressed as the piston moved to the TDC,had relatively higher temperatures and thus lowered the HCemissions. Advancing the injection timing by 5� (from 20� to25� CA BTDC) caused lower HC emissions, which reducedby 36.34% for B5 at 20 N m load.

NOxEmissions.The conversion of nitrogen and oxygen toNOx is generated by the high combustion temperaturesoccurring within the burning fuel sprays and is controlledby local conditions. NOx is a collective term used to refer tonitric oxide (NO) and nitrogen dioxide (NO2). NOx emis-sions form in the high-temperature burned gas region, whichis non-uniform, and the formation rates are highest at theregions close to stoichiometry.31 The variation of NOx withthe engine load for different fuel blends is presented inFigure 6. NOx emissions increasedwith the increasing engineload because of the increasing combustion temperature, asshown in Figure 8. At the ORG injection timing, while theNOx emission was determined to be 133 ppm with B50 at 20N m load, it was 19 ppm at 5 N m.

As seen in Figure 6, theNOx emissions generally increasedwith the increasing fraction of COME in the fuel blend. It isobvious that the increased NOx with the use of COME andits blends is a result of increasing oxidation. The oxygencontent of COME is an important factor in the high NOx

formation levels, because the oxygen content of COMEprovides high local peak temperatures and a correspondingexcess of air.32-34 As indicated in Figure 8, the combustiontemperature increased with the increasing amount of COMEin the fuel blend. For all engine loads, NOx emissions forB100, B50, B20, and B5 increased by 65.25, 48.1, 39.82, and12.47% compared to those of B0, respectively.

Figure 7 illustrates the effect of fuel injection timing onNOx emissions. Retarding the fuel injection timing caused adecrease in the ignition delay and cylinder gas temperature.Consequently, the NOx concentration tended to be less.35

The exhaust gas temperatures obtained in the experimentsare shown in Figure 9, which confirmed this statement.Retarding the injection timing by 5� (from 20� to 15� CABTDC) caused the NOx emission to reduce by 28.11% forB20 at 20 N m load. However, advancing the fuel injectiontiming increased the peak cylinder pressure because of thelonger ignition delay. Higher peak cylinder pressures re-sulted in higher peak temperatures. As a consequence, theNOx concentration started to rise. Advancing the injectiontiming by 5� (from 20� to 25� CA BTDC) caused the NOx

emission to increase by 8.59% for B20 at 20 N m load.Smoke Opacities. Because of the heterogeneous nature of

diesel combustion, there is a wide distribution of fuel/airratios within the cylinder. Smoke emissions are attributed toeither fuel-air mixtures that are too lean to auto-ignite orsupport a propagating flame or fuel-air mixtures that aretoo rich to ignite. The soot formation mainly takes place inthe fuel-rich zone at high temperatures and pressures, espe-cially within the core region of each fuel spray, and is causedby high-temperature decomposition.36 The variation ofsmoke with the engine load for different fuel blends isdepicted in Figure 10. The formation of smoke stronglydepends upon the engine load. When the load is increased,

Figure 5. HC emissions versus injection timing for the fuels at20 N m engine load.

Figure 6. NOx emissions versus engine load for the fuels at ORGinjection timing.

Figure 7. NOx emissions versus injection timing for the fuels at20 N m engine load.

(31) Ilkilic, C. Emission characteristics of a diesel engine fueled withby 25% sunflower oil methyl ester and 75% diesel fuel blend. EnergySources 2009, 31 (6), 480–491.(32) Beatrice, C.; Bertoli, C.; D’Alessio, J.; Del Giacomo, N.;

Lazzaro, M.; Massoli, P. Experimental characterization of combustionbehavior of new diesel fuels for low emission engines. Combust. Sci.Technol. 1996, 120 (1-6), 335–355.(33) Theodoros, C. Z.; Dimitrios, T. H. DI diesel engine performance

and emissions from the oxygen enrichment of fuels with variousaromatic content. Energy Fuels 2004, 18 (3), 659–666.(34) Ren, Y.; Huang, Z. H.; Jiang, D.M; Liu, L. X.; Zeng, K.; Liu, B.;

Wang, X. B. Combustion characteristics of a compression-ignitionengine fuelled with diesel-dimethoxy methane blends under variousfuel injection angles. Appl. Therm. Eng. 2006, 26 (4), 327–337.

(35) Scholl, K.W.; Sorenson, S. C. Combustion of soybean oil methylester in a direct injection diesel engine. SAE Tech. Pap. 930934, 1993.

(36) Yoshiyuki,K.; Changlin,Y.;Kei,M. Effects of fuel properties oncombustion and emission characteristics of a direct injection dieselengine. SAE Tech. Pap. 2000-01-1831, 2000.

2680

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

more fuel is injected into the combustion chamber, and thiscauses an increase in smoke formation.34 The results ob-tained in this study supported this statement. At the ORGinjection timing, while smoke opacity was measured to be94% with B5 at 20 N m load, it was found as 65% at 5 N m.

As seen in Figure 10, smoke opacity has a tendency todecrease with the increasing fraction of COME in the fuelblend. The oxygen content of the COME, which enablesmore complete combustion even in the regions of the com-bustion chamber with fuel-rich diffusion flames, promotesthe oxidation of the already formed soot.37,38 For all engineloads, smoke opacity for B100, B50, B20, and B5 diminishedby 21.26, 12.43, 4.87, and 3.63% compared to those of B0,respectively.

The effect of injection timing on smoke opacity can be seeninFigure 11.Retarding the fuel injection timing increased thesmoke opacity. This is due to increasing the fraction ofdiffusive combustion while retarding the fuel injection tim-ing.39,40 Retarding the injection timing by 5� (from 20� to 15�CABTDC) caused the smoke emission to increase by 3.22%

for B20 at 20 Nm load. Advancing the injection timing leadsto higher temperatures during the expansion stroke andmore time in which oxidation of the soot particles occurs.Advancing the injection timing by 5� (from 20� to 25� CABTDC) caused the smoke emission to decrease by 6.51% forB20 at 20 N m load.

CO2 Emissions. The CO2 emission is produced by com-plete combustion of fuel. Ideally, combustion of a HC fuelshould produce only CO2 and water (H2O). The concentra-tions have an opposite behavior when compared to the COconcentrations and because of the improvement in thecombustion process.41 The variation of CO2 with the engineload for different fuel blends is shown in Figure 12. Asexpected, the CO2 emission increased with the increasingload. The main reason of increasing CO2 with an increasingload is more fuel injected into the engine. The other reasonsare the increasing combustion temperature and oxidizationrates.11 At the ORG injection timing, while CO2 emissionwas measured to be 13.32% with B20 at 20 N m load, it wasfound as 11.61% at 5 N m.

As seen in Figure 12, the CO2 emission increased with theincreasing fraction of COME in the fuel blend. For all engineloads, CO2 emissions increased by 9.25, 4.12, and 1.13% forB100, B50, and B20 and decreased by 1.32% for B5 com-pared to those of B0, respectively. The figure shows thatmostof the carbon converts to CO2 under all engine load condi-tions, except 5 N m. While the test engine was running with

Figure 8. Exhaust temperatures versus engine loads for the fuels atORG injection timing.

Figure 9. Exhaust temperatures versus injection timing for the fuelsat 20 N m engine load.

Figure 10. Smoke opacity versus engine load for the fuels at ORGinjection timing.

Figure 11. Smoke opacity versus injection timing for the fuels at20 N m engine load.

(37) Lapuerta, M.; Armas, O.; Rodriguez-Fernandez, J. Effect ofbiodiesel fuels on diesel engine emissions. Prog. Energy Combust. Sci.2008, 34 (2), 198–223.(38) Huang, Z. H.; Lu, H. B.; Jiang, D. M.; Zeng, K.; Liu, B.; Zhang,

J. Q.;Wang, X. B. Performance and emissions of a compression ignitionengine fueled with diesel/oxygenate blends for various fuel deliveryadvance angles. Energy Fuels 2005, 19 (2), 403–410.(39) Shuai, S.; Abani, N.; Yoshikawa, T.; Reitz, R. D.; Park, S. V.

Evaluation of the effects of injection timing and rate-shape on diesel lowtemperature combustion using advanced CFD modeling. Fuel 2009, 88(7), 1235–1244.(40) Buyukkaya, E.; Cerit, M. Experimental study of NOx emissions

and injection timing of a low heat rejection diesel engine. Int. J. Therm.Sci. 2008, 56 (2), 97–103.

(41) Sayin, C.; Kilicaslan, I.; Canakci, M.; Ozsezen, A. N. An experi-mental study of the effect of octane number higher than engine require-ment on the engine performance and emissions.Appl. Therm. Eng. 2005,25 (8-9), 1315–1324.

2681

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

COME, the air/fuel equivalence ratio was lower than that ofdiesel fuel at all engine loads. Probably, at 5 N m, a higherBSFC amount, viscosity, and density of COME,which led topoor injection characteristics relative to diesel fuel, affectedthe complete combustion reaction of COME. The CO2

emission of the COME and its blends was measured lowerfor 5 N m but higher for 10, 15, and 20 N m than those ofdiesel fuel. The shorter ignition delay and higher boilingpoint of COME increased the combustion duration. Thisbehavior became more obvious at the high engine loads.Additionally, the CO2 formation depends upon the car-bon-hydrogen ratio of the fuel. Stoichiometrically, combus-tion of aHC fuel should produce only CO2 andwater (H2O),as mentioned above. The relative proportion of these twodepends upon the carbon/hydrogen ratio in the fuel. In thisstudy, these ratios are about 1:1.78 for diesel fuel and 1:1.93for COME (see Table 3). As seen in the eqs 1 and 2, if thecomplete combustion reactions are established, diesel fuelreleases 3.18 kg ofCO2/kg of fuel andCOMEreleases 2.79 kgof CO2/kg of fuel. Thus, CO2 emissions from an engine canbe reduced by reducing the carbon content of the fuel per unitenergy.

The complete combustion reaction for COME is

C18:08H34:86O2 þ 25:79ðO2 þ 3:79N2Þ f 18:08CO2

þ 17:43H2Oþ 96:97N2 ð1ÞThe complete combustion reaction for diesel fuel is

C14:16H25:21 þ 20:46ðO2 þ 3:76N2Þ f 14:16CO2

þ 12:60H2Oþ 76:92N2 ð2ÞThe effect of injection timing onCO2 emissions is shown in

Figure 13. Retarding the fuel injection timing caused a lowercombustion temperature, as depicted in Figure 9, and thisreduced the chemical speed. Thus, a decrease in the CO2

emissions occurred.42 Retarding the injection timing by 5�(from 20� to 15� CA BTDC) caused the CO2 emission toreduce by 3.12% for B100 at 20 N m load. CO2 emissionsincreased with advancing the injection timing for all fuelblends. Advancing the injection timing by 5� (from 20� to 25�CA BTDC) caused the CO2 emission to increase by 2.13%for B100 at 20 N m load.

BSFC. BSFC is defined as the ratio of the fuel consump-tion to the brake power. The variation of BSFCwith the loadfor different fuels is presented in Figure 14. For all fuel

blends, BSFC decreased with the increase in the engine load.One possible explanation for this reduction could be the higherpercentage increase in the brake power with load as comparedto the fuel consumption. At the ORG injection timing, whileBSFCwasmeasured to be 346 g kW-1 h-1 with B50 at 20Nmload, it was found to be 474 g kW-1 h-1 at 5 N m.

As seen in the Figure 14, the BSFC was slightly increasedwith the increased COME percentage in the fuel blend. Itshould be noted that the lower heating value (LHV) of COMEis 14% lower than that of diesel fuel. With the increase in theCOMEpercentage in the fuel blends, theLHVof the fuel blenddecreased. The BSFC increased when the COME percentagewas increased in the fuel blend compared to that of diesel fuel.On the other hand, COME is an oxygenated fuel and leads tomore complete combustion; hence, BSFC may reduce. It isclear from the figure that the LHV is more effective than theoxygen content with regard to increasingBSFC.43On average,BSFC for all engine loads for B100, B50, B20, and B5 boostedby 23.75, 18.27, 13.27, and 2.25% compared to those of B0,respectively.

The effect of fuel injection timing on BSFC is depicted inFigure 15. When the injection timing was retarded andadvanced 5� CA BTDC compared to ORG injection timing,BSFC increased by 16.19 and 6.47% for B5, respectively.

Figure 12. CO2 emissions versus engine load for the fuels at ORGinjection timing. Figure 13. CO2 emissions versus injection timing for the fuels at

20 N m engine load.

Figure 14. BSFC versus engine load for the fuels at ORG injectiontiming.

(42) Sayin, C.; Canakci, M. Effects of injection timing on the engineperformance and exhaust emissions of a dual-fuel diesel engine. EnergyConvers. Manage. 2009, 50 (1), 203–213.

(43) Ozsezen, A. N.; Canakci, M.; Sayin, C. Effects of biodiesel fromused frying palm oil on the performance, injection, and combustioncharacteristics of an indirect injection diesel engine. Energy Fuels 2008,22 (2), 1297–1305.

(44) Lombardini. Engine technical specification; Turkey, 2000 (inTurkish).

(45) EKO Biodiesel. Product specification; Turkey, 2009 (in Turkish).(46) T€upras-. Product specification; Turkey, 2009 (in Turkish).

2682

Energy Fuels 2010, 24, 2675–2682 : DOI:10.1021/ef901451n Sayin et al.

MinimumBSFCwas obtained at ORG injection timing for allfuel blends. The fuel consumption atORG injection timingwasthe lowest, and it increased at advanced and retarded injectiontimings. Advancing the injection timing causes the maximumpressure to increase andoccurs beforeTDC in the compressionstroke. This reduces the maximum pressure occurring duringthe expansion stroke and torque output. On the other hand,with retarding the injection timing, the combustion will delayand result in a reduced effective pressure to dowork during theexpansion stroke. Thus, for both cases, fuel consumption perpower output increased.19

Conclusion

In this study, the effects of fuel injection timing on theexhaust emission characteristics of a diesel engine have beeninvestigated when the engine was fueled with COME-dieselblends. From the present paper, the following conclusions aresummarized: (1) COME is an oxygenated fuel and leads tomore complete combustion. It has a high cetane number andlow aromatic content when compared to diesel fuel.When thetest enginewas fueledwithCOMEor its blends, the amount ofsmoke, HC, and CO emissions reduced. However, the com-bustion temperature increased with the increasing amount ofCOME in the fuel blend. This led to a higher quantity of NOx

formation. (2) In terms of injection timing, the test resultsdemonstrated that, with the advanced injection timing, thesmoke opacity and the emissions of CO and HC decreased,while NOx and CO2 emissions increased. When the injectiontiming was advanced, the CO emission decreased because ofthe improved reaction between the fuel and oxygen. Thiscaused an increase in the CO2 emissions. Advancing the

injection timing caused an earlier start of combustion relativeto theTDC.Because of this,HCemissions decreased andNOx

emissions increased. Advancing the injection timing gave thebest results for the smokeopacity and the emissions ofCOandHC for B100. On the other hand, retarding the injectiontiming presented the minimum results of NOx and CO2

emissions for B0 and B100, respectively. (3) Increasing theCOME ratio in the fuel blend led to an increase in BSFC. Thisis probably the result of theLHVof theCOME,which is lowerthan that of the diesel fuel. TheORG injection timing gave thebest results for BSFC compared to the retarded and advancedinjection timings. When the injection timing is advanced, theignition delay will be longer and the speed of the flame will beshorter. These two lead to the reduction in the maximumcylinder pressure and engine power output. Thus, fuel con-sumption per power output will augment. On the other hand,retarded injection timing means later combustion, and there-fore, cylinder pressure increases only when the cylinder vo-lume becomes higher rapidly and results in a reduced effectivepressure to do work.

Acknowledgment. This study was supported by the ScientificResearch Project Commission of Marmara University underGrant BSE-075/131102.

Nomenclature

B0 = 100% dieselB5 = 5% COME plus 95% PBDF (volumetric)B20 = 20% COME plus 80% PBDF (volumetric)B50 = 50% COME plus 50% PBDF (volumetric)B100 = 100% COMEBSFC= brake-specific fuel consumption (g kW-1 h-1)BTDC= before top dead centerBTE = brake thermal efficiencyCA= crank angleCI = compression ignitionCO= carbon monoxideCOME= canola oil methyl esterCO2 = carbon dioxideDI = direct injectionLHV = lower heating value (kJ/kg)NOx = nitrogen oxideORG = originalTDC = top dead centerHC= hydrocarbonWCO= waste cooking oilppm = particulates per millionrpm = revolutions per minute

Figure 15. BSFC versus injection timing for the fuels at 20 N mengine load.