Crystallization and polymorphism of 1,3-acyl-palmitoyl-rac-glycerols

Article abstract of DOI:10.1007/s11746-011-1769-0

Crystallization and melting behavior, small-angle X-ray scattering, X-ray powder diffraction and infra-red absorbance were measured for nine 1,3-acyl-palmitoyl-rac-glycerols (1,3-acetoyl-, -butyroyl-, -hexanoyl-, -octanoyl-, -decanoyl-, -lauroyl, -myristoyl- and -oleoyl-palmitoyl-rac-glycerol and 1,3-dipalmitoyl-glycerol). All but one of the prepared 1,3-diacylglycerols (1,3-DAG) were β-stable with 1,3-acetoyl-palmitoyl-rac-glycerol being the exception (β-stable). Small-angle X-ray scattering indicates that molecules in β-tending diacid 1,3-DAG adopt a herringbone-type configuration similar to monoacid 1,3-DAG. In this configuration acyl chains of the same length associate and regular chain-end matching between terminal methyl groups delineate lamellae. In contrast, molecules in crystalline 1,3-acetoyl-palmitoyl- rac-glycerol are oriented similar to those of 1(3)-monoacylglycerol. Interestingly, DSC curves indicate five of the nine diacid compounds have meta-stable forms-suggesting these forms are quite common for diacid 1,3-DAG. Meta-stable forms are observed in the melting curve when the difference in length between acyl chains is large (1,3-acetoyl-, -butyroyl- and -hexanoyl-palmitoyl-rac-glycerol), and in the crystallization curve when the difference is moderate (1,3-decanoyl- and -lauroyl-palmitoyl-rac-glycerol).

Full text of DOI:10.1007/s11746-011-1769-0

J Am Oil Chem Soc (2011) 88:1113–1123
DOI 10.1007/s11746-011-1769-0
ORIGINAL PAPER
Crystallization and Polymorphism of 1,3-Acyl-Palmitoyl-rac-
Glycerols
•
R. John Craven Robert W. Lencki
Received: 2 September 2010 / Revised: 8 December 2010 / Accepted: 17 January 2011 / Published online: 18 February 2011
Ó AOCS 2011
Abstract Crystallization and melting behavior, small-
angle X-ray scattering, X-ray powder diffraction and infra-red
absorbance were measured for nine 1,3-acyl-palmitoyl-rac-
glycerols (1,3-acetoyl-, -butyroyl-, -hexanoyl-, -octanoyl-,
-decanoyl-, -lauroyl, -myristoyl- and -oleoyl-palmitoyl-rac-
glycerol and 1,3-dipalmitoyl-glycerol). All but one of the
prepared 1,3-diacylglycerols (1,3-DAG) were b-stable with
1,3-acetoyl-palmitoyl-rac-glycerol being the exception
(b0-stable). Small-angle X-ray scattering indicates that
molecules in b-tendingdiacid 1,3-DAG adopt a herringbone-
type configuration similar to monoacid 1,3-DAG. In this
configuration acyl chains of the same length associate and
regular chain-end matching between terminal methyl groups
delineate lamellae. In contrast, molecules in crystalline 1,3-
acetoyl-palmitoyl-rac-glycerol are oriented similar to those
of 1(3)-monoacylglycerol. Interestingly, DSC curves indi-
cate five of the nine diacid compounds have meta-stable
forms—suggesting these forms are quite common for diacid
1,3-DAG. Meta-stable forms are observed in the melting
curve when the difference in length between acyl chains is
large (1,3-acetoyl-, -butyroyl- and -hexanoyl-palmitoyl-
rac-glycerol), and in the crystallization curve when the
difference is moderate (1,3-decanoyl- and -lauroyl-palmitoyl-
rac-glycerol).
Introduction
1,3-Diacylglycerol (1,3-DAG) play a key role in a number
of industrially important systems. For example, they are
components of natural fats and oils (e.g. palm oil, milk fat)
and are known to influence their nucleation, crystal growth
and polymorphism [1–4]. DAG are also used as food
emulsifiers and as intermediates in the preparation of spe-
cific structured lipids and pharmaceutical products [5]. In
addition, replacing dietary TAG with 1,3-DAG reportedly
leads to weight loss and improves blood cholesterol levels
due to a difference in the way the molecules are metabo-
lized [6]. To further develop 1,3-DAG-based products for
these applications, a detailed understanding of their physi-
cal characteristics is required.
The phase behavior of DAG is quite complex and
therefore of particular interest. While 1,2-diacylglycerols
crystallize in a and b0 polymorphs, the majority of 1,3-
DAG are b-tending—crystallizing in either high-melting
b1 or low-melting b2 forms [7]. There are however
exceptions; for example, diacid 1,3-DAG containing both
short- and long-chain acyl groups (e.g. 1,3-acetoyl-stea-
royl-rac-glycerol) are reportedly b0-tending [8].
The complete melting behavior for monoacid 1,3-DAG
of long-chain FA (viz. 1,3-dipalmitin and 1,3-distearin) has
been determined [9] and the melting points (b1 and b2
forms) of many other monoacid 1,3-DAG have been
reported [10]. In addition, the conformation of the most
stable form for 1,3-dilaurin has been ascertained by single-
crystal X-ray studies [11, 12]. By comparison, diacid 1,3-
DAG have received far less attention—perhaps because
they are not commercially available and are a challenge to
prepare. In general, only polymorphic tendency (b- or b0-
tending) and melting point of the highest-melting form
have been reported for diacid 1,3-DAG [10].
Keywords 1,3-Diacylglycerol ꢀ Physical chemistry ꢀ
X-ray powder diffraction ꢀ Small-angle X-ray scattering ꢀ
Infra-red spectroscopy ꢀ Polymorphism
R. J. Craven ꢀ R. W. Lencki (&)
Department of Food Science, University of Guelph,
Guelph, ON N1G 2W1, Canada
e-mail: rlencki@uoguelph.ca
123

1114
J Am Oil Chem Soc (2011) 88:1113–1123
Even for the most remarkable of these molecules (e.g.
exceptions: solketal was used as a starting material; shorter
reflux times (*5 h) were employed; and free 1(3)-MAG
was produced using an acidic cation exchange resin [19].
The resulting MAG and DAG were purified by recrystal-
lization or flash chromatography.
1,3-acetoyl-palmitoyl-rac-glycerol), which display excep-
tional polymorphism (b0-tending), complete crystallization
and melting behavior has yet to be analyzed by DSC—
although it has been studied by dilatometry [8]. This is not
surprising since most of the published data was collected in
the 1940s and 1950s, prior to the first commercial DSC
(1964). Likewise, literature values are questionable
because, at the time, methodologies for determining purity
of starting materials and final products were not fully
developed [13].
Gas Chromatography
Trimethylsilyl-derivatives were analyzed on a 25 m 9
0.25 mm polarizable capillary column (Quadrex, Wood-
bridge, CT, USA) [15]. The column was housed in a
Hewlett Packard 5890 (Agilent, Palo Alto, CA, USA) GC
equipped with FID and on-column inlet. The inlet pressure
for the carrier gas (hydrogen) was set to 15 psi, cool on-
column injection was employed and the detector was held
at 370 °C. After sample injection, the oven was held at
60 °C for 2 min, then the temperature was increased to
250 °C at 35 °C/min, and finally the temperature was
increased to 300 °C at 4 °C/min.
Thus, the first goal of this work was to characterize 1,3-
acyl-palmitoyl-rac-glycerols using current, consistent and
recognized methods. To this end, X-ray powder diffraction
(XRD), small-angle X-ray scattering (SAXS), infra-red
spectroscopy (IR) and differential scanning calorimetry
(DSC) were used to examine 1,3-acyl-palmitoyl-rac-gly-
cerols (acetoyl-, butyroyl-, hexanoyl-, octanoyl-, decanoyl-,
lauroyl-, myristoyl- and oleoyl-palmitoyl-rac-glycerol and
1,3-dipalmitoyl-glycerol). After placing this new data in
context (by comparing with existing literature), the role of
acyl chain length in the crystallization behavior of 1,3-
acyl-palmitoyl-rac-glycerols could be examined by com-
paring results for compounds within the series.
Differential Scanning Calorimetry
Crystallization and melting curves were determined using a
Q1000 DSC (TA Instruments, New Castle, DE, USA) that
was calibrated following the manufacturer’s recommen-
dations with a pure indium standard and a matched pair of
sapphires. The sample cell was purged with dry nitrogen
flowing at 25 mL/min. Hermetically sealed alodined alu-
minum DSC pans containing 4–8 mg of the sample were
processed and heat flow was measured relative to an empty
sealed DSC pan. Results were analyzed using Universal
Analysis software version 4.2E (TA instruments).
Materials and Methods
Unless noted otherwise, reagents, chemicals and enzymes
were purchased from Sigma-Aldrich (Mississauga, ON,
Canada) and were of the highest practical grade; solvents
were purchased from Fisher Scientific (Ottawa, ON, Can-
ada) and were HPLC grade. The declared purity of free
fatty acids (FFA) was: hexanoic (6:0) and myristic (14:0)
acids &99.5%, palmitic (16:0) acid C96%, lauric (12:0)
acid C95% and oleic (18:1) acid C90% purity. The
declared purity of acid chlorides was: butyroyl (4:0) and
octanoyl (8:0) chloride C99%, decanoyl (10:0), lauroyl and
palmitoyl chloride &98% and hexanoyl chloride &97%
purity. Solketal (isopropylidene glycerol) and acetic
anhydride were both 98% purity. The vinyl ester of pal-
mitic acid donated by Japan Vam & Poval (Osaka, Japan)
was C96% purity.
Samples were prepared by transferring solid DAG to
the sample pan without melting; the initial melt of these
samples was measured (at 5 °C/min). This provided the
melting data (Te, Tp and DHf) for the high-melting form
of each compound. Afterwards, samples were analyzed by
DSC at three different heating/cooling rates (2.5, 5 and
10 °C/min) because DSC scans collected at several rates
are useful for detecting polymorphism [20]. Universal
Analysis software was used to determine the extrapolated
onset temperature (Te), peak temperature (Tp) and
enthalpy (DHf) of both melting and crystallization. Mea-
surements obtained from melting curves are considered
more reliable than those arising from crystallization
because the latter requires undercooling and is delayed by
the need for nucleation [21]. The preferred measurement
of melting temperature (Tm) is the extrapolated onset of
melting temperature (Te) because it has been shown to
vary the least with heating rate. Te is the temperature at
which a tangent through the linear portion of the leading
edge of the peak intersects with the baseline [22]. In the
event that Te is not available, as is the case with peaks
Synthesis
Diacid 1,3-DAG were prepared by reacting 1(3)-MAG with
acid chlorides and triethylamine in the presence of
4-dimethylaminopyridine [14, 15]. Monoacid 1,3-DAG
were prepared by mixing glycerol with a fatty acid vinyl
ester in the presence of immobilized lipase B from Candida
antarctica [16]. 1(3)-MAG precursors were synthesized
following published methods [17, 18] with the following
123

J Am Oil Chem Soc (2011) 88:1113–1123
1115
that merge due to polymorphism, Tp is an acceptable
approximation [21, 22].
software version 6.5 (Materials Data, Livermore, CA,
USA).
Infra-Red Spectroscopy
Small Angle X-Ray Scattering
Data was collected using a Shimadzu IRPrestige-12 FTIR
(Kyoto) equipped with a Pike MIRacleTM ATR sample
stage (Madison, WI, USA). Data was analyzed using the
Shimadzu IR Solution 1.30 software package.
Samples were ground into a fine powder and sealed
between sheets of Kapton. Samples were held at 20 °C in
a temperature-controlled stage while data was recorded
using a Nanostar SAXS (Bruker AXS, Madison, WI,
USA) with CuKa radiation and a High Star wire-type
detector. Results were analyzed using Bruker AXS soft-
ware version 4.1.26.
X-Ray Powder Diffraction
Solid samples were ground into fine powders and applied
to sample slides with a 0.2 mm deep depression (Rigaku,
Tokyo Japan). An aftermarket temperature controller uti-
lizing Peltier cooling (Electron Dynamics, Southampton,
UK) maintained the sample stage at 20 °C. Powder dif-
fraction data was recorded using a Rigaku Multiflex pow-
der X-ray diffractometer with a CuKa radiation (k =
Table 2 XRD results for 1,3-acyl-palmitoyl-rac-glycerols
˚
Short spacings (A)
FA
Experimental values
4.38wa
4.57s
2:0
4.14vs
3.80sa
3.82vs
3.77sa
4.48m
3.79s
4.53ma
4.52ma
3.76m
3.70w
3.74vs
3.74m
3.71vs
3.87m
3.72s
3.61wa
3.56m
˚
4:0
1.5406 A [23]) and a scintillation detector. Samples were
scanned from 2 to 30 degrees at 2 degrees/min. Results
were analyzed and major peaks were identified using Jade
6:0
4.62w
4.56s
8:0
10:0
12:0
14:0
16:0
18:1
4.60vs
4.58vs
4.63vs
4.63vs
4.64vs
3.67s
3.67s
Table 1 DSC melting data for 1,3-acyl-palmitoyl-rac-glycerols
3.87m
3.73wb
3.67wa
FA
2:0
Te ( °C) Tp ( °C) DHf (kJ/mol) Literature ( °C)
b0 25.33
9.90
31.55
29.58
40.61
38.63
35.67
26.61
50.18
42.92
53.29
49.36
60.96
57.30
61.44
56.89
73.11
71.75
42.83
40.64
–
43.6a
35.0a
42.4a
39.6a
41.0b
33.5b
–
Literature values
4:0
6:0
8:0
b1 38.80
b2 34.91
b1 33.36
b2 12.95
b1 47.99
b2 37.83
67.44
61.04
–
2:0
4.38s
4.14vs
4.11vs
3.85vs
3.8vs
3.54ms
3.80w
3.76vs
3.76vs
[8]
4.34w
4.59vs
4.63vs
3.54m
[24]
[8]
4:0
–
12:0
[27]
[9]
88.42
78.39
88.62
84.99
108.08
97.02
118.91
78.79
129.05
121.32
120.29
112.21
16:0 by polymorph:
b1
b2
4.64vs
4.68vs
4.55s
4.55s
3.82s
3.90s
3.74s
3.78s
10:0 b1 47.95
b2 44.96
54.5c
3.69s
a
Shoulder
12:0 b1 58.84
b2 53.80
*59d
56d
*66e
b
Low-hump
14:0 b1 59.77
b2 52.31
*64f
*44h
Table 3 FTIR data for carbonyl-stretch absorption band of 1,3-acyl-
palmitoyl-rac-glycerols in high-melting form
16:0 b1 70.04
b2 69.46
72.0g
70.0g
46e
FA
Wavenumber (cm-1
)
18:1 b1 42.11
b2 36.51
2:0
–
1,732
1,732
1,732
1,732
1,732
1,732
1,730
1,732
1,730
4:0
1,707
1,709
1,709
1,713
1,709
1,715
1,709
1,707
a
[8]
6:0
b
[24]
8:0
c
[28]
10:0
12:0
14:0
16:0
18:1
d
[27]
e
[29]
f
[30]
g
[9]
h
69% pure [31]
123

1116
J Am Oil Chem Soc (2011) 88:1113–1123
Results
a
5
In the following paragraphs, physical data relating to each
of the prepared compounds is detailed. The synthesis,
purification and analysis of these compounds, along with
NMR results and TLC Rf has been discussed in a prior
publication [15]. The most-thermodynamically stable
polymorph was obtained by storing each sample at -30 °C
for at least 1 week, and was evaluated by DSC, XRD and
FTIR (Tables 1, 2 and 3). DSC melting data for the b1 and
b2 forms (or b0 where they occur) is reported for melts at
5 °C/min. Crystallization and melting behavior was col-
lected at 2.5, 5 and 10 °C/min by DSC to detect and
evaluate polymorphism—these plots are reproduced here.
Finally, SAXS results for all compounds are compared and,
based on the observed trends, the molecular arrangement
within the unit cell is determined for 1,3-acyl-palmitoyl-
rac-glycerols.
3
1
-1
-20
0
20
40
60
80
Exo Up
Temperature (°C)
b
3
1,3-Acetoyl-Palmitoyl-rac-Glycerol (2:0-OH-16:0)
1,3-Acetoyl-palmitoyl-rac-glycerol (2:0-OH-16:0) is
a
pliable, waxy white solid with texture similar to ParafilmÒ
at room temperature. The product was isolated from the
crude product directly by flash chromatography recovering
21.5% yield for a product that is 84.97% pure (by GC).
Difficulties encountered while preparing diacid 1,3-DAG
of high purity have been discussed in detail previously
[15]. The main peaks along with additional peaks due to
polymorphism are visible in DSC heating and cooling
curves (Fig. 1). The melting point of the b0 form
(Te = 25.33 and Tp = 31.55 °C) and an additional peak
(Tp = 11.8 °C) were determined by DSC (Table 1). Melt-
ing points in the literature for the b0 form (43.6 and
41.0 °C) [8, 24] are a great deal higher than this. However,
these discrepancies can most likely be attributed to
1
-1
-20
Exo Up
0
20
40
60
80
Temperature (°C)
Fig. 1 DSC curves for a crystallization and b melting of 1,3-acetoyl-
palmitoyl-rac-glycerol. Plots in each graph were measured at 10 °C/
min (top), 5 °C/min (middle) and 2.5 °C/min (bottom). Dashed line
indicates first melt of high-melting form at 5 °C/min
Fig. 2 X-ray powder
diffraction results for high-
melting forms of: a 1,3-acetoyl-
palmitoyl-rac-glycerol (2:0-
OH-16:0) b0 form and b 1,3-
butyroyl-palmitoyl-rac-glycerol
(4:0-OH-16:0) b1 form
123

J Am Oil Chem Soc (2011) 88:1113–1123
1117
limitations of preparative and analytical methodologies and
to differences in the purity of available starting materials in
previous studies. The XRD results for the current study are
almost identical to those reported in previous studies for
the b0 form (Fig. 2; Table 2) [8, 24]. Interestingly, the b0
form has only one IR absorption peak in the carbonyl
stretching region indicating hydrogen-bonding in the b0
form is uniform (Fig. 3; Table 3). In addition, two meth-
ylene rocking vibrations (719.5 and 729 cm-1) indicate
methylene groups in adjacent acyl chains are packed in an
orthorhombic perpendicular arrangement (i.e. b0-form)
[25].
1,3-Butyroyl-Palmitoyl-rac-Glycerol (4:0-OH-16:0)
1,3-Butyroyl-palmitoyl-rac-glycerol (4:0-OH-16:0) is a
loose, white, talc-like powder at room temperature. Pure
product was isolated by flash chromatography of the crude
product with recoveries of 24.0% (yield) for a product that
is 92.94% pure (by GC). The main peaks along with an
additional peak due to polymorphism are visible in DSC
heating and cooling curves (Fig. 4). The melting points of
both b2 and b1 forms (Te = 34.91 and 38.80 °C corre-
spondingly) were determined by DSC, as was the melting
point of the meta-stable form (Tp = 27.99 °C) (Table 1).
Melting points of 42.4 (form I), 39.6 (form II) and 29.4 °C
(form III) have been reported previously for a sample that
was approximately 96% pure by both melting point and
hydroxyl value [8]. Melting points for forms II and III
correspond well with those for b1 and b2 in this work,
suggesting the reported melting point for form I is actually
due to a contaminant. This would not be surprising given
the quality of starting materials and analytical methodol-
ogies available at the time. The main peaks in the XRD
˚
spectra for the b1 form (4.57, 3.80 and 3.74 A) are similar
˚
to those previously reported (4.59, 3.83 and 3.76 A) [8]
Fig. 3 FTIR for high-melting forms of: a 1,3-acetoyl-palmitoyl-
rac-glycerol (2:0-OH-16:0) b0 form; b 1,3-hexanoyl-palmitoyl-rac-
glycerol (6:0-OH-16:0) b1 form; and c 1,3-dipalmitoyl-glycerol
(16:0-OH-16:0) b1 form
(Fig. 2; Table 2).
1,3-Hexanoyl-Palmitoyl-rac-Glycerol (6:0-OH-16:0)
1,3-Hexanoyl-palmitoyl-rac-glycerol (6:0-OH-16:0) is a
fine white powder that becomes creamy when worked at
room temperature. It was isolated from the crude product
by flash chromatography with recovery approaching 60%
yield for a product that is 86.05% pure (by GC). Purifica-
tion of prepared diacid 1,3-DAG is confounded by
numerous factors as outlined previously [15]. DSC heating
and cooling curves were difficult to interpret due to the
polymorphic behavior of this compound (Fig. 4). The
melting points of both b2 and b1 forms (Tp = 26.61 and
Te = 33.36 °C, correspondingly) were determined by DSC
and extensive polymorphism was observed, confounding
DSC determination of DHf and Te for the b2 form (Fig. 4d;
Table 1). The main peaks in the XRD spectra for the b1
˚
form (4.62, 3.82 and 3.74 A) are similar to those previously
reported (Table 2). A CAS number has yet to be assigned
for this compound indicating synthesis and characterization
has not been reported previously.
1,3-Octanoyl-Palmitoyl-rac-Glycerol (8:0-OH-16:0)
1,3-Octanoyl-palmitoyl-rac-glycerol (8:0-OH-16:0) is a
clean, white waxy solid at room temperature. The pure
product (91.63% by GC) was isolated by flash chroma-
tography from a fraction obtained by recrystallization from
123

1118
J Am Oil Chem Soc (2011) 88:1113–1123
acetone (51% yield for recrystallized solid). The use of
melt crystallization to isolate 8:0-OH-16:0 from its posi-
tional isomers has been disclosed in a patent; however,
actual melting point data for this compound has not been
reported [26]. The main peaks are visible in DSC heating
and cooling curves (Fig. 5). The melting points of both b2
and b1 forms (Te = 37.83 and 47.99 °C correspondingly)
were determined by DSC and no extra polymorphs were
observed (Table 1). The main peaks in the XRD spectra for
(62% yield for recrystallized solid). A number of factors
make the isolation of prepared diacid 1,3-DAG difficult as
discussed previously [15]. The main peaks along with
additional peaks due to polymorphism are visible in DSC
heating and cooling curves (Fig. 5). The melting points of
both b2 and b1 forms (Te = 44.96 and 47.96 °C corre-
spondingly) were determined by DSC (Table 1). Consid-
ering its purity, the b1 form’s Te is close to the previously
reported melting point (determined by Kofler bench) of
54.5 °C (99% purity) [28]. The main peaks in the XRD
spectra for the high-melting b form (4.60, 4.48, 3.87 and
˚
the high-melting b1 form (4.56, 3.77 and 3.71 A) (Table 2)
are similar to those previously reported for the b form of
˚
˚
4:0-OH-16:0 (4.59, 3.83 and 3.76 A) [8] and the b form of
3.67 A) are similar to those of other prepared 1,3-DAG
˚
12:0-OH-16:0 (4.63, 3.80 and 3.76 A) [27].
(Table 2).
1,3-Decanoyl-Palmitoyl-rac-Glycerol (10:0-OH-16:0)
1,3-Lauroyl-Palmitoyl-rac-Glycerol (12:0-OH-16:0)
1,3-Decanoyl-palmitoyl-rac-glycerol (10:0-OH-16:0) is a
white solid with a texture similar to bar soap. Product
(73.98% by GC) was isolated by flash chromatography
from a fraction obtained by recrystallization from acetone
1,3-Lauroyl-palmitoyl-rac-glycerol (12:0-OH-16:0) is a
fine, slightly grainy white powder at room temperature. A
45% yield of[99% pure (by GC) product was attained by
two successive recrystallizations. The main peaks along
a
c
15
10
8
6
10
5
4
2
0
0
-5
-20
-2
-20
0
20
40
60
80
0
20
40
60
80
Temperature (°C)
Exo Up
Exo Up
Temperature (°C)
8
6
4
2
0
5
3
1
b
d
-2
-20
-1
-20
0
20
40
60
80
0
20
40
60
80
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
Fig. 4 DSC curves for a crystallization and b melting of 1,3-
butyroyl-palmitoyl-rac-glycerol and c crystallization and d melting of
1,3-hexanoyl-palmitoyl-rac-glycerol. Plots in each graph were
measured at 10 °C/min (top), 5 °C/min (middle) and 2.5 °C/min
(bottom). Dashed line indicates first melt of high-melting form at
5 °C/min
123

J Am Oil Chem Soc (2011) 88:1113–1123
1119
with additional peaks due to polymorphism are visible in
DSC heating and cooling curves (Fig. 6). The melting
points of both b2 and b1 forms (Te = 53.80 and 58.84 °C,
correspondingly) were determined by DSC (Table 1). This
data compares favorably with values reported previously
(56 °C for Form II—i.e. b2 and 59–59.5 °C for Form I—
i.e. b1) [27] (Table 1). The main peaks in the XRD spectra
were observed (Table 1). Literature melting points reported
for this compound are 65.5–66.5 °C [29] and 63.5–64 °C
[30] (Table 1). While these melting points are somewhat in
agreement, a purer sample would have had a higher melt-
ing point and consequently better agreement. The main
peaks in the XRD spectra for the high-melting b form
˚
(4.63, 4.53, 3.87 and 3.67 A) are similar to those reported
˚
for the b1 form (4.58, 3.79 and 3.72 A) are similar to those
˚
previously reported (4.63, 3.80 and 3.76 A) [27] (Table 2).
in the literature for 16:0-OH-16:0 (4.64, 4.55, 3.82 and
˚
3.74 A) [9] (Table 2).
1,3-Myristoyl-Palmitoyl-rac-Glycerol (14:0-OH-16:0)
1,3-Dipalmitoyl-glycerol (16:0-OH-16:0)
1,3-Myristoyl-palmitoyl-rac-glycerol (14:0-OH-16:0) is a
white, slightly grainy solid with a light and fluffy flour-like
texture at room temperature. Product (72.62% purity by
GC) was purified by two successive recrystallizations. The
main peaks are visible in DSC heating and cooling curves
(Fig. 6). The melting points of both b2 and b1 forms
(Te = 52.31 and 59.77 °C, correspondingly) were deter-
mined by DSC and no extra peaks due to polymorphism
1,3-Dipalmitoyl-glycerol (16:0-OH-16:0) is a light and
fluffy white powder with a texture similar to flour at room
temperature. This product was purified ([99% by GC) by
two successive recrystallizations from hexane/ethyl acetate
(97:3, v/v) for an overall yield of 35%. The main peaks are
visible in DSC heating and cooling curves (Fig. 7). The
tops of the exothermic peaks in the crystallization curves
are distorted by the melt and subsequent recrystallization of
a
20
15
10
5
c
15
10
5
0
0
-5
-20
-5
-20
0
20
40
60
80
0
20
40
60
80
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
10
12
10
8
b
d
8
6
4
2
0
6
4
2
0
-2
-20
-2
-20
0
20
40
60
80
0
20
40
60
80
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
Fig. 5 DSC curves for a crystallization and b melting of 1,3-
octanoyl-palmitoyl-rac-glycerol and c crystallization and d melting of
1,3-decanoyl-palmitoyl-rac-glycerol. Plots in each graph were
measured at 10 °C/min (top), 5 °C/min (middle) and 2.5 °C/min
(bottom). Dashed line indicates first melt of high-melting form at
5 °C/min
123

1120
J Am Oil Chem Soc (2011) 88:1113–1123
the sample caused by rapid evolution of heat. The melting
points of both b2 and b1 forms (Te = 69.46 and 70.04 °C,
correspondingly) were determined by DSC and no extra
peaks due to polymorphism were observed (Table 1). This
melting data agrees with literature values (Tp = 70 and
72 °C, correspondingly); however, the enthalpies of fusion
(DHf = 121.32 kJ/mol for b2 and 129.05 kJ/mol for b1) are
larger than in previous reports (109.4 and 113.9 kJ/mol
correspondingly) [9] (Table 1). The main peaks in the
XRD spectra for the high-melting b form (4.63, 4.52 and
80:20) (46% yield for recrystallized solid). The main peaks
are visible in DSC heating and cooling curves (Fig. 7). The
melting points of both b2 and b1 forms (Te = 36.51 and
42.11 °C correspondingly) were determined by DSC and
no extra peaks due to polymorphism were observed
(Table 1). This melting point is much lower than those
previously reported for the b1 form: 46 °C [29] and
44–44.5 °C (69% purity) [31]. However, given the relative
quality of starting materials and analytical methodologies
available in the past, our sample is probably a better rep-
resentative (95.22% by GC). The main peaks in the XRD
˚
3.73 A) are similar to those previously reported (4.64, 4.55,
˚
3.82 and 3.74 A) [9] (Table 2).
˚
spectra for the b1 form (4.64, 3.76 and 3.67 A) are similar
to those previously reported for other 1,3-DAG (Table 2).
1,3-Oleoyl-Palmitoyl-rac-Glycerol (16:0-OH-18:1)
1,3-Oleoyl-palmitoyl-rac-glycerol (16:0-OH-18:1) is
a
Discussion
fine, grainy white powder at room temperature. Pure
product (95.22% by GC) was isolated by flash chroma-
tography from a fraction that was obtained by two suc-
cessive recrystallizations (1st acetone, 2nd hexane/ethanol;
The most thermodynamically stable crystal form for 1,3-
acyl-palmitoyl-rac-glycerols, with one exception, is b1.
The exception, 1,3-acetoyl-palmitoyl-rac-glycerol is
c
15
a
14
10
5
10
6
0
2
-2
-5
0
20
40
60
80
100
0
20
40
60
80
100
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
b 10
d
8
6
8
6
4
4
2
2
0
0
-2
-2
-4
-4
0
0
20
40
60
80
100
20
40
60
80
100
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
Fig. 6 DSC curves for a crystallization and b melting of 1,3-lauroyl-
palmitoyl-rac-glycerol and c crystallization and d melting 1,3-
myristoyl-palmitoyl-rac-glycerol. Plots in each graph were measured
at 10 °C/min (top), 5 °C/min (middle) and 2.5 °C/min (bottom).
Dashed line indicates first melt of high-melting form at 5 °C/min
123

J Am Oil Chem Soc (2011) 88:1113–1123
1121
b0-tending. Differences in polymorph were confirmed by
both XRD and IR measurement (Figs 2, 3; Tables 2, 3).
Meta-stable polymorphs were observed: (1) in the melting
curves for three of the prepared 1,3-DAG (i.e. 1,3-ace-
toyl-, -butyroyl- and -hexanoyl-palmitoyl-rac-glycerol);
and (2) in the crystallization curves for two of the pre-
pared 1,3-DAG (i.e. 1,3-decanoyl- and -lauroyl-palmitoyl-
rac-glycerol). For large differences in acyl chain length
(case 1 above), the metastable form crystallizes and per-
sists until it is melted—at which point, a more stable
polymorph is formed. For smaller differences in acyl
chain length (case 2 above), the metastable form crys-
tallizes but does not persist and is not detected in the
melting curve. This behavior is a reflection of differences
in free energy between meta-stable and stable forms as
well as differences in the activation energy between these
forms. Interestingly, meta-stable forms occur far more
frequently than was originally thought (five of the nine
cases presented here).
Melting points for 1,3-acyl-palmitoyl-rac-glycerols
increased with increasing chain length for saturated FA
chains (2:0–16:0) excluding 1,3-hexanoyl-palmitoyl-rac-
glycerol. Similarly, the hexanoyl derivative in a series of
1,2-dipalmitoyl-3-acyl-sn-glycerols (2:0–16:0) was the
lowest melting in the series [32]. The melting behavior in
that case, however, was extremely different since the
melting points increased as the acyl chain length both
increased and decreased from 6:0. This phenomenon was
attributed to polymorphic and polytypic differences within
the series. In the current case, polymorphic and polytypic
behavior (determined by SAXS discussed below) was far
more uniform and thus a simpler trend was observed.
However, the melting curve of 1,3-hexanoyl-palmitoyl-
rac-glycerol demonstrated remarkable polymorphism and,
considering its relative purity, the melting point determined
for the high-melting form may indeed be underestimated.
Experimental results from small angle X-ray scattering
measurements on high-melting forms of 1,3-acyl-
a
c
20
25
20
15
10
5
15
10
5
0
0
-5
-5
-20
0
20
40
60
80
100
0
20
40
60
80
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
20
15
10
5
b
d
12
7
2
0
-5
0
-3
-20
20
40
60
80
100
0
20
40
60
80
Exo Up
Temperature (°C)
Exo Up
Temperature (°C)
Fig. 7 DSC curves for a crystallization and b melting of 1,3-
dipalmitoyl-glycerol and c crystallization and d melting 1,3-oleoyl-
palmitoyl-rac-glycerol. Plots in each graph were measured at
10 °C/min (top), 5 °C/min (middle) and 2.5 °C/min (bottom). Dashed
line indicates first melt of high-melting form at 5 °C/min
123

1122
J Am Oil Chem Soc (2011) 88:1113–1123
palmitoyl-rac-glycerols are listed in Table 4. Also listed
are values calculated using the equation:
d ¼ 1:268CN þ 7:07
ð1Þ
where, d is the long spacing and CN is the carbon number.
Equation (1) was derived by linear regression (R2 = 0.9998)
using long-spacing data for monoacid 1,3-DAG of 18:0,
16:0, 14:0, 12:0 and 10:0 FA (52.8, 47.5, 42.6, 37.4 and
˚
32.5 A, respectively) [33]. For the most part, the long
spacings of diacid 1,3-DAG determined experimentally
were very closeto these calculated values (viz. 1,3-butyroyl-,
-hexanoyl-, -octanoyl-, -decanoyl- and -lauroyl-palmitoyl-
rac-glycerols). Indicating acyl chains of the same length
associate and regular chain-end matching between terminal
methyl groups delineates the lamellae (Fig. 8).
Not all experimentally determined long spacings were
well-matched with calculated values (viz. 1,3-acetoyl-,
-myristoyl-. -palmitoyl- and -oleoyl-palmitoyl-rac-glyc-
erol). Discrepancies are due to differences in structure
(oleoyl is cis unsaturated, acetoyl is very short-chain),
polymorph (1,3-acetoyl-palmitoyl-rac-glycerol is b0-sta-
ble), tempering or purity. The oleic acid derivative has an
Fig. 8 Orientation of acyl chains in b1-form of 1,3-decanoyl-
palmitoyl-rac-glycerol (based on [11]). View down the a-axis with
b- and c-axis vertical and horizontal, respectively, along the page
˚
experimental long spacing of 45.96 A which is much less
diacid 1,3-DAG. Molecules in crystalline 1,3-acetoyl-pal-
mitoyl-rac-glycerol appear to be arranged in a manner
similar to 1(3)-monopalmitin as suggested by Feuge and
Lovegren [8]. Hypothetical addition of the appropriate
length for a single or double layer of acetyl chains (2.54 or
˚
5.07 A, respectively) to the long spacing for 1-mono-
˚
palmitin (45.8 A) yields calculated long spacings of either
than the calculated value based on saturated monoacid 1,3-
˚
DAG (50.17 A). This is not unexpected since 1,3-DAG
with cis unsaturated chains yield smaller long spacings
than their saturated counterparts. For example, long spac-
ings for the b1 forms of 1,3-distearin and 1,3-diolein are
˚
52.8 and 39.3 A, respectively [33, 34]. Likewise, the acetic
acid derivative has an experimental long spacing of
˚
49.85 A which is much greater than the calculated value
˚
48.34 or 50.87 A. Both these values are well within range
˚
of the experimental value (49.85 A) indicating there may
˚
˚
(29.89 A), although a minor peak does occur at 31.07 A.
This indicates the majority of 1,3-acetoyl-palmitoyl-rac-
glycerol molecules are oriented differently than other
be full, partial or no interdigitation of acetoyl chains on
adjacent molecules. Furthermore, the aforementioned long
spacing is for a 2L polytype and assumes the angle between
adjacent 1,3-acetoyl-palmitoyl-rac-glycerol molecules will
be the same as for adjacent 1(3)-monopalmitin molecules.
Table 4 SAXS results for 1,3-acyl-palmitoyl-rac-glycerols
FA
Experimental
˚
d (A)
Calculateda
˚
d (A)
Literature
˚
d (A)
Acknowledgments Financial support for this project was provided
by Dairy Farmers of Ontario, Ontario Centres of Excellence and
National Sciences and Engineering Research Council of Canada; we
are grateful for their generosity. Thanks also to Japan VAM &
POVAL Co., Ltd. for their donation of vinyl esters of fatty acids.
2:0
49.85b
32.21
34.74
37.55
39.75
42.63
42.84
45.25
45.96
29.89
32.42
34.96
37.49
40.03
42.56
45.10
47.63
50.17
41.0c, 50.46d
65.46d
4:0
6:0
8:0
10:0
12:0
14:0
16:0
18:1
a
42.7e
References
1. Chawla P, deMan JM (1994) Effect of temperature cycling on the
crystalline form, size and textural properties of margarine fats.
J Food Lipids 1:313–314
2. Hernqvist L, Herslof B, Larsson K, Podlaha O (1981) Polymor-
phism of rapeseed oil with a low content of erucic acid and
possibilities to stabilize the b0-crystal form in fats. J Sci Food
Agric 32:1197–1202
d = 1.268 CN ? 7.07
b
c
d
e
˚
Minor peak at 31.07 A
[24]
[8]
3. Siew W-L, Ng W-L (1999) Influence of diglycerides on crys-
tallization of palm oil. J Sci Food Agric 79:722–726
[27]
123

J Am Oil Chem Soc (2011) 88:1113–1123
1123
4. Smith PR, Cebula DJ, Povey MJW (1994) The effect of lauric-
based molecules on trilaurin crystallization. J Am Oil Chem Soc
71:1367–1372
5. Xu X, Skands ARH, Adler-Nissen J (2001) Purification of spe-
cific structured lipids by distillation: effects on acyl migration.
J Am Oil Chem Soc 78:715–718
6. Christophe AB (2005) Structural effects on adsorption, metabo-
lism, and health effects of lipids Inc. In: Shahidi F (ed) Bailey’s
industrial oil and fat products. John Wiley, Hoboken, pp 535–553
7. Larsson K (1986) Physical properties—structural and physical
characteristics. In: Gunstone FD, Harwood JL, Padley FB (eds)
The lipid handbook. Chapman and Hall, London, pp 321–384
8. Feuge RO, Lovegren NV (1956) Dilatometric properties of some
butyropalmitins, butyrostearins and acetopalmitins. J Am Oil
Chem Soc 33:367–371
9. Shannon RJ, Fenerty J, Hamilton RJ, Padley FB (1992) The
polymorphism of diglycerides. J Sci Food Agric 60:405–417
10. Gunstone FD, Harwood JL, Dijkstra AJ (2007) The lipid hand-
book dictionary (CD) in The Lipid Handbook. CRC Press, Boca
Raton
20. Bernstein J (2002) Polymorphism in molecular crystals. Oxford
University Press, Oxford
21. Hohne GWH, Hemminger WF, Flammersheim HJ (2003) Dif-
ferential scanning calorimetry. Springer, New York
22. Laye PG (2002) Differential thermal analysis and differential
scanning calorimetry. In: Haines PJ (ed) Principles of thermal
analysis, calorimetry. Royal Society of Chemistry, Cambridge,
pp 55–93
23. Cullity BD (1978) Elements of X-ray diffraction. Addison-
Wesley, Reading
24. Martin JB, Lutton ES (1972) Preparation and phase behavior of
acetyl monoglycerides. J Am Oil Chem Soc 49:683–687
25. Chapman D (1965) The structure of lipids by spectroscopic and
X-ray techniques: with a chapter on separation techniques
including thin layer and gas liquid chromatography. Methuen,
London
26. De Groot WTM (1972) 1,3-diglycerides by isomerization of 1,2-
diglycerides. West Germany Patent: 2156091
27. Sidhu SS, Daubert BF (1946) X-ray investigation of glycerides
vs. diffraction analyses of synthetic diacid diglycerides. J Am
Chem Soc 68:2603–2605
28. Mank APJ, Ward JP, van Dorp DA (1976) A versatile flexible
synthesis of 1,3-diglycerides and triglycerides. Chem Phys Lipids
16(2):107–114
29. Daubert BF, Longenecker HE (1944) Unsaturated synthetic gly-
cerides III. Unsaturated symmetrical mixed diglycerides. J Am
Chem Soc 66:53–55
30. Verkade PE, van der Lee J, Meerburg W (1937) The synthesis of
glycerides with the help of trityl compounds III. Triacid glyce-
rides (in German). Recueil des Travaux Chemique des Pays-Bas
56:365–374
11. Hybl A, Dorset D (1971) The crystal structure of the 1,3-
diglyceride of 11-bromoundecanoic acid. Acta Crystallogr A
B27:977–986
12. Larsson K (1963) The Crystal Structure of the 1,3-diglyceride of
3-thiadodecanoic acid. Acta Crystallogr A 16:741–748
13. Nutter B (2009) Nu-Chek Prep, Inc. Catalog 2009/2010, Elysian
14. Gaffney PRJ, Reese CB (2001) Synthesis of naturally occurring
phosphatidylinositol 3, 4, 5-triphosphate [PtdIns(3, 4, 5)P3] and
its diastereomers. J Chem Soc Perkin Trans 1:192–205
15. Craven RJ, Lencki RW (2010) Preparation of diacid 1,3-diacyl-
glycerols. J Am Oil Chem Soc 87(11):1281–1291
16. Halldorsson A, Magnusson CD, Haraldsson GG (2003) Chemo-
enzymatic synthesis of structured triacylglycerols by highly
regioselective acylation. Tetrahedron 59:9101–9109
17. Quinn JG, Sampugna J, Jensen RG (1967) Synthesis of 100-gram
quantities of highly purified mixed acid triglycerides. J Am Oil
Chem Soc 44:439–442
18. Jensen RG, Pitas RE (1976) Synthesis of some acylglycerols and
phosphoglycerides. In: Paoletti R, Kritchevsky D (eds) Advances
in Lipid Research. Academic Press, New York, pp 214–247
19. Yu CC, Lee Y-S, Cheon BS, Lee SH (2003) Synthesis of glycerol
monostearate with high purity. Bull Korean Chem Soc 24:
1229–1231
31. Hartman L (1957) Glyceride synthesis by direct esterification.
J Chem Soc 3572–3575
32. Kodali DR, Atkinson D, Redgrave TG, Small DM (1984) Syn-
thesis and polymorphism of 1,2-dipalmitoyl-3-acyl-sn-glycerols.
J Am Oil Chem Soc 61:1078–1084
33. Howe RJ, Malkin T (1951) An X-ray and thermal examination of
the glycerides. Part XI. The 1:2-diglycerides, and further obser-
vations on 1:3-diglycerides. J Chem Soc 2663–2667
34. Daubert BF, Lutton ES (1947) X-ray diffraction analyses of
synthetic unsaturated monoacid diglycerides. J Am Chem Soc
69:1449–1451
123