13C NMR quantification of mono and diacylglycerols obtained through the solvent-free lipase-catalyzed esterification of saturated fatty acids

Article abstract of DOI:10.1002/mrc.3814

In the present investigation, we studied the enzymatic synthesis of monoacylglycerols (MAG) and diacylglycerols (DAG) via the esterification of saturated fatty acids (stearic, palmitic and an industrial residue containing 87% palmitic acid) and glycerol in a solvent-free system. Three immobilized lipases (Lipozyme RM IM, Lipozyme TL IM and Novozym 435) and different reaction conditions were evaluated. Under the optimal reaction conditions, esterifications catalyzed by Lipozyme RM IM resulted in a mixture of MAG and DAG at high conversion rates for all of the substrates. In addition, except for the reaction of industrial residue at atmospheric pressure, all of these products met the World Health Organization and European Union directives for acylglycerol mixtures for use in food applications. The products were quantified by ¹³C NMR, with the aid of an external reference signal which was generated from a sealed coaxial tube filled with acetonitrile-d3. After calibrating the area of this signal using the classical external reference method, the same coaxial tube was used repeatedly to quantify the reaction products. Copyright

Full text of DOI:10.1002/mrc.3814

Research Article
Received: 14 September 2011
Revised: 16 February 2012
Accepted: 25 February 2012
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/mrc.3814
13C NMR quantification of mono and
diacylglycerols obtained through the
solvent-free lipase-catalyzed esterification
of saturated fatty acids
Jane Luiza Nogueira Fernandes,a Rodrigo Octavio Mendonça Alves de Souzab
and Rodrigo Bagueira de Vasconcellos Azeredoa*
In the present investigation, we studied the enzymatic synthesis of monoacylglycerols (MAG) and diacylglycerols (DAG) via the
esterification of saturated fatty acids (stearic, palmitic and an industrial residue containing 87% palmitic acid) and glycerol in a
solvent-free system. Three immobilized lipases (Lipozyme RM IM, Lipozyme TL IM and Novozym 435) and different reaction
conditions were evaluated. Under the optimal reaction conditions, esterifications catalyzed by Lipozyme RM IM resulted in
a mixture of MAG and DAG at high conversion rates for all of the substrates. In addition, except for the reaction of industrial
residue at atmospheric pressure, all of these products met the World Health Organization and European Union directives for
acylglycerol mixtures for use in food applications. The products were quantified by 13C NMR, with the aid of an external
reference signal which was generated from a sealed coaxial tube filled with acetonitrile-d3. After calibrating the area of this
signal using the classical external reference method, the same coaxial tube was used repeatedly to quantify the reaction
products. Copyright © 2012 John Wiley & Sons, Ltd.
Keywords: esterification; acylglycerols; lipases and quantitative 13C NMR
Among several basic requirements, a suitable standard for NMR
quantification must produce a sharp signal, preferably a singlet, to
Introduction
Partial acylglycerols, such as monoacylglycerols (MAG) and
diacylglycerols (DAG), are well-known biodegradable, biocom-
patible, non-toxic and non-ionic surfactants that are widely used
in food, pharmaceutical and industrial applications.[1,2] Compared
with traditional chemical glycerolysis, lipases demand milder reac-
tion conditions, which minimize energy costs, allow better reaction
control and provide higher-quality products.[3] Several studies have
shown that partial acylglycerols can be produced from unsaturated
fatty acids by lipase-catalyzed reactions.[4–7] However, to the best of
our knowledge, very few studies have explored enzymatic esterifica-
tion from saturated fatty acids (SFA) in a solvent-free system.[8–12]
In recent years, a growing number of scientific reports on the
successful application of 1H NMR and 13C NMR to the quantifica-
tion of esterification and trans-esterification products have been
published.[13–18] Compared with traditional chromatographic-
based techniques (GC and HPLC),[19] NMR offers the following
advantages: (i) sample preparation is easy, (ii) the technique is
non-destructive and non-invasive, (iii) multiple target analytes
in a mixture can be directly quantified without isolation, (iv)
additional structural information is readily available in a single
analysis and (v) a reference of the compound of interest is not
required.
avoid peak overlap. For both internal (the standard compound is
co-dissolved with the analyte) and external (the standard is added
to a coaxial insert) calibration methods, the peaks corresponding to
the calibration compound and the analyte must not overlap.[21,22]
Although peak overlap can be circumvented by altering the external
calibration method (the spectrometer response can be calibrated
with standards that are independently measured in separate
precision tubes), this approach does not guarantee that the analyte
and reference will be analyzed and processed under identical
conditions.
Alternatively, the benefits of both methods can be combined,
and a generic liquid compound in a sealed coaxial insert, such
as regular or deuterated solvent, can be used as a suitable inter-
nal reference for quantification. After calibrating the signal from
the internal reference against an external standard, the coaxial
insert can be switched from one tube to another after each
*
Correspondence to: Rodrigo Bagueira de Vasconcellos Azeredo, Laboratório de
Aplicações da RMN, Departamento de Química Orgânica, Instituto de Química,
Universidade Federal Fluminense, Rua Outeiro de São João Batista s/nꢀ,CEP:
24020-150, Niterói, Rio de Janeiro, Brasil. E-mail: rbagueira@vm.uff.br
Quantification by NMR, sometimes referred to as qNMR is
based on the premise that the signal produced by exciting a
nucleus from a fully relaxed state is directly proportional to the
number of molecules that contain the nucleus of interest. qNMR
can be performed by comparing the area of a selected analyte
peak with a standard compound at a known concentration.[19,20]
a Laboratório de Aplicações da RMN, Departamento de Química Orgânica,
Instituto de Química, Universidade Federal Fluminense, Rua Outeiro de São
João Batista s/nꢀCEP: 24020-150, Niterói, Rio de Janeiro, Brazil
b Laboratório de Biocatálise e Síntese Orgânica, Departamento de Química
Orgânica, Instituto de Química, Universidade Federal do Rio de Janeiro, Rua
Athos da Silveira Ramos 149, CT, Bloco ACEP: 21941-909, Rio de Janeiro, Brazil
Magn. Reson. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.

J. L. Nogueira Fernandes, R. Octavio Mendonça Alves de Souza and R. Bagueira de Vasconcellos Azeredo
NMR analysis, allowing accurate and robust quantification. Based
Calibration
on this technique, a simple method for the 13C NMR quantifica-
tion of monoacylglycerols and diacylglycerols produced via
solvent-free, lipase-catalyzed esterification of SFA was developed,
and an external reference signal was generated from a sealed
coaxial tube filled with acetonitrile-d3.
Calibration of the reference signal was performed using six
samples of L-menthol (≥99%, 0.106 Â 10À3–0.731 Â 10À3 mol/ml
in a CDCl3 stock solution). To assess the stability of the system,
three calibration samples were prepared and analyzed prior to
the two product quantification runs. Using the quantitative 13C
NMR spectra, the calibration model was established according
to the linear relationship Conc = b0 + b1*(Area/Arearef), where Conc
is the calibration sample concentration (mol/ml), and Area and
Arearef are the areas of the calibration and reference peaks,
respectively. The CH group of menthol at 49.2 ppm was arbitrarily
selected as the calibration peak. The intercept (b0) and angular
coefficient (b1) were estimated by least squares regression using
Origin ver.6.0 software (OriginLab, USA). Unlike the NMR
precision tubes, the stem coaxial inserts are not manufactured
with absolute rigid tolerances; thus, the developed calibration
model can only be applied to the insert employed in the present
study. Otherwise, a new calibration model must be constructed.
Methodology
Lipase-catalyzed reactions
Solvent-free esterification reactions were evaluated by varying
four parameters, including (i) the type of lipase[23–27]: Novozym
435, Lipozyme RM IM or Lipozyme TL IM; (ii) the type of free fatty
acids (FFA): stearic acid (C18 : 0), palmitic acid (C16 : 0) or an
industrial residue (containing 87% palmitic acid); (iii) the reaction
time: 3 or 6 h and (iv) vacuum application: no vacuum or 3 mm
Hg. The FFA industrial residue and all of the lipases used in the
present study were kindly donated by Agropalma A/S and
Novozymes A/S, respectively. The remaining reagents and
solvents described herein were analytical grade and were
acquired from a local distributor (Vetec, Brazil).
The substrate mixture, which consisted of FFA (0.025 mol for
stearic and palmitic acid and 0.022 mol for the industrial
residue, relative to palmitic acid), glycerol (0.0375 mol) and lipase
(1.0 g), was introduced into a jacket reactor under magnetic
agitation, which was maintained at 65ꢀC (for stearic acid and the
industrial residue) or 60ꢀC (for palmitic acid) via water circulation.
All reactions were carried out in duplicate. At the end of the
reaction period, the resultant mixture was dissolved in dichloro-
methane, filtered to remove the catalyst and washed with dis-
tilled water (3Â 15ml) to remove residual glycerol. The organic
phase was extracted with dichloromethane (3 Â 15 ml), and the
solvent was removed under reduced pressure. The products
were dried over anhydrous sodium sulfate and were stored in
a refrigerator in an amber flask. The reactions were performed
in parallel under the same conditions, except for the presence
of the enzyme.
Quantification
Samples from each esterification product were prepared from the
same CDCl3 stock solution at a concentration of approximately
30% (m/v) and were analyzed by quantitative 13C NMR in two
different runs, which were separated by a 30-day period. The acylgly-
cerols were identified using the C-2 groups along the glycerol
backbone, which resonates in the characteristic spectral region from
68 to 76 ppm, according to Gunstone[15]: 1-MAG= 70.37; 2-MAG=
75.02; 1,2-DAG = 72.24; 1,3-DAG = 68.35 and TAG = 69.02 ppm. After
integrating the internal reference peak and the 13C NMR peaks of
the glycerol backbone of MAG, DAG and TAG, the concentrations
were predicted with the previously prepared calibration model. In
the same manner, the residual FFA concentrations were also mea-
sured using the carboxylic peak[30] at 179 ppm. To verify the reliability
of the method, the FFA concentrations obtained by 13C NMR were
compared with those measured by NaOH titration.
Acidity by titration
The FFA concentration was determined by titration, which was
executed according to a modified Ca 5a-40 method.[31] Approxi-
mately 0.5 g of the sample was diluted in 50 ml of an ethanol/ethyl
ether 1 : 1 (v/v) mixture and was titrated with a NaOH solution
(0.05 M) using phenolphthalein as the end-point indicator. The acidity
indicates the percentage of fatty acids remaining at the end of the
reaction or the degree of esterification.
13C NMR quantification
To shorten the relaxation time, a CDCl3 stock solution of Cr(AcAc)3
(0.1 M) and 0.1% TMS was used to prepare standards and samples.
The standard and sample solutions were transferred from
volumetric flasks to the annular space between the stem coaxial
insert (no. WGS-0BL) and the 10-mm precision NMR tube (no.
513-7PP-7, Wilmad-LabGlass, USA). The volume of liquid added to
the coaxial insert and sample tube is not important because the
tube extends well past the sensitive area of the receiver coil.[28,29]
All deuterated solvents, standards and relaxation agents were
obtained from Sigma Aldrich.
Data acquisition and processing
NMR experiments were performed on a 11.75-Tesla VNMRS
(Varian, USA) spectrometer equipped with two 10-mm channel
broadband probes for 15 N-31P{1H-19 F}. Quantitative 13C NMR
spectra were recorded at 35 ꢀC without spinning the sample,
and an inverse-gated decoupling sequence[32] with a recycle
delay of 3.2 s (more than five times the longest T1) and a
WALTZ16 proton decoupling scheme were employed. Free-
induction decays (FIDs) were collected using 42.779 data points
over a spectral width of 220 ppm, and a 90ꢀ pulse was applied
for 12 ms until a signal-to-noise level of 200 (relative to the larger
signal between 68 to 76 ppm) was obtained. The samples were
identically positioned, individually tuned and matched before
each experiment. Prior to multiplication by a 5.0-Hz exponential
function and Fourier transformation, the FID was zero-filled to
143.360 points. The phase and baseline were corrected manually.
Internal reference
A solution of Cr(AcAc)3 (0.1 M) in acetonitrile-d3 (99.8%) was
placed in the coaxial insert, and the 10-mm sample tube was
fitted to the insert prior to 13C NMR analysis. The C ꢁ N group
of acetonitrile-d3, which appeared at 118.7 ppm, produced a
non-overlapping, constant-amplitude reference signal on each
spectrum. After NMR analysis, the coaxial insert was carefully
rinsed with acetone, completely dried and switched to another
sample tube.
wileyonlinelibrary.com/journal/mrc
Copyright © 2012 John Wiley & Sons, Ltd.
Magn. Reson. Chem. (2012)

Absolute quantification of mono- and diacylglycerol by 13C NMR
The peak areas were measured by Voigt function deconvolution
using MestReNova ver.6.0.2 (Mestrelab, Spain).
compound for acylglycerol quantification via internal and external
(by coaxial inserts) calibration because its C-1 peak (CH at
71.3 ppm) lies in the middle of the glycerol backbone region.
In addition to the CH group at 49.2 ppm, other menthol 13C
signals were also evaluated (data not shown); however, the
results were virtually identical. When the 13C NMR spectra are
acquired under quantitative conditions, the peak area must be
equivalent regardless of which peak is selected.
Results and Discussion
13C NMR quantification
Internal reference calibration
After collecting quantitative 13C NMR spectra of the six menthol
calibration samples, which were split in two different runs sepa-
rated by a 30-day period, a calibration model was developed by
performing a linear regression on the peak areas ratios from
menthol (CH at 49.2 ppm) and acetonitrile-d3 (C ꢁ N at
118.7 ppm) versus the menthol concentration, as shown in Fig. 1.
Although the spectrometer configuration was switched several
times between the two runs (the probes were changed and the
system was recabled), the correlation coefficient (R2 = 0.9997,
p = 0.0001) for the regression was high, which indicates that the
response for the two runs was linear, even after 30 days. The
ruggedness of the configurational changes is an interesting
feature, considering the context of most NMR facilities, in which
multipurpose spectrometers cannot be scheduled for long
uninterrupted periods. The regression statistics also showed that
the intercept of the calibration model was not significantly different
from 0 (b0 = À0.0028 Æ (0.0054)), which is a good indication that
the adopted baseline correction and line fitting procedure were
suitable for avoiding instrumental artifacts.[33]
Acetonitrile-d3 was chosen as an internal reference instead of
non-deuterated acetonitrile solely because it is readily available
in the solvent inventory of most NMR laboratories. However, a
regular solvent is preferable as an internal reference for 13C
NMR quantification because, in addition to reducing sensitivity,
peak splitting due to scalar coupling between 2H and 13C can
interfere with the peak of interest.
Regarding the external calibration standards, even those that
produce too many NMR peaks are suitable for calibrating the
internal reference signal. Aside from the purity of the standard,
which must be analytical grade, the only prerequisite is that the
peaks of the standard must not overlap with those of the internal
reference. For example, menthol is not the most appropriate
Quantification performance evaluation
After collecting and processing the 36 13C NMR spectra of the
esterification products, the peaks were integrated relative to FFA,
and their concentrations (mol/ml) were calculated by applying
the internal reference and respective peak area ratios (Area/Arearef
)
to the previously developed calibration model. To verify the reliabil-
ity of this method, the FFA concentrations (m/v) obtained by 13C
NMR were compared with the reference values measured by NaOH
titration, as shown in Fig. 2.
In addition to the high correlation coefficient (R2 = 0.9945,
p = 0.0001) of the linear regression of the response curve, the
intercept and slope were not significantly different from 0
(b0 = À0.1079 Æ (0.6202)) and 1 (b1 = 0.9922 Æ (0.0130)), respec-
tively, suggesting that the NMR response was not biased (by addi-
tive or proportional systematic errors) compared with the titration.
Regarding the accuracy of NMR, the root-mean-square error, a
measurement of the average difference between the predicted
and measured response, suggested that the FFA concentration
can be measured with an average error of 1.8%, which can be
extrapolated to estimate the quantification error of acylglycerols.
The performance of NMR was compatible with the expected
analytical errors involved in the preparation of samples that are
difficult to handle, such as SFA derivatives, which require extra
care to ensure sample homogeneity and avoid losses.
The overall performance of NMR measurements also suggested
that the quantitative experimental parameters were well adjusted,
especially the relaxation delays, which were selected to ensure 13C
was detected from a completely relaxed spin state (at least 99%).
Synthesis evaluation
From the 13C NMR spectra of the esterification products, the
peaks were unambiguously assigned and integrated relative to
0.8
linear fit
100
0.7
linear fit
95% confidence bands
90
95% confidence bands
0.6
1st run
80
2nd run (30 days later)
RMSE=1.8%
70
0.5
60
50
40
30
20
0.4
0.3
0.2
0.1
R2=0.9997
0.5
10
R2=0.9945
0.0
0.0
0.1
0.2
0.3
0.4
0
Area/Arearef
0
10 20 30 40 50 60 70 80 90 100
FFAtitration (%)
Figure 1. Regression between menthol (CH at 49.2ppm) and acetonitrile-
d5 (C ꢁ N at 118.7 ppm). 13C NMR peak area ratios and menthol solution
concentrations were acquired in two different runs, which were separated
by a 30-day period. The resultant internal reference calibration model was
Conc= À0.0028(Æ0.0054) + 1.4035(Æ0.0165)*(Area/Arearef).
Figure 2. Regression between the FFA concentrations measured by
NaOH titration (FFAtitration) and predicted by NMR (FFANMR). The resultant
model response curve was FFANMR = À0.1079(Æ0.6202) + 0.9922(Æ0.0130)
*FFAtitration
.
Magn. Reson. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/mrc

J. L. Nogueira Fernandes, R. Octavio Mendonça Alves de Souza and R. Bagueira de Vasconcellos Azeredo
the C-2 groups on the glycerol backbone of the acylglycerols.
The MAG, DAG and TAG concentrations (mol/ml) were calculated
by computing the internal reference and respective C-2 peak area
ratios (Area/Arearef) and applying the results to the previously
developed calibration model. Table 1 summarizes the acylglycerol
quantification data according to the enzyme and reaction
Figure 3 presents a typical spectrum of acylglycerols produced
from lipase-catalyzed esterification and an expansion of the
glycerol backbone region, where the MAG, DAG and TAG C-2
peaks were clearly observed.
Figure 3. Typical 13C NMR spectrum of a mixture of acylglycerols. The insert shows the C-2 region of acylglycerols, which was used for quantification.
Table 1. Lipase-catalyzed esterification reaction of pure stearic acid, palmitic acid and industrial fatty acid residue containing 87% palmitic acid,
excluding Lipozyme TL IM
Acylglycerol composition (%)a
MAG DAG
DAG/MAG
Conversion (%)
Substrate (lipase type)
Conditions
TAG
Vacuum
Time (h)
1-
2-
1,2-
1,3-
Stearic acid (C18 : 0)
(Novozym 435)
No
No
3
6
3
6
3
6
3
6
20
27
25
22
35
29
29
34
1
1
0
0
1
1
9
1
14
15
1
35
34
54
49
35
35
33
38
20
14
12
19
3
2.3
1.8
2.2
2.4
1.1
1.4
1.2
1.2
91
92
92
95
80
80
88
85
Yes
Yes
No
5
(Lipozyme RM IM)
5
No
6
9
Yes
Yes
14
6
3
5
Palmitic acid (C16 : 0)
(Novozym 435)
No
No
3
6
3
6
3
6
3
6
20
17
19
16
34
32
38
36
0
0
0
0
0
1
1
2
6
3
7
9
5
7
5
7
53
54
44
41
37
34
43
43
9
14
12
16
5
3.0
3.2
2.7
3.2
1.2
1.2
1.2
1.3
88
83
82
81
82
85
92
95
Yes
Yes
No
(Lipozyme RM IM)
No
10
4
Yes
Yes
6
Fatty acid residue
(Novozym 435)
No
No
3
6
3
6
3
6
3
6
7
11
23
26
26
25
32
32
0
0
1
1
1
1
1
0
8
16
12
11
9
59
51
35
32
28
30
43
38
14
14
13
17
9
10
6
88
92
83
87
74
80
88
92
Yes
Yes
No
2
1.6
1.4
1.5
1.4
1.5
(Lipozyme RM IM)
No
10
4
14
8
Yes
Yes
12
10
aDetermined by 13C qNMR.
wileyonlinelibrary.com/journal/mrc
Copyright © 2012 John Wiley & Sons, Ltd.
Magn. Reson. Chem. (2012)

Absolute quantification of mono- and diacylglycerol by 13C NMR
conditions for reactions with conversion rates greater than 20%. All
reactions with Lipozyme TL IM were below this limit, and the
corresponding results are not reported in Table 1.
and lipases, respectively. We would also like to acknowledge
two anonymous reviewers for their technical and editorial
comments, which improved the paper.
Regarding the FFA conversion, Novozym 435 and Lipozyme
RM IM showed superior performance. This result is in good
agreement with that of Freitas et al.,[8] who found that the
conversion degree of Lipozyme RM IM (86%) in the esterification
of lauric acid (C12 : 0) was superior to Lipozyme TL IM (45%). To
our knowledge, reports describing the application of Novozym
435 to the solvent-free synthesis of partial acylglycerols from
SFA have not been previously published.
Aside from the natural predominance of 1-MAG over 2-MAG,
1,3-DAG was the main diester produced under all of the reaction
conditions. This behavior was expected because the enzymes are
1,3-specific; however, the DAG-to-MAG ratio also suggested
that Novozym 435 was more specific for diester formation. In
general, Novozym 435 also presented a 1,3-DAG-to-1,2-DAG
ratio that was equal to or greater than that of Lipozyme RM
IM, indicating that the regioselectivity of the studied reaction
system was higher.
References
[1] D. G. Hayes. J. Am. Oil Chem. Soc. 2004, 81, 1077–1103.
[2] F. Camacho, A. Robles, P. A. González, B. Camacho, L. Esteban, E.
Molina. Appl. Catal. Gen. 2006, 301, 158–168.
[3] F. Blasi, L. Cossignani, M. S. Simonetti, P. Damiani. Enzyme Microb.
Technol. 2007, 41, 727–732.
[4] C. E. Martinez, J. C. Vinay, R. Brieva, C. G. Hill Jr., H. S. Garcia. Appl.
Biochem. Biotechnol. 2005, 125, 63–75.
[5] M. Linder, N. Kochanowski, J. Fanni, M. Parmentier. Process Biochem.
2005, 40, 273–279.
[6] V. Triphathi, R. Trivedi, R. P. Singh. J. Oleo Sci. 2006, 555, 65–69.
[7] I.-H. Kim, S.-M. Lee. J. Food Sci. 2006, 71(7), C378–C382.
[8] L. Freitas, V. H. Perez, J. C. Santos, H. F. de Castro. J. Braz. Chem. Soc.
2007, 18, 1360–1366.
[9] M. A. M. da Silva, V. C. Medeiros, M. A. P. Langone, D. M. G. Freire.
Appl. Biochem. Biotechnol. 2003, 108, 757–768.
[10] T. Sakiyama, T. Yoshimi, A. Miyake, M. Umeoka, A. Tanaka, S. Ozaki, K.
Nakanishi. J. Biosci. Bioeng. 2001, 91, 27–32.
Considering the application of acylglycerol mixtures in the food
industry, Lipozyme RM IM provided the best performance under
the reaction conditions evaluated in the present study. Except for
the reaction of industrial residue at atmospheric pressure, all of the
esterifications catalyzed by Lipozyme RM IM resulted in a final
mixture that fulfilled the World Health Organization (WHO) and
European Union (EU) directives.[34,35] According to the WHO and
EU, acylglycerol mixtures used as food emulsifiers must contain at
least 70% MAG + DAG and 30% MAG and less than 10% of glycerol
and TAG.
[11] M. A. P. Langone, M. E. Abreu, M. J. C. Rezende, G. L. Sant’Anna Jr.
Appl. Biochem. Biotechnol. 2002, 98, 987–996.
[12] R. Rosu, M. Yasui, Y. Iwasaki, T. Yamane. J. Am. Oil Chem. Soc. 1999,
76, 839–843.
[13] V. M. Mello, F. C. C. Oliveira, W. G. Fraga, C. J. Nascimento, P. A. Z.
Suarez. Magn. Reson. Chem. 2008, 46, 1051–1054.
[14] L. F. Cabeça, L. V. Marconcini, G. P. Mambrini, R. B. V. Azeredo, L. A.
Colnago. Energy Fuel 2011, 25, 2696–2701.
[15] F. D. Gunstone. Chem. Phys. Lipids 1991, 58, 219–224.
[16] G. Knothe. J. Am. Oil Chem. Soc. 2000, 77, 489–493.
[17] P. R. C. Neto, M. S. B. Caro, L. M. Mazzuco, M. G. Nasciment. J. Am. Oil
Chem. Soc. 2004, 81, 1111–1114.
[18] S. Ng. J. Am. Oil Chem. Soc. 2000, 77, 749–754.
[19] R. J. Wells, J. M. Hook, T. S. Al-Deen, D. B. Hibbert. J. Agric. Food Chem.
2002, 50, 3366–3374.
[20] J. K. Satyarthi, D. Srinivas, P. Ratnasamy. Energy Fuel 2009, 23, 2273–2277.
[21] T. Beyer, D. Bernd, U. Holzgrabes. Bioanal. Rev. 2010, 2, 1–22.
[22] T. Rundlof, M. Mathiasson, S. Bekiroglu, B. Hakkarainen, T. Bowden, T.
Arvidsson. J. Pharm. Biomed. Anal. 2010, 52, 645–651.
[23] R. A. Sheldon. Adv. Synth. Catal. 2007, 349, 1289–1307.
[24] R. C. Rodrigues, R. F. Lafuente. J. Mol. Catal. B: Enzym. 2010, 66,
15–32.
Although some of the reaction conditions resulted in the
formation of products with specifications that did not fulfill the
prerequisites for use in foodstuffs, good conversion rates were
obtained, and the products can be employed in non-food appli-
cations, such as cosmetic creams, pharmaceutical preparations,
cement and concrete additives, polishes, shoe waxes and plastics.
Conclusions
[25] J. Svensson, P. Adlercreutz. Eur. J. Lipid Sci. Technol. 2011, 113,
1258–1265.
A mixture of MAG and DAG was obtained with high conversion via
lipase-catalyzed esterification of SFA under solvent-free reaction con-
ditions. Due to the simplicity and efficiency of the proposed method-
ology, this method can be used as an alternative to alkaline-catalyzed
chemical glycerolysis to obtain MAG and DAG on an industrial scale.
We demonstrated that the utilization of a constant natural
reference signal from an analytical grade solvent in a sealed
coaxial insert, combined with classical external calibration, was
an extremely simple and reliable approach for routine NMR
quantification. Methods such as electronic reference to access
in vivo concentrations and quantification by artificial signal also
use artificial internal reference signals for quantification;
however, dedicated hardware and software are required.[36–38]
Regarding the quantification of acylglycerol products, 13C NMR
can compete with GC and HPLC, as highlighted in the introduction.
However, in general, quantitative 13C NMR is more time-consuming
and expensive than chromatographic-based techniques.
[26] F. Heffner, T. Norin. Chem. Pharm. Bull. 1999, 47, 591–600.
[27] J. Uppenberg, M. T. Hansen, S. Patkar, T. A. Jones. Structure 1994, 2,
293–308.
[28] T. J. Henderson. Anal. Chem. 2002, 74, 191–198.
[29] I. W. Burton, M. A. Quilliam, J. A.Walter, Anal. Chem. 2005, 77,
3123–3131.
[30] G. Vlahov, A. A. Giuliani, P. D. Re. Anal. Methods 2010, 2, 916–923.
[31] AOCS, Official Methods and Recommended Practices of the
American Oil Chemists’ Society, AOCS, Champaign, 2004.
[32] R. Freeman, H. D. W. Hill, R. Kaptein. J. Magn. Reson. 1972, 7, 327–329.
[33] P. Soininen, J. Haarala, J. Vepsalainen, M. Niemitz, R. Laatikaainen.
Anal. Chim. Acta 2005, 542, 178–185.
[34] European Parliament and Council Directive No. 95/2/EC on Food
Additives other than Colours and Sweeteners, Office for Official
Publications of the European Communities. Available from http://
ec.europa.eu/food/fs/sfp/addit_flavor/flav11_en.pdf (accessed on
11 January 2011).
[35] EFEMA, E471: Mono- and diglycerides of fatty acids, in EFEMA Index
of Food Emulsifiers (5th edn) (Ed: European Food Emulsifier Manufac-
tures Association), 2009, pp. 51–54. Available from www.emulsifiers.
org (accessed on 11 January 2011).
[36] S. H. Schweizer, N. D. Zanche, M. Pavan, G. Mens, U. Sturzenegger, A.
Henning, P. Boesiger. NMR Biomed. 2010, 23, 406–413.
[37] F. Ziarelli, S. Viel, S. Caldarelli, D. N. Sobieski, M. P. Augustine. J. Magn.
Reson. 2008, 194, 307–312.
Acknowledgements
We are grateful to CAPES (Coordenação de Aperfeiçoamento de
Pessoal de Nível Superior) for a research grant and Agropalma
A/S and Novozymes A/S for donating the industrial FFA residue
[38] R. D. Farrant, J. C. Hollerton, S. M. Lynn, S. Provera, P. J. Sidebottom, R.
J. Upton. Magn. Reson. Chem. 2010, 48, 753–762.
Magn. Reson. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/mrc