Separation of acylglycerides obtained by enzymatic esterification using solvent extraction

Article abstract of DOI:10.1007/s11746-013-2374-1

New avenues to add value to glycerol are currently being explored. One of them is the synthesis of structured lipids through glycerol esterification. In this work we have analyzed the recovery and purification of dicaprin obtained by esterification of glycerol with capric acid (C) in heptane, mediated by Lipozyme RM IM. This is an intermediate step to obtain lipids MLM. In the first stage, the diglyceride synthesis MGM (being G a central HC-OH) was carried out. When M = C, the diglyceride is CGC. Recovery of the diglyceride CGC is required to carry out the esterification of the sn-2 position with palmitic acid (P), thus obtaining the triglyceride CPC. Different solvents were evaluated using Ecofac 1.0 (a molecular design software solvent) through a theoretical approach to explore the best solvents for the acylglycerides separation. Then, the performance of the selected solvents to separate dicaprin from mono and tricaprin was experimentally studied in a liquid-liquid extraction process. Previously, the remaining fatty acid had been neutralized. With liquid-liquid extraction in three simple steps, using ethanol/water, 94 % of the dicaprin obtained by enzymatic esterification was recovered with a purity of 89 % (wt%). It was also possible to obtain dicaprin with a purity of 97 % but with a yield of 56 %.

Full text of DOI:10.1007/s11746-013-2374-1

J Am Oil Chem Soc (2014) 91:261–270
DOI 10.1007/s11746-013-2374-1
ORIGINAL PAPER
Separation of Acylglycerides Obtained by Enzymatic
Esterification Using Solvent Extraction
•
Daniel Alberto Sanchez Gabriela Marta Tonetto
•
´
Marıa Lujan Ferreira
´
´
Received: 30 April 2013 / Revised: 11 September 2013 / Accepted: 14 October 2013 / Published online: 7 December 2013
ꢀ AOCS 2013
Abstract New avenues to add value to glycerol are cur-
rently being explored. One of them is the synthesis of
structured lipids through glycerol esterification. In this
work we have analyzed the recovery and purification of
dicaprin obtained by esterification of glycerol with capric
acid (C) in heptane, mediated by Lipozyme RM IM. This is
an intermediate step to obtain lipids MLM. In the first
stage, the diglyceride synthesis MGM (being G a central
HC–OH) was carried out. When M = C, the diglyceride is
CGC. Recovery of the diglyceride CGC is required to carry
out the esterification of the sn-2 position with palmitic acid
(P), thus obtaining the triglyceride CPC. Different solvents
were evaluated using Ecofac 1.0 (a molecular design
software solvent) through a theoretical approach to explore
the best solvents for the acylglycerides separation. Then,
the performance of the selected solvents to separate dica-
prin from mono and tricaprin was experimentally studied in
a liquid–liquid extraction process. Previously, the remain-
ing fatty acid had been neutralized. With liquid–liquid
extraction in three simple steps, using ethanol/water, 94 %
of the dicaprin obtained by enzymatic esterification was
recovered with a purity of 89 % (wt%). It was also possible
to obtain dicaprin with a purity of 97 % but with a yield of
56 %.
Introduction
Structured lipids (SL) containing a mixture of medium-
chain fatty acids (MCFA) and long-chain fatty acids
(LCFA) in the same molecule have been designed for
medical and nutritional purposes. The placement of a long
(L) chain fatty acid at the sn-2 position of glyceryl carbons,
particularly if it is saturated, increases the digestibility of
the MLM lipid, being in the sn-1 and sn-3 positions a
medium (M) chain fatty acid. These SL are reported to be
helpful to the immune system and to improve lipid clear-
ance, having application in the prevention of obesity and its
associated medical conditions [1–3].
The most studied process for obtaining MLM is acid-
olysis between long-chain triacylglycerols (TAG) and
MCFA in a one-step reaction. The most selective catalysts
are some 1,3-specific lipases [4–9]. However, this method
requires an important post-reaction purification because the
desired MLM triglyceride is obtained with other secondary
and undesired products, such as mono and diglycerides,
fatty acids and other triglycerides. Hydrolysis is a sec-
ondary reaction very difficult to avoid using commercial
lipases. In addition, acyl migration is also present, with the
generation of minor amounts of glycerol and MMM
triglyceride.
Keywords Separation ꢀ Fatty acids ꢀ Acylglycerides ꢀ
Extraction ꢀ Liquid–liquid
Several authors have reported a two-step process [10–
13]. In the first step, 2-monoacylglycerols (2-MAG) are
obtained (from vegetable oils or TAG with the desired fatty
acids) by enzymatic alcoholysis using 1,3-specific lipases.
In the second step, MCFA are added in the extreme posi-
tions of 2-MAG by enzymatic esterification with 1,3-spe-
cific lipases. This step can also be carried out with a novel
heterogeneous catalyst.
´
D. A. Sanchez ꢀ G. M. Tonetto ꢀ M. L. Ferreira (&)
´
Planta Piloto de Ingenierıa Quımica (PLAPIQUI), Universidad
Nacional del Sur, CONICET, Camino La Carrindanga Km 7, CC
´
The recovery of the reaction intermediate has also been
studied. Esteban et al. [14] described a process of
´
717, 8000 Bahıa Blanca, Argentina
e-mail: mlferreira@plapiqui.edu.ar
123

262
J Am Oil Chem Soc (2014) 91:261–270
Adsorption of Glycerol on Silica gel
extraction. However, they searched for the recovery of
monoglycerides, separating the ethyl esters. A similar
procedure was described by Irimescu et al. [15] but using a
highly toxic solvent. Hita et al. [9] included capric acid
neutralization as a first stage. However, they did not study
the appropriate concentration of the KOH solution because
the evaluated sample contained only triglycerides.
The enzymatic synthesis of 1,3-diacylglycerols (1,3-
DAG) can be the first step to the desired MLM structured
triglyceride. The metabolic characteristics of DAG are
interesting too, and we recently researched them. Results of
clinical studies showed that DAG can be considered to be
effective in preventing obesity [16, 17] and reduces the
postprandial increase in serum chylomicron TAG levels
[18, 19].
One gram of glycerol and 2 g of silica gel were mechani-
cally mixed until total adsorption on the solid.
Lipase-Catalyzed Esterification
Esterification of glycerol was performed in 10-mL flasks
which were kept in a thermostatic bath with temperature
control and magnetic stirring. The reaction time was 6 h.
One hundred and ten milligrams of capric acid was dis-
solved in 3 mL of n-heptane, then 250 mg of glycerol
adsorbed onto 500 mg of silica were added. This silica
acted as a water-removing surface. When the reaction
mixture reached 60 ꢁC, 50 % of the total amount of
enzyme to use (time 0) was added. The remaining 50 % of
the biocatalyst was introduced at 3 h of reaction. The total
enzyme dosage was 40 mg.
In this work we studied the separation of the reaction
products of the synthesis of 1,3-diacylglycerol, specifically
1,3-dicaprin (CGC). The same was carried out by enzy-
matic esterification of glycerol with capric acid (C) using
as a biocatalyst an immobilized form of a lipase from
Rhizomucor miehei, specific for sn-1 and sn-3 positions
(Lipozyme RM IM). However, to maximize obtaining a
specific triglyceride, it is necessary to purify the reaction
intermediate, since in the first stage, besides specific gen-
erated diglyceride monocaprin, nonspecific diglycerides
(generated by acyl migration) and tricaprin (CCC) are also
obtained.
In all reactions, once the set time was reached the flask
was subjected to 15 min of ultrasound with the aim of
desorbing capric acid and reaction products that could be
adsorbed on the silica. Then the content of the flask was
filtered and the solid material separated. From recovered
liquid phase 100 lL of sample was taken and stored at
-20 ꢁC to analyze its composition by gas chromatography.
The remaining volume was used to examine the separation
of the acylglycerides.
Different solvents were evaluated using a computer
program to select those solvents that allow the dicaprin
separation from the mixture, but are also of low toxicity.
The purpose is the synthesis of products for nutritional
purposes therefore toxic solvents must be avoided. Once
the solvents were selected, the separation of acylglycerols
present in the reaction products was carried out, previously
performing the neutralization of un-reacted capric acid
with a solution of KOH.
Evaluation of Solvents
To study the feasibility of separation of acylglycerols nine
solvents were evaluated using the software Ecofac 1.0
(software of solvent molecular design) based on UNIFAC
interaction parameters, designed by Cismondi and Brignole
[20–22].
The log P is the log of the solubility of the un-ionized
compound in octanol vs in un-ionized water and known
as the distribution ratio. n-Heptane, chloroform, diethyl
ether, t-butanol, propanol, acetic acid, acetone, ethanol
and methanol (ordered in their log P values) were eval-
uated in a process of liquid–liquid extraction at 300 K.
Being glycerol polar and capric acid non-polar their
solubilities in different solvents are very different. This
step was an attempt to explore and confirm the best
solvent for the glycerol esterification reaction considering
mainly the solubility of the compounds present in the
reaction media.
Experimental
Materials
Lipozyme RM IM, which is a commercial form of the 1,3-
specific lipase from R. miehei immobilized by adsorption
on a macroporous anion exchange phenolic resin Duolite
586-A, was kindly donated by Novo Nordisk A/S (Brasil).
Glycerol, n-heptane, acetone, isopropyl ether and silica gel
were supplied by Cicarelli Laboratorios. Capric acid, 1,2,4-
butanetriol and silylation reagents were obtained from
Fluka. Monocaprin, dipalmitin, trilaurin and tricaprin were
supplied by Sigma-Aldrich. Absolute ethanol was supplied
by Dorwil and phenolphthalein and pyridine were provided
by Anedra S.A. All products were of analytical grade.
The Unifac (UNIversal Function Activity Coefficient)
group contribution method [23] and the corresponding
parameters were used for the calculation. This group con-
tribution approach enables one to assume that the mixture
does not consist in molecules but in limited functional
groups of which the interaction parameters are known [24].
123

J Am Oil Chem Soc (2014) 91:261–270
263
c1
B;S MWA
c1A;S MWB
The software enables the performance evaluation of
solvents for the separation of two components, one of
which is extracted with the solvent and the other one is the
main component.
b ¼
ð1Þ
Solvent power is defined as:
1 MWA
The separation mixtures analyzed were:
Sp ¼
ð2Þ
c1A;S MWB
•
•
•
Monocaprin extraction with solvent (from dicaprin and
tricaprin mixtures respectively),
Solute distribution coefficient is defined using the
folowing equation:
Dicaprin extraction with solvent (from monocaprin and
tricaprin mixtures respectively), and
c1
A;B MWB
Tricaprin extraction with solvent (being dicaprin or
monocaprin the main component in the solution).
m ¼
ð3Þ
c1A;S MWS
The parameters that allowed us to evaluate the perfor-
Where
mance of the solvents using Ecofac were selectivity, sol-
vent power and solute distribution coefficient.
MWA is the molecular weight of the compound A (to be
extracted with the solvent)
Selectivity is defined by the following equation:
Fig. 1 Separation stages for
liquid–liquid extraction
MWB
123

264
J Am Oil Chem Soc (2014) 91:261–270
Table 1 Liquid–liquid extraction
Table 3 Liquid–liquid extraction
Solvent
Selectivity Solvent power Distribution coefficient
Solvent
Selectivity Solvent power Distribution coefficient
Methanol
Acetic acid
Ethanol
2.65
1.48
2.61
1.63
0.25
3,595.36
3,139.38
1,183.92
806.76
730.83
513.71
240.40
181.18
0.77
120.28
105.02
39.61
26.99
24.45
17.19
8.04
Chloroform
Diethyl ether
Acetone
108.48
34.16
3.99
26,078.22
6,189.71
2,918.95
2,116.78
1,356.12
493.70
383.22
90.96
42.89
31.11
19.93
7.25
Propanol
Acetone
Acetic acid
Methanol
Propanol
0.67
0.38
tert-Butanol 1.25
Chloroform 0.01
Diethyl ether 0.03
n-Heptane 0.01
0.61
Ethanol
0.38
454.20
6.67
6.06
tert-Butanol
n-Heptane
0.80
409.83
6.02
0.03
81.01
62.44
0.92
Compound to be recovered: monocaprin; major component in the
residual mixture: dicaprin
Compound to be recovered: dicaprin; major component in the residual
mixture: monocaprin
Table 2 Liquid–liquid extraction
Table 4 Liquid–liquid extraction
Solvent
Selectivity Solvent power Distribution coefficient
Solvent
Selectivity Solvent power Distribution coefficient
Methanol
Acetic acid
Ethanol
26.05
6.58
20.95
0.00
8.50
0.23
5.15
0.00
0.00
3,595.36
3,139.38
1,183.92
941.32
806.76
730.83
513.71
240.40
0.77
620.19
541.54
204.22
162.38
139.17
126.07
88.61
Chloroform
0.03
26,078.22
6,189.71
2,918.95
2,116.78
1,356.12
493.70
533.59
126.65
59.72
43.31
27.75
10.10
9.29
Diethyl ether 0.11
Acetone
0.91
4.44
9.83
5.20
8.04
Diethyl ether
Propanol
Acetic acid
Methanol
Propanol
Ethanol
Acetone
tert-Butanol
Chloroform
n-Heptane
454.20
41.47
tert-Butanol 4.11
n-Heptane 0.06
409.83
8.39
0.13
62.44
1.28
Compound to be recovered: monocaprin; major component in the
residual mixture: tricaprin
Compound to be recovered: dicaprin; major component in the residual
mixture: tricaprin
is the molecular weight of the compound B (main
compound of the residual mixture product)
Yield is defined using
DAGi
MWS is the molecular weight of the solvent
c1A;S
c1B;S
c1A;B
is the infinite dilution activity coefficient of
compound A in the solvent
YDAG
¼
ꢁ 100
ð5Þ
DAGf
is the infinite dilution activity coefficient of
compound B in the solvent
where is
is the infinite dilution activity coefficient of
compound A in the compound B
YDAG is yield in dicaprin (Recovered) related to the
initial amount of dicaprin
DAGi
DAGf
initial dicaprin
is
is
Purity (P) was calculated by the following equation:
DAG
recovered dicaprin
P ¼
ꢁ 100
ð4Þ
ðDAG þ TAG þ MAG þ FAÞ
Capric Acid Neutralization
where
DAG
After filtering the contents of the flask and obtaining the
sample for chromatographic analysis, non-reacted capric
acid was neutralized. A volume was withdrawn (3 mL)
and an equal volume of an aqueous solution of KOH was
added. Different KOH concentrations (1, 0.5, 0.25, 0.1
and 0.05 M) were studied to obtain one basic solution that
allows the total acid neutralization but minimizes the
is dicaprin in the ethanolic
phase
DAG ? TAG ? MAG ? FA is total DAG, tricaprin
(TAG), monocaprin
(MAG) and fatty acid
(FA) in the ethanolic
phase
123

J Am Oil Chem Soc (2014) 91:261–270
265
Statistical Analysis of Experimental Error
The percentage error was calculated as.
removal of the compound of interest, in this case,
dicaprin.
After adding the appropriate volume of alkaline
solution, the sample was subjected to intense agitation
and then centrifuged at 8,000 rpm for 15 min. The hep-
tane phase was recovered and 100 lL were withdrawn to
analyze by gas chromatography. The remaining volume
of sample was used for the separation of the
acylglycerides.
Average of different measures ꢂ Individual values
Error % ¼
ꢁ 100
Average value
The average error is the summation of all errors (from
replicates) divided by the total number of experiments.
Reactions were performed in triplicate. The average
error found in the acid titration step was 2.3 %. The
average error between all the chromatographic injections,
including the calibration step, was 1.6 %. All the injections
were performed in duplicate. The neutralization step was
performed in duplicate. The average error in the determi-
nation of the remnant acid in the heptane phase was 3.4 %.
The amount of capric acid that remains after the neutral-
ization step in the organic phase is very low (and this low
concentration increases the relative error of the titration
step).
Liquid–Liquid Extraction
To the solution of n-heptane containing acylglycerides, an
equal volume of an ethanol/water (90/10 v/v) is added. The
small proportion of water allowed phase separation, which
would otherwise not occur. The sample was subjected to
intense agitation allowing contact between the phases and
then centrifuged for 20 min at 8,000 rpm.
The heptane phase was extracted and 200 lL of this
was taken for gas chromatographic analysis. The alco-
hol phase is maintained at -20 ꢁC. This procedure with
the heptanes phase was performed 39 as shown in
Fig. 1.
Table 5 Liquid–liquid extraction
Then, the three recovered ethanolic phases (Fig. 1) were
combined and subjected to an extraction with an equal
volume of n-heptane which eliminated the triglyceride
fraction.
Solvent
Selectivity Solvent power Distribution coefficient
Chloroform
Diethyl ether
Acetone
3,842.57
301.43
4.37
–
–
54,612.23
3,194.51
1,106.07
477.13
138.02
99.73
3,968.04
232.11
80.36
34.67
10.03
7.25
Finally, the two heptane phases were combined and
subjected to a new extraction process with ethanol/water
solution as shown in Fig. 1.
n-Heptane
Acetic acid
Methanol
tert-Butanol
Propanol
1,434.83
0.15
0.04
0.19
Gas Chromatography Analysis
0.12
94.96
6.90
Samples were diluted with pyridine and silylated with N-
methyl-N-(trimethylsilyl) trifluoroacetamide (MSTFA).
The analysis of samples was performed in a PerkinElmer
AutoSystem XL gas chromatograph equipped with on-
column injection, a flame ionization detector (FID) and a
high temperature capillary column ZB-5HT Inferno
(15 m 9 0.32 mm, with an ID of 0.10 lm), using H2 as
gas carrier. The detector temperature was maintained at
380 ꢁC. The initial column temperature was held at 50 ꢁC
for 1 min, elevated to 180 ꢁC at a rate of 15 ꢁC/min, then
the temperature was increased to 230 ꢁC at a rate of 7 ꢁC/
min, further increased up to 370 ꢁC at a rate of 10 ꢁC/min,
and finally held there for 5 min. This method allowed the
detection of capric acid and of all the acylglycerides
present in the sample in \25 min [25]. Results are the
average of two injections with an average relative error
lower than 2 %. The determination of elution times of
reactants and products was performed with high purity
standards.
Ethanol
0.05
56.51
4.11
Compound to be recovered: tricaprin; major component in the
residual mixture: monocaprin
Table 6 Liquid–liquid extraction
Solvent
Selectivity Solvent power Distribution coefficient
Chloroform
Diethyl ether
Acetone
35.42
8.82
1.09
17.71
0.23
0.10
0.24
0.19
0.12
–
–
54,612.23
3,194.51
1,106.07
477.13
138.02
99.73
617.73
36.13
12.51
5.40
1.56
1.13
1.07
0.64
n-Heptane
Acetic acid
Methanol
tert-Butanol
Propanol
94.96
Ethanol
56.51
Compound to be recovered: tricaprin; major component in the
residual mixture: dicaprin
123

266
J Am Oil Chem Soc (2014) 91:261–270
The overall separation process was performed by
mixture solution is identified as ‘‘residual mixture’’ from
here thereafter.
duplicate at similar conditions. The average error of the
first, second and third step were 1.8, 1.3 and 2.1 % (with a
global error for a whole procedure including the three steps
of 5.2 %).
Table 2 shows the results of the evaluation of solvents
that are suitable for the separation of mono- and tricaprin.
Again, solvents highlighted with shadow gray are the best
candidates for this separation. Methanol, acetic acid, eth-
anol and propanol are the best solvents for extraction of
monocaprin, separating it of remaining acylglycerides. Of
these solvents, ethanol would be most appropriate due to its
low toxicity.
Results and Discussion
Solvent Selection
Tables 3, and 4 show the results obtained when evalu-
ating the solvent extraction of dicaprin from mono- and
tricaprin, respectively.
Using Ecofac 1.0, for selectivity, solvent power, and solute
distribution coefficient values were obtained for each sep-
aration and proposed solvent. Based on these results, we
selected the most suitable combination of solvents to
maximize separation.
Among the tested solvents there is not a single solvent
that allows the extraction of the dicaprin, separating it from
other acylglycerols.
Table 1 shows the results obtained by the evaluation of
the separation between monocaprin and dicaprin, per-
forming solvent extraction of the monoglyceride. Solvents
marked with shadow gray would perform an adequate
liquid–liquid extraction. The resulting mixture or residual
Tables 5, and 6 present the theoretical results obtained
with Ecofac 1.0 to evaluate the performance of solvents for
extraction of tricaprin from monocaprin and dicaprin.
Chloroform, diethyl ether, acetone and n-heptane are pre-
sented as possible solvents to extract the triglyceride, n-
Fig. 2 Typical chromatogram
of the reaction products and
unconsumed reactants in
enzymatic esterification of
glycerol and capric acid using
Lipozyme RM IM. IS Internal
calibration standards.
Experimental conditions:
glycerol: 250 mg (adsorbed on
silica gel), temperature: 60 ꢁC
and enzyme: 40 mg, reaction
time: 6 h
Table 7 Performance of different KOH solutions for neutralization of capric acid and extraction of monocaprin
KOH concentration
Acid neutralization (%)
Monocaprin extraction (%)
Dicaprin loss (%)
Tricaprin loss (%)
1 M
100
99
98
93
40
99
97
92
82
38
53
7
38
5
0.5 M
0.25 M
0.1 M
0.05 M
5
2
3
0
0
0
123

J Am Oil Chem Soc (2014) 91:261–270
267
heptane being the most convenient because of its low
toxicity.
A typical chromatogram of the reaction products and un-
reacted substrates is shown in Fig. 2.
According to the results obtained with the software the
extraction procedure was designed using combinations of
n-heptane and ethanol.
Acid Capric Neutralization
With n-heptane, triglyceride can be extracted and acyl-
glycerides remain into the ethanolic phase. After the sep-
aration of the tricaprin, di- and monocaprin have to be
separated using these same solvents. In the following sec-
tions the proposed separation is studied to verify the sol-
vent-extraction using only heptane and ethanol. Relative
volumes of sample and solvents and number of stages of
treatment will be analyzed as important parameters in the
general procedure.
The final esterification mixture contains monocaprin,
dicaprin, tricaprin and unreacted capric acid. All these
compounds are dissolved in the reaction solvent, n-
heptane. Purification of dicaprin, obtained as the product
of esterification was performed by extraction with sol-
vents. Unreacted acid is previously neutralized with an
aqueous solution of KOH. Table 7 shows the experi-
mental performance of the solutions of different
concentrations.
As shown in Table 7, the hydroxide solution not only
allows the neutralization of the acid and the removal of it
from the heptane phase but it also allows the removal of the
monoglyceride.
Identification of Products
The products of the enzymatic esterification of glycerol
(GGG) with capric acid (C) are monocaprin (CGG) and
1,3-dicaprin (CGC), but also products with capric acid in
the sn-2 position are found. Acyl migration makes it pos-
sible to obtain two additional dicaproylglycerol isomeric
configurations: 1,2-dicaprin (CCG) and 2,3-dicaprin
(GCC), which are undesirable reaction products. When
these are not desirable, the dicaproylglycerols are esteri-
fied, tricaprin (CCC) is generated as the product. The used
capillary column differentiates positional isomers. Based
on the method for determination of mono- and diglycerides
by gas chromatography reported by Bruschweiler and
Dieffenbacher [26], their identification in the reaction
media was possible.
The KOH solution should be able to neutralize all the
acid, and to minimize losses of interest products, in this
case the diglyceride. With this objective, two washing steps
were performed with an alkaline solution of concentration
0.1 M. The first stage allowed the neutralization and sep-
aration of 93 % of acid and 82 % monoglyceride with 3 %
loss of diglyceride and no triglyceride extraction. The
second stage allowed the total removal of 99.5 % acid and
95 % of monocaprin. In this neutralization process 4.5 %
of dicaprin and 4 % of tricaprin were lost.
Figure 3 shows the chromatograms of the initial sample
and the sample after the neutralization with 0.1 M aqueous
solution of KOH.
Fig. 3 Overlay chromatograms
comparing esterification
reaction sample and the sample
after two-stage neutralization
with KOH solution with a
concentration of 0.1 M. The
initial sample contained 52 mg
of capric acid, 4 mg of
monocaprin, 28 mg of dicaprin
and 12 mg tricaprin
123

268
J Am Oil Chem Soc (2014) 91:261–270
Fig. 4 Results for the first
extraction stage
the separation process described in ‘‘Liquid–Liquid
Extraction’’.
In the first extraction step 94 % of the initial dicaprin
(115 mg) with a purity of 89 % was recovered. To increase
the purity of the extracted diglyceride, the ethanolic phases
obtained were combined, and washed with an equal volume
of n-heptane. At this stage, 47 mg of dicaprin was lost, but
its purity had increased to 97 %.
A third separation step was carried out with the aim of
recovering a higher proportion of dicaprin, and with high
purity. At this stage 45 mg of dicaprin was recovered,
however, the purity was only 80 %. Using the overall
procedure, 113 mg of the diglyceride was recovered with
89 % purity. This experimental finding does not represent
any improvement comparing it to the process in a single
step.
Fig. 5 Results for the second extraction stage
Liquid–Liquid Extraction
Figures 4, 5, and 6 show the performance of each of the
separation steps.
An initial sample containing 122 mg of dicaprin and 74 mg
of tricaprin, in a total volume of 12.5 mL, was subjected to
123

J Am Oil Chem Soc (2014) 91:261–270
269
Fig. 6 Results for the third
extraction stage
There is a close relationship between the covered dica-
prin fraction and the dicaprin purity. Carrying out only the
first stage it was possible to recover 94 % of the dicaprin
present in the sample after neutralization with a purity of
89 %. When looking for a higher purity product, two
separation steps allowed a recovery of 56 % of the dicaprin
with a purity of 97 %. A product with 93 % purity is
obtained at the third extraction stage but only 81 % of the
dicaprin present at the start of the extraction is recovered.
recovered with a purity of 93 %. An extra step did not
improve purity or yield. Considering the overall process,
performing neutralization and only the first extraction stage
it was possible to recover 90 % of the initial dicaprin with
89 % purity.
Acknowledgments The authors acknowledge the financial support
of CONICET and the PICT 2010 Bicentenario 0788 (ANPCyT). The
authors also are grateful to Dr Esteban Brignole for useful advice.
References
Conclusions
1. Jandacek RJ, Whiteside JA, Holcombe BN, Volpenhein RA,
Taulbee JD (1987) The rapid hydrolysis and efficient absorption
of triglycerides with octanoic acid in the 1 and 3 positions and
long-chain fatty acid in the 2 position. Am J Clin Nutr
45:940–945
2. Ikeda I, Tomari Y, Sugano M, Watanabe S, Nagata J (1991)
Lymphatic absorption of structured glycerolipids containing
medium-chain fatty acids and linoleic acid, and their effect on
cholesterol absorption in rats. Lipids 26:369
3. Porsgaard T, Høy CE (2000) Lymphatic transport in rats of
several dietary fats differing in fatty acid profile and triacyl-
glycerol structure. J Nutr 130:1619–1624
4. Akoh CC, Huang KH (1995) Enzymatic synthesis of structured
lipids: transesterification of triolein and caprylic acid. J Food
Lipids 2:219–230
5. Shimada Y, Sugihara A, Maruyama K, Nagao T, Nakayama S,
Nakano H, Tominaga Y (1996) Production of structured lipids
containing docosahexaenoic and caprylic acids using immobi-
lized Rhizopus delemar lipase. J Ferment Bioeng 81:299–303
6. Shimada Y, Suenaga M, Sugihara A, Nakai S, Tominaga Y
(1999) Continuous production of structured lipid containing lin-
olenic and caprylic acids by immobilized Rhizopus delemar
lipase. J Am Oil Chem Soc 76:189–193
Separation of acylglycerols obtained as a product of the
esterification of glycerol with capric acid mediated by
Lipozyme RM IM with a simple process using solvents of
low toxicity has been carried out. Dicaprin recovery was
possible, separating it from mono- and tricaprin.
A preliminary selection of solvents was made with
Ecofac 1.0 and the solvents chosen performed well for
liquid–liquid extraction. A preliminary step of the extrac-
tion was the neutralization of unreacted capric acid. A two-
stage process was applied with a 0.1 M KOH solution,
removing 99.5 % of the acid and 95 % of the monocaprin,
with losses of 4.5 and 4 % of di-and tricaprin, respectively.
With a separation procedure in three simple steps, it is
possible to recover 94 % of the dicaprin with 89 % purity.
If higher purity is required, an extra step can increase it to
97 % but with a considerable decrease in dicaprin yield. An
intermediate condition is achieved by performing one step
of the third stage, in which 81 % of the dicaprin is
123

270
J Am Oil Chem Soc (2014) 91:261–270
7. Xu X, Balchen S, Høy CE, Adler-Nissen J (1998) Pilot batch
16. Yasukawa T, Yasunaga K (2001) Nutritional functions of dietary
diacylglycerols. J Oleo Sci 50:427–432
production of specific-structured triacylglycerols by lipase-cata-
lyzed interesterification in a laboratory-scale continuous reactor.
J Am Oil Chem Soc 75:301–308
17. Maki KC, Davidson MH, Tsushima R, Matsuo N, Tokimitsu I,
Umporowicz DM, Dicklin MR, Foster GS, Ingram KA, Anderson
BD, Frost SD, Bell M (2002) Consumption of diacylglycerol oil
as part of a reduced-energy diet enhances loss of body weight and
fat in comparison with consumption of a triacylglycerol control
oil. Am J Clin Nutr 76:1230–1236
18. Watanabe H, Nagao T, Yasukawa T, Shimasaki H (2001) Studies
on the role of dietary diacylglycerols in human nutrition. World
Rev Nutr Diet 88:155–162
19. Tomonobu K, Hase Y, Tokimitsu I (2006) Dietary diacylglycerol in
a typical meal suppresses postprandial increases in serum lipid
levels compared with dietary triacylglycerol. Nutrition 22:128–135
20. Cismondi M, Brignole EA (2001) ECOFAC —Computer aided
solvent design and evaluation in environmental problems, based
on group contribution methods with association. Comput Aided
Chem Eng 9:375–380
´
´
8. Camacho Paez B, Robles Medina A, Camacho Rubio F, Gonzalez
Moreno P, Molina Grima E (2002) Production of structured tri-
glycerides rich in n-3 polyunsaturated fatty acids by the acidol-
ysis of cod liver oil and caprylic acid in a packed-bed reactor:
equilibrium and kinetics. Chem Eng Sci 57:1237–1249
´
´
9. Hita E, Robles A, Camacho B, Ramırez A, Esteban L, Jimenez
´
´
MJ, Mun˜ıo MM, Gonzalez PA, Molina E (2007) Production of
structured triacylglycerols (STAG) rich in docosahexaenoic acid
(DHA) in position 2 by acidolysis of tuna oil catalyzed by lipases.
Process Biochem 42:415–422
10. Schmid U, Bornscheuer UT, Soumanou MM, McNeill GP, Sch-
mid RD (1998) Optimization of the reaction conditions in the
lipase-catalyzed synthesis of structured triglycerides. J Am Oil
Chem Soc 75:1527–1531
11. Soumanou MM, Bornscheuer UT, Schmid RD (1998) Two-step
enzymatic reaction for the synthesis of pure structured triacyl-
glycerides. J Am Oil Chem Soc 75:703–710
12. Irimescu R, Furihata K, Hata Y, Iwasaki Y, Yamane T (2001)
Utilization of reaction medium-dependent regiospecificity of C.
antarctica lipase (Novozym 435) for the synthesis of 1 3-dica-
pryloyl-2-docosahexanoyl (or eicosapentanoyl) glycerol. J Am
Oil Chem Soc 78:285–289
21. Gani R, Brignole EA (1983) Molecular design of solvents for
liquid extraction based on UNIFAC. Fluid Ph Equilib 13:331–340
22. Cismondi M, Diaz MS, Espinosa S, Brignole EA (2003)
Molecular solvent design and near critical solvents optimization
with ECOFAC. Latin American applied research, 33
23. Fredenslund A, Jones RL, Prausnitz JM (1975) Group-contribu-
tion estimation of activity coefficients in nonideal liquid mix-
tures. AIChE J 21:1086–1099
13. Irimescu R, Furihata K, Hata Y, Iwasaki Y, Yamane T (2001)
Two-step enzymatic synthesis of docosahexaenoic acid-rich
symmetrically structured triacylglycerols via 2-monoacylglyce-
rols. J Am Oil Chem Soc 78:743–748
24. Gmehling J (1995) From UNIFAC to modified UNIFAC to PSRK
with the help of DDB. Fluid Ph Equilib 107:1–29
25. Buchgraber M, Ulberth F, Anklam E (2004) Interlaboratory
evaluation of injection techniques for triglyceride analysis of
cocoa butter by capillary gas chromatography. J Chromatogr A
1036:197–203
26. Bru¨schweiler H, Dieffenbacher A (1991) Determination of mono-
and diglycerides by capillary gas chromatography: results of a
collaborative study and the standardized method. Pure App Chem
63:1153–1162
´
´
´
14. Esteban L, Mun˜ıo MM, Robles A, Hita E, Jimenez MJ, Gonzalez
PA, Camacho B, Molina E (2009) Synthesis of 2-monoacylgly-
cerols (2-MAG) by enzymatic alcoholysis of fish oils using dif-
ferent reactor types. Biochem Eng J 44:271–279
15. Irimescu R, Iwasaki Y, Hou CT (2002) Study of TAG ethanolysis
to 2-MAG by immobilized Candida antarctica lipase and syn-
thesis of symmetrically structured TAG. J Am Oil Chem Soc
79:879–883
123