Pb

MIC

a

ARRAA

KDPEHB

1

sasd2auam

be[Ib

of

k

h0

Industrial Crops and Products 59 (2014) 96–104

Contents lists available at ScienceDirect

Industrial Crops and Products

jo u r n al homep age: www.elsev ier .com/ locate / indcrop

erformance, emissions, and heat losses of palm and jatrophaiodiesel blends in a diesel engine

.J. Abedin ∗, H.H. Masjuki, M.A. Kalam ∗, A. Sanjid, S.M. Ashrafur Rahman,.M. Rizwanul Fattahentre for Energy Sciences, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

r t i c l e i n f o

rticle history:eceived 22 January 2014eceived in revised form 30 April 2014ccepted 4 May 2014vailable online 27 May 2014

eywords:

a b s t r a c t

After the successful implementation of B5, 5% palm (Elaeis guineensis) based biodiesel, in Malaysia on June1, 2011, the Malaysian government is now looking to phase out B5 by replacing it with B10 or even a higherblending ratio. Being non-edible feedstock, jatropha (Jatropha curcas) can play a vital role along with theexisting palm oil. This experiment was conducted in a four-cylinder diesel engine fuelled with B5, 10%,and 20% blends of palm (PB10 and PB20) and jatropha (JB10 and JB20) biodiesel and compared with fossildiesel at full load and in the speed range of 1000 to 4000 RPM. The brake power was decreased on average

iesel engineerformancemissioneat lossiodiesel

2.3% to 10.7% while operating on 10% and 20% blends of palm and jatropha biodiesel. An average of 26.4%BSFC increment was observed for PB20 and JB20 blends. An average of 30.7% carbon monoxide (CO) and25.8% hydrocarbon (HC) emission reductions were found for 20% blends. On average, the nitrogen oxides(NOx) emission is decreased by 3.3% while operating on PB10 and PB20 blends, whereas it is increasedby 3.0% while operating on JB10 and JB20 blends.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Climate change, limited efficiency of automotive engines andtringent antipollution laws imposed by governments have created

stimulus to explore more efficient engines with acceptable emis-ions level. In this context, biodiesel can be a promising solutionue to its comparable properties with fossil diesel fuel (Abedin et al.,013; Jayed et al., 2009). It is non-toxic, non-flammable, biodegrad-ble, and also environment-friendly. In addition, biodiesel can besed either pure or blended with fossil diesel fuel at any proportions

nd can be burnt by existing diesel engines without any furtherodifications (Mofijur et al., 2013; Rahman et al., 2013).

Abbreviations: BSFC, brake specific fuel consumption [g/KW h]; B5, 5% palmiodiesel + 95% diesel; B10, 10% biodiesel (any) + 90% diesel; FAME, fatty acid methylster; FFA, free fatty acid; LPG, liquefied petroleum gas; LHV, lower heating valuekJ/kg]; PB10, 10% palm biodiesel + 90% diesel; PB20, 20% palm biodiesel + 80% diesel;C, internal combustion; JB10, 10% jatropha biodiesel + 90% diesel; JB20, 20% jatrophaiodiesel + 80% diesel; RPM, revolution per minute.∗ Corresponding authors at: Department of Mechanical Engineering, Universityf Malaya, 50603 Kuala Lumpur, Malaysia. Tel.: +60 1 82697028/+60 3 79676863;ax: +60 3 79675317/+60 3 79674448.

E-mail addresses: joynul06me@siswa.um.edu.my (M.J. Abedin),alam@um.edu.my (M.A. Kalam).

ttp://dx.doi.org/10.1016/j.indcrop.2014.05.001926-6690/© 2014 Elsevier B.V. All rights reserved.

Oilseeds, in particular palm oil, are the main (about 73%)agricultural product of Malaysia. According to the United StatesDepartment of Agriculture (USDA), Malaysia is the World’s secondlargest producer (32.7%) and exporter (40.0%) of palm oil, behindIndonesia (production, 53.3%, and export, 49.5%) (United StatesDepartment of Agriculture (USDA), 2013). As a non-edible feed-stock, Jatropha curcas has drawn the attention of the Malaysiangovernment. Few projects on jatropha cultivation have started and57,601 hectares of jatropha are expected by 2015 (Mofijur et al.,2012). Palm oil based B5 biodiesel first rolled out in central regionsof Malaysia on June 1, 2011, and then nationwide in early 2013. B5is now available at 247 BHPetrol stations in Kuala Lumpur and con-sumes 1.03 million liters of palm-oil biodiesel each month, whichsaves nearly 12.4 million liters of fossil diesel fuel consumption peryear (Paultan.org, 2011). After the successful implementation of B5,the Malaysian government is now looking to go beyond 5% (i.e., 10%or even higher) by mid-2014 (Mysinchew.com, 2013).

Most of the articles published regarding palm and jatrophabiodiesel are based on single-cylinder engines and are subjectedto biodiesel properties, engine performance, and emission anal-yses. It is reported that in a four-cylinder diesel engine (Canakci

et al., 2009) at full load and different engine speeds using variouspalm biodiesel blends (5%, 10%, 20%, and so on), the power is almostsame as diesel for 10% biodiesel and decreased with the increase ofthe blend ratio. The same article also reported that BSFC is higher

M.J. Abedin et al. / Industrial Crops an

Nomenclature

C specific heat [kJ/kg K]m- mass flow rate [kg/s]N engine speed [rev/s]P power [kW]Q heat [kJ]T temperature [K]

Subscriptsa airb brakeexh exhaustf fuelg gasoil lubricating oils suppliedw water

fCbp2lhftcaNfpBtacaa2aefSiewvvicctaibdto

aaf

un unaccounted

or palm biodiesel compared to diesel fuel. They also found lowerO, HC, and smoke emissions and higher NOx emission for palmlends. In a six-cylinder engine (Ozsezen and Canakci, 2011) usingalm biodiesel, it is observed that the power is decreased about.5% and the BSFC is increased about 7.5%. The article also reported

ower CO (86.89%), HC (14.29%), and smoke (67.65%) emissions andigher NOx (22.13%) emission compared to diesel fuel. In a jatropha-

uelled three-cylinder engine (Sahoo et al., 2009), it is reported thathe brake power is slightly increased and the BSFC is much higherompared to diesel at full load and different engine speeds. It islso reported that HC and smoke emissions are lower and CO andOx emissions are higher for jatropha blends. In a jatropha-fuelled

our-cylinder engine (Manickam et al., 2009), it is reported that theower is decreased except at optimised injection timing, and theSFC is always higher at full load and different engine speeds. Inhe same literature, it is observed that all emissions except NOx

re lower compared to diesel fuel. However, according to single-ylinder engine reports, the results are similar (lower brake powernd higher BSFC) for palm (Ndayishimiye and Tazerout, 2011; Ngnd Gan, 2010; Sharon et al., 2012) and jatropha (Chauhan et al.,012; Ganapathy et al., 2011; Mofijur et al., 2013) biodiesel, though

few exceptions are also found (Agarwal and Agarwal, 2007; Huangt al., 2010; Tan et al., 2012). Similar emission results are reportedor single-cylinder engines, but two researchers (Jindal et al., 2010;undaresan et al., 2007) reported higher CO emission while operat-ng on jatropha biodiesel blends, and a few researchers (Kinoshitat al., 2003; Ng et al., 2012) reported slightly lower NOx emissionhile operating on palm biodiesel blends. According to the pre-

iously cited articles, the lower calorific value, higher kinematiciscosity, and density of biodiesel cause poor fuel spray and atom-sation which results in lower brake power, and higher BSFC. Theontradictory results are attributed to the complete combustionharacteristics of biodiesel for having more oxygen molecules inheir structure. This factor is also responsible for the lower COnd HC emissions of biodiesel. Researchers claimed that the highn-cylinder combustion temperature associated with oxygenatediodiesel is the main reason for higher NOx emission. The contra-ictory emission results, like lower NOx emission, are attributed tohe lower in-cylinder temperature due to the lower calorific valuef biodiesel.

An IC engine is a complex of machinery and instrumentation,ll of which must work together as a whole. It can be considereds a thermodynamic (open) system, which is an effective conceptor understanding the thermodynamic behaviour of the system. It

d Products 59 (2014) 96–104 97

is linked to the idea of “control volume”, a space enclosing the sys-tem and surrounded by an imaginary surface often known as the“control surface” (Fig. 1). If one can identify all the mass and energyflows into and out of a system, then it is possible to visualise theinside of that system by drawing up a thermal balance sheet ofthe inflows and outflows (Abedin et al., 2013). Heat loss analysisand thermal balancing are investigated by using hydrogen-gasoline(Yüksel and Ceviz, 2003), alcohol-diesel blends (Ajav et al., 2000),LPG fuel (i.e., liquefied petroleum gas) (Özcan and Söylemez, 2006),and biodiesel (Canakci and Hosoz, 2006; Debnath et al., 2013). Theyreported that all the heat losses are higher except exhaust heat losswhile using biodiesel. The reason is attributed to the promotion ofbetter combustion of biodiesel. The reason for lower exhaust heatloss is attributed to the lower concentration of HC and CO emissionsin the exhaust gas for biodiesel (Canakci and Hosoz, 2006).

The objective of this experiment is to find out the variations inengine performance and emissions as well as heat loss characteris-tics while operating on diesel, B5, and 10%, and 20% blends of palmand jatropha biodiesel.

2. Materials and methods

2.1. Biodiesel production

Crude vegetable oils were converted to biodiesel by applying(a) acid esterification and (b) base trans-esterification processes.Crude palm oil only needed base trans-esterification process butcrude jatropha oil required both steps because the acid value wasgreater than 4 mg KOH/gm. Methanol was used as solvent withH2SO4 and KOH for the first and second process, respectively. Usingacid catalyst, free fatty acid (FFA) level of crude jatropha oil wasreduced about 1–2%. A Favorit jacket reactor of 1 l capacity wasused with IKA Eurostar digital stirrer and Wiscircu water batharrangement. One liter crude oil, 200 ml methanol, and 0.5%·H2SO4(v/v) were mixed in a flask for acid-catalysed esterification. Themixture was constantly stirred at 700·RPM, 50 ◦C to 60 ◦C tempera-ture, and maintained at atmospheric pressure by circulating hotwater through the jacket. A sample of 5 ml was taken from theflask at 10 min intervals to determine the FFA level. After com-pleting the acid esterification process, the product was pouredinto a separating funnel where H2SO4 and excess alcohol includingimpurities were moved to the top. Then the top layer was sepa-rated and the lower layer was collected for base trans-esterification.The same experimental setup was used for the alkaline-catalysedtrans-esterification process. Meanwhile, 1% KOH (w/w) dissolvedin 25%·methanol (v/v) was poured into the flask. Then the mixturewas stirred at 700 RPM and at 70 ◦C temperature. The mixture washeated and stirred for 3 h and again poured into a separating funnelwhere it formed two layers. The lower layer contained glycerol andimpurities, and the upper layer was methyl ester of crude vegetableoil. The lower layer was discarded and the yellow upper layer waswashed with hot distilled water (100%·v/v) and stirred gently toremove the remaining impurities and glycerol. Then the biodieselwas taken in an IKA RV10 rotary evaporator to reduce the moisturecontent. Finally, moisture was absorbed using sodium sulphate andthe final product was collected after filtration.

2.2. Fatty acid (FA) composition

The fatty acid compositions of palm and jatropha biodiesel weretested using the MPOB test method (Malaysian Palm Oil Board

(MPOB), 2005). A specified amount (0.02 g) of biodiesel was dilutedwith 1.5 ml hexane in a small vial. A Perkin-Elmer GC equipped witha flame ionisation detector (FID) was charged with 2 �l of dilutedsample for FAC analysis. Identification of each peak was done by

98 M.J. Abedin et al. / Industrial Crops and Products 59 (2014) 96–104

ume s

caaetctfeT1euew2r

2

bbrPt

TF

Fig. 1. IC engine control vol

omparing it with an external standard reference mixture of fattycid methyl esters. The concentrations of the identified peaks weredded as an absolute value. The percentage of each fatty acid methylster (FAME) was calculated based on this value. Table 1 showshe FA composition of palm and jatropha biodiesel. Palm biodieselonsists of twelve components; seven saturated fatty acid esters,hree mono-unsaturated fatty acid esters and two polyunsaturatedatty acid esters. Among them, methyl oleate is the dominatingster (44.2%), with methyl palmitate (38.1%) at second position.hus, it contains 44.1% saturated, 44.6% mono-unsaturated and1.3% polyunsaturated methyl esters. Jatropha biodiesel consists ofight ester components; four saturated fatty acid esters, two mono-nsaturated fatty acid esters and two polyunsaturated fatty acidsters. Among them, methyl oleate is the dominating ester (41.8%),ith methyl linoleate (32.9%) at second majority. Thus, it contains

4.3% saturated, 42.6% monounsaturated and 33.1% polyunsatu-ated methyl esters.

.3. Fuel property measurement

Four different biodiesel blends (i.e., 10% and 20% of eachiodiesel) were prepared for this experiment. The 10% blends will

e denoted as PB10 and JB10 for palm and jatropha biodiesel,espectively. On the other hand, the 20% blends will be presented asB20 and JB20 for palm and jatropha biodiesel, respectively, wherehe numerical value denotes biodiesel percentage in the blends. The

able 1atty acid methyl ester (FAME) compositions of the tested biodiesels.

FAME Structure Molecular mass Formula

Methyl laurate 12:00 214.34 CH3(CH2)10CMethyl myristate 14:00 242.4 CH3(CH2)12CMethyl palmitate 16:00 270.45 CH3(CH2)14CMethyl palmitoleate 16:01 268.43 CH3(CH2)5CHMethyl stearate 18:00 298.5 CH3(CH2)16CMethyl oleate 18:01 296.49 CH3(CH2)7CHMethyl linoleate 18:02 294.47 CH3(CH2)3(CMethyl linolenate 18:03 292.46 CH3(CH2CH

Methyl archidate 20:00 326.56 CH3(CH2)18CMethyl eicosenoate 20:01 324.54 CH3(CH2)16CMethyl behenate 22:00 354.61 CH3(CH2)20CMethyl lignocerate 24:00 382.66 CH3(CH2)22CSaturation

Mono-unsaturation

Poly-unsaturation

Unsaturation

a Compositions are presented as wt.%

howing all energy flows [2].

quality of biodiesel depends upon the feedstock quality, chemicalcomposition of feedstock, production process, storage and hand-ling process. Crude palm and jatropha oil were provided by ForestResearch Institute Malaysia (FRIM). Diesel and B5 (5% palm + 95%diesel) biodiesel were bought by the authors from a BHPetrol oil fill-ing station near the University of Malaya. The oils were converted tothe biodiesel, and the properties were measured at the Departmentof Mechanical Engineering, University of Malaya, Malaysia.

Saponification Number (SN), Iodine Value (IV), and CetaneNumber (CN) of the produced biodiesel were calculated usingequation, SN =

∑(560 × Ai/MMi), IV =

∑(254 × DBi × Ai/MMi), and

CN = [46.3 + (5458/SN) − (0.225 × IV)], respectively. Where Ai is thepercentage of each component, DBi is the number of double bonds,and MMi is the molecular mass of each component (RizwanulFattah et al., 2014). Table 1 is used for the calculation. The mea-sured fuel properties of all the tested fuels are listed in Table 2along with equipment specifications.

2.4. Experimental setup and procedure

The experiment was performed in a four cylinder diesel engine.The schematic diagram of the test bed and engine specifications are

given in Fig. 2 and Table 3, respectively.

Table 3 Technical specification of the tested engine The enginewas loaded with a water-cooled passive eddy-current dynamome-ter. It was operated at full load condition and varying speeds

Palm biodiesela Jatropha biodiesela

OOCH3 0.3 0OOCH3 1 0.1OOCH3 38.1 17.7

CH(CH2)7COOCH3 0.2 0.8O2CH3 4.1 6.4

CH(CH2)7COOCH3 44.2 41.8H2CH CH)2(CH2)7COOCH3 11 32.9CH)3(CH2)7COOCH3 0.3 0.2OOCH3 0.4 0.1H CHCOOCH3 0.2 0OOCH3 0.1 0OOCH3 0.1 0

44.1 24.344.6 42.611.3 33.155.9 75.7

M.J. Abedin et al. / Industrial Crops and Products 59 (2014) 96–104 99

ayout

ftdapKmemcemBmp

3

(t

u

Q

W

Q

mr

P

w

f

Q

wrc

tpcca

fuels. The maximum brake power observed is 39 kW for diesel fuelat the rated speed (3000 RPM). An average of 2.3% to 10.7% brakepower reduction was observed for 10% to 20% biodiesel additionof both palm and jatropha at all speeds. Researchers (Buyukkaya,

Fig. 2. Schematic l

rom 1000 to 4000 RPM at an interval of 500 RPM. REO-dCA sys-em software has been used as an engine test bed controller andata acquisition system. It provides full test sequence control withutomatic and manual data logging options. The transducer boxrovides 16 K/J thermocouple input channels, where a total of eight-type thermocouples were used at eight different locations toeasure the temperature of cooling water, lubricating oil, and

xhaust. It also provides the facility of fuel flow, oil flow, and air floweasurements. Rotameters have been used to measure the engine

ooling water flow. The whole engine test including the exhaustmissions were performed at the ‘Heat Engine Lab’ of the Depart-ent of Mechanical Engineering, University of Malaya, Malaysia. A

osch (BEA- 350) exhaust gas analyser was used for the measure-ent of engine emissions. The specifications of the analyser are

rovided in Table 4.The uncertainty level of NOx emission for B5 is ±1.63% (Table 5).

. Theory of heat loss calculation

From Fig. 1, if we consider the IC engine as a control volumesurrounded by control surface), than the energy flows from and tohe engine can be expressed by the equations as follows:

The steady flow 1st law of thermodynamics for this control vol-me will be (Heywood, 1988)

s = Pb + Qw + Qoil + Qexh + Qun (1)

here Qs is the supplied fuel energy and is given by

s = mf × QLHV (2)

˙ f and QLHV are the mass flow rate and calorific value of the fuel,espectively.

The engine brake power (Pb) can be computed by

b = 2 × � × N(

rev/s)

× T (N m) × 10−3 (3)

here N and T are the engine speed and torque, respectively.The cooling water heat loss (Qw) can be calculated by using the

ollowing equation:

w = mw × Cw × �Tw (4)

here mw and Cw are the mass flow rate and specific heat of water,espectively and �Tw is the temperature difference between theooling water inlet and outlet.

Now, if we can measure the required heat to increase theemperature of the total mass (air + fuel) from the outside air tem-

erature (Ta) to the exhaust gas temperature (Tg), then we canompute the exhaust heat loss (Qexh) from the engine. The spe-ific heat of air at mean exhaust gas temperature is assumed as theverage specific heat (Cg) of the exhaust gases for exhaust heat loss

of the test engine.

calculation (Özcan and Söylemez, 2006). The equation for exhaustheat loss calculation will be

Qexh = (mf + ma) × Cg ×(

Tg − Ta)

(5)

Like cooling water heat loss, the lubricating oil heat loss (Qoil)can be found by using

Qoil = moil × Coil × �Toil (6)

where moil and Coil are mass flow rate and specific heat of lubri-cating oil, respectively, and �Toil is the temperature differencebetween the lubricating oil inlet and outlet. Finally, the unac-counted heat loss (Qun) can be estimated by applying subtractionrule

Qun = Qs − (Pb + Qw + Qexh + Qoil) (7)

4. Results and discussion

4.1. Performance characteristics

4.1.1. Engine brake power (Pb)The usable power delivered by an engine is known as brake

power (Pb). This is the final output power of the engine, which isthe product of engine torque and angular speed (Eq. (3)). Fig. 3shows the variation of brake power of the tested biodiesel blendswith varying engine speeds at full load. The brake power increasedsteadily with the increase of engine speed to a maximum point forall fuel blends, then falls slightly at the end. Diesel fuel resulted inhigher brake power throughout the speed range compared to other

Fig. 3. Engine brake power (Pb) variations for all tested blends.

100 M.J. Abedin et al. / Industrial Crops and Products 59 (2014) 96–104

Tab

le

2M

easu

red

fuel

pro

per

ties

of

all t

he

test

ed

blen

ds.

Prop

erti

es

Den

sity

aC

alor

ific

valu

eV

isco

sity

bFl

ash

poi

nt

Sap

onifi

cati

onn

um

ber

Iod

ine

valu

eC

etan

en

um

ber

FFA

c

Un

it

kg/m

3M

J/kg

cSt

◦ C

–

gI2/1

00

g

–

%D

iese

l

830.

4

45.2

3.61

77.0

–

–

48.0

d–

Palm

biod

iese

l

855.

5

39.8

4.56

167.

3

196.

4

58.0

61.0

1.8

Jatr

oph

a

biod

iese

l

869.

2

40.0

4.75

180.

0

192.

6

93.8

53.5

9.0

B5

832.

6

44.4

3.80

80.3

–

–

– –

PB10

833.

0

44.0

3.93

86.0

–

–

– –

PB20

836.

4

43.3

4.10

95.0

–

–

– –

JB10

836.

2

44.2

4.25

89.7

–

–

–

–JB

20

839.

2

43.6

4.36

99.0

–

–

–

–Te

stin

g

equ

ipm

ent

SVM

3000

(An

ton

Paar

, UK

)C

2000

basi

cca

lori

met

er

(IK

A, U

K)

SVM

3000

(An

ton

Paar

, UK

)Pe

nsk

y-m

arte

ns

NPM

440

(Nor

ma

Lab,

Fran

ce)

Cal

cula

ted

usi

ng

Tabl

e

1C

alcu

late

du

sin

g

Tabl

e 1

Cal

cula

ted

usi

ng

Tabl

e

1C

alcu

late

d

byti

trat

ion

Test

met

hod

and

spec

ifica

tion

lim

its

AST

M

D40

52

AST

M

D24

0

AST

M

D44

5(1

.9–6

.0)

AST

M

D93

(>13

0◦ C

)

not

spec

ified

inA

STM

D67

51n

ot

spec

ified

inA

STM

D67

51m

in. 4

7

inA

STM

D67

51–

aA

t

15◦ C

.b

At

40◦ C

.c

Cru

de

oil.

dPr

ovid

ed

by

sup

pli

er.

Table 3Technical specification of the tested engine.

Parameter Specification

Engine type In-line four cylinderBore 92 mmStroke 96 mmDisplacement 2.5 l (2476 cm3)Compression ratio 21:1Power 55 kW at 3000 RPM

Torque 142 N m at 2000 RPMFuel injection system Distributor type jet pump (indirect injection)

2010; Mofijur et al., 2013; Sahoo et al., 2009) attributed the reasonsto lower calorific value, higher viscosity and density of biodieselthan fossil diesel. Biodiesel contains higher oxygen content thanfossil diesel which results in lower calorific value. Density andviscosity have great influence on engine performance. Higher den-sity and kinematic viscosity influence the fuel atomisation processwhich slows down the fuel-air mixing rate. Poor atomisation rateresults in uneven combustion, hence lower brake power (Ozsezenand Canakci, 2011). Cetane number of biodiesel also greatly influ-ence the combustion process hence engine brake power. Highercetane number makes auto-ignition easily and gives shorter igni-tion delay which means combustion starts earlier for higher cetanenumber fuel. Shorter ignition delay period ensures higher peakin-cylinder pressure and higher heat release rate at pre-mixed com-bustion phase. Therefore, high cetane number is desirable for bettercombustion and high brake power (Candeia et al., 2009; Kannanet al., 2011; Tesfa et al., 2013). Table 1 ensures palm biodiesel hashigher saturation level, saponification number and cetane numberthan the jatropha biodiesel. Palm biodiesel exhibits slightly lowercalorific value than the jatropha biodiesel. But it was observedthat the kinematic viscosity and density of jatropha biodiesel weremuch higher than the palm biodiesel. Due to higher kinematic vis-cosity, density and lower cetane number, jatropha biodiesel blendsshowed an average of 5.3% lower brake power than the palmbiodiesel blends.

4.1.2. Brake specific fuel consumption (BSFC)Fig. 4 shows the nature of BSFC for all the tested fuel blends with

varying engine speeds. Fuel properties like density, viscosity, andcalorific value clearly influence BSFC (Qi et al., 2010). For example,fuel is injected on a volumetric basis in the combustion chamberand the density being much higher for biodiesel than diesel fuelindicates that the fuel injection pump is injecting more biodieselmass as compared to that of diesel mass for same power output(Ozsezen and Canakci, 2011). Again, higher kinematic viscosity of

biodiesel may cause poor atomisation of the fuel, hence poor mix-ing with air resulting in higher BSFC (Ganapathy et al., 2011). Dieselfuel has the highest calorific value, the lowest viscosity, and den-sity among all the tested blends. JB20 blend has the highest viscosity

Fig. 4. BSFC of all tested fuel blends.

M.J. Abedin et al. / Industrial Crops and Products 59 (2014) 96–104 101

Table 4Exhaust gases analysis specifications.

Gas analyser Emissions Method Maximum limit Accuracy

BOSCH BEA-350 CO Non dispersive infrared 10 vol.% ±0.001 vol.%CO2 Non dispersive iHC Flame ionisationNOx Electro-chemica

Table 5NOx emissions of B5 with engine speed.

Speed (RPM) NOx emission (ppm) Uncertainty (%)

1000 124 ±2.421500 180 ±1.672000 195 ±1.542500 216 ±1.853000 236 ±1.69

aerwp1aiK

4

4

iptstrttabemiCbdc

3500 263 ±1.144000 272 ±1.10Total uncertainty level ±1.63

nd density among all the tested blends. Clearly, diesel fuel and JB20xhibited the lowest and highest BSFC throughout the speed range,espectively. An average of 5.0%, 19.0%, and 26.4% BSFC incrementsere observed for the addition of 5%, 10%, and 20% biodiesel (bothalm and jatropha), respectively. Palm blends showed an average of4.4% lower BSFC compared to jatropha blends due to lower densitynd viscosity in spite of having slightly lower calorific value. Sim-lar results are reported by other researchers (Buyukkaya, 2010;inoshita et al., 2007; Mofijur et al., 2013).

.2. Emission characteristics

.2.1. CO emissionCO is formed during the combustion process whenever charge

s burned with an insufficient air supply with low in-cylinder tem-erature (Rizwanul Fattah et al., 2014). CO emission decreased withhe increase of biodiesel percentage, especially at higher enginepeeds. Diesel fuel resulted in the highest CO emission among all theested blends (Fig. 5). An average of 20.0% and 30.7% CO emissioneductions were observed for 10% and 20% biodiesel blends, respec-ively. CO emission of jatropha blends was slightly higher comparedo palm blends. This can be attributed to the higher viscositynd density of jatropha biodiesel. Higher viscosity, and density ofiodiesel result in inadequate fuel–air mixing, especially at lowerngine speeds, and CO emission increases with inadequate fuel–airixing. At lower engine speeds, the in-cylinder combustion rate

s relatively poor. Incomplete combustion is a reason for higher

O emission at lower engine speeds (An et al., 2012). This effectecomes less significant at higher engine speeds, when the cylin-er temperature is high. At higher engine speeds, the in-cylinderombustion is more complete. The oxygen-rich biodiesel results

Fig. 5. Carbon monoxide (CO) emission for all tested blends.

nfrared 18 vol.% ±0.001 vol.% detector 9999 ppm ±1 ppml transmitter 5000 ppm ±1 ppm

in more complete combustion of the blends. As a result, the trendis more dominant at higher engine speeds (Chauhan et al., 2012;Mofijur et al., 2013). A higher cetane number exhibits a shorterignition delay and allows a longer combustion duration. Then theoxygen content of biodiesel comes into play, which enhances thecombustion process (Ozsezen and Canakci, 2011). Lower viscosityand density along with higher cetane number and oxygen contentare the reasons of lower CO emissions (on average 8.8%) of palmblends compared to jatropha blends.

4.2.2. HC emissionLocally over-lean and over-rich fuel-air mixtures are the two

major sources of HC emission in heterogeneous combustion envi-ronment of diesel engines. At over-lean condition, the mixture istoo lean to auto ignite or to support a propagating flame and at over-rich condition, the mixture is too rich to ignite or support a flame(Heywood, 1988). As a result incomplete combustion occurs dur-ing these two conditions. Over-leaning is a dominant mechanismat lower engine speeds and over-rich mixture is dominant at higherengine speeds (Dhar and Agarwal, 2014). Oxygenated biodieselreduces the deficiency of oxygen during over-rich condition athigher engine speeds. At higher speeds, palm blends showed lowerHC emission than jatropha blends (Fig. 6) (Rahman et al., 2013).Higher cetane number also reduces HC emission (Buyukkaya, 2010;Ozsezen and Canakci, 2011). Besides higher engine speed ensuresbetter mixing of fuel and air, hence better combustion (Rahmanet al., 2013). Over-leaning depends on the amount of fuel injectedand mixing rate with air during the ignition delay period. Higherignition delay increases HC emission. As higher cetane numbershortens the ignition delay period, palm blends showed slightlylower HC emissions than jatropha blends at lower engine speedswhen over-leaning is dominant. Another reason can be the higherviscosity and density of jatropha biodiesel which causes uneven andincomplete combustion. An average of 17.5% and 25.8% HC emis-sion reductions were observed for using 10% and 20% blends of bothpalm and jatropha biodiesel, respectively. Diesel fuel showed thehighest HC emission, and the palm blends showed 2.2% (average)lower HC emission compared to jatropha blends.

4.2.3. NOx emissionStudies investigating the effect of biodiesel on NOx emission

reported mixed responses, ranging from increased NOx emis-sion for using biodiesel at all operating points to decreased NOx

Fig. 6. Hydro carbon (HC) emission characteristics.

102 M.J. Abedin et al. / Industrial Crops and Products 59 (2014) 96–104

e2tldori2tTotnMcnsiShe2tNaleT2owDbIc

4

4

cBdbteHe

the brake power increased. Similarly, all the heat losses increasedwith engine speed, but the increment was not same for all. Both heatlosses were slightly higher for palm blends compared to jatrophablends. It indicates that the palm blends have higher combustion

Fig. 9. Water heat loss (Qw) characteristics.

Fig. 7. NOx emission characteristics of all tested blends.

mission at all operating points (Lapuerta et al., 2008; Sun et al.,010). The higher NOx emission of biodiesel is mainly attributedo better combustion due to high oxygen content. But according toiterature review, the NOx emission behaviour doesn’t completelyepend on a change in a single fuel property, rather it is the resultf a number of coupled mechanisms whose effects may tend toeinforce or cancel one another under different conditions, depend-ng on operating conditions and fuel characteristics (Mueller et al.,009; Sun et al., 2010). NOx emission increases considerably withhe increase of double bounds (Sun et al., 2010; Tsolakis et al., 2007).he unsaturation of FAME is strongly correlated with the increasef NOx emission. The higher the unsaturation level of methyl estershe higher the ignition delay. Because unsaturation lowers cetaneumber which results in higher NOx emission (Kannan et al., 2011;ueller et al., 2009). On the other hand, pure saturated fatty acid

ompounds show a considerable reduction in NOx emission. Theumber of carbon single bonds in the molecule reflects degree ofaturation. The iodine number is a measure of saturation, where lowodine number indicates high saturation (increased single bonds).everal studies correlate the degree of saturation to NOx emission;ighly saturated fuel molecules produce lower amounts of NOx

mission (Kinoshita et al., 2003; McCormick et al., 2001; Sun et al.,010). Kalam et al. (2011) explained that the higher viscosity leadso a bigger droplet size and higher NOx emission. It is reported thatOx emission increases with the increase of fuel density (Ozsezennd Canakci, 2011). Therefore, higher cetane number along withower viscosity, and density are the main reasons of lower NOx

mission of palm biodiesel compared to jatropha biodiesel (Fig. 7).he NOx emission is decreased by 3.3% for the addition of 10% to0% palm biodiesel, whereas it is increased by 3.0% for the additionf 10% to 20% jatropha biodiesel. A moderate level of NOx emissionas observed for diesel fuel compared to palm, and jatropha blends.iesel fuel exhibits better viscosity, and density compared to palmiodiesel but the cetane number is much lower than palm biodiesel.

n comparison with jatropha biodiesel, diesel fuel exhibits a loweretane number but the viscosity, and density are much lower.

.3. Heat loss analysis

.3.1. Supplied fuel energy (Qs)The supplied fuel energy (Qs) is the amount of fuel entering the

ombustion chamber of an engine multiplied by its calorific value.iodiesel has higher density and lower calorific value than fossiliesel (Mofijur et al., 2013). The mass flow rate is higher for alllends, but the volume flow rate is same since it is controlled by

he fuel injector. Biodiesel increment decreased the supplied fuelnergy of the blend as it lowered the calorific value (Canakci andosoz, 2006). The supplied fuel energy for diesel fuel was the high-st since its calorific value was the highest among all the tested

Fig. 8. Variation of supplied fuel energy with RPM.

blends. As the speed increased, more fuel was required to burnin the combustion chamber. As a result, the supplied fuel energyincreased with the increase of engine speed (Fig. 8).

4.3.2. Water heat loss (Qw) and lubricating oil heat loss (Qoil)Both the water heat loss (Qw) and the lubricating oil heat loss

(Qoil) increased with the increase of biodiesel percentage in theblends at all speeds (Figs. 9 and 10). As explained earlier, biodieselpromotes better fuel combustion; hence, the in-cylinder combus-tion temperature, pressure, and heat release rate were higher whenthe engine is running on biodiesel. The water heat loss and thelubricating oil heat loss were increased on average 0.8–3.5% and1.0–4.7%, respectively for 10% biodiesel increment (both palm andjatropha). These two heat losses were increased on average 6% to9.5% for 20% blends at all speeds. This trend of heat loss is supportedby other researchers (Debnath et al., 2013; Tat, 2011). It is clear thatwith the increase of engine speed, both the supplied fuel energy and

Fig. 10. Lubricating oil heat loss (Qoil) analysis for all tested blends.

M.J. Abedin et al. / Industrial Crops an

ebseSHtw

4

pccbrhdaLAtgMlb

teftd(

5

i

(

(

Fig. 11. Exhaust heat loss (Qexh) of all tested blends.

fficiency, hence higher brake power, compared to the jatrophalends (Fig. 3). Despite the higher heat losses, the palm blendshowed higher brake power because the extra heat losses are recov-red by the lower exhaust heat loss and unaccounted heat loss.ome researchers (Abedin et al., 2013; Aydin, 2012; Canakci andosoz, 2006) have suggested providing ceramic insulation around

he cylinder walls and head to reduce these two heat losses, whichill eventually increase the exhaust heat loss.

.3.3. Exhaust heat loss (Qexh)Diesel fuel resulted in the highest exhaust heat loss (Qexh) com-

ared to other blends (Fig. 11). This can be attributed to the highoncentration of CO and HC emissions associated with diesel fuelompared to biodiesel blends (Canakci and Hosoz, 2006). The palmlends resulted in lower CO and HC emissions compared to jat-opha blends, and here, the palm blends have shown lower exhausteat loss. From Fig. 11, it is also clear that the biodiesel incrementecreased the exhaust heat loss. This trend of exhaust heat loss canlso be explained by the exhaust gas temperature characteristics.ower exhaust gas temperature signifies lower exhaust heat loss.

decreasing exhaust gas temperature trend was observed withhe increase of biodiesel percentage. A similar trend of exhaustas temperature is reported by other researchers (An et al., 2012;uralidharan et al., 2011; Rahman et al., 2013). The exhaust heat

oss was reduced on average 1.5% to 8.0% for the addition of 10%iodiesel.

Finally, the unaccounted heat loss (Qun) was computed by sub-racting the summation of all heat losses from the supplied fuelnergy. It covers mostly the convection and radiation heat lossesrom the cylinder walls and also other unknown heat losses fromhe engine (Abedin et al., 2013). The trend of this heat loss was notefinite. Similar results for unaccounted heat loss are reported inÖzcan and Söylemez, 2006) and (Yüksel and Ceviz, 2003).

. Conclusions

The following conclusions are drawn based on this experimentalnvestigation.

1) The brake power was decreased on average 2.3% to 10.7% whileoperating on 10% to 20% blends. The palm blends produced anaverage of 5.3% higher brake power than the jatropha blends.An average of 5.0%, 19.0%, and 26.4% BSFC increments wereobserved for the addition of 5%, 10%, and 20% biodiesel (bothpalm and jatropha), respectively. The palm blends provided anaverage of 14.4% lower BSFC compared to jatropha blends. Vis-cosity, density, and cetane number of the fuels played vital roles

in engine performance.

2) An average of 20% CO and 17.5% HC emission reductions wereobserved for 10% blends. And an average of 30.7% CO and 25.8%HC emission reductions were observed for 20% blends. The palm

d Products 59 (2014) 96–104 103

blends produced an average of 8.8% and 2.2% lower CO and HCemissions, respectively than the jatropha blends. Higher cetanenumber i.e., shorter ignition delay and higher oxygen contentof biodiesel are the main reasons of lower CO and HC emissionscompared to diesel fuel. Viscosity and density also influencethese emissions. At higher engine speeds, these emissions wereconsiderably lower.

(3) The NOx emission is decreased by 3.3% and increased by 3.0% forthe addition of 10% to 20% palm and jatropha biodiesel, respec-tively. Diesel fuel exhibited a moderate level of NOx emission.It was observed that higher saturation (higher cetane number),lower viscosity and density are the most desirable propertiesof biodiesel for lower NOx emission.

(4) On average, both the water heat loss and lubrication oil heat losswere increased in the range of 0.8% to 4.7% and 6% to 9.5% forusing 10% and 20% blends, respectively. A decreasing exhaustheat loss trend was observed for biodiesel addition. Overall, thepalm blends provided better heat loss characteristics than thejatropha blends.

In conclusion, biodiesel can be used up to 20%, especially thepalm biodiesel without any major modification of the engine.

Acknowledgment

The authors would like to acknowledge University of Malaya,Kuala Lumpur, Malaysia for financial support through HighImpact Research Grant entitles: “Development of Alterna-tive and Renewable Energy Career (DAREC)” grant number:UM.C/HIR/MOHE/ENG/60.

References

Abedin, M.J., Masjuki, H.H., Kalam, M.A., Sanjid, A., Rahman, S.M., Masum, B.M.,2013. Energy balance of internal combustion engines using alternative fuels.Renewable Sustainable Energy Rev. 26, 20–33.

Agarwal, D., Agarwal, A.K., 2007. Performance and emissions characteristics of jat-ropha oil (preheated and blends) in a direct injection compression ignitionengine. Appl. Therm. Eng. 27, 2314–2323.

Ajav, E.A., Singh, B., Bhattacharya, T.K., 2000. Thermal balance of a single cylin-der diesel engine operating on alternative fuels. Energy Convers. Manage. 41,1533–1541.

An, H., Yang, W.M., Chou, S.K., Chua, K.J., 2012. Combustion and emissions character-istics of diesel engine fueled by biodiesel at partial load conditions. Appl. Energy99, 363–371.

Aydin, H., 2012. Combined effects of thermal barrier coating and blending withdiesel fuel on usability of vegetable oils in diesel engines. Appl. Therm. Eng.51, 623–629.

Buyukkaya, E., 2010. Effects of biodiesel on a DI diesel engine performance, emissionand combustion characteristics. Fuel 89, 3099–3105.

Canakci, M., Hosoz, M., 2006. Energy and exergy analyses of a diesel engine fuelledwith various biodiesels. Energy Sources, B: Econ., Plann. Policy 1, 379–394.

Canakci, M., Ozsezen, A.N., Arcaklioglu, E., Erdil, A., 2009. Prediction of performanceand exhaust emissions of a diesel engine fueled with biodiesel produced fromwaste frying palm oil. Expert Syst. Appl. 36, 9268–9280.

Candeia, R., Silva, M., Carvalho Filho, J., Brasilino, M., Bicudo, T., Santos, I.,Souza, A., 2009. Influence of soybean biodiesel content on basic properties ofbiodiesel–diesel blends. Fuel 88, 738–743.

Chauhan, B.S., Kumar, N., Cho, H.M., 2012. A study on the performance and emissionof a diesel engine fueled with Jatropha biodiesel oil and its blends. Energy 37,616–622.

Debnath, B.K., Sahoo, N., Saha, U.K., 2013. Thermodynamic analysis of a variablecompression ratio diesel engine running with palm oil methyl ester. EnergyConvers. Manage. 65, 147–154.

Dhar, A., Agarwal, A.K., 2014. Performance, emissions and combustion characteris-tics of Karanja biodiesel in a transportation engine. Fuel 119, 70–80.

Ganapathy, T., Gakkhar, R.P., Murugesan, K., 2011. Influence of injection timingon performance, combustion and emission characteristics of Jatropha biodieselengine. Appl. Energy 88, 4376–4386.

Heywood, J.B., 1988. Internal combustion engine fundamentals. McGraw-Hill, NewYork, NY.

Huang, J., Wang, Y., Qin J.-b. Roskilly, A.P., 2010. Comparative study of performanceand emissions of a diesel engine using Chinese pistache and jatropha biodiesel.Fuel Process. Technol. 91, 1761–1767.

1 rops an

J

J

K

K

K

K

L

M

M

M

M

M

M

M

M

N

N

N

Distribution+++++++++++++++++++++++++++++++++++++++++++++++++++&

04 M.J. Abedin et al. / Industrial C

ayed, M.H., Masjuki, H.H., Saidur, R., Kalam, M.A., Jahirul, M.I., 2009. Environmentalaspects and challenges of oilseed produced biodiesel in Southeast Asia. Renew-able Sustainable Energy Rev. 13, 2452–2462.

indal, S., Nandwana, B., Rathore, N., Vashistha, V., 2010. Experimental investigationof the effect of compression ratio and injection pressure in a direct injectiondiesel engine running on jatropha methyl ester. Appl. Therm. Eng. 30, 442–448.

alam, M.A., Masjuki, H.H., Jayed, M.H., Liaquat, A.M., 2011. Emission and perfor-mance characteristics of an indirect ignition diesel engine fuelled with wastecooking oil. Energy 36, 397–402.

annan, G., Karvembu, R., Anand, R., 2011. Effect of metal based additive on per-formance emission and combustion characteristics of diesel engine fuelled withbiodiesel. Appl. Energy 88, 3694–3703.

inoshita, E., Hamasaki, K., Jaqin, C., 2003. Diesel combustion of palm oil methylester. In: SAE Technical Paper 2003-01-1929.

inoshita, E., Myo, T., Hamasaki, K., Nishi, S., 2007. Combustion characteristics ofdiesel engine with coconut oil ethyl ester. In: SAE Technical Paper 2007-01-2021.

apuerta, M., Armas, O., Rodriguez-Fernandez, J., 2008. Effect of biodiesel fuels ondiesel engine emissions. Prog. Energy Combust. Sci. 34, 198–223.

alaysian Palm Oil Board (MPOB), 2005. Determination of fatty acid composition(FAC) as their methyl esters by capillary column gas chromatography. MPOBTest Method 2005, 310–317.

anickam, M., Kadambamattam, M., Abraham, M., 2009. Combustion characteristicsand optimization of neat biodiesel on high speed common rail diesel enginepowered SUV. SAE Technical Paper 2009-01-2786.

cCormick, R.L., Graboski, M.S., Alleman, T.L., Herring, A.M., Tyson, K.S., 2001. Impactof biodiesel source material and chemical structure on emissions of criteriapollutants from a heavy-duty engine. Environ. Sci. Technol. 35, 1742–1747.

ofijur, M., Masjuki, H.H., Kalam, M.A., Atabani, A.E., 2013. Evaluation of biodieselblending, engine performance and emissions characteristics of Jatropha curcasmethyl ester: Malaysian perspective. Energy 55, 879–887.

ofijur, M., Masjuki, H.H., Kalam, M.A., Hazrat, M.A., Liaquat, A.M., Shahabuddin, M.,Varman, M., 2012. Prospects of biodiesel from jatropha in Malaysia. RenewableSustainable Energy Rev. 16, 5007–5020.

ueller, C.J., Boehman, A.L., Martin, G.C., 2009. An experimental investigation ofthe origin of increased NOx emissions when fueling a heavy-duty compression-ignition engine with soy biodiesel. SAE Int. J. Fuels Lubr. 2, 789–816.

uralidharan, K., Vasudevan, D., Sheeba, K.N., 2011. Performance, emission andcombustion characteristics of biodiesel fuelled variable compression ratioengine. Energy 36, 5385–5393.

ysinchew.com, 2013. Malaysia Set to Implement B5 or B10 Biodiesel Schemeby Mid-2014, http://www.mysinchew.com/node/82837 (date accessed:September 21, 2013).

dayishimiye, P., Tazerout, M., 2011. Use of palm oil-based biofuel in the inter-nal combustion engines: performance and emissions characteristics. Energy 36,1790–1796.

g, H.K., Gan, S., 2010. Combustion performance and exhaust emissions from the

non-pressurised combustion of palm oil biodiesel blends. Appl. Therm. Eng. 30,2476–2484.

g, J.-H., Ng, H.K., Gan, S., 2012. Characterisation of engine-out responses from alight-duty diesel engine fuelled with palm methyl ester (PME). Appl. Energy 90,58–67.

d Products 59 (2014) 96–104

Özcan, H., Söylemez, M.S., 2006. Thermal balance of a LPG fuelled, four stroke SIengine with water addition. Energy Convers. Manage. 47, 570–581.

Ozsezen, A.N., Canakci, M., 2011. Determination of performance and combustioncharacteristics of a diesel engine fueled with canola and waste palm oil methylesters. Energy Convers. Manage. 52, 108–116.

Paultan.org, 2011. B5 Biodiesel Debuts in Kuala Lumpur, at 247 Stations,http://paultan.org/2011/10/06/b5-biodiesel-debuts-in-kuala-lumpur-at-247-stations/ (date accessed: September 21, 2013).

Qi, D.H., Chen, H., Geng, L.M., Bian, Y., 2010. Experimental studies on the combus-tion characteristics and performance of a direct injection engine fueled withbiodiesel/diesel blends. Energy Convers. Manage. 51, 2985–2992.

Rahman, S.M.A., Masjuki, H.H., Kalam, M.A., Abedin, M.J., Sanjid, A., Sajjad, H., 2013.Production of palm and Calophyllum inophyllum based biodiesel and investiga-tion of blend performance and exhaust emission in an unmodified diesel engineat high idling conditions. Energy Convers. Manage. 76, 362–367.

Rizwanul Fattah, I.M., Masjuki, H.H., Kalam, M.A., Mofijur, M., Abedin, M.J., 2014.Effect of antioxidant on the performance and emission characteristics of adiesel engine fueled with palm biodiesel blends. Energy Convers. Manage. 79,265–272.

Sahoo, P.K., Das, L.M., Babu, M.K.G., Arora, P., Singh, V.P., Kumar, N.R., Varyani, T.S.,2009. Comparative evaluation of performance and emission characteristics ofjatropha, karanja and polanga based biodiesel as fuel in a tractor engine. Fuel88, 1698–1707.

Sharon, H., Karuppasamy, K., Soban Kumar, D.R., Sundaresan, A., 2012. A test on DIdiesel engine fueled with methyl esters of used palm oil. Renewable Energy 47,160–166.

Sun, J., Caton, J.A., Jacobs, T.J., 2010. Oxides of nitrogen emissions from biodiesel-fuelled diesel engines. Prog. Energy Combust. Sci. 36, 677–695.

Sundaresan, M., Chandrasekaran, S., Porai, P.T., 2007. Analysis of combustion, per-formance and emission characteristics of blends of methyl esters of jatropha oil(MEJ) in DI diesel engine. SAE technical Paper 2007-32-0066.

Tan, P.-q., Hu, Z.-y., Lou, D.-m., Li, Z.-j., 2012. Exhaust emissions from a light-dutydiesel engine with jatropha biodiesel fuel. Energy 39, 356–362.

Tat, M.E., 2011. Cetane number effect on the energetic and exergetic efficiencyof a diesel engine fuelled with biodiesel. Fuel Process. Technol. 92, 1311–1321.

Tesfa, B., Mishra, R., Zhang, C., Gu, F., Ball, A., 2013. Combustion and performancecharacteristics of CI (compression ignition) engine running with biodiesel.Energy 51, 101–115.

Tsolakis, A., Megaritis, A., Wyszynski, M., Theinnoi, K., 2007. Engine performance andemissions of a diesel engine operating on diesel–RME (rapeseed methyl ester)blends with EGR (exhaust gas recirculation). Energy 32, 2072–2080.

United States Department of Agriculture (USDA), 2013. Palm Oil: World Sup-ply and Distribution, http://www.fas.usda.gov/psdonline/psdReport.aspx?hidReportRetrievalName=Table+11%3a+Palm+Oil%3a+World+Supply+and+

hidReportRetrievalID=710&hidReportRetrievalTemplateID=8 (date accessed:September 21, 2013).

Yüksel, F., Ceviz, M.A., 2003. Thermal balance of a four stroke SI engine operating onhydrogen as a supplementary fuel. Energy 28, 1069–1080.