Energy and Power En gi neering, 2011, 3, 339-342
doi:10.4236/epe.2011.33042 Published Online July 2011 (http://www.SciRP.org/journal/epe)
Marula Oil and Petrodiesel: A Comparative Performance
Analysis on a Variable Compression Ignition Engine
Jerekias Gandure, Clever Ketlogetswe
Department of Mechanical Engineering, University of Botswana, Gaborone, Botswana
Received April 20, 2011; revised May 12, 2011; accepted May 17, 2011
Abstract
The quest for biofuel production and use in Botswana is driven by factors including volatile oil prices, need
for fuel security, potential for job creation, potential reduction in greenhouse gas emissions, and economic
diversification. In line with national efforts to come up with energy sources that are both environmentally
friendly and sustainable, this work was carried out to compare performance properties of native crude marula
(Sclerocarya birrea) seed oil and petrodiesel fuel on a variable compression engine test rig with automatic
data acquisition set up. Parameters such as engine torque, brake power and specific fuel consumption were
measured at different loads for the two fuels. The results indicate that engine performance when powered
with crude marula oil compares favourable with those for petrodiesel. Optimum numerical values for engine
torque, brake power and specific fuel consumption were 28.2 Nm, 6.27 W and 0.34 g/kWh respectively for
petrodiesel, and 22.7 Nm, 6.6 W, 0.33 g/kWh respectively for crude marula oil. The engine performance was
also analysed for same parameters, namely , engine torque, brake power and specific fuel consumption when
powered using the same fuels over a range of compression ratios while the load was fixed at 80%. Optimum
numerical values for engine torque, brake power and specific fuel consumption were 27.2 Nm, 3.67 W and
0.59 g/kWh respectively for petrodiesel, and 26.3 Nm, 3.6 W, 0.34 g/kWh respectively for crude marula oil.
The results indicate that compression ratio of 16:1 yields optimum engine performance in terms of engine
torque and brake power for both fuels under review. However, marula oil fuel recorded smooth steady in-
crease in performance profile across all compression ratios which out-performs petrodiesel on lower com-
pression ratios for engine torque and brake power, and is largely better than petrodiesel on fuel consumption.
Keywords: Performance, Marula Oil, Petrodiesel
1. Introduction
The expedition for biofuel production and use in Bot-
swana derives from volatile oil prices, need for fuel se-
curity, potential for job creation, potential reduction in
greenhouse gas emissions, and economic diversification.
The ambition to establish national energy self-reliance
and to develop alternatives to finite fossil fuel resources
have resulted in the development of fuel technologies
that are based on the use of renewable agriculture based
materials as feedstocks. In the case of renewable fuels
for compression ignition (diesel) eng ines, the majority of
efforts to date have focused on biodiesel, which consists
of alkyl esters of fatty acids found in agricultural acyl-
glycerol - based fats and oils. Biodiesel has been shown
to give engine performance that is generally comparable
to that of conventional diesel fuel while reducing engine
emissions of particulates, hydrocarbons and carbon mo-
noxide [1-3]. Biodiesel can be produced from any mate-
rial that contains fatty acids, bonded to other molecules
or present as free fatty acids. As a result various vegeta-
ble fats and oils, animal fats, waste greases, and edible
oil processing wastes can be used as feedstocks for bio-
diesel production. The choice of feedstock is based on a
number of factors including availability, cost, govern-
ment support and performance as a fuel [4].
This work sought to establish performance properties
of marula oil as a potential feedstock for biodiesel pro-
duction in Botswana. The results are expected to provide
indicators which can be used to stimulate rapid devel-
opment of biodiesel using indigenous feedstocks in
Botswana. Marula tree is indigenous to most parts of the
Southern African Development Community. In Bot-
swana, for example, it is widely distributed all over the

J. GANDURE ET AL. 340
country and concentrated in the north eastern part. The
tree grows in warm and dry climatic conditions, and
produce oval fruits that turn pale yellow when ripe as
demonstrated in Figure 1.
The fruit presented in Figure 1 consists of a hard
woody seed covered by pulp and juice which makes the
fleshy part of the fruit. The hard seed contains mostly
two oil rich nuts (kernel) which can be eaten as a snack.
In Botswana the kernel oil is currently being used to
make cosmetic ointments. The fact that the marula tree
grows in drier parts where common oil seeds cannot
thrive has stimulated interest in the use of marula oil as
substitute for petroleum diesel fuel. This situation h as led
to the evaluation of marula nut oil as a potential source
of vegetable oil for biodiesel production.
2. Materials and Methods
2.1. Marula Oil
The crude marula oil under review is sourced from
Kgetsi Ya Tsie, a Community Trust whose mandate is to
promote economic and social empowerment of rural
women in the Tswapong Hills of Eastern Botswana. The
Community Trust extracts marula oil mostly for cosmetic
markets in Europe and America.
2.2. Engine Performance Analysis
The engine performance test was conducted on a variable
compression engine test rig. The test rig is water cooled,
four-stroke diesel engine that is directly coupled to an
electrical dynamometer. In addition to the conventional
engine design, the engine incorporates variable compres-
sion design features which allow th e compressio n ratio to
be varied from 5:1 to 18:1.
To ensure that engine operating conditions were re-
produced consistently as any deviation could exert an
Figure 1. Ripe marula fruit.
overriding influence on performance results, the dyna-
mometer speed control set points were maintained within
±4 revs/min of the desired engine speed. The experi-
mental work began with engine run on D100. This was
done to determine the engine’s operating parameters
which constitute the b aseline that was compared with the
subsequent case when B100 (crude marula oil applied to
function as 100% biodiesel) was used to run the engine.
At the point of fuel change, the engine was allowed to
run using the fuel under review for approximately 15
minutes in order to stabilise before readings were re-
corded.
3. Results and Discussions
3.1. Engine Performance Analysis with Varying
Load
Marula oil was neutralised using NaOH to remove acid-
ity and minimise corrosion effects on engine parts. The
oil was also heated to a temperature of approximately
68˚C to lower its viscosity and to provide a reaction tem-
perature for faster conversion to biodiesel. The fuel was
then used to run a variable compression ignition engine
in order to test the engine performance, prior to its trans-
esterification. The results recorded for using raw marula
oil were compared with the results for petroleum diesel
under similar operating conditions on the basis of engine
torque, engine brake power, and specific fuel consump-
tion for compression ratio 16:1. The experimental data
were recorded as discussed in section 2.2, leading to the
results presented in Figures 2 to 4.
There are several clear findings to be drawn from the
data presented in Figures 2 to 4. Firstly, the results indi-
cate that the engine torque, brake power, and specific
fuel consumption recorded for D100 and B100 for opera-
tion condition under review (compression ratio 16:1)
compares favourably well. The data in Figure 2 show a
steady increase in engine torque for D100 and B100 with
corresponding increase in engine load from 30 to 60%.
However the same data show that between the engine
load 30% to 50%, B100 recoded relatively high engine
torque compared with D100. However, the data also de-
pict that as the engine load increase from 60% the data
recorded for D100 shows a steady increase in engine
torque while the data for B100 shows a slight decrease
with increase in engine load. The profile suggests that
the use of raw marula oil as fuel in IC engine gives better
engine torque at full engine load.
The results in Figures 3 and 4 also demonstrate that
the data recorded for B100 compares favourably well
with that for D100. The maximum variation in brake
power of 0.63 W was recorded at 90% engine load, while

J. GANDURE ET AL.341
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 2. Engine torque profile for marula oil and petro-
leum diesel fuel at compression ratio of 16:1.
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 3. Engine brake power profile for marula oil and
petroleum diesel fuel at compression ratio of 16:1.
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 4. Engine specific fuel consumption profile for ma-
rula oil and petroleum diesel fuel at compression ratio of
16:1.
the minimum variation of 0.06 W was recorded at 60%
of engine load, with D100 recording 5.06 W. The trends
presented in Figure 3 suggest that the optimum com-
pression ignition engine performance using raw marula
oil occurs at 60% engine load. The data shown in Figure
4 for the specific fuel consumption recorded for D100
and B100 reinforces the above observations. One of the
most discernible trends connected to Figure 4 is that the
variations in specific fuel consumption recorded for
D100 and B100 between 30% and 60% do not show any
significant difference for the fuels under review. The
minimum variation of specific fuel consumption between
30% and 60% engine load is 0.01 g/kWh, while a maxi-
mum of 0.17 g/kWh was recorded at the engine load of
90%. Overally, the result in Figure 4 points out that raw
marula oil is a potential feedstock for biodiesel produc-
tion in Botswana .
3.2. Engine Performance Analysis with Varying
Compression Ratio
This section presents results for engine performance at a
fixed load of 80% with compression ratio varying from
13:1 to 17:1. The results are a comparison between D100
and B100. The analysis aims to establish the influence of
compression on the engine performance using raw ma-
rula oil and compare such trends with those recoded for
D100 under similar operating conditions. The experi-
mental data were recorded and results presented in Fig-
ures 5 to 7.
The results presented in Figures 5 and 6 indicate that
profiles for engine torque and brake power are largely
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 5. Engine torque profile for marula oil and petro-
leum diesel fuel at 80% load.
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 6. Engine brake power profile for marula oil and
petroleum diesel fuel at 80% load.
B100 = Marula oil fuel; D100 = Pet rodiesel
Figure 7. Engine specific fuel consumption profile for ma-
rula oil and petroleum diesel fuel at 80% load.

J. GANDURE ET AL. 342
similar. The best performance for marula oil fuel was
found to be at compression ratio of 17:1 for engine
torque with a valu e of 2 6.3 Nm, and compression ratio of
16:1 for brake power with a value of 3.61 W. Similarly,
best performance for petrodiesel was found to be at
compression ratio of 16:1 for both engine torque and
brake power with a values of 27.2 Nm 3.7 W respec-
tively. These results imply in part that compression ratio
16:1 yields optimum performance results for engine
torque and brake power for petrodiesel and marula oil
fuels. Petrodiesel is marginally better than marula oil at
this compression ratio. For lower compression ratios
particular ly 14:1 and 15:1 , marula o il fuel is significantly
better than petrodiesel for both engine torque and brake
power. Moreover, unlike petrodiesel, marula oil fuel has
a smooth rising performance profile across all compres-
sion ratios which suggest smooth running of the engine
at any set compression ratio. The results in Figure 7
show that engine specific fuel consumption of marula oil
fuel is largely lower than that of petroleum diesel. The
fuel consumption at the optimum compression ratio of
16:1 is almost the same for the two fuels, with petro-
diesel being marginally better by 0.1 g/kWh.
4. Conclusions
A comparative experimental study to analyse the per-
formance of marula oil fuel and petroleum diesel was
carried out. From the experimental results, it can be con-
cluded that:
1) Marula oil has properties that enable it to function
as a biofuel in variable compression diesel engines. This
suggests that transesterifying marula oil under standard
conditions should produce biodiesel of international
quality standard.
2) Since crude marula oil has proved to perform well
as a fuel, it may not be necessary to incur costs through
processing the oil to produce biodiesel. Instead, the oil
may be modified through additives to enhance perform-
ance.
3) The compression ratio of 16:1 yields optimum per-
formance results for engine torque and brake power for
petrodiesel and marula oil fuels.
4) The performance of variable compression diesel
engine using raw marula oil is close to that using petro-
diesel fuel, suggesting that marula oil is a potential in-
digenous fee ds t ock for biodiesel pro duct i o n in Botswana.
5. Acknowledgements
We acknowledge support of the University of Botswana,
Kgetsi Ya Tsie Community Trust for providing marula
oil, and the Ministry of Wildlife, Tourism and Environ-
ment for granting a research permit for this work.
6. References
[1] M. S. Graboski and R. L. McCormick, “Combustion of
Fat and Vegetable Oil Derived Fuels in Diesel Engines,”
Progress in Energy and Combustion Science, Vol. 24, No.
2, 1998, pp. 125-164.
doi:10.1016/S0360-1285(97)00034-8
[2] P. M. Adriano, C. T. Richard, M. Luciano, C. Jonas and F.
A. Dejanira, “Biodegradability of Diesel and Biodiesel
Blends,” African Journal of Biotechnology, Vol. 7, No. 9,
2008, pp. 1323-1328.
[3] S.-M. Chen, Y.-J. Huang, S.-C. Chuang and H.-H. Yang,
“Effects of Biodiesel Blending on Particulate and Poly-
cyclic Aromatic Hydrocarbon Emissions in Nano/Ul-
trafine/Fine/Coarse Ranges from Diesel Engine,” Aerosol
and Air Quality Research, Vol. 9, No. 1, 2009, pp. 18-31.
[4] M. J. Haas, A. J. McAloon, W. C. Yee and T. A. Foglia,
“A Process Model to Estimate Biodiesel Production
Costs,” Bioresource Technology, Vol. 97, No. 4, 2006, pp.
671-678. doi:10.1016/j.biortech.2005.03.039