Bangladesh J. Sci. Ind. Res.

42(3), 287-298, 2007

Extraction of Gasohol Grade Ethanol from Aqueous

Solution using Gasoline as Solvent
M. A. Rahman, M. Asadullah,* M. S. Rahman, M. N. Nabi and M. A. K. Azad

Department of Applied Chemistry and Chemical Technology

University of Rajshahi, Rajshahi 6205, Bangladesh

Abstract
Liquid-liquid phase equilibria of ethanol - water - gasoline (‘iso-octane’, ‘octane’
‘petrol’, ‘kerosene’ and ‘diesel’) were experimentally determined and the ternary
phase diagrams of the systems were studied with an end in view to develop methods
of producing ‘gasohol’ by a once through operation. Binodal curves of the different
systems were, therefore, obtained to give quantitative ideas about the extent of extrac-
tion of alcohol from aqueous solution using gasoline. Phase equilibrium data, distri-
bution coefficient and separation factor were obtained experimentally for ethanol-
water-gasoline systems. The order of the heterogeneous region of the solvents is iso-
octane .‘octane’ .‘petrol’ < ‘kerosene’ < ‘diesel’. The equilibriation and settling time
for ethanol - water - solvent systems are about 10 and 20 min, respectively. The sol-
ubility of ethanol in the solvents is in the order of iso-octane .‘octane’.‘petrol’ >
‘kerosene’ > ‘diesel’. The distribution coefficient increases with increasing ethanol
concentration in the feed; however, to the contrary the separation factor decreases.
The distribution co-efficients are lower than 1 and the separation factors are greater
than 10 for all of the solvents.

Key words : Gasohol, solvent, solvent extraction, gasoline and liquid-liquid equilibria.

Introduction

According to the projections from the World fossil fuel sources is threatening to the ener-
Energy Council and others world's estimated gy depended future generation. Therefore,
recoverable oil reserves of 1 trillion barrels alternative sources of energy are seriously
and natural gas reserves of 5000 trillion cubic sought (Tomishig and Asadullah, 2005).
feet (Tcf) should both be available through Possibilities of production of hydrocarbons
out the 21st century and proved recoverable as substitute of natural gas and gasoline has
reserves of coal of 984 billion tons at the been studied by hydrogenation of carbon
present rate of production could be available dioxide, CO2 (Reverse Water Gas Synthesis,
for more than 200 years. This limitation of RWGS) (Choi et al. 1996) and reforming of
*Correspondence should be made : [email protected]; [email protected]
288 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

carbon monoxide, CO (Fischer-Tropsch tion may be avoided by using gasoline as sol-

Synthesis, FT) have been studied (Zhang et vents. In the present work possibility of
al. 2002 and Wang et al. 2003). A valuable extraction of ethanol from its dilute aqueous
renewable source of energy could be the bio- solution using gasoline as solvent has been
mass (agricultural wastes, forestry wastes studied.
and city wastes) whose pyrolysis and gasifi-
cation may yield liquid and gaseous products Experimental
which may be used as diesel and natural gas
Materials: Gasolines commercially known
alternative (Bridgwater 1996). Unfortu-nate-
as ‘octane’, ‘petrol’, ‘diesel’ and ‘kerosene’
ly, high technology is involved in the afore-
were obtained from the local gas station and
said process due to unavoidable use of het-
characterized by determining their boiling
erogeneous catalysts and enormous volumes
ranges, densities and refractive indices as fol-
of research are yet to be done before these
lows.
processes could be exploited commercially.
Gasoline Boiling Density, Refractive
Petroleum dependence for gasoline may be
reduced to some extent by using mixture of range, O C g/cm3 index, ηD20
ethanol and gasoline. A mixture of 10 vol % Iso-octane 99.5 0.693 1.3843
ethanol and 90 vol % gasoline popularly ‘Octane’ 150-180 0.701 1.4250
known as ‘Gasohol’ is in use commercially ‘Petrol’ 168-200 0.729 1.4110
as a good substitute of gasoline (Pimentel ‘Diesel’ 250-320 0.820 1.4536
2003). But the conventional method of distil- ‘Kerosene’ 200-260 0.795 1.4387
lation and azeotropic distillation for the
Dilute ethanol solutions were prepared from
recovery of ethanol to an almost water free
rectified spirit and water.
state as required for ‘gasohol’ from a renew-
able source like fermentation broth is an Equipment : Electric shaker (Model: 1866;
energy intensive operation; energy require- Jencons Scientific Ltd., England), Thermo-
ment is about 27,000 Btu/USgallon of static water bath (Model: SB-3; Grants
ethanol (Tomishig and Asadullah 2005). Instruments Ltd., England) and Gas
Liquid-liquid extraction of ethanol from its Chromatograph (Model: Shimadzu GC-14B)
dilute solutions could be a cheap means for were used in this work.
separation and purification of ethanol and a
number of solvents have been tried for this Procedures : All the extractions were carried
purpose (Rahman et al. 1995) but again the out by shaking two phases (aqueous and
problem of separation of solvent from organic) contained in closed Erlenmeyer
ethanol is encountered here. Solvent separa- flasks. The flask was tightly closed and fixed
 Rahman, Asadullah, Rahman, Nabi and Azad 289

to a shaft of a electrically driven flask shaker. solvent to construct the binodal curve on the
Then the flask was placed in a thermostatic solvent side.
bath. The temperature range of the thermo-
static bath was 0O C to 100O C with a con- The samples from each phase were taken at
troller accuracy of ± 0.1O C. The shaker had the end of each experiment and analyzed by
a maximum capacity of eight flasks to be means of a Shimadzu GC-14B gas chromato-
shaken at a time. The shaking speed was con- graph equipped with a flame ionization
trolled by a regulator. detector. A 30m × 0.25mm i.d. Alltech
capillary column (coated with ATTM-5,
For extraction process, precise liquid-liquid 30m×0.2mm i.d.) was used isothermally for
equilibrium data is required. The cloud-point component separation. The column and
method was used for producing the data for detector temperatures were maintained at 120
the binodal curve of the ternary system. For and 250O C, respectively. For the quantitative
this experiment 10 cm3 of water was taken measurement, n-heptane was used as an
into a 125 cm3 closed Erlenmeyer flask and internal standard. The amount of each
solvent (gasoline) was added from a burette component was measured in weight, which is
and agitated till the solution started to appear then converted to weight fraction.
turbid. The solution was kept undisturbed
for about 2 h at a constant temperature For the separation of ethanol from water, the
(25± 0.1O C). The amount of solvent added feed to ethanol ratio and equilibrium contact
was recorded as the maximum solubility of time were optimized. For the optimum feed
the solvent in the water and gave the first to ethanol ratio, the experiments were
point of the binodal curve on the base line. performed using the solution of 20 cm3 water
The appearance of the turbidity indicated the and 10 cm3 ethanol in four different 250 cm3
beginning of the formation of the second closed Erlenmeyer flasks. The different
phase, the solvent layer. Therefore, the fur- amounts of solvents were added into the
ther addition of a small amount of solvent flasks and the flasks were shaken for about 4
gave a heterogeneous mixture. Then ethanol h and allowed to settle for about 2 h at
was added from a burette until the first 25±0.1O C. For the determination of optimum
appearance of distinct clear homogeneity. contact time, the flasks were shaken for
This gave another point of binodal curve on different times from 1 to 30 h. The samples of
the triangular diagram. This procedure was aqueous and organic phases were analyzed
repeated to construct the binodal curve on the by gas chromatograph, Shimadzu GC-14B.
aqueous side. Same procedure was applied
starting with an initially measured quantity of
290 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

Results and Discussion and 1e. From this figure it is seen that the
binary systems of water-solvents are immis-
Determination of the composition of organic cible and the binary systems of ethanol-
and aqueous phases solvents are miscible at temperature, (25 ±
0.1O C). It is also seen that the miscibility of
The equilibrium extraction of ethanol from
water in solvent is larger than that of solvent
dilute aqueous solution using commercial
in water. The miscibility of solvent in water
gasoline- ‘octane’ ‘petrol’ ‘kerosene’ and
is very low. The miscibility of solvent - water
‘diesel’ was carried out and compared with
binary systems decreases with increasing
that of analytical grade iso-octane at 25 ± 0.1O
number of carbon atoms in the chain of
C. The compositions of the organic and aque-
Hydrocarbon solvents. It is also observed
ous phases experimentally determined are
that the ascending order of heterogeneous
given in Table I. The concentrations of
region of the binodal curves for the solvents
ethanol and solvents were determined by gas
can be presented as ‘octane' . iso-octane >
chromatography and the concentration of
‘kerosene’ > ‘diesel’.
water was determined by using a triangular
coordinate plot with one side as a base; the Effect of contact time on ethanol extraction
left vertex represents 100 % water, the right
vertex represents 100 % solvent, and the top The effect of contact time on ethanol extrac-
vertex represents 100 % solute. Other percent tion has been studied at temperature,
lines for the respective components are laid (25±0.1O C). These results are summarized in
off at proportional distances along the Table II. Distribution coefficient of ethanol
altitude joining each vertex with the opposite (KD) has been plotted against phase contact
side; the length of this altitude is taken as time for ethanol-water-solvent systems in
100 %. The sum of the vertical distances to Figure 2. It is seen from this figure that the
the three sides from any point within the value of the distribution coefficient of
triangle is also equal to 100 %, from geomet- ethanol increases with increasing the phase
ric considerations; thus the plotting of the contact time and attains a maximum value at
values for the percentages of any two 10 min. So the equilibrium time can be taken
compounds of a ternary system determines a as 10 min, but 30 min have been allowed to
point in the triangle, fixes the percentage of ensure equilibration. It is further seen that
the third component, and checks the corre- dispersed phases separated into two distinct
sponding arithmetic. phases within 20 min, after shaking, but 30
min have been allowed to ensure setting.
The binodal curves of ethanol-water-solvent
systems are plotted in Figures 1a, 1b, 1c, 1d
 Rahman, Asadullah, Rahman, Nabi and Azad 291

Table I. Compositions of equilibrium phases of ethanol - water - hydrocarbon systems at

30 ± 0.1O C
Solvent Composition of water rich phase, (wt %) Composition of solvent rich phase, (wt%)
Water Solvent Ethanol Water Solvent Ethanol
‘Octane’ 100.0 0 0 0.0 100.0 0.0
74.0 0.3 25.7 0.4 81.8 17.8
57.6 0.5 41.9 1.0 73.0 26.0
46.8 0.7 52.5 1.6 62.5 35.9
35.8 1.7 62.5 3.3 43.4 53.3
29.8 2.4 67.8 4.3 34.4 61.3
20.4 5.2 74.4 6.1 24.4 69.5
16.2 7.5 76.3 7.3 19.3 73.4
13.8 8.8 77.3 8.3 16.7 75.0
12.4 9.9 77.7 10.0 13.4 76.6
Iso-octane 100.0 0.0 0.0 0 100.0 0
47.2 0.6 52.2 2.4 41.4 56.2
34.8 1.4 63.8 4.5 30.7 64.8
29.0 2.4 68.6 6.7 23.1 70.2
24.4 3.3 72.3 9.8 16.8 73.4
18.0 7.4 74.6 11.8 13.4 74.8
14.8 10.1 75.1 - - -
‘Petrol’ 100 0 0 0 100.0 0
73.2 0.2 26.6 0.6 81.7 17.7
57.6 0.5 41.9 1.0 71.3 27.7
53.5 0.8 45.7 1.5 56.8 41.7
42.2 1.8 56.0 3.6 43.6 52.8
28.6 2.1 69.3 4.7 34.7 60.6
24.6 3.6 71.8 6.8 25.1 68.1
20.0 5.8 74.2 9.7 17.7 72.6
17.3 8.0 74.7 11.5 14.2 74.3
15.8 9.2 75.0 13.1 12.0 74.9
‘Kerosene’ 100.0 0 0 0 100.0 0
36.3 0.1 63.6 0.2 47.4 52.4
29.1 0.2 70.7 0.6 42.7 56.7
25.3 0.4 74.3 0.9 35.5 63.6
18.2 1.4 80.4 2.7 21.3 76.0
15.1 2.3 82.6 3.9 15.6 80.5
10.9 5.1 84.0 7.2 9.3 83.5
‘Diesel’ 100.0 0.0 0.0 0 100 0
26.1 0.1 73.8 0.2 31.8 68.0
22.6 0.2 77.2 0.4 28.9 70.7
20.4 0.3 79.3 0.6 25.2 74.2
14.5 1.2 84.3 1.9 15.9 82.2
11.7 1.9 86.4 2.9 11.9 85.2
8.4 4.1 87.5 5.5 7.5 87.0
292 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

Ethanol Ethanol

20 80 20 80

P
40 40
a 60 60

60 40 60 40
P

80 20 80 20
b

Water 20 40 60 80 iso-octane Water 20 40 60 80 1-Octane

Ethanol
Ethanol

P
20 80
20 80

40 60
40 60

c
60 40 60 40

20 80
d 20
80

Water 20 40 60 80 Diesel Water 20 40 60 80 Petrol

Ethanol

20 80
P

40 60

e
60 40

80 20

Water 20 40 60 80 Kerosene

Fig. 1. Binodal curve and tie lines for ethanol-water-solvent systems at 25 ± 0.1O C.
a: ethanol-water-iso-octane; b: ethanol-water-1-octane; c: ethanol-water-diesel;
d: ethanol-water-petrol and e: ethanol-water-kerosene.
 Rahman, Asadullah, Rahman, Nabi and Azad 293

Table II. Effect of contact time on the composition of organic and aqueous phases

Solvent Phase Composition of Composition of organic Distribution

contact organic phase, (wt %) phase, (wt %) coefficient
time,(min) of ethanol
XW XS XE XW XS XE KD
Octane 0.50 0.08 97.12 2.80 51.30 0.80 47.90 0.058
2.00 0.09 97.01 2.90 54.50 0.70 44.80 0.065
5.00 0.10 96.90 3.00 55.90 0.60 43.50 0.069
10.00 0.10 96.80 3.10 56.70 0.60 42.70 0.073
15.00 0.10 96.80 3.10 56.70 0.60 42.70 0.073
20.00 0.10 96.80 3.10 56.70 0.60 42.70 0.073
25.00 0.10 96.80 3.10 56.70 0.60 42.70 0.073
30.00 0.10 96.80 3.10 56.70 0.60 42.70 0.073
Iso- 0.50 0.08 99.02 0.90 77.80 0.30 21.90 0.041
octane 2.00 0.09 98.91 1.00 80.30 0.20 19.50 0.051
5.00 0.10 98.80 1.10 82.80 0.10 17.10 0.064
10.00 0.10 98.70 1.20 83.40 0.10 16.50 0.073
15.00 0.10 98.70 1.20 83.40 0.10 16.50 0.073
20.00 0.10 98.70 1.20 83.40 0.10 16.50 0.073
25.00 0.10 98.70 1.20 83.40 0.10 16.50 0.073
30.00 0.10 98.70 1.20 83.40 0.10 16.50 0.073
‘Petrol’ 0.500 0.09 98.21 1.70 49.20 0.90 49.90 0.034
2.00 0.10 97.70 2.20 80.3 0.80 47.60 0.046
5.00 0.11 97.50 2.30 82.80 0.70 45.00 0.051
10.00 0.11 97.50 2.39 83.40 0.70 43.30 0.055
15.00 0.11 97.50 2.39 83.40 0.70 43.30 0.055
20.00 0.11 97.50 2.39 83.40 0.70 43.30 0.055
25.00 0.11 97.50 2.39 83.40 0.70 43.30 0.055
30.00 0.11 97.50 2.39 83.40 0.70 43.30 0.055
‘Keros 0.50 0.03 99.47 0.50 80.14 0.06 19.80 0.025
ene’ 2.00 0.04 99.31 0.60 81.95 0.05 18.00 0.033
5.00 0.04 99.28 0.65 83.06 0.04 16.90 0.038
10.00 0.04 99.28 0.68 83.85 0.04 16.11 0.042
15.00 0.04 99.28 0.68 83.85 0.04 16.11 0.042
20.00 0.04 99.28 0.68 83.85 0.04 16.11 0.042
25.00 0.04 99.28 0.68 83.85 0.04 16.11 0.042
30.00 0.04 99.28 0.68 83.85 0.04 16.11 0.042
Diesel 0.500 0.01 99.74 0.25 81.05 0.05 18.90 0.013
2.00 0.02 99.72 0.26 82.26 0.04 17.70 0.015
5.00 0.02 99.71 0.27 82.97 0.03 17.00 0.016
10.00 0.02 99.70 0.28 83.70 0.03 16.27 0.017
15.00 0.02 99.70 0.28 83.70 0.03 16.27 0.017
20.00 0.02 99.70 0.28 83.70 0.03 16.27 0.017
25.00 0.02 99.70 0.28 83.70 0.03 16.27 0.017
30.00 0.02 99.70 0.28 83.70 0.03 16.27 0.017
294 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

Fig. 2. Effect of contact time on distribution coefficient of ethanol-water-solvent systems.

♦)1- octane, ( ) iso-octane, (S) petrol, (×) kerosene and (*) diesel.
(♦

Compositions of coexisting phases in the respectively. Similarly the distribution coeffi-

ethanol-water-solvent systems cient of water KW is the ratio of concentration
of water in organic and aqueous phases,
The equilibrium tie line data of the, ethanol- respectively. These results are given in Table
water-iso-octane, ethanol-water-‘octane’, III. These data allowed us to draw the corre-
ethanol-water- ‘petrol’, ethanol-water-‘kero- sponding equilibrium distribution curves in
sene’, and ethanol-water-‘diesel’ systems are Figure 3 and equilibrium separation factor in
determined at (25± 0.1O C). The composi- Figure 4. Figure 3 shows that the concentra-
tions of the coexisting phases are presented tion of ethanol in organic phase increases
in Table III. The distribution coefficient and with increasing concentration of ethanol in
separation factor between the coexisting liq- aqueous phase.
uid phases have been calculated, separation
factor is determined numerically from the tie Since the concentration of ethanol in iso-
line data because it is the ratio of distribution octane, ‘octane’, ‘petrol’ ‘diesel’, and
coefficient of ethanol to the distribution coef- ‘kerosene’ is lower than that in water - rich
ficient of water. The distribution coefficient phase; water has stronger affinity for ethanol
of ethanol KD is the ratio of concentration of than iso-octane, ‘octane’, ‘petrol’, ‘diesel’,
ethanol in organic and aqueous phases, and ‘kerosene’. From Figure 1 it is seen that
Table III. Compositions of coexisting phases for separation factor in the ethanol-water-solvent systems.
Solvent Ethanol Composition of initial Composition of Composition of aqueous Distribution Distribution Separation
conc. in mixtures, (wt %) organic phase, (wt %) phase, (wt %) coefficient of coefficient of factor
feed ethanol water
wt % Water 1-octane Ethanol YW YS YE XW XS XE KD = YE/XE KW = YW/XW α= KD/KW
Octane 44.1 38.7 30.8 30.5 0.10 96.8 3.1 56.7 0.60 42.70 0.073 0.002 40.56
61.3 29.6 23.6 46.8 0.15 95.00 4.85 39.00 1.30 59.7 0.081 0.004 21.32
70.3 24.00 19.10 56.90 0.20 92.70 7.10 29.80 2.40 67.80 0.105 0.007 15.67
76.00 20.2 16.00 63.80 0.24 90.20 9.56 23.50 4.00 72.50 0.132 0.010 12.94
97.80 17.40 13.80 68.80 0.30 87.70 12.00 19.50 5.70 94.80 0.160 0.015 10.67
Iso- 16.4 52.9 36.7 10.4 0.10 98.7 1.2 83.4 0.1 16.5 0.073 0.0012 60.83
octane 28.3 47.9 33.2 18.9 0.1 97.8 2.1 72.8 0.2 27.0 0.078 0.0014 55.71
37.2 43.8 30.3 25.9 0.2 96.6 3.2 63.7 0.4 35.9 0.089 0.003 29.67
44.1 40.3 27.9 31.8 0.2 95.5 4.3 56.7 0.6 42.7 0.101 0.004 25.25
49.7 37.3 25.9 36.8 0.3 93.0 6.7 51.2 0.7 48.1 0.139 0.006 53.17
Petrol 44.2 39.7 28.9 31.4 0.11 97.5 2.39 56.0 0.7 43.3 0.055 0.002 27.50
61.3 30.2 22.0 47.8 0.14 95.9 3.96 38.5 1.9 59.6 0.066 0.004 18.33
70.3 24.4 17.8 54.8 0.17 94.8 5.03 29.3 2.1 68.6 0.073 0.0058 12.59
75.9 20.5 14.9 68.6 0.23 92.4 7.37 23.5 4.0 72.5 0.102 0.0098 10.41
79.8 17.6 12.8 69.6 0.27 90.6 9.13 19.1 6.4 74.5 0.123 0.0140 8.79
Kero- 16.5 50.2 39.9 9.9 0.04 99.28 0.68 83.85 0.04 16.11 0.0422 0.00048 87.91
sene
Rahman, Asadullah, Rahman, Nabi and Azad

28.3 45.7 36.3 18.0 0.05 98.48 1.47 72.35 0.05 27.60 0.0533 0.00069 77.24
37.2 41.9 33.3 24.8 0.06 97.60 2.34 63.44 0.06 36.50 0.0641 0.00095 67.47
44.2 38.7 30.7 30.6 0.07 96.79 3.14 56.73 0.07 43.20 0.0727 0.00123 59.10
49.7 35.9 28.6 35.5 0.09 95.11 4.80 51.42 0.08 48.50 0.0989 0.00175 56.51
Diesel 16.5 49.6 40.6 9.8 0.02 99.7 0.28 83.7 0.03 16.27 0.0172 0.0002 86.0
28.3 45.2 37.0 17.8 0.03 99.07 0.90 72.3 0.04 27.66 0.0325 0.004 81.25
37.3 41.4 34.0 24.6 0.04 98.5 1.46 62.9 0.05 37.05 0.0394 0.0006 65.67
44.2 38.3 31.4 30.3 0.05 97.49 2.46 56.4 0.06 43.54 0.0565 0.0009 62.78
49.7 35.6 29.2 35.2 0.06 96.48 3.46 51.0 0.07 48.93 0.0707 0.0012 58.92
295
296 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

0.2

0.16

0.12
KD

0.08

0.04

0
0 50 100 150
Ethanol concentration in feed, wt%

Fig. 3. Effect of ethanol concentration in feed on distribution co-efficient of ethanol.

♦)1- octane, ( ) iso-octane, (S) petrol, (×) kerosene and (*) diesel.
(♦

♦)1- octane, ( )
Fig. 4. Effect of ethanol concentration in feed on separation factor of ethanol. (♦
iso-octane, (S) petrol, (×) kerosene and (*) diesel.
 Rahman, Asadullah, Rahman, Nabi and Azad 297

the two - liquid - phase regions of ethanol - of the separation factor is petrol < ‘octane’ <
water - iso-octane, ethanol - water - ‘octane’ iso-octane < ‘kerosene’ .‘diesel’.
and ethanol - water - ‘petrol’ are almost the
same. The concentration of ethanol in iso- Conclusions
octane-rich phase is higher than those in
The binary systems of water - hydrocarbon
‘octane’ and ‘petrol’-rich phase respectively;
solvents are partially or completely immisci-
and the negative gradients of the tie lines of
ble, and the binary systems of solvent -
Figures 1a and 1d are sharper than those of
ethanol are miscible at temperature, (25 ±
Figure 1b. This shows the polarity order is
0.1)O C. Therefore, these solvents may be
1-octane > iso-octane > ‘petrol’ > ‘kerosene’
used as effective solvents for the extraction
> ‘diesel’ in order of strength. Therefore, the
of ethanol from aqueous solution with an
increase in number of carbon atoms in the
opportunity of their subsequent separation
chain of solvents results in wider heteroge-
from the product. The order of the heteroge-
neous zones and in decreasing solubility of
neous region of the solvents in the binodal
the ethanol in the organic phase. The solubil-
curve is ‘octane’ . iso-octane . ‘petrol’ <
ity of ethanol in different higher hydrocar-
‘kerosene’ < ‘diesel’. The equilibrium time
bons is found in the order of ‘octane’ > iso-
for ethanol - water - solvent systems is about
octane > ‘petrol’ > ‘kerosene’ > ‘diesel’.
10 min and the settling time is about 20 min.
It is seen from the Figure 3 that the distribu- The solubility order of ethanol in the solvents
tion coefficient of ethanol (KD) increases is ‘octane’ > iso-octane > ‘petrol’ >
with increasing concentration of ethanol in ‘kerosene’ > ‘diesel’. The distribution coeffi-
the feed solution and large values of distribu- cient increases with increasing ethanol con-
tion coefficient of ethanol are desirable since centration in the feed; however, the separa-
less solvent will then be required for the tion factor is completely opposite. The distri-
extraction. The order of the KD value for dif- bution co-efficient is lower than 1 for all sol-
ferent hydrocarbon solvents is iso-octane > vents and that of the separation factor is
‘octane’ > ‘petrol’ > ‘kerosene’ > ‘diesel’. greater than 10 for all of the solvents.
This indicates that ethanol has preferential Therefore, it is concluded that gasoline com-
solubility in solvents as desired in the extrac- mercially sold as ‘octane’, ‘petrol’ and
tion process. The separation factor decreases ‘diesel’ in the local gas station can extract
with increasing concentration of ethanol in ethanol from its dilute aqueous solutions and
the feed solution as shown in Figure 4. The this will enable one to produce ‘gasohol’ by a
separation factors are considerably greater once through operation without involving
than 10 for all the solvents studied. The order separation of solvent.
298 Extraction of Gasohol Grade Ethanol from Aqueous 42(3) 2007

References Tomishig, K. and Asadullah, M. (2005) In

Progress in Catalysis Research. Edited by
Bridgwater, A.V. (1996) An Introduction to Fast Bevy, L. P. (Nova Science Publishers, lnc.,
Pyrolysis of Biomass for Fuels and New York) pp 1.
Chemicals, Fast pyrolysis of biomass: A
hand book, edited by Bridgwater, A.V. Wang, C.Xu. and L. Wang, Q. (2003) Review of
Czernik, S. Diebold, J. Meier, D. Oasmaa, Directly Producing Light Olefins via CO
A. Peacocke, C. Piskorz, J. Radlein, D. hydrogenation. J. Nat. Gas Chem. 12 : 10-
(CPL Scientific Publishing Services 15.
Limited, Newbury) pp 1-19.
Zhang, Y. Jacobs, G. Sparks, D.E. Dry, M. E. and
Choi, P.H. Jun, K.W. Lee, S.J. Choi, M.J. and Lee, Davis, B.H. (2002) CO and CO2 hydrogena-
K.W. (1996) Hydrogenation of carbon diox- tion study on supported cobalt Fischer-
ide over alumina supported Fe-K catalysts. Tropsch synthesis catalysts. Catal. Today 71
Catal. Lett. 40(1-2), 115-118. (3) : 411-415.

Pimentel, D. (2003) Natural Resources Research. Received : December 03, 2006;

2nd Ed. (Springer, Amsterdam), pp127. Accepted : August 22 , 2007

Rahman, M.A. Rahman, M.S. and Asadullah, M.

(1995) Production of fuel grade ethanol
from aqueous solution by liquid-liquid
extraction using vegetable oils as solvents.
Indian J. Chem. Tech., 2, 90-92.