Fuel 117 (2014) 450–457

Contents lists available at ScienceDirect

Fuel

journal homepage: www.elsevier .com/locate / fuel

Effect of the degree of unsaturation of biodiesel fuels on the exhaustemissions of a diesel power generator

0016-2361/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.fuel.2013.09.028

Abbreviations: ASTM, American Society for Testing and Materials; US EPA, United States Environment Protect Agency; GC, gas chromatograph; FID, flame iodetector; LHV, lower heating value (kJ/kg); ppm, parts per million; rpm, revolution per minute; CO, carbon monoxide; HC, hydrocarbons; NOx, nitrogen oxides; BSspecific fuel consumption (g/kW h); BTE, brake thermal efficiency (%); IN, iodine number (g of I2/100 g); ULSD, ultra-low sulphur diesel; POME, palm oil methyl estercottonseed oil methyl ester; WFOME, waste fish oil methyl ester.⇑ Tel.: +90 488 2173675; fax: +90 488 2157201.

E-mail address: saltun72@yahoo.com

S�ehmus Altun ⇑Department of Mechanical Education, Technical Education Faculty, Batman University, 72060 Batman, Turkey

h i g h l i g h t s

� The effect of the degree of unsaturation of biodiesels on diesel emissions was investigated.� Biodiesels resulted in smoke opacity, with an increase in BSFC compared to ULSD.� Saturated biodiesel had highest cetane number and lowest adiabatic flame temperature which was good to reduce NOx emission.� The cetane number and adiabatic flame temperature appear to be the key properties that determined the emissions.

a r t i c l e i n f o

Article history:Received 1 August 2013Received in revised form 6 September 2013Accepted 10 September 2013Available online 25 September 2013

Keywords:BiodieselIodine numberDegree of unsaturationDiesel emissions

a b s t r a c t

In this work, three biodiesel fuels with iodine numbers ranging from 59 to 185 were tested in a direct-injection diesel engine powered generator set at constant speed of 1500 rpm under variable load condi-tions to investigate the effect of the degree of unsaturation of biodiesel fuels, which are quantified by theiodine number, on the performance and exhaust emissions of a diesel engine. The increase in unsatura-tion involved a decrease in cetane number, and therefore, allowed for the maximization of the effect ofthe cetane number, while other properties, such as oxygen content, heating value, and viscosity, variedwithin a small range. Experimental results showed that biodiesel fuels resulted in lower emissions ofnitrogen oxides, carbon monoxide, and smoke opacity, with some increase in emissions of unburnedhydrocarbons. With their low energy content, neat biodiesel fuels resulted in an increase in fuel con-sumption compared to the conventional diesel fuel (ultra-low sulphur diesel). The degree of unsaturationof biodiesel fuels had effects on engine emissions via its effect on the cetane number and adiabatic flametemperature while engine performance was not significantly affected by the type of biodiesel fuel or itsdegree of unsaturation. The biodiesel having lowest iodine number had highest cetane number, and low-est density and adiabatic flame temperature, which was good to reduce NOx emissions, as it agreed withexperimental results. Additionally, more unsaturated biodiesel fuels showed higher NOx emissions,smoke opacity, and lower HC emissions. It can be said that cetane number and adiabatic flame temper-ature are responsible for such results.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

As a consequence of the increasing concern about environmen-tal pollution and more stringent regulations on exhaust emissions,reduction in engine emissions have become a major subject in en-gine studies. In addition, much effort has been made to reducedependence on the petroleum as it is obtained from limited re-

serves. These concerns have led to much research on alternativerenewable fuels in the last decade. Among the proposed alternativefuels, biodiesel has received much attention for using in diesel en-gines in recent years [1,2]. Biodiesel is typically produced bytransesterification reaction of vegetable oils or animal fats withmethanol in the presence of a catalyst to yield glycerin and methylesters, and it can be used directly or blended with fossil diesel fuels

nizationFC, brake; CSOME,

S�. Altun / Fuel 117 (2014) 450–457 451

in diesel engines with no or little modifications. However, somediesel engine manufacturers do not allow using neat biodiesel orits blends instead of fossil diesel fuels. Warranties are often appliedto only biodiesel that fulfills the ASTM D 6751-03 for USA and EN14214 for European Union standards [3,4], and in Europe in con-centrations below 7%, as specified by standard EN-590.

Although the fuel properties of biodiesel can vary with produc-tion technologies and feedstock, it has some fuel advantages overfossil diesel fuel, including renewability, biodegradability, betterlubricity, and high flash point and oxygen content [5]. Lapuertaet al. [6] concluded in their review study that using biodiesel fuelsin diesel engines result in reductions in CO, HC and particulateemissions due to the higher oxygen content of biodiesel comparedto fossil diesel fuel. However, in the case of nitrogen oxides (NOx),despite the highest consensus lies in an increase of NOx emissionswith biodiesel, some contradictory trends have also been reported.Some of the researchers found increases in NOx emissions; someothers did not find any differences between diesel and biodieselfuels, and others found decreases in NOx emissions while using bio-diesel. This is mainly attributed to differences in physicochemicalproperties of biodiesel, its effect on injection timing, ignition delay,adiabatic flame temperature, radiative heat loss, and other com-bustion phenomena, also varies with engine technology and oper-ating conditions [7,8]. However, several researchers indicate thatNOx formation in biodiesel–fueled diesel engines is highly depen-dent on the degree of unsaturation and cetane number of biodiesel[9,10]. In general, higher degrees of saturation correlate with high-er cetane numbers, and saturated biodiesel fuels (i.e. without dou-ble bonds) produce lower amounts of NOx than unsaturated fuels[11]. In a study done by US EPA (United States Environment ProtectAgency) [12], it was confirmed the direct relationship between NOx

emissions and molecular unsaturation. It was observed in thestudy that on average, biodiesel obtained from soybean oil pro-vided a 15% increase in NOx emissions as compared to those withfossil diesel fuel, rapeseed oil based biodiesel provided a 12% in-crease, while biodiesel produced from animal fats led to only a3% increase. The study of Wyatt et al. [13] also showed that the ani-mal fat-based biodiesel fuels, which are highly saturated, had low-er NOx emission levels than did the soybean oil based biodiesel fuel(unsaturation ones). In the same way, Wang et al. [14] found ameasurable reduction in NOx emissions with biodiesel obtainedfrom high oleic soybean oil compared to normal soybean oil basedbiodiesel which contains 25% oleic acid in its fatty acid composi-tion. In another study by Ng et al. [15], NO emission was foundto decrease with increasing palm and coconut oil biodiesels con-tent in the fuel when compared with fossil diesel, opposite to theemissions of the highly unsaturated ones (soybean oil based bio-diesel), where NO level are higher in relation to that of fossil diesel.

Table 1Speciation of tested fuels.

Fatty acid type Carbon chain Palm o

Linoleic C18:2 11.8155Palmitic C16:0 39.7834Oleic C18:1 43.5639Stearic C18:0 3.1628Miristic C14:0 1.3810Palmitoleic C16:1 –Trideconoic C13:0 –Linolenic C18:3 0.2933Eicosatetraenoic C20:4 –Eicosapentaenoic C20:5 –Docosapentaenoic acid C22:5 –Docosahexaenoic C22:6 –Others –Total unsaturated 55

However, the effect of the degree of unsaturation of biodiesel onopacity and particulate emissions is much less known. There isno consensus about whether or not the degree of unsaturation ofbiodiesel fuels affects smoke opacity and/or particulate emissions.Although a few authors found a slight dependency [16], there arealso others who found no effect of the biodiesel feedstock on ex-haust opacity and particulate emissions [17].

In this work, effect of the degree of unsaturation of biodieselfuels, which are quantified by the iodine number, on the character-istics of the engine was investigated experimentally in a diesel en-gine powered generator set. Biodiesel fuels were produced fromdifferent feedstock such as palm oil, cottonseed oil and waste an-chovy fish oil via transesterification process, and consequentlytheir iodine values were different (from 59 to 185). The increasein unsaturation involved a decrease in cetane number, and there-fore, allowed for the maximization of the effect of the cetane num-ber, while other properties, such as oxygen content, heating value,and viscosity, varied within a small range.

2. Experimental section

2.1. Test fuel characterization

A conventional diesel fuel (Ultra-Low Sulphur Diesel (ULSD))was provided from Shell fueling station located in Batman, Turkey,and used as the reference fuel in this work. Biodiesel fuels wereproduced from different feedstock such as palm oil, cottonseedoil and waste anchovy fish oil through transesterification reactionusing methanol in the presence of an alkali catalyst. Biodiesel fuelswere selected to have different values of iodine number, from 59 to185. Among the biodiesels tested, palm oil is the most saturatedone, and consequently, it has a high cetane number and low den-sity. Cottonseed oil is available abundantly in Turkey, and wasteanchovy fish oil is also taken into account as a promising alterna-tive feedstock for biodiesel production [18]. In addition to these,they have more unsaturated fatty acids than palm oil. For transe-sterification reaction, methanol and sodium hydroxide with purityof 98% were provided from Refinery and Petro-Chemistry labora-tory at Batman University, Batman, Turkey. After solving the NaOHcatalyst in methanol at room temperature in a magnetic stirrer, themoisture-free oils were added to the reaction tank to start thetransesterification reaction. The mixture was agitated throughoutat 60 �C. After glycerol separation, the ester phase was washedwith warm distilled water. After washing process, the methyl esterwas subjected to a heat at 110 �C to remove excess alcohol andwater, and then filtered. To determine the fatty acid profile, GCanalyses were carried out on a Shimadzu GC2010 plus Gas

il Cottonseed oil Waste anchovy fish oil

57.1 4.4320.9 20.2017.9 19.712.43 4.20.65 6.710.46 6.590.29 –0.18 1.64– 0.79– 10.41– 0.82– 21.58– 1.5375 69

Table 2Fuel properties.

Properties ULSD Palm oil Cottonseedoil

Waste anchovyfish oil

Chemical formula – C18.10H34.89O2 C18.49H34.77O2 C19.07H33.87O2

% C (wt.) 86.5 76.40 76.85 77.59% H (wt.) 13.5 12.36 11.98 11.57% O (wt.) 0 11.24 11.07 10.84IN (g of I2/100 g) 0 59 115 185MW (kg/kmol) 212 284 289 295Cetane number 51 68 56 52LHV (MJ/kg) 42 37.23 37.25 37.30Density (kg/m3) 835 876 884 892Kin. Viscosity

(mm2/s)2,8 4.763 4.33 4.435

Flash point (�C) 55 165.2 180.6 182.4CFPP (�C) �15 +11 +6 +4Stoichiometric air/

fuel ratio– 12.55 12.48 12.43

Table 3The specification of the engine.

Model of engine 1.4 diesel Inter heavy-duty

Engine type Four strokes, direct injection, water-cooled andnaturally aspirated

Injection pump Mechanically controlled in-line typeNumber of cylinder and

arrangement3-In line

Bore � stroke 80 mm � 90 mmCompression ratio 22.5:1Rated speed 1500 rpmRated power 12.5 kWAftertreatment None

Generator detailsPower output 9.6 kWFrequency 50 HzTension 230/400 V

452 S�. Altun / Fuel 117 (2014) 450–457

Chromatograph equipped with a flame ionization detector (FID).The GC uses a 30 m � 0.25 mm ID DB-23 capillary column (filmthickness of 0.25 lm). The fatty acid composition is given in Ta-ble 1. Palm oil biodiesel (denoted as POME) is defined by havinga balance between saturated (39.7% C16:0) and monounsaturated(43.5% C18:1) methyl esters while cottonseed oil biodiesel (de-noted as CSOME) is predominantly unsaturated (75%), having sig-nificant contents of diunsaturated (17.9% C18:2) and especiallytriunsaturated (57.1% C18:3) methyl esters. On the other hand,waste anchovy fish oil (denoted as WFOME) is also predominantlyunsaturated (69%) and it has more polyunsaturated methyl esters.

3 cylinders DI diesel engine

Fuel reservoir

Generator

Control panel Ga

Load bank

Electronic scale

Fuel

Air

Fig. 1. Schematic diagram of

Some properties of biodiesel fuels studied here are shown in Ta-ble 2. Density, kinematic viscosity, flash point and CFPP were mea-sured, whereas other properties were calculated or obtained fromsuppliers. Density and kinematic viscosity were measured at15 �C and 40 �C, respectively. Carbon and Hydrogen values forULSD was taken from Ref. [19] where these values were taken fromShell Global Solutions (UK). Cetane number, viscosity, heating va-lue and density are the main physical properties of the fuel that af-fect the diesel combustion process. These properties are affected bythe degree of unsaturation to different extents, as inferred from Ta-ble 2. For instance, the cetane number, which reflects the ignitionquality of the fuel, is higher in the case of POME (high saturatedfuel) compared to cottonseed oil and waste anchovy fish oil. In thisstudy, there was a good correlation between density and the de-gree of unsaturation. The biodiesels having highest density, indescending order, are: WFOME, CSOME, and POME. The lowerheating values (LHV) of the saturated and unsaturated fuels variedwithin a narrow range. The kinematic viscosity of the test fuelsslightly decreased with the degree of unsaturation.

2.2. Experimental equipment

The experimental study was carried out in a three-cylinder,four-stroke, naturally aspired and direct-injection diesel enginepowered generator, whose main characteristics are shown in Ta-ble 3. A bank of electric resistances was used to supply the load ap-plied to the generator. Fig. 1 illustrates the schematic diagram ofthe experimental set-up. The generator is linked up directly tothe engine from the flywheel of the engine to the clutch of the gen-erator. The engine is fitted with an in-line, three-cylinder fuelinjection pump coupled to a variable-speed mechanical governor.Several temperatures such as engine oil, coolant and exhaust gastemperatures were measured to monitor and control the engineoperation. Fuel consumption was determined by weighing fuelused for a period of time on an electronic scale with a precisionof 0.1 g. Therefore, the engine was modified in order to have a fuelconsumption control by gravity, changing the original fuel tank bya remote tank, which was placed on a balance. In each test, fuelconsumption, opacity and exhaust gas emissions, such as nitrogenoxides (NOx), carbon monoxide (CO) and total unburned hydrocar-bons (HC) were measured. In this study, gaseous emissions weremeasured by using a gas analyzer (Capelec Cap 3200) and the opac-ity level was monitored by means of Capelec smoke opacity tester.Measurement devices were calibrated, and their lines were cleanedbefore the experiments. The accuracy of the measurements anduncertainties in calculated results are shown in Table 4. The adia-batic flame temperature shows exclusively the influence of the

s emissions analyzer

Opacity meter

Air filter

Exhaust out To atmosphere

Air

the experimental set up.

Table 4The accuracy of the measurements and the uncertainties in the calculated results.

Parameter Accuracy

Load ±1%Speed ±15 rpmTime ±0.1sTemperatures ±1%HC ±1 ppmCO2 ±0.1%CO ±0.001%O2 ±0.01%NOx ±1 ppmSmoke ±0.1%Calculated results UncertaintyBSFC ±2.5% max.BTE ±2.5% max.

S�. Altun / Fuel 117 (2014) 450–457 453

molecular structure of the fuel on NOx emission, regardless of otherthermodynamical conditions which may change in combustionchamber. For the calculation of the adiabatic flame temperature,a chemical equilibrium model [20] which takes 29 chemical spe-cies into consideration was used. The initial conditions of 80 barand 900 K were selected as representative of the conditionsreached around the top dead center in the combustion chamberof a diesel engine. The constant-pressure condition (which permit-ted to calculate the adiabatic temperature from enthalpy balance)was preferred to the constant-volume condition (associated tointernal energy) consistently with the local (rather than bulk) char-acter of the NO formation mechanisms.

2.3. Test conditions

The test engine was run with a preliminary investigation of theengine running on a diesel reference fuel (ULSD), and then thesame procedure was repeated for other test fuels by keeping thesame operating conditions. Except for the reference fuel, three bio-diesel fuels, i.e. POME, CSOME and WFOME were also tested. Addi-tionally, a blend of reference fuel with biodiesel fuels in percentageof 20% by volume (denoted as POME_20, CSOME_20 andWFOME_20) were tested, too. The engine was considered to be atsteady state when the exhaust temperature reached a constanttemperature. Before starting a new experiment, the lines of themeasurement equipments were cleaned to remove residues ofthe previous test. A hand pump was used to drain fuel in betweenfuel changes. After that, the engine was warmed with the new fuelfor at least 30 min to cleanup any remains of the previously usedfuel from the fuel system. No modification on the engine or fuelinjection system was applied, and test engine had no aftertreat-ment system. The tests are performed using each of the conven-tional and alternative fuels with the engine working at a speed of1500 rpm and at different loads. Different from automotive diesel

200

400

600

800

1000

0 2 4 6 8

BSF

C (g

/kW

h)

Power output (kW)

ULSDPOMECSOMEWFOME

Fig. 2. Break specific f

engines, diesel engine powered generators operate under differentload and constant speed conditions; that is why such operatingconditions were selected for the tests in this study. Since the gasmixture in the engine cylinder has a longer residence time withhigh combustion temperatures under low speed and high load con-ditions, it is expected that the effect of biodiesel on NOx emissionswill be more significant under these conditions; hence, it is also ofinterest to investigate the effect of biodiesel on NOx emissions un-der such conditions [21]. Therefore, NOx emissions vs. the degree ofunsaturation were evaluated under low speed and high load condi-tions. This test condition yielded the minimum air/fuel ratio andmaximum smoke opacity among the collection of tested steadyoperation conditions.

3. Results and discussion

Experimental results obtained for brake specific fuel consump-tion, brake thermal efficiency, HC, CO, NOx emissions and smokeopacity are presented in this section. Results obtained for testedfuels as a function of load are shown in first figures, and then datawere reorganized and presented in two different figures to studythe effect of the biodiesel content and its degree of unsaturation.

3.1. Engine performance

A decreasing trend in brake specific fuel consumption (BSFC)was observed when increasing the load for all tested fuels, asshown in Fig. 2. This is mainly due to lesser contribution of themechanical losses to the total power output at high load condi-tions. At higher loads, the conversion of heat energy to mechanicalwork increases with the rise in combustion temperature that leadsto a decrease in BSFC with respect to the load [22]. In the same fig-ure, it can be seen that there is an increase in BSFC values of purebiodiesel fuels in comparison with ULSD, and the increase is almostproportional to the differences in lower heating value of the fuels.In the case of blended fuels containing 20% biodiesel, BSFC was inbetween those of their components corresponding to the blendingproportions [23]. The reason for higher BSFC with biodiesel fuels isevidently because of their lower heating value when reference die-sel fuel and biodiesel fuels with significant differences in heatingvalue are compared. To maintain the same power output at equalspeed, more biodiesel has to be injected into combustion chamberdue to its low energy content compared with fossil diesel [23].When brake thermal efficiency is examined, it will be clearly seenthat the efficiency of ULSD, biodiesel fuels and intermediate blendsare very similar as can be observed in Fig. 4 since brake thermalefficiency is a function of BSFC and LHV of the fuel for a constantpower output. BSFC values for tested fuels were reorganized inFig. 3 to further study the effect of the biodiesel content (Fig. 3a)

200

400

600

800

1000

0 2 4 6 8

BSF

C (g

/kW

h)

Power output (kW)

ULSDPOME_20CSOME_20WFOME_20

uel consumption.

0

5

10

15

20

25

0 2 4 6 8

BTE

(%)

Power output (kW)

ULSDPOMECSOMEWFOME

0

5

10

15

20

25

0 2 4 6 8

BTE

(%)

Power output (kW)

ULSDPOME_20CSOME_20WFOME_20

Fig. 4. Brake thermal efficiency.

0

4

8

12

0 20 40 60 80 100

(BSF

C/B

SFC

ULS

D_ 1

).10

0 (%

)

Biodiesel content in fuel (%v/v)

POMECSOMEWFOME

4

6

8

10

12

14

40 60 80 100 120 140 160 180 200

(BSF

C/B

SFC

ULS

D_1

).100

(%)

IN (g of I2 /100g) for pure biodiesel fuels

(a) (b)

Fig. 3. Increase of BSFC for blends (a) and for pure biodiesel fuels (b).

454 S�. Altun / Fuel 117 (2014) 450–457

and it was calculated as the average of the seven operation modestested for that fuel, and the effect of iodine number of purebiodiesel was also shown in Fig. 3b. It was found that the increasein BSFC was linear with the biodiesel content in the fuel blend, upto about 11% in the case of WFOME, very close to the difference ofheating values between ULSD and biodiesel fuels, as shown in Ta-ble 2. The biodiesels that result in the highest fuel consumption, indescending order, are these: WFOME, CSOME, and POME; and thusit can be said that BSFC slightly increased with increasing degree ofunsaturation characterized by iodine number (IN) as shown inFig. 3b. Differences in brake thermal efficiency (BTE) between pureor blended biodiesel fuels and ULSD are shown in Fig. 4, and itshows similar BTE values except for higher loads where ULSD has

-3

-1

1

3

5

7

0 20 40 60 80 100

(BTE

/BTE

ULS

D-1

). 100

(%)

Biodiesel content in fuel (%v/v)

POMECSOMEWFOME

(a)

Fig. 5. Difference of BTE and for blends

slightly low efficiency. Moreover, it can be seen in Fig. 5a that thereis no linear change in BTE with biodiesel content, and also in Fig. 5bwhere BTE slightly decreased by increasing the degree ofunsaturation.

The results indicated that the performance was not significantlyaffected by the type of biodiesel fuel or its degree of unsaturation,and thus this allows for an expressive analysis of the effect of thedegree of unsaturation of biodiesel fuels on exhaust emissions.

3.2. Gaseous emissions

Fig. 6 shows the total unburned hydrocarbons (HC) emissionsfor ULSD and the biodiesel fuels of various origins. Biodiesel fuels

0

1

2

3

4

5

6

7

40 60 80 100 120 140 160 180 200

(BTE

/BTE

ULS

D-1

).100

(%)

IN (g of I2/100g) for pure biodiesel fuels

(b)

(a) and for pure biodiesel fuels (b).

0

50

100

150

200

0 2 4 6 8

HC

Em

issi

ons

(ppm

)

Power output (kW)

ULSDPOMECSOMEWFOME

0

50

100

150

200

0 2 4 6 8

HC

Em

issi

ons

(ppm

)

Power output (kW)

ULSDPOME_20CSOME_20WFOME_20

Fig. 6. Unburned hydrocarbon emissions.

50

60

70

80

90

40 60 80 100 120 140 160 180 200

(HC

/HC

ULS

D-1

).100

(%)

IN (g of I2/100g) for pure biodiesel fuels

0

25

50

75

100

0 20 40 60 80 100

(HC

/HC

ULSD

-1).

100

(%)

Biodiesel content in fuel (%v/v)

POMECSOMEWFOME

(a) (b)

Fig. 7. Difference of HC emissions and for blends (a) and for pure biodiesel fuels (b).

600

700

800

900

1000

1100

1200

NO

xEm

issi

ons

(ppm

)

ULSD POMEPOME20 CSOMECSOME20 WFOMEWFOME20

Fig. 8. NOx emissions for fuels tested.

S�. Altun / Fuel 117 (2014) 450–457 455

tested result in higher HC emissions than ULSD. This is an unex-pected result as HC emissions are usually found to significantlydecrease with biodiesel. The observed increases in HC emissionscould probably be associated with excessive viscosity and low vol-atility of pure biodiesel fuels, which lead to problems in fuel atom-ization and evaporation, as compared with ULSD, in the case thattest engine is a low-injection-pressure one. Jindal et al. [24] also re-ported that HC emissions tend to increase with increase in com-pression ratio and also on reduction in injection pressure. On theother hand, a decrease in HC emissions was observed as the degreeof unsaturation of the biodiesel fuels increased, as shown in Fig. 7.This result is in agreement with the findings by Karavalakis et al.[25] who obtained lower THC emissions when the degree of unsat-uration of the biodiesel fuel was increased. It was reported in theirstudy that the lower volatility of the saturated blends may lead toincomplete vaporization and combustion of the fuel, therebyincreasing HC emissions. Exhaust CO emissions for all tested fuelswere very small in the range, and there appeared a slight increasewith the increase in load, so they were not plotted in any Figure.However, the palm-based biodiesel produced the lowest CO emis-sions among the biodiesels tested. The difference in CO emissionwith different biodiesels is likely to be a combined effect of oxygencontent and cetane number as the composition of the palm-basedbiodiesel, which corresponds to a high cetane number due to itshigher content of linear hydrocarbons. This result seems to agreewith a study conducted by Wu et al. [26], indicating that CO emis-sions decrease as the saturation level is increased.

The NOx emissions for ULSD and the biodiesel fuels of variousorigins are plotted in Fig. 8 for high load conditions, where theyyielded the minimum air/fuel ratio and maximum smoke opacityamong the collection of steady operation conditions. Results pre-

sented in Fig. 8 show that the use of biodiesel fuels leads to a slightdecrease in NOx emissions, especially in the case of highly satu-rated biodiesel as a consequence of their higher cetane number,opposite to trends generally observed (see Table 2). NOx-decreas-ing trend when using more saturated biodiesel fuels has also beenreported in other works [27,28]. The reduction of aromatic contentin the fuel also contributes to decrease NOx emissions, as reportedby Kalligeros et al. [29]. Several researchers indicated that there isa strong relationship between iodine number, which is a good indi-cator for the degree of unsaturation, and NOx emissions [30–32].This effect can also be more easily observed in Fig. 9. As the iodinevalue of POME is lower than those of CSOME and WFOME, andhence decrease in NOx emissions is expected. The differences in

-16

-14

-12

-10

40 60 80 100 120 140 160 180 200

(NO

x/N

Ox U

LSD-1

).100

(%)

IN (g of I2/100g) for pure biodiesel fuels

-16

-14

-12

-10

-8

-6

-4

-2

0

0 20 40 60 80 100

(NO

x/N

OxU

LSD-

1). 1

00 (%

)

Biodiesel content in fuel (%v/v)

POMECSOMEWFOME

(a) (b)

Fig. 9. Difference of NOx emissions and for blends (a) and for pure biodiesel fuels (b).

2620

2640

2660

2680

2700

2720

2740

2760

0.9 0.95 1 1.05 1.1 1.15 1.2

POME

CSOME

WFOME

Relative fuel/air ratio

Adia

batic

flam

e te

mpe

ratu

re (K

)

Fig. 10. Adiabatic flame temperature at constant pressure of the tested fuels fortypical initial conditions.

0

20

40

60

80

Smok

eop

acity

(%)

ULSD POMEPOME20 CSOMECSOME20 WFOMEWFOME20

Fig. 11. Smoke opacity for fuels tested.

-35

-30

-25

-20

-15

-10

-5

0

0 20 40 60 80 100

(Sm

oke/

Smok

e ULS

D-1

). 100

(%)

Biodiesel content in fuel (%v/v)

POMECSOMEWFOME

(a)

Fig. 12. Difference of opacity for blends

456 S�. Altun / Fuel 117 (2014) 450–457

the adiabatic flame temperature, which is shown in Fig. 10, couldbe pointed out to explain the observed reduction. The adiabaticflame temperature was calculated from a 29 species equilibriummodel, which is presented in detail in experimental equipmentsection, for a wide range of local relative fuel/air ratios and for typ-ical initial conditions indications. Fig. 10 shows that the adiabaticflame temperature increases slightly as the degree of unsaturationincreases and reaches maximum levels around equivalence ratiosof 1.065 in all cases. POME is the most saturated one (lowest IN)and it has lowest adiabatic flame temperature, which is good to re-duce NOx emissions. In the opposite, WFOME is most unsaturated(IN = 185), and it has highest adiabatic flame temperature. Knotheet al. [33] also pointed out that the adiabatic flame temperaturewas responsible for a reduction in NOx emissions for the saturatedesters.

3.3. Smoke opacity

Smoke opacity level is a measure of the relative amount of lightextinction of a light beam inciting perpendicularly to the exhaustgas stream, and is an indicator of soot emissions in the exhaust.It was observed that increasing contents of biodiesel in the fuel de-creased the opacity of the exhaust gas, and it increased with in-crease in load for all tested fuels, as shown in Fig. 11. This isbecause of the oxygen content of biodiesel fuels increases the oxy-gen present in fuel rich regions of the combusting fuel spray,reducing the amount of soot formed [34]. The reduction of aro-matic compounds in the fuel and the lower stoichiometric needof air can be given as other reasons. Fig. 12a shows that, smokeopacity decreased consistently with the decrease in aromatic com-pounds and the increase in oxygen content in the fuel blends. The

-35

-30

-25

-20

40 60 80 100 120 140 160 180 200

(Sm

oke/

Smok

e ULS

D-1)

.100

(%)

IN (g of I2/100g) for pure biodiesel fuels

(b)

(a) and for pure biodiesel fuels (b).

S�. Altun / Fuel 117 (2014) 450–457 457

effect of the iodine number of the biodiesel fuel in the smokeopacity can be observed from Fig. 12b. As the iodine number wasincreased, increase in opacity was found, and then it decreased.Unsaturated functional groups contribute to increased sooting ten-dency, as reported by Barrientos et al. [35], other researchers havealso found decreases in PM emissions and smoke opacity with in-creased unsaturation [16]. However, biodiesel fuels showing highexhaust temperature (such as those with highest IN) continuedsuch trend by promoting soot oxidation. On the other hand, Grabo-ski et al. [36] did not found any correlation either with the chainlength or with the unsaturation level. They argued that PM emis-sions depend on the oxygen content, which is almost constantfor every biodiesel or pure ester. However, there was some depen-dence on the density and cetane number. When the biodiesel den-sity was higher than 895 kg/m3 or the cetane number lower than45 the PM emissions increased considerably [36]. Choi et al. [37]also concluded that the effect of the composition and structureon PM emissions is negligible as compared to the oxygen content,which was acknowledged as the main factor causing PM emissionreductions.

4. Conclusions

From the experimental work carried out to investigate the effectof the degree of unsaturation of biodiesel fuels on engine charac-teristics, the following conclusions can be drawn. In comparisonwith ULSD, biodiesel fuels showed higher BSFC throughout allthe power range investigated, and similar BTE values. Biodieselfuels reduced emissions of oxides of nitrogen and carbon monox-ide, and smoke opacity, with the magnitude of these decreasesdepending upon the biodiesel fuel origin. HC emissions were high-er when the biodiesel fuels were used instead of ULSD, with typesof biodiesel producing similar results. Results obtained showedthat there was no significant effect of the degree of unsaturationon the engine performance while the effect on emissions was clear.More unsaturated biodiesel fuels showed higher NOx emissions,and lower HC emissions. Although increasing contents of biodieselin the fuel decreased the opacity of the exhaust gas, there was notfound a linear changing with degree of unsaturation. According tothe experimental results obtained in this work, the cetane numberand adiabatic flame temperature appear to be the key propertiesthat determined the exhaust emissions of the biodiesel fuels.

Acknowledgements

This work was partially supported by the Unit of Scientific Re-search Projects of Batman University, Turkey with Research ProjectNo. BAP-2010-MF-01. I am thankful to Prof. Dr. Magín Lapuertafrom University of Castilla La-Mancha, Ciudad Real, Spain for hisvaluable contribution and advices, and also the Council of HigherEducation of Turkey (YÖK) for supporting the authors’ stay at theUniversity of Castilla La-Mancha.

References

[1] Hulwan DB, Joshi SV. Performance, emission and combustion characteristic ofa multicylinder DI diesel engine running on diesel–ethanol–biodiesel blends ofhigh ethanol content. Appl Energy 2011;88:5042–55.

[2] Altun S, Öner C. Gaseous emission comparison of a compression–ignitionengine fueled with different biodiesels. Int J Environ Sci Technol2013;10:371–6.

[3] EMA. Engine manufacturers issue technical statement on use of biodiesel fuel,7 March 2003.

[4] Özsezen AN, Çanakçı M, Türkcan A, Sayın C. Performance and combustioncharacteristics of a DI diesel engine fuelled with waste palm oil and canola oilmethyl esters. Fuel 2009;88:629–36.

[5] Lin C-Y, Li R-J. Fuel properties of biodiesel produced from the crude fish oilfrom the soapstock of marine fish. Fuel Process Technol 2009;90:130–6.

[6] Lapuerta M, Armas O, Rodríguez-Fernández J. Effects of biodiesel fuels ondiesel engine emissions. Prog Energy Combust Sci 2008;34:198–223.

[7] Hoekman SK, Robbins C. Review of the effects of biodiesel on NOx emissions.Fuel Process Technol 2012;96:237–49.

[8] Sun J, Caton JA, Jacobs TJ. Oxides of nitrogen emissions from biodiesel-fuelleddiesel engines. Prog Energy Combust Sci 2010;36:677–95.

[9] Fernando S, Hall C, Jha S. NOx reduction from biodiesel fuels. Energy Fuels2006;20:376–82.

[10] McCormick RL et al. Impact of biodiesel source material and chemicalstructure on emissions of criteria pollutants from a heavy-duty engine.Environ Sci Technol 2001;35(9):1742–7.

[11] Bakeas E, Karavalakis G, Stournas S. Biodiesel emissions profile in moderndiesel vehicles. Part 1: Effect of biodiesel origin on the criteria emissions. SciTotal Environ 2011;409:1670–6.

[12] Assessment and Standards Division (Office of Transportation and Air Quality ofthe US Environmental Protection Agency). A comprehensive analysis ofbiodiesel impacts on exhaust emissions. EPA420-P-02-001; 2002.

[13] Wyatt VT, Hess MA, Dunn RO, Foglia TA, Hass MJ, Marmer WN. Fuel propertiesand nitrogen oxide emission levels of biodiesel produced from animal fats. JAm Oil Chem Soc 2005;82:585–91.

[14] Wang PS, Van Gerpen JH, Thompson J, Clemente T. Improving the fuelproperties of soy biodiesel. ASAE Paper No. 2007-076091; 2007.

[15] Ng J-H, Ng HK, Gan S. Engine-out characterisation using speed–load mappingand reduced test cycle for a light-duty diesel engine fuelled with biodieselblends. Fuel 2011;90:2700–9.

[16] Lapuerta M, Armas O, Rodríguez-Fernández J. Effect of the degree ofunsaturation of biodiesel fuels on NOx and particulate emissions. SAE paper2008-01-1676; 2008.

[17] Graboski MS, McCormick RL, Alleman TL, Herring AM. The effect of biodieselcomposition on engine emissions from a DDC series 60 diesel engine. NationalRenewable Energy Laboratory. NREL/SR-510-31461; 2003.

[18] Behcet R. A comparative study on anchovy fish oil, anchovy fish oil methylester and diesel fuels in a diesel engine. Energy Ed Sci Technol Part A2011;27(2):313–22.

[19] Pinzi S, Rounce P, Herreros JM, Tsolakis A, Dorado MP. The effect of biodieselfatty acid composition on combustion and diesel engine exhaust emissions.Fuel 2013;104:170–82.

[20] Lapuerta M, Hernández JJ, Armas O. Kinetic modeling of gaseous emissions in adiesel engine. SAE Paper 2000-01-2939; 2000.

[21] Zhang Y, Boehman AL. Impact of biodiesel on NOx emissions in a common raildirect injection diesel engine. Energy Fuels 2007;21:2003–12.

[22] Sayın C, Gümüs� M, Çanakçı M. Influence of injector hole number on theperformance and emissions of a DI diesel engine fueled with biodieseledieselfuel blends. Appl Therm Eng 2013;61:121–8.

[23] Lapuerta M, Armas O, Harnandez JJ, Tsolakis A. Potential for reducingemissions in a diesel engine by fuelling with conventional biodiesel andFischer–Tropsch diesel. Fuel 2010;89:3106–13.

[24] Jindal S, Nandwana BP, Rathore NS, Vashistha V. Experimental investigation ofthe effect of compression ratio and injection pressure in a direct injectiondiesel engine running on Jatropha methyl ester. Appl Therm Eng2010;30:442–8.

[25] Karavalakis G, Stournas S, Bakeas E. Light vehicle regulated and unregulatedemissions from different biodiesels. Sci Total Environ 2009;407:3338–46.

[26] Wu F, Wang J, Chen W, Shuai S. A study on emission performance of a dieselengine fueled with five typical methyl ester biodiesels. Atmos Environ2009;43:1481–5.

[27] Öner C, Altun S. Biodiesel production from inedible animal tallow and anexperimental investigation of its use as alternative fuel in a direct injectiondiesel engine. Appl Energy 2009;86:2114–20.

[28] Cecrle E et al. Investigation of the effects of biodiesel feedstock on theperformance and emissions of a single-cylinder diesel engine. Energy Fuels2012;26:2331–41.

[29] Kalligeros S et al. An investigation of using biodiesel/marine diesel blends onthe performance of a stationary diesel engine. Biomass Bioenergy2003;24(2):141–9.

[30] Puhan S, Saravanan N, Nagarajan G, Vedaraman N. Effect of biodieselunsaturated fatty acid on combustion characteristics of a DI compressionignition engine. Biomass Bioenergy 2010;34:1079–88.

[31] Benjumea P, Agudelo JR, Agudelo AF. Effect of the degree of unsaturation ofbiodiesel fuels on engine performance. combustion characteristics, andemissions. Energy Fuels 2011;25:77–85.

[32] Canakci M, Van Gerpen JH. Comparison of engine performance and emissionsfor petroleum diesel fuel, yellow grease biodiesel, and soybean oil biodiesel.Transac ASAE 2003;46(4):937–44.

[33] Knothe G, Sharp CA, Ryan III TW. Exhaust emissions of biodiesel, petrodiesel,neat methyl esters, and alkanes in a new technology engine. Energy Fuels2006;20:403–8.

[34] Graboski MS, Mccormick RL. Combustion of fat and vegetable oil derived fuelsin diesel engine. Prog Energy Combust Sci 1998;24:125–64.

[35] Barrientos EJ, Lapuerta M, Boehman AL. Group additivity in soot formation forthe example of C-5 oxygenated hydrocarbon fuels. Combust Flame2013;160:1484–98.

[36] http://www.nrel.gov/vehiclesandfuels/npbf/pdfs/31461.pdf.[37] Choi CY, Bower GR, Reitz RD. Effects of biodiesel blended fuels and multiple

injections on D.I. diesel engines. SAE paper 1997-970218.