Content uploaded by Habis Al-Zoubi
Author content
All content in this area was uploaded by Habis Al-Zoubi on Nov 25, 2017
Content may be subject to copyright.
Chemical Engineering Journal 76 (2000) 61–65
Short communication
Extraction of jojoba oil by pressing and leaching
M.K. Abu-Arabi1,∗, M.A. Allawzi, H.S. Al-Zoubi, A. Tamimi
Department of Chemical Engineering, Jordan University of Science and Technology, Irbid, Jordan
Received 9 October 1998; received in revised form 8 June 1999; accepted 31 August 1999
Abstract
Jojoba oil extraction by pressing alone, pressing followed by leaching, and leaching alone were investigated. The extraction process by
first and second pressing followed by leaching gave about 50% by weight oil with reference to total seed, which is in agreement with what
has been reported previously. The extraction by leaching process was carried out using different solvents. These solvents were; hexane,
benzene, toluene, petroleum ether, chloroform, and isopropanol. Hexane, benzene, and petroleum ether gave the highest yield (all about
50% by weight oil with reference to total seed), but when cost is considered, petroleum ether is recommended as the best solvent to leach
jojoba oil. The yield obtained in this work for leaching by hexane and benzene are 3–5% and about 10% for isopropanol more than those
reported in the literature. Traces of solvent remained with the extracted oil after simple distillation followed by a second stage distillation
via a Rotavapour apparatus. These traces slightly affected some of the oil properties such as pour point and flash point. ©2000 Elsevier
Science S.A. All rights reserved.
1. Introduction
Jojoba (Simmondsia chinensis) is unique among plants
in the fact that its product (seeds) contains about 50% by
weight oil, which is more than twice the amount in soybeans
and somewhat more than in most oilseed crops. The oil is
composed mainly of straight chain monoesters in the range
of C20–C22 as alcohol and acids, with two double bonds,
one at each side of the ester bond [1–3].
Interest in jojoba oil stems from its unusual properties
that differ from all known seed oils. The complete absence
of glycerin makes it a liquid wax, not fat. Jojoba oil has
been evaluated for suitability in many applications such as
cosmetics, pharmaceuticals, lubricants, food, electrical insu-
lators, foam control agents, high-pressure lubricants, heat-
ing oil, plasticizers, fire retardants, and transformer oils plus
others [1–3].
Different methods, similar to those applied to other
oilseeds, have been used for extraction of jojoba oil from
the seeds [3–6]. Those methods are mainly mechanical
pressing, mechanical pressing followed by leaching (solvent
extraction), or leaching only. Some of the oilseeds require
pretreatment/preparation such as cleaning, dehulling, crush-
∗Corresponding author. Fax: +968-697-107
1On leave at The Middle East Desalination Research Center, P.O. Box
21, Al Khuwair, Postal Code 133, Sultanate of Oman.
ing, flaking, cooking, etc. before the extraction process.
Hexane is the solvent most commonly used in the leaching
process because of its relatively low cost and low toxicity.
Other organic solvents such as benzene, alcohol, chloroform
are also used. Water as a solvent was evaluated to extract
oil from soybean, but gave low yield and high potential of
oil microbiological contamination [6–8].
Knoepfler et. al [9] used carbon tetrachloride, benzene,
hexane, heptane, isopropyl alcohol and tetrachloroethylene
to leach jojoba oil. They first cracked jojoba seeds into
8–12 pieces by passing them through corrugated crush-
ing rolls. They were then chopped into flakes of 0.010in.
(0.25mm) average thickness by passing them through a pair
of smooth rolls. A Soxhlet extractor was used to extract the
oil from the flakes. They found that carbon tetrachloride,
benzene, hexane, and heptane extracted between 45–47%
by weight (based on total seed). Spadaro et al. [10] in-
vestigated the conditions required for material-preparation
and extraction to get efficient leaching of oil from jojoba
seeds using a filtration-extraction process that was previ-
ously used for other oilseeds. The seeds were subjected to;
flaking to different thickness (0.004–0.01in.), cooking in
a mixer type cooker, crisping by evaporative cooling for
20min., slurrying the cooked material with solvent. The
following filtration-extraction characteristics were studied;
flake thickness, moisture content (which was varied be-
tween 5–20%) of flakes during cooking, rerolling of cooked
1385-8947/00/$ – see front matter ©2000 Elsevier Science S.A. All rights reserved.
PII: S1385-8947(99)00119-9
62 M.K. Abu-Arabi et al./Chemical Engineering Journal 76 (2000) 61–65
flakes, and extraction temperature on mass velocity and
extraction efficiency using commercial-grade hexane and
heptane as solvents. Tests with heptane showed that in-
creasing the extracting temperature from 80F (26.7◦C) to
140F (60◦C) and/or increasing the moisture content up to
10%, increased the extraction efficiency by about 3%. Per-
forming the extraction at 140F and increasing the moisture
content up to 20%, increased the extraction efficiency by
less than 1%. Actually, their data shows no improvement
with increasing the moisture content above 10%. For the
80F extracting temperature, increasing the moisture content
to 10, 15, and 20%, increased the efficiency by 2.6, 2.9, and
3.1%, respectively. They reported that the adequate mass
velocity, defined as pounds of miscella filtrate per hour per
square foot of filter area, for commercial application should
be greater than 2000. For the 140F extracting temperature,
when the moisture content was increased to 10%, the mass
velocity dropped to 1617. For the 80F, the mass velocity
dropped below 2000 when the moisture content was in-
creased to 20%. Rerolling was found not required if the
flakes are rolled initially to a thickness of 0.004in.. Based
on their results, they concluded that both heptane and hex-
ane are suitable as solvents for commercial extraction of
jojoba flakes as there are no significant differences in mass
velocity and extraction efficiency obtained with the two
solvents. However, they recommended using hexane rather
than heptane, because it is more readily available, cheaper
and has a lower boiling point, which facilitates its removal
from the products.
They also conducted three experiments to determine
the filtration-extraction characteristics of uncooked jojoba
flakes. The results were erratic in that the mass velocity var-
ied from about 200 and 1600 and extraction efficiency from
96.5 to 97.8. So they concluded that omission of the cooking
step is not recommended in any commercial operations.
Rawles [11] reported on mechanical seeds grinding and
pressing done at the Western Regional Research Center
(WRRC), Albany, California, and at the San Carlos Apache
Indian Reservation during 1972–1977. Many problems were
encountered during the grinding operations that were tested.
As the mill heated up, a sticky, mud-like meal formed that
plugged the grinder. They were able to solve these prob-
lems by freezing the seed before grinding, but this is very
expensive.
The effect of moisture and temperature on the efficient op-
eration of a Rosedowns press, made in England, was inves-
tigated. The best results were obtained when operating be-
tween 175–190F (80–88◦C), and with 3–4% moisture con-
tent. The operation of the press required very careful moni-
toring to avoid loss of oil and plugging. An extraction rate
of 38.2% was obtained, and the oil content remaining in
the meal was 17–20%. A Hander press, made in Japan, was
used for pressing at the San Carlos reservation. Experience
with it indicated that, feeding 20% hulls by weight with the
seed improved the extraction efficiency. Double pressings
were necessary to bring the extraction rate up to 42% from
35–39% in a single press. Also preheating was required to
get this extraction rate. The remaining meal, after double
pressing, contained 9–10% oil. Ruiz et al. [12] also used the
same model Hander press and concluded that moisture con-
trol during the feeding process was critical. Feeding with
4% moisture content, they were able to extract 80% of the
original oil in one pass and up to 94% in two passes.
Miller et al. [13] reported on the mechanical rendering
of jojoba oil by grinding and pressing the seeds. The de-
hulled seeds were grounded at room temperature by using
8in. Bauer single-disk attrition mill modified by the addi-
tion of two L-shaped case wipers to the rotating disk to pre-
vent plugging. Pressing was done by passing preheated feed
(80–90F) through a Rosedowns press. Pressing with about
4% moisture content gave 31.4% oil yield (based on total
seed weight).
Spadaro and Lambou [14] investigated mechanical ex-
traction and solvent-extraction of jojoba oil. The mechani-
cal extraction (cold hydraulic pressing) was performed by
cracking the seeds into 6–10 pieces, and then flaking to
about 0.025 in. thick. The flakes were charged to a six-stack,
pilot-plant model hydraulic press and pressed for 50min at
4400 lb ram pressure. They collected 40 lb of oil from 130 lb
of seed (30.8% oil). In the solvent-extraction, six solvents
were used: carbon tetrachloride, benzene, isopropyl alcohol,
heptane, hexane, and tetrachloroethylene. The process was
carried out, after cracking and flaking the seeds, in a Soxhlet
extractor. A total of 20–24 solvent passes were used. Ex-
traction was done at the boiling point of the used solvent.
Separation of oil from solvent was conducted under vacuum
at 3–6mm Hg for 2 h. The stripped oil was dried in a vac-
uum oven at 105◦C for 2h. They reported only the results
of three of the solvents, which they considered to have the
most potential. Their results as well as others’ results are
shown in Table 2.
Lanzani et al. [15] have reported that a wet process tech-
nology was applied to jojoba seeds to obtain oil and detox-
ified protein meal. The yield of oil was less than 20% by
weight with reference to total seed.
The objectives of this work are: (i) to study oil extraction
from jojoba seeds grown in Jordan by pressing alone without
pretreatment, pressing followed by leaching with hexane,
and leaching with different solvents, (ii) to investigate the
chemical and physical properties of the pressed oil and the
oil leached by hexane, and (iii) to find out if there are any
differences between them.
2. Equipment and experimental procedures
2.1. Pressing only
To obtain the jojoba oil by pressing, a manually oper-
ated hydraulic press type (P/N, 15.011), made by SPECAC
Limited (UK), with variable load (0–15) metric tons was
used. For each run, about 80g of whole jojoba seeds were
M.K. Abu-Arabi et al./Chemical Engineering Journal 76 (2000) 61–65 63
placed in a cylindrical container (6.52cm inside diameter),
then were subjected to the press load. The amount of oil
collected was considered as the weight of oil extracted. A
second pressing was done on some samples by removing it
from the cylindrical container after the first press, breaking
the disk–like residue and then putting it again in the cylin-
drical container and subjecting it to a second pressing. Both
the first and second pressing were done at room temperature.
2.2. Pressing followed by leaching
In this case, the jojoba seeds were subjected to first
and second pressing as mentioned above, then the sample
was taken out of the cylindrical container and crushed to
<1.0mm average size. The crushed sample was leached
by hexane using Soxhlet extractor. The total amount of oil
extracted is the sum of that obtained by first and second
pressing and from the leaching step.
2.3. Leaching only
A Soxhlet extractor was employed for the leaching ex-
periment. For each run, 30g of crushed jojoba seeds was
charged into the Soxhlet extractor and 150ml of organic sol-
vent was used. The solvents used were; hexane, petroleum
ether, isopropanol, benzene, toluene, and chloroform, which
were laboratory grade. Leaching was carried out at the boil-
ing point of each solvent until a clear liquid was obtained
from the jojoba, which indicated complete leaching of the
leachable oil. The Soxhlet extraction process took about
18 h. The extracted phase (oil and solvent) was then distilled
in two stages to separate the oil and solvent. The first stage
was a simple distillation followed by a second stage, which
was a Rotavapour apparatus. A vacuum pump was attached
to the Rotavapour apparatus to ensure complete removal of
the solvent. The oil produced by this method was compared
with the pure pressed oil by measuring their properties.
3. Results and discussion
First, the amount of oil extracted by pressing was found
as a function of the press load. Fig. 1 shows the oil yield
(by weight with reference to total seed) with pressure. The
amount extracted increases exponentially as the pressure in-
creases, but as the pressure is increased to 35.4MPa, it starts
to level off indicating that this is about the required load to
get the maximum amount extracted by pressing. This type
of data is needed for the mechanical design of any press
to be used in large-scale production. Further pressing runs
in this study were subjected to 35.4MPa pressure. The ex-
perimental data was fitted to a third polynomial to give the
following equation with R2=0.987.
% oil =0.0009 P3−0.0761 P2+2.7086P
where Pis the pressure.
Fig. 1. Jojoba oil yield (wt.% with reference to total seed) as a function
of press load.
The results obtained from first and second pressing and
followed by leaching using hexane as a solvent are shown in
Table 1. The first pressing gave about 35.4%, while the sec-
ond pressing gave about 8.4% by weight oil. Both presses
were subjected to 35.4 MPa pressure. Leaching process gave
about 6.7% by weight oil. As can be seen, there is a slight
difference in the percentage of oil recovered for the different
trails. This is expected, because there will be a slight differ-
ence in the amount of oil present in each seed and the seeds
have different sizes. Table 2 shows the results of pressing
obtained in this work and those reported in the literature,
which are in agreement.
3.1. Leaching process
The leaching process depends mainly on the chemical
structure of the solvent and the kind of solute that will be ex-
tracted from solid material. This probably follows the prin-
ciple ‘like dissolves like’. This does not mean that both
the solute and the solvent have to be chemically similar
but rather have similar functional groups. Solvents are usu-
ally classified by the number of functional groups present
in their molecules, which affect the interaction of either or
both types of physical and chemical interactions between
the solute and the solvent [16].
According to the above principles, jojoba oil extraction
depends on the kind of organic solvent and its structure.
Jojoba oil is a nonpolar ester compound with a very long
straight chain structure. Therefore, any solvent that has a
similar structure (and nonpolar compound) will leach more
oil. Table 3 shows the yield of different solvents used in this
work and those reported in the literature. Solvents’ proper-
ties and their cost [17] are also shown in Table 3. This work
results of the jojoba oil yield are the average of four runs.
Hexane, petroleum ether, and benzene gave a high percent
of yield because they are nonpolar hydrocarbon compounds.
Chloroform, which is polar compound, leached a lower per-
64 M.K. Abu-Arabi et al./Chemical Engineering Journal 76 (2000) 61–65
Table 1
Yield of jojoba oil after first and second pressing followed by leaching process using hexane.
No. of Mass of Mass of oil after Mass of oil after Mass of Leached Total oil,
trial seeds, g first pressing, g (% oil) second pressing, g (% oil) oil, g (% oil) g (% oil)
1 90.0 34.1 (37.9) 8.0 (8.9) 4.3 (4.8) 46.4 (51.5)
2 90.0 31.4 (34.9) 7.1 (7.9) 7.3 (8.1) 45.8 (50.9)
3 90.0 30.3 (33.7) 7.7 (8.6) 6.3 (7.0) 44.3 (49.2)
Average 90.0 31.9 (35.4) 7.6 (8.4) 6.0 (6.7) 45.5 (50.6)
SD 1.95 0.46 1.53 1.08
Table 2
Pressing results
Source % oilaType of press No. of pressing Pretreatment or treatment processes
This work 35.4 hydraulic (manually operated) single none
43.8 hydraulic (manually operated) double breaking the disk-like residue from first press
Rawles [11] 38.2 Rosedowns press single preheating, hulls and moisture contents were adjusted
35–39 Hander press single preheating, hulls and moisture contents were adjusted
40–42 Hander press double preheating, hulls and moisture contents were adjusted
43 Hander press triple preheating, hulls and moisture contents were adjusted
Ruiz et al. [12] 80–94bExpeller Hander EX-100 single flaking, preheating, hulls contents were adjusted, moisture
content=4%
50–70bExpeller Hander EX-100 single flaking, preheating, hulls contents were adjusted, moisture
content=6%
Spadaro and Lambou [14] 30.8 hydraulic press (pilot-plant) single preheating, cracking, flaking
Miller et al. [13] 31.4 Rosedowns press single grinding, preheating, moisture content = 4%
aBased on total seed weight.
bThis is % extracted from original oil, the whole oil content in the seeds is not reported.
Table 3
A comparison between the organic solvents used in the leaching process
Solvent Structure Refractive Boiling Specific Color No. of Cost of the % Leached oil
index (25◦C) point (◦C) gravity leached oil solvent ($/L)
(G/ml) (ASTM) This work Literature values
Hexane C6H14 1.3723 68.7 0.670 0.5 25.0 52 48.8 [14], 46.9 [9]
Petroleum ether a mixture of 1.3787 60–100 0.656 L 2.0 11.2 50 –
(ligroin) hydrocarbona
Benzene C6H61.4972 80.1 0.874 L 1.5 28.0 49.3 45.5 [9]
Chloroform CHCl31.4459 61.2 1.490 L 2.0 25.8 32.5 –
Isopropanol CH3HCOHCH31.3772 82.2 0.785 1.5 19.0 45 36.1 [14], 35.1 [9]
Toluene C6H5CH31.4941 110.5 0.867 3.5 17.0 44.8 –
Carbon tetrachloride CCl4–76.5 1.584 – – – 45.7 [9]
Tetrachloroethylene C2Cl4– 121.1 1.620 – – – 42.3 [9]
Heptane C7H16 –98.4 0.684 – – – 48.1 [14], 46.5 [9]
aMainly hexane and heptane.
centage of jojoba oil than the above solvents. Isopropanol
and toluene leached about the same amount. The yield ob-
tained in this work for hexane, benzene, and isopropanol are
higher than those reported in the literature. This could be
due to the solvents’ purity and the time of leaching. The re-
sults obtained in this work for hexane, petroleum ether, and
benzene show that these solvents are able to get the same
(or slightly more in the case of hexane) the amount obtained
by first and second pressing followed by leaching.
Petroleum ether is the best solvent used in this research
since its cost is relatively low and it leached a high per-
centage of oil, while hexane gave the highest yield but
its cost is relatively high. Benzene gave the same yield
as that of petroleum ether, but its cost is the highest one.
Toluene and isopropanol can be ranked third with regard
to the amount of oil leached. Toluene has a high boiling
point so it needs more heat to vaporize in any distillation
recovery process compared to the other solvents. The oil
leached by toluene has different color number, as shown
in Table 3, from those obtained by other solvents which
probably means that toluene may extract materials from
the seeds other than the oil, such as sugar, pigments, etc.
Chloroform leached the least amount of oil, also its cost
is relatively high. Additionally, chloroform has high spe-
cific gravity that resulted in the flotation of the jojoba meal
during the leaching process. Overall, petroleum ether can
be classified as the best solvent based on the foregoing
discussion.
M.K. Abu-Arabi et al./Chemical Engineering Journal 76 (2000) 61–65 65
Table 4
A comparison between the properties of pressed and leached oil by hexane
Characteristics Observed value of Observed value of
(ASTM No.) pressed jojoba oil leached jojoba oil
(literature value) oil
Flash point -open cup, ◦C
(D 92) 275 (295) 267.0
Aniline point, ◦C (D 611) 52.9 (N/A) 52.7
Pour point, ◦C (D 97) 8.0 (9.0)a6.0
TAN, mg KOH/g (D 974) 0.36 (N/A) 2.89
TBN, mg KOH/g (D 2896) 1.0 (N/A) 1.0
Ash content, wt.% (D 482) 0.0 (0.0)a0.0
Color number (D 1500) 1.0 (1.5) 0.5
Refractive index (D 1218) 1.4593 (1.465) 1.4600
Viscosity, Cst (D 446)
at 40◦C 24.75 (24.92)a26.16
at 100◦C 6.43 (6.43)a6.57
Viscosity index (D 2270) 233 (233)a233
aThese values were taken from reference [18], while the other liter-
ature values were taken from reference [2].
Jojoba oil produced from the leaching process by organic
solvents is different from that produced by the pressing pro-
cess since the leached oil is not pure (i.e., some of the or-
ganic solvents do not separate completely from oil during
distillation process). In Table 4, the properties of jojoba oil
leached by hexane and the pressed oil are shown. The pour
point of the leached oil is 2◦C less than that of the pressed
oil meaning that some long chain paraffin’s were extracted
less by hexane, and therefore the pour point of the leached
oil would be reduced. Flash point is another indicator of the
presence of traces of hexane because the flash point of hex-
ane is much less than that of jojoba oil. Thus, traces of hex-
ane will reduce the flash point of the leached oil compared
to that of the pressed oil. The other properties did not sig-
nificantly change. Further treatment to remove these traces
from the oil is needed, if it is to be used in cosmetics, phar-
maceuticals or any other products that may come in contact
with living tissues.
4. Conclusions
The yield of jojoba oil (with reference to total seed) that
can be obtained by pressing is about 44% by weight and by
first and second pressing followed by leaching is about 50%
by weight. The pressure required in the hydraulic press is
about 35.4MPa to obtain the maximum amount by press-
ing. The six organic solvents used in the leaching process
leached different amounts depending on the type of solvent
(polar or nonpolar) and its structure. Hexane leached the
highest amount of oil followed by Petroleum ether, ben-
zene, isopropanol, toluene, and then chloroform. When cost
is considered as a parameter, petroleum ether is the best sol-
vent to be used in the leaching process. Further investigation
is needed to determine the percentage of solvent recovery
and how much losses take place to be able to really decide
which is the best solvent. Traces of solvents remained with
the oil after simple distillation followed by a second stage
distillation via a Rotavapour apparatus. These traces affect
some of the oil properties.
References
[1] J. Wisniak, J. Hillel, O. Katz, Hold up and extraction characteristics
of jojoba meal, J. Am. Oil Chem. Soc. 64(9) (1987) 1352–1354.
[2] National Research Council, Jojoba: New Crop for Arid Lands, New
Material for Industry, National Academy Press, Washington, D.C.,
1985
[3] J. Wisniak, in: The Chemistry and Technology of Jojoba Oil,
American Oil Chemists’ Society, Champaign, IL, USA, 1987.
[4] D.K. Salunkhe, J.K. Chavan, R.N. Adsule, S.S. Kadam, in: World
Oilseeds: Chemistry, Technology, and Utilization, Van Nostrand
Reinhold, New York, 1992.
[5] D. Kimber, D.I. Mcgregor (Eds.), in: Brassica Oilseeds: Production
and Utilization, CAB International, Wallingford, England, 1995.
[6] J.R. Wilcox (Ed.), Soybeans: Improvement, Production, and Uses,
2nd edn., American Society of Agronomy, Madison, Wisconsin,
1987.
[7] J.T. Lawhon, L.J. Manak, K.C. Rhee, K.S. Rhee, E.W. Lusas,
Combining aqueous extraction and membrane isolation techniques
to recover protein and oil from soybeans, J. Food Sci. 46 (1981)
912–916.
[8] L.A. Johnson, E.W. Lusas, Comparison of alternative solvents for
oil extraction, J. Am. Oil Chem. Soc. 60 (1983) 229–242.
[9] N.B. Knoepfler, E.J. McCourtney, L.J. Molaison, J.J. Spadaro, J.
Am. Oil Chem. Soc. 36 (1959) 644.
[10] J.J. Spadaro, P.H. Eaves, E.A. Gastrock, J. Am. Oil Chem. Soc. 37
(1960) 121.
[11] L.R. Rawles, in: Proceedings of the Third International Conference
on Jojoba and its Uses, University of California, Riverside, CA,
September 1978, pp. 279–283
[12] A. Ruiz, E. Flores, M.A. Kuljasha, in: Proceedings of the Third
International Conference on Jojoba and its Uses, University of
California, Riverside, CA, September 1978, pp. 285–289.
[13] R.E. Miller, R.H. Edwards, G.O. Kohler, J. Am. Oil Chem. Soc. 54
(1979) 556.
[14] J.J. Spadaro, M.G. Lambou, in: Proceedings of the First International
Conference on Jojoba and Its Uses, Tucson, Ariz., Office of Arid
Lands Studies College of Earth Sciences, University of Arizona, June
1972.
[15] A. Lanzani, P. Bondioli, A. Brillo, M. Cardillo, E. Fedeli, A. Ponzetti,
G. Pieralisi, A wet process technology applied to jojoba seed to
obtain oil and detoxified protein meal, J. Am. Oil Chem. Soc. 68(10)
(1991) 772–774.
[16] C. Reichardt, in: Solvent and Solvent Effects in Organic Chemistry,
VCH Weinheim, Germany, 1990.
[17] A. Bader, D. Harvey, J. Nagarkatti, in: Aldrich, Aldrich Chemical
Company, USA, 1990–1991.
[18] R.P. Bisht, G.A. Sirasankaram, V. Bhatia, Additive properties of
jojoba oil for lubricating oil formulations, Wear 161 (1993) 193–197.
