Diffusion rate analysis in palm kernel oil extraction using different extraction solvents

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 639
DIFFUSION RATE ANALYSIS IN PALM KERNEL OIL EXTRACTION
USING DIFFERENT EXTRACTION SOLVENTS
Egbuna S.O1
, Ozonoh. M2
, Aniokete. T .C3
1, 2, 3
Department of chemical engineering, Enugu State University of Science and Technology, Enugu, Nigeria.
Abstract
This study investigated the rate of diffusion on the extraction of palm kernel oil using various solvents, namely, n-hexane, Benzene,
Trichloroethylene, and Carbondisulfide. The solvents and the extracted palm kernel oil were characterized using the AOCS standard
methods, and the results are shown in tables 1 and 2. The palm kernel used in the study was obtained from Akpugo in Nkanu West
local Government Area of Enugu State. The effects of extraction time and temperature as well as quantity and nature of solvent were
investigated as reported in figures 1-3. The characterized PKO was used to determine parameters used in establishing the Diffusivity
(6.47 x 10-9
m2
/s), which in turn was used to calculate the rate of diffusion,(4.00 x 107
kmole/m2
.s), and the coefficient of mass transfer,
(3.2 x 10-5
kmloe/m2
.s), all of which conformed to the standard values, [12]..
Keywords: Diffusivity, Diffusionrate, Mass transfer Coeff, Extraction, Palm kernel oil, solvents, characterization
----------------------------------------------------------------------***------------------------------------------------------------------------
1. INTRODUCTION
Vegetable oils are triglycerides of fatty acid extracted from
plants. Such oils have been part of human culture for
millennia, (4,000-year old “kitchen” unearthed in Indiana)[1].
The term “vegetable oil” can be narrowly defined as referring
only to substances that are liquid at room temperature, or
broadly defined[1] without regard to the substances state of
matter at a given temperature[3].Hence vegetable oils that are
solid at room temperatures are sometimes called vegetable
fats. Vegetable oils are primarily extracted from seeds and
nuts, even though it can be extracted from other parts of the
plant.
Since ancient times human beings have known how to extract
oils from their various natural sources and make them fit for
their own use. The natives in the tropical regions of the world
have long extracted these oils after drying the nuts and seeds
under the sun, 4,000-year-old “Kitchen unearthed in Indiana”
Archaeo News, January 26, 2006, Retrieved 30 - 06 - 2013[1].
These oils were consumed raw, since very little or no
treatment was made order than filtration and/or decantation.
Mahatta[2] observed that olive oil was extracted as far back as
3000 BC in ancient Egypt. Vegetable oils are used as
ingredient or component in many manufactured products,
including soap, candle, perfumes, cosmetics, margarine,
shortening, hydraulic fluid biodiesel and lubricants[4].
A large number of articles have been published on extraction
of oils from their mother seeds and nuts using various
methods, and characterization of the extracted oils. [5] and [6],
noted that mechanical method of extraction, termed crushing
or pressing offers advantage in terms of purity over the solvent
extraction method. The oil is conditioned by bringing the
kernel to optimum condition of moisture in order to obtain the
maximum yield. Roasting is one of the most important steps in
the extraction of palm kernel oil. It serves to regulate the
moisture content, render the cell walls porous, rupture the cell
walls by generating steam within the cell, and coagulating the
protein so that filtration is made easy. The meal obtained by
grinding the kernel must not be too wet or too dry so as to
prevent imperfect extraction of the oil or poor yield.
Solvent extraction which is essentially a separation process
based on apparent equilibrium steps, on the other hand offers
the advantage of penetrating action on the cell of the
prepared oil seeds to improve yield tremendously, [7] and is
less expensive. However, the effects of diffusion of the oil
from the main bulk of the oil seed of the ruptured cells into the
solvent in liquid phase, and mass transfer coefficient, have
received little attention.[8], have studied the equilibrium
condition of expression after a constant pressure has been
maintained until no further flow of oil occurred. They
considered the quantity of oil expressed as the difference
between the quantity originally present and that which
remained in the cake after expression. They noted that an
increase in pressure on a system of expressible material
considered as a fractional increase over the previous pressure
causes a proportional increase in the bulk of the solid portion
of the system. Hence.
skd
P
dP
 = 






V
1
dk 1

__________________________________________________________________________________________
or
 
Pk
V1d
dP
 2
The equation is integrated to get,
LogP =
V
k
K

 3
Where K, k and k’
, are constant, depending on the nature of
material, and on the expression conditions. ρs is bulk density
of the solid portion of the system, P, the pressure, and V,
specific volume of the system based on the solid content.
Many factors affect the rate at which the oil diffuses into the
palm kernel grit, including time of exposure, nature and
quantity of solvent, and temperature of the system.
A succinct review of published literatures on liquid – liquid
extraction explained the important influence of solvent nature
on extraction. [9], observed that direct solvent extraction is
used for low oil content, (< 20% oil) seed such as soya bean,
rice bran and dry milled corn germ. [10] experimented on a
homogeneous oil impregnated material consisting of thin
platelets of uniform thickness with two phases as the total
surface area based on simple diffusion and observed that the
theoretical rate of extraction is given by;
24R
Dθ2/1)(2
0n
0n
21)(2n
1
2π
8
E ne 

 
   4
Where E is the fractional total oil unextracted at the end of
time, θ; R is one-half of the plate thickness (ft); and D, and
diffusion coefficient in ft2
/hr. For π = 0, equation reduces to,















 
24R
Dθn/2e
2n
8
E 5
Or
Log10
2R
oD
1.070.091E  6
Trebal,, [11] gave an expression to be used to determine the
diffusivity of a liquid in another, considered to be stagnant as,
 
0.6
BμV
T0.5
BM)18(117.3X10
AD

 7
Where DA - Diffusivity of oil, T – absolute temperature, MB
is molecular weight of solvent, VB- molar volume of solvent
 is association factor of solvent, and  - viscosity .
The rate at which the glyceride of fatty acid diffuses into
hexane is given by as; [12]
 A2A1avLMB12
AB
A XX
M)(XΖZ
D
N 








8
Where ρ - density of oil, and  BX LM
is log mean of mole
fraction of the oil and solvent. Perry and Chilton, [13], also
gave the mass transfer coefficient as;
 A2XAIXAKAN  9
Fan, [14] observed two solvent extraction methods, solution
extraction, in which oil is extracted from ruptured cells, and
diffusion extraction, in which the oil is extracted from
unruptured cells. This gives reason for the variation of
diffusion coefficient with time.
Treybal,, [15], also gave an expression from which efficiency
can be estimated, namely;
hTρ
KM
)43.7(101
1
oE

 10
In the present work, the extraction was carried out using N-
hexane, Benzene, Trichloroethylene, and Carbon disulphide,
as solvents. The efficiency of these solvents in terms of yield,
suitability in terms of quality, and kinetics of the extraction in
terms of diffusion rate were all investigated.
2. EXPERIMENTAL METHODS
The experimental work was aimed at obtaining data, on a
practical basis, that would be used in the evaluation of the
diffusion rate. The effects of time of extraction, type and
quantity of solvent used, nature of the solvent and temperature
of the system on the efficiency of extraction, were
determined.
The fundamental method of experiment, in this respect, is to
contact the prepared oil bearing kernel with appropriate
solvent in order to extract its oil content to a specified
minimum level. The oil-bearing seeds were pretreated to make
their oils extractable. The oil - solvent mixture (Miscella),
was separated into pure components by distillation. Traces of
solvent entrained in the meal were removed by
desolventization, and the desolventized meal was dried with
steam, and the last traces of solvent removed by stripping.

__________________________________________________________________________________________
2.1 Extraction of Palm Kernel Oil (PKO)
This experiment was carried out with a view to determining
the amount of extractable oil in the representative samples of
PKO grit, The following parameters were determined; Percent
weight of drained solution (extract), percent weight of the
cake, amount of oil in the extract, amount of oil retained in the
cake, and quantity of solvent used in the extraction. The
equipment and materials used in the extraction experiment
include: 600ml glass beakers, laboratory glass distillation
column, heating mantle, water circulating tube, flasks,
solvents and palm kernel. The properties of the solvents used
are given in table 1[16]
Table 1 Characterization of Solvents used in the Extraction
Oil Seed Specific
Gravity at
15o
C
Specific
heat per
litre
Latent
heat per
litre
Vapour
pressure at
20o
C(mm)
Toxicity
limit, at
20o
C
(mg/m3
)
Boiling
point,
o
C
N - Hexane 0.680 0.358 54 88 10.80 68.60
Benzene 0.700 0.360 56 80 10.80 60.70
Carbon disulfide 1.292 0.310 112 298 1.50 46.26
Trichloroethylene 1.489 0.327 84 70 11.00 87.00
Sample Preparation: l00g of PKO seed, were cleaned,
roasted in seed cooker, and ground.
Experimental procedure: The experiment was carried out by
using the method of [17],as follows:
100g of palm kernel grit, was charged into each of 600cm3
beaker in separate experiments, and 100cm3
of n - hexane
gradually added with stirring. The mixture was filtered and the
filtrate (the miscella) was collected at interval and recycled.
This process was repeated many times, without a fresh solvent
charge. The miscella was collected in a beaker at the last
stage. This process was repeated for each of the solvents, and
the results reported in figures 1-3
Calculation:
G
WW
W 12
m

 11
Where Wm - percent weight of oil extracted, W2 - the weight
of extract, and,W1 - the weight of solvent, G-weight of palm
kernel grit used.
2.2 Determination of Residual Oil in Cake
The aim is to determine the amount oil that remains in the
cake after extraction. This is useful in sizing extraction
equipment. The equipment and materials used include;
separating funnel, stop watch, retort stand, micro boiling flask
for distillation, thermometer, heating mantle, glass wool, palm
kernel grit sample and solvents.
Experimental procedure: Glass wool was inserted into a
separating funnel to plug the liquid outlet and form,
somewhat, a false bottom. 100g of the oil-bearing material
(palm kernel grit) was placed at the base of the funnel, and the
latter was stoppered with rubber bung which has hole in the
center where glass tube of 15cm long was inserted to extend
into the mass of the sample. This tube served as both solvent
delivery as well as stirrer. The separating funnel was clamped
to the retort stand, and 50ml of n-hexane was added gradually
over the bed of the sample, with occasional stirring using the
glass tube. Equal volume samples of extract were withdrawn
at 20 minutes interval by means of a pre-weighed micro -
boiling flask which was connected to the micro-distillation
unit which distilled the oil hexane solution (miscella), at 20°C
over a heating mantle. Solvent was recovered by
condensation and the boiling flask was cooled in a
desiccator and reweighed. Temperature was varied between 20
and 50o
C, Time , 20 to 120 min. and quantity of solvent from
50 to 250ml.
Extraction time: The aim was to determine the time
required to reduce the oil content of the given oil seed sample
to minimum. In the experiment, temperature and weight of oil
seed were kept constant while time of extraction and amount
of solvent were varied proportionately according to the
following sequence: 20 min to 50ml; 40min to 100ml; 60min
to 150ml; etcetera. The amount of oil grit used was 125gms
and the solvents were used in turn. The result are given in
figure 1.
Quantity of solvent: The aim of this experiment was to
determine the amount of solvent required in order to obtain the
lowest possible level of oil in the residue. Extraction time was
kept constant at 60min, and temperature was maintained
constant during the test. The results are given in figure 2
Solvent Temperature: This test was carried out to determine
the temperature of solvent that would give the best result. The

__________________________________________________________________________________________
condition under which it was done were: Solvent quantity,
250ml; sample used 125gms and extraction time, 120min.
Temperature was varied from 20 to 50o
C. The results are
given in figure 3
Nature of solvent: This was aimed at ascertaining the
dissolving power of the solvent. The temperature of the
system was kept at 45o
C, and Extraction time was 4hours,
while other conditions of weight and solvent quantity
remained constant.
2.3 Characterization of Extracted Palm Oil:
The raw palm kernel oil was characterized so as to determine
its physical and chemical properties, and hence the level and
extent of refining required in order to stabilize the quality of
the final product. The properties of the oil that were
determined include: color, melting point, specific gravity,
refractive index, moisture content, (Ppm), FFA(%),
PV(m.eq/kg), AV(m.eq/kg), Fe(Ppb). These were done by
using the American Oil Chemists Society (AOCS), standard
test methods.
2.3.1 Specific Gravity:
The Method of the AOCS, Cc 5 – 25, [18]was used. Specific
gravity of a raw oil sample is the ratio of density of the oil
relative to that of water. It is a measure of the level of purity of
an oil sample, since it involves conversion of volume into
weight. The aim is to determine the level of adulteration in a
given sample of oil.
Procedure: 100g of the raw oil sample was heated to a
temperature of 50o
C and weighed. It was then introduced into
a known weight of density bottle and the bottle reweighed.
The oil was removed and the density bottle washed thoroughly
with detergent and rinsed with distilled water to avoid
contamination. It was then refilled to the same level with
water and the weight taken. The specific gravity was then
calculated as;
Sp.gr.=
C5atwaterofvolumeequalofDensity
C50atoilofDensity
o
o
12
2.3.2 Refractive Index:
The Method of the AOCS, Cc 7 – 25, was used. Refractive
Index is used to determine the angle by which a beam of light
is bent when passing through a thin film of melted fat or oil. It
helps to check the quality in terms of purity of the raw oil
sample.
Procedure: 5 drops of melted raw oil sample were placed on
the face of a prism block. The latter was closed and tightened
firmly. The temperature was maintained at 40o
C. The
illuminating light in the refractometer was adjusted such that
the field dwindling line was coincident with the cross line. At
this setting, the reading on the gauge indicated the refractive
index of the oil sample.
2.3.3 Moisture Content:
The Method of the AOCS, Cc 4 – 25, was used. Moisture in
raw oil sample causes hydrolysis of the sample, and this in
turn aggravates rancidity and oxidation at high temperatures.
The result of oxidation is the formation of hydro - peroxides
and other cleavage products like aldehydes and ketones.
Experiment is aimed at determining the amount of water
contained in a given sample of raw oil as impurity.
Procedure: 10g of the raw oil sample was weighed into a
beaker containing stirring rod. The beaker, with the stirring
rod, was weighed empty and also with the oil sample. Heating
was done using the Bunsen burner, and drops of acetone were
gradually added to accelerate evaporation of moisture. The
system was maintained at a maximum temperature of 105o
C.
The complete evaporation of water was indicated when no
more bubbles were observed. The beaker, with its content, was
allowed to cool and was reweighed. The difference in weight
showed the weight of water that has evaporated, and it was
calculated as;
Moisture(%)= 100x
sampleoilcrudeofWeight
MoistureofWeight
13
2.3.4 Phosphorous:
ALFA – LAVAL[19] , was used. The phosphorous in the oil
sample was determined by ashing. The phosphate obtained
was transferred into phosphomolybdate which was reduced to
a blue - coloured compound. The concentration of the blue
compound was determined by comparison with blue coloured
glass disks.
Procedure: 5g raw oil sample was weighed into a platinum
dish, and 0.5g calcium oxide added and both ashed. The ash
was dissolved in 10 - 15 cc of 2N hot dilute hydrochloric acid,
and filtered into a 100cc volumetric flask. The dish was
washed into the volumetric flask, filtered, and made up to
l00cc. A blank experiment was similarly prepared, but with no
oil sample present. 5cc of the filtrate was taken in a tube and
2cc and 1 cc of molybdate and hydroquinone solution added in
that order. The mixture was allowed to stand for 5 minute for
the green phosphate colour to develop. 2cc of
carbonate/sulphate solution was quickly added and stirred,
(CO2 evolved). Both the test experiment and its blank were
placed in the comparator against a uniform light for
comparison. The result was reported as; Phosphorous (Ppm)

__________________________________________________________________________________________
10000vxG
0.3265xVxA
14
Where A - comparator scale reading (Ppm), V - volume of ash
solution, v - volume of ash solution taken for the colour
development, and G - weight of oil sample.
2.3.5 Determination of Color:
The Method of the AOCS, Ca 20 – 25, was used. Color
pigments present in raw Palm kernel oil include; chlorophyll
and gossypol. The aim is to find the amount of colour pigment
present in the oil sample.
Procedure: Lovibond Tintometer with 1-inch cell was used
for the analysis of colour, and the latter read in terms of red or
yellow colour band that matched the colour of the oils.
2.3.6 Free Fatty acid:
ALFA – LAVAL , [19],was used. Free fatty acid results from
chemical or enzymatic hydrolysis of the fatty acid glycerides.
Its presence in oil sample is a measure of the quality of the
oils, unrefined or refined. The aim of the experiment was to
determine the degree of enzymatic hydrolysis of triglycerides
of the fatty acid that has occurred.
Procedure: 2.8ml of oil of an unknown FFA content was
measured into a conical flask and diluted with 25ml of
ethanol. A drop of phenolphthalein was added. This was
titrated against 0.1N sodium hydroxide until a permanent pink
colour was registered, and the results recorded as;
10W
NxMxV
FFA(%) 15
Where, N - Normality of NaOH; V - Volume of NaOH, W -
Weight of oil sample, M - Molecular weight of oil sample
used.
2.3.7 Oxidation products
Kharasch et al method, adapted from [19], was used. When an
unsaturated fatty acid chain reacts with air at room
temperature, (a process known as autoxidation),
hydroperoxides are formed. At high temperature, these
peroxides break down to hydrocarbons, aldehydes and
ketones. These cleavage products impart odour and flavor to
oil and must be removed.
2.3 8 Determination of Peroxide Value;
This is a measure of primary oxidation whose product is
hydrocarbons. These hydrocarbons are further oxidized to
water, which causes rancidity of the oil on storage. The aim is
to determine the degree of primary oxidation which the oil has
passed through.
Procedure: 30ml of chloroform - glacial ethanoic acid
mixture in the volume ratio of 1:2 was transferred to a conical
flask connected to a reflux condenser. The mixture was then
heated to boiling and the vapor condensed in the lower part of
a jacketed tube. When the reflux became steady, about 1.6ml
of potassium iodide was added from the top of the condenser.
The precipitate of KI was dissolved by adding 5 drops of
water. The mixture was heated for 5 minutes and 2ml of the
oil was pipetted into the mixture through the top of the
condenser also. The pipette was rinsed with 2ml of chloroform
into the boiling mixture, and boiling continued for 5 minutes.
50ml of distilled water was added, and 2ml of the sample was
then titrated with 0.02N thiosulphate solution, using starch
solution as indicator. The result was reported as;
PV =
G
1000xNxV
16
Where V - vol. of thiosulphate used (ml), N - normality of
thiosulphate solution, and G - vol. of sample (ml).
2.3.9 Determination of Anisidine value:
This measures the amount of secondary oxidation in a raw
sample of oil. Its products are aldehydes and ketones, whose
oxidation induces higher rancidity effect to the oil. The
experiment is aimed at determining the level of secondary
oxidation which the oil has passed through.
Equipment /Materials and Procedure; These are the same
as for the PV, except that the temperature at which these
cleavage products were formed was higher
2.3.10 Determination of Iodine value:
Wij’s method, adapted from[19] , was used. It is a measure of
the degree of unsaturation of fat or oil. It is the number of
gram of iodine which combines with the unsaturated bonds in
100g of fat. The aim is to determine the degree of unsaturation
of fatty acid in the oil.
Procedure: 0.5g of oil sample was weighed into a 250ml
stoppered bottle, and 15ml of chloroform was added to
dissolve the oil. 25ml of Wij’s solution was added and mixture
was kept in a dark place and allowed to stand for 30minutes.
At the end of the period, 20ml of 15% KI solution was added.
This was shaken vigorously and the inside of the stoppered
bottle was washed with 1ml of boiled and cooled water. This
was titrated against a standard 0.1N sodium thiosulphate
solution. The reagent was added with constant shaking until
the yellow colour of the iodine almost disappeared. 2ml of 1%
starch solution indicator was added before the titration was

__________________________________________________________________________________________
continued. The end point was reached when the blue colour of
iodine disappeared. A blank experiment was also carried out
with equal proportion of the Wij’s solution. The iodine Value
was calculated as;
W
100x0.1269xNS)(B
IV

 17
Where B - Volume of the standard Na2SO3 S - Volume of
the standard Na2SO3 required for sample, N - Normality of
the standard sodium thiosulphate solution, W - Weight of Oil
sample in gram.
2.3.11 Determination of Iron (Fe)
Method of Cock and Rede, [20] , was used. Iron is a metal
element which, with copper, induces oxidation of the
unsaturated fats and oils at the double bond. Removal of iron
will reduce the rate of oxidation reaction at the high
temperature of deodorization. The aim is to determine the
degree at which oxidation of the fatty acid is expected in raw
palm kernel oil.
Procedure: 0.2mg Fe stock solution was pipetted into 100ml
conical flask. 10ml of 10% hydroxylamine hydrochloride was
added. The solution was diluted with water and mixed. 10ml
of 0.25% phenolphthalein indicator was added and allowed to
stand for 15 minutes and diluted to the mark. Using about 5ml
test tube, the transmittance was read in spectrophotometer 20
at 510 UV light. The results of characterization experiments
are presented in table 5
2.4 Diffusion parameters
2.4.1 Diffusivity
Equations 7, 8, and 9 were used to determine the diffusivity,
diffusion rate, and mass transfer coefficient, respectively.
Using 7, MB was calculated from the normal boiling point of
the oil which is oleic or luaric acid. For oleic and lauric acids
with the molecular formulae of C17H33COOH, and
C11H23COOH, the molecular weight is, 282Kg/Kmole and
241 Kg/Kmole respectively. Temperature, T=333.15K.
Viscosity of pure substances is given by, μ = 0.324 bρt ,
[21] [22], where the normal boiling point of oil tb = 225.31K,
and density of oil at this point ρtb = 0.87g/cm3
, μ =
0.324(0.87)0.5
= 0.000302Kg/m.s Molar volume VB, of the oil
at the normal boiling point is obtained by summing up the
atomic volume of elements that make up the compound,
C11H23COOH, Treybal, (1980), and the value was found to be
0.2812m3
/Kmole. Hence, DA = 4.0 x 10-9
m2
/s
2.4.2 Diffusion Rate
The rate at which the fatty acid diffuses into the solvent is
given in equation 8. Now DA = 4.0 x 10-9
m2
/s, MHexane =
86 Kg/Kmole, Moil = 241Kg/Kmole. It shall be assumed that
the interphase is small and taken to be 0.001mm. The
concentration of oil in solution of hexane at the meal side is
assumed to be about 2%, the residual value, and 1% on the
other side. Oil is glycerol ester of fatty acid, called
triglycerides. As such, it may be assumed that the fatty acid
composition of PKO is predominantly that of lauric acid.
Hence we may use the density of glycerol ester as presented
by Perry and Chilton, (1973), to compute the densities of 2%
and 1% palm kernel oil respectively to be used in computing
the rate of diffusion of oil into the solvent.
At 333.15K, density of 2% and 1% solution of oil as lauric
acid is 1000.4Kg/m3
, and 998Kg/m3
, respectively, ( Perry and
Chilton,1973). Now mole fraction of oil in one side of the
interphase
xA1 = 0.0072, xB1 = 0.9927.
Molecular weight of mixture
M1 = 87.123Kg/Kmol
3
1
1
Kmole/m483.11
123.87
4.1000
M
ρ






Similarly, xA2 = 0.0275, xB1 = 0.9725.
3
2
2
Kmole/m216.4
74.236
998
M
ρ






Average molecular weight of the mixture
3Kmole/m849.7
M
ρ









av
Logmean of mole fraction,
  9823.0
9725.0
9927.0
ln
9725.09927.0
xB 

LM
This gave
.s2Kmole/m710x47.6 AN

__________________________________________________________________________________________
2.4.3 Mass Transfer Coefficient
The mass transfer coefficient is given by Eq. 9 as ;
NA = Kx(xA1 - xA2), [11],
where Kx - Mass transfer coefficient in the liquid phase.
Now Eq. 8 is,
NA =
 
 A2A1
avLmB12
AB
xx
M
ρ
xzz
D







Combining Eqs 8and 9, we have;
Kx=
avLmB12
AB
M
ρ
)(xzz
D







= .s2Kmole/m510x2.3 
Similar calculations were carried out for other solvents and
results presented in Table 3.
3. DISCUSSION
3.1 Extraction of Palm Kernel Oil From Oil Nuts
Ashton, [23], observed that a number of solvents can be used
to leach out oil from the prepared seed. He noted that
efficiency of extraction depends on a number of factors,
including, extraction time, quantity of solvent, temperature of
the system and the nature of solvent used.
Extraction time is a fundamental factor that affects the
efficiency of oil extraction. This was explained by the
observed relationship between the time of extraction and the
residue oil in the cake. This relationship is given in Fig.1.
Fig. 1 Effect of extraction time on Oil Extraction
From the figure, it is noted that greater amount of oil was
extracted during the first 30min of extraction. An extremely
long extraction time is required to obtain a residue oil content
of less than 1 percent. Every type of solvent behaves
differently during the extraction process. The extraction time
relate to quantity of oil extracted by a linear function down to
a residual oil content of about 5%, below which the function
has a curvilinear nature as shown.
3.1.1 Quantity of Solvent
Fig.2 shows the effect of solvent quantity on extraction
efficiency at constant temperature and time
Fig.2 Effect of quantity of solvent on Oil Extraction
From the figure, it was observed that efficiency of extraction
first increases with increase in solvent and after reaching a
limit, the increase becomes marginal. The quantity of solvent
used to reduce the residue oil content to less than 1% varies
with the type of solvent as shown.
In Fig.3, the result of the effect of temperature on efficiency of
solvent extraction is presented.

__________________________________________________________________________________________
Fig.3 Effect of Temperature on Oil Extraction
3.2 Properties of the extracted palm oil
3.2.1 Nature of Solvent
Figs. 1-3 also show the effect of solvent nature on the
extraction process, that is , their dissolving power. The
following were kept constant during the experiment; amount
of palm kernel grit processed, time of extraction, solvent
quantity and extraction temperature. From the figures, it was
found that hexane and Benzene have the same solvent
efficiency. Carbondisulfide has a higher solvent efficiency
than Hexane and Benzene, but lower efficiency than
trichloroethylene. It may be erroneously concluded that
trichloroethylene has the highest extraction efficiency and
hence the most suitable, but quality of oil is also an important
factor which must be given due consideration
The results of characterization experiments are shown in table
2 for the extracted oils. The result showed that with increase
in solvent temperature, the efficiency of extraction is
improved.
From Table 2, it may be observed that colour of the extracted
oil, when extraction was done with Hexane and Benzene, was
light yellow compared with deep yellow obtained with
Carbondisulfide and Trichloroethylene. This colour position
was because, the latter have very high efficiency of extraction
than the former, but also higher degree of impurities like
phosphatide in the extracted oil. Odour is characteristic palm
kernel oil odour, but taste is of palm kernel oil. The specific
gravity increases in the order H-B-C-T, but marginal with
Hexane and Benzene than Carbondisulfide and
Trichloroethylene. The same judgement passed for the melting
point, moisture, refractive index, Phosphorous, Iron, FFA,
Peroxide and Anisidine values, and, of course Iodine value.
Lovibond Tintometer Red colour reading confirmed the
physical appearance where Hexane and Benzene extracted oils
were found to be lighter in colour than Carbondisulfide and
Trichloroethylene.

__________________________________________________________________________________________
Table 2 Physio-Chemical Properties of the palm Oil extracted with Hexane, Benzene, Carbon disulfide and Trichloroethylene
3.2 Diffusion parameters
Diffusivity, Diffusion rate, and Mass Transfer
Coefficient.
The diffusion parameters for the diffusion of Palm oil from
meal into the solvent were calculated from Eqs 7, 8 and 9. The
viscosity used in Eq 7 was given by Perry, (1973), as
bt 324.0 = 0.000302Kg/m.s. The calculated
diffusivity of 4.00 x
10-9
m2
/s was used in Eq. 8 to determine the rate of diffusion
which was found to be 4.67 x 10-7
Kmole/m2
.s. The mass
transfer coefficient was obtained by comparing Eq 9 with 8
and was found to be 3.20 x 10-5
Kmole/m2
.s for Hexane.
Similar calculations were carried out and the results of table 3
obtained.
Table 3 Diffusion rate, Diffusivity and Mass Transfer Coefficient for the diffusion of Palm oil into solvents
From the table, it is observed that Hexane and Benzene have
almost the same diffusion rate which is lower than those of
Carbondisulfide and Trichloroethylene. This explains why
extraction is more with the latter than the former. However,
diffusivity of Benzene is marginally higher than that of
Hexane. This shows that the penetrating power of Benzene is
more than that of Hexane, but lower than Carbondisulfide and
Trichloroethylene. Since Extraction rate depends on time,
Temperature , quantity and nature of solvents used, it then
follows that same is proportional to diffusion rate. Diffusivity
is inversely proportional to viscosity, but directly proportional
to Temperature, other factor being constant. Extraction is
essentially a diffusion process and the amount of oil extracted
depends on the rate of diffusion of the oil into the solvent.
Diffusion rate of liquids is given in Eq 8 as;
 21 AXAX
avmz
D
AN 







Where D – Diffusivity, m2
/s
Properties n- Hexane Benzene Carbondisulfide Trichlorothylene
Colour (Physical Appearance) Light Yellow Light Yellow Light Yellow Light Yellow
Odour Characteristic of
PKO
Characteristic of
PKO
Characteristic of
PKO
Characteristic of PKO
Taste PKO taste PKO taste PKO taste PKO taste
Specific Gravity 0.8752 0.8757 0.8756 0.8759
Melting Point(°C) 29 29 30 31
Moisture (%) 1.12 1.12 1.13 1.135
Refractive Index 1.4400 1.4401 1.4402 1.4403
Free Fatty Acid (%) 4.2 4.3 4.35 4.38
Lovibond Red Unit (1" Cell) 4.2 4.3 4.53 4.54
Anisidine Value (M.eq/kg) 6.5 6.5 6.95 7.25
Peroxide Value (M.eq/kg) 2.45 2.43 2.44 2.48
Phosphorous (Ppm) 6.5 6.51 6.52 6.54
Iron (Ppb) 3.5 3.45 3.52 3.48
Saponification Value 240 241 242 243
Iodine Value 22 22 23 24
Diffusion Parameters Hexane Benzene Carbondisulfide Trichloroethylene
Diffusion rate (10-7
) Kmol/m2
.s 4.00 4.10 4.50 4.90
Diffusivity(10-9
)m2
/s 6.47 6.72 6.78 7.00
Mass Trans. Coef (10-5
) Kmol/m2
.s 3.20 3.50 3.80 4.00

__________________________________________________________________________________________
It follows that rate of diffusion, NA is proportion to
Diffusivity, D. But Diffusivity foe liquid substances is given
by;
6.0)(
5.0)(
AV
TM
D


 . When the two equations are combined
we obtain,
 216.0)(
5.0)(
AXAX
avmzAV
TM
AN 









,
from which we obtain,
 21 AXAX
z
TD
AN 

 ,
where
D
avMzAV
M





 


6.0)(
5.0)(
However, (ZA1-XA2) is fixed based on the extracted oil, then
TDAN  , where, )21( AXAXDD  . It therefore
means that rate of diffusion (extraction) of oil is proportional
to temperature, and hence the higher the temperature the
higher the extracted oil.
CONCLUSIONS
The results of our experiments have shown that the various
solvents used in the extraction of oil from their oil seeds, have
different performance capacities.
The results (Figs 1-3) show that Carbondisulfide is the most
efficient in terms of yield, followed by Trichloroethylene ,
when the purity of the final product is not the major factor of
consideration. N-hexane and Benzene are ideal for extraction
when oil of premium quality is required. Infact, they are
needed in the extraction of oils for edible purposes. The results
also showed that time, and temperature of extraction, as well
as, the nature and quantity of solvent used play a vital role on
the extraction yield.
REFERENCES
[1] 4,000-year old “kitchen” unearthed in Indiana. Archaeo
News. January 26, 2006. Retrieved 30 – 12 – 2011.
[2] Mahatta, T.L (1975), Technology of refining oils and fats,
Small Business Publications (SBP) Building 4//45 Roop
Nagar, Delhi.
[3] Anderson A.J.C. (1962), “Refining of oil for edible
purposes,” 2nd
ed. Pergamon Press London.
[4] Pahl G, (2005), Biodiesel growing a new energy
economy,White River Junction VT; Chelsea Green Publishing
Co.
[5] Janet Bachmann, Oil seed Processing for Small – Scale
Producers. http://www.attra.org/attar-pub/oilseed.html.
Retieved 2012 – 10 - 11
[6] “Kalu (Oil presser)” Banglapedia. Retrieved 12 – 11 -
2012.
[7] Rydberg, J., Musikas, C.,Choppin, G.R., (1992), Principles
and practice of solvent extraction, Marcel Dekker
[8] Gurnhan and Masson (1975), Industrial Engineering
Chemistry, 38, 1309
[9]Davie, and Vincent, (1980), Extraction of vegetable oils
and fats, Simon Rosedown Ltd, Cannon Str. Hull, U K.
[10]Tray and Bilbe,((1987), Solvent Extraction of vegetable
oils, Chem. Eng. (N.Y),54,5,139.
[11] Treybal, R. (1980), Liquid – Liquid extraction, McGraw
– Hill New York.
[12] Bird, B.R., Warren, E.S ., and Edwin, N.L (2003),
Transport Phenomena, John Wiley and Sons New York
[13] Perry, R.H., and Chilton, C.H., (1993), Chemical
Engineering Handbook, 5th
ed. McGraw – Hill New York.
[14] Fan and Co workers, (1987)Industrial Engineering
Chemistry, 38,905.
[15] Treybal, R. (1980), Liquid – Liquid extraction, McGraw
– Hill New York.
[16] Mahatta, T.L (1987), Technology of refining oils and fats,
Small Business Publications (SBP) Building 4//45 Roop
Nagar, Delhi.
[17] Walter, L.B. ,James, T.B. (1975), Vegetable Oil
Extraction, Theory and Procedure, Vicka Publishing House,
PVI Ltd, New Delhi.
[18] AOCS Official Method Cd 8 - 53. Surplus peroxide
Value Acetic acid – Chloroform method Definition; AOCS
Cold Spring Harbour, New York, NY, U.S.A
[19] ALFA-LAVAL (1978), Analysis of Fats and Oils,
Laboratory manual, Sweden.
[20] Cocks, L.V and Rede, V.C, (1966), Laboratory Handbook
for Oil and Fat Analysis, Academic Press London.
[21] Perry, R.H., and Chilton, C.H., (1973), Chemical
Engineering Handbook, 5th
ed. McGraw – Hill New York.
[22] Egbuna S.O., (2005), The Fundamentals of Mass
Transfer, Sege Ventures, Pp. 8 – 22, ISBN 978-37147-9-1.
[23] Ashton, N.F, (1983), Handbook of Chemistry, 2nd
ed.
W.H Freeman and Company, New York, Pp280-480.

Diffusion rate analysis in palm kernel oil extraction using different extraction solvents

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Diffusion rate analysis in palm kernel oil extraction using different extraction solvents (20)

More from eSAT Journals (20)

Recently uploaded (20)

Diffusion rate analysis in palm kernel oil extraction using different extraction solvents