Kinetics of cinnamoyl glycerol formation

Article abstract of DOI:10.1007/s11746-007-1189-3

The esterifications of glycerol with cinnamic acid, 2-methoxy cinnamic acid, and 4-methoxy cinnamic acid were investigated in batch reactions. Conversions of over 50% were achieved for cinnamic acid and 4-methoxy cinnamic acid within 8 h. After 24 h, conv

Full text of DOI:10.1007/s11746-007-1189-3

J Am Oil Chem Soc (2008) 85:221–225
DOI 10.1007/s11746-007-1189-3
ORIGINAL PAPER
Kinetics of Cinnamoyl Glycerol Formation
Ronald A. Holser
Received: 18 July 2007 / Revised: 26 November 2007 / Accepted: 11 December 2007 / Published online: 4 January 2008
Ó AOCS 2008
Abstract The esterifications of glycerol with cinnamic
acid, 2-methoxy cinnamic acid, and 4-methoxy cinnamic
acid were investigated in batch reactions. Conversions of
over 50% were achieved for cinnamic acid and 4-methoxy
cinnamic acid within 8 h. After 24 h, conversions of over
80% were obtained for cinnamic acid and 4-methoxy cin-
namic acid. Conversions of 33 and 40% were observed for
2-methoxy cinnamic acid after 8 and 24 h, respectively.
These reactions were modeled by a system of sequential
first-order rate expressions. Kinetic parameters were esti-
mated from experimental data fit to the model equations.
and exhibits such favorable properties as low toxicity, low
volatility, and water solubility that are suitable for the
development of environmentally benign products and
processes. Recent work examined the esterification of
glycerol with cinnamic acid via the Mitsunobu reaction [1–
3]. However, this involved the use of protecting groups to
prevent the formation of unwanted products. While suc-
cessful, this approach requires additional synthetic steps to
block reaction sites and recover product.
Cinnamoyl esters are commonly used as organic
ultraviolet (UV) filters in sunscreens and cosmetic
formulations [4, 5]. Compounds such as octyl methoxy-
cinnamate (OMC) and 2-ethylhexyl-p-methoxycinnamate
are typical and the esters of longer chain fatty acids,
hydroxy fatty acids, and epoxy fatty acids have also been
prepared [6–8].
Keywords Co-products Á Synthesis Á Ester Á
Reaction rates
Introduction
Reports have appeared concerning the risk associated
with the absorption of such lipid compounds from sun-
screen formulations [9–11]. The risk focuses on the
bioactivity of these compounds and their potential to act as
endocrine disruptors. Clinical investigations also confirmed
the transdermal properties of these compounds [12, 13]. In
one study with human test subjects exposed for 1 week the
endocrine effects of common sunscreens were measured,
however, the long-term effects of repeated applications are
not known [13].
The current surplus of glycerol generated by the increase in
fatty methyl ester production for fuel use, i.e., biodiesel,
provides a versatile substrate for the synthesis of biobased
products. Glycerol possesses a high degree of functionality
Disclaimer The use of trade, firm, or corporation names in this
publication is for the information and convenience of the reader.
Such use does not constitute an official endorsement or approval by
the United States Department of Agriculture or the Agricultural
Research Service of any product or service to the exclusion of others
that may be suitable.
Preparation of the more hydrophilic cinnamoyl esters of
glycerol were investigated as alternative UV filters with
limited transdermal properties. The associated risk would
be significantly reduced with such a hydrophilic com-
pound. The reactions of glycerol with cinnamic acid, 2-
methoxy cinnamic acid, and 4-methoxy cinnamic acid
were studied to explore the structural and positional
influence of the substituent methoxy group on reaction rate,
selectivity, and conversion.
R. A. Holser (&)
Richard Russell Research Center, Agricultural Research Service,
United States Department of Agriculture,
950 College Station Rd, Athens, GA 30605, USA
e-mail: Ronald.Holser@ars.usda.gov
123

222
J Am Oil Chem Soc (2008) 85:221–225
Experimental
Materials
software. Strong absorbances in the 3,200–3,600 cm-1
region were observed for hydroxyl groups and near
1,700 cm-1 for ester linkages.
Glycerol (99.9%), cinnamic acid (97%), 2 methoxy cin-
namic acid (98%), 4 methoxy cinnamic acid (99%), toluene
(HPLC grade), and methanol (HPLC grade) were pur-
chased from Sigma–Aldrich Chemicals, (St. Louis, MO).
The catalyst p-toluenesulfonic acid (ptsa), 99%, was pur-
chased from Fisher Scientific Co. (Pittsburgh, PA). All
materials were used as received.
GC-MS
Mass spectra of silylated samples were collected using the
Agilent 6890N gas chromatograph equipped with the 5973
mass selective detector operated in EI mode. Separations
were achieved on the HP-5ms column, 30 m 9 0.25 mm
ID 9 0.25 lm film thickness. Helium was used as the
carrier gas with a linear velocity of 35 cm/s. The oven
temperature was programmed from 120 to 240 °C at
10 °C/min with an initial 2 min hold and a final 10 min
hold. The inlet was heated to 230 °C and set for splitless
injections with a 1 lL injection volume. The detector
source was heated to 230 °C and the detector quadrupole
was heated to 150 °C. Data were collected and processed
via Chemstation software.
Esterification
Reactions were performed in 250-mL glass vessels fitted
with a condenser and Dean Stark trap to remove the water
of reaction. A magnetic stirrer bar was placed into the glass
vessel and 2 g quantities of glycerol, and either cinnamic
acid, 2-methoxy cinnamic acid, or 4-methoxy cinnamic
acid were added. The reactants were mixed with 100 mL
toluene and 100 mg ptsa catalyst while heating to reflux
temperature (110 °C). The reaction mixtures were sampled
periodically for the first 8 h and then after 24 h. Samples
were diluted in methanol for chromatographic analysis. All
reactions were replicated and analyses were performed in
duplicate.
Molecular Modeling
Solubility parameters were calculated using Molecular
Modeling Pro, Norgwyn Montgomery Software, Inc.,
North Wales, PA.
High Performance Liquid Chromatography (HPLC)
Results and Discussion
Samples were analyzed with the Agilent 1100 HPLC using
a diode array detector (Agilent Technologies, Inc., Palo
Alto, CA). Separations were achieved on a C18 column
measuring 150 mm 9 4.6 mm with 5 lm diameter pack-
ing (Alltech Associates, Inc., Deerfield, IL). Compounds
were eluted from the column with isocratic methanol at
0.5 mL/min. The UV absorbance was monitored at 288 nm
and spectra were obtained by scanning peaks in the range
190–400 nm. Data were collected and processed via
Chemstation software (Agilent Technologies, Inc., Palo
Alto, CA).
The reaction of glycerol with cinnamic acid, 2-methoxy
cinnamic acid, and 4-methoxy cinnamic acid proceeded at
reflux conditions. Products were characterized by FTIR and
GC-MS. The infrared spectra displayed a strong absor-
bance at 1,700 cm-1 which indicated the presence of the
ester function. The shift from a more typical absorbance at
1,740 cm-1 is attributed to the double bond conjugated
with the aromatic ring structure. A pair of strong sharp
bands was observed at 1,172 and 1,116 cm-1 consistent
with the C–O stretching mode. Analysis by GC-MS iden-
tified the molecular ions of the monoester and diester
products. The location of the ester linkages at the primary
alcohol sites of glycerol were determined from the frag-
mentation pattern of the trimethyl silyl derivatives [14].
The progress of the reaction was followed by the disap-
pearance of the acid and the formation of products. This was
easily measured by the strong UV absorbance of the aro-
matic group. Plots of concentration versus time are shown
for the esterification of cinnamic acid and glycerol in Fig. 1.
The monoester was produced rapidly in the first 3 h before
the appearance of the diester product. After 8 h of reaction
time the cinnamic acid has decreased to 41.9% of the starting
Infrared Spectra (IR)
Infrared spectra were obtained with a Thermo Nicolet,
Nexus FT-IR 470 spectrometer, using the ZnSe ATR
accessory. Samples, 10–20 mg, were dissolved in 2-mL
volumes of diethyl ether and a drop was placed onto the
ZnSe crystal. The solvent was allowed to evaporate before
the spectra were collected. Spectral data were collected
over the range 600–4,000 cm-1 and processed by Omnilab
123

J Am Oil Chem Soc (2008) 85:221–225
223
Fig. 3 Concentration versus time plot for the reaction of 2-methoxy
cinnamic acid and glycerol
Fig. 1 Concentration versus time plot for the reaction of cinnamic
acid and glycerol
concentration with production of 39.1% of the monoester
and 19% of the diester. Formation of the 4-methoxy and the
2-methoxy cinnamoyl glycerol esters followed similar
trends with the 4-methoxy cinnamic acid decreasing to
47.5% and the 2-methoxy cinnamic acid decreasing to
67.5% of their initial concentrations (Figs. 2, 3).
The composition of the reaction mixtures after 8 and
24 h are compared in Table 1. Conversions of 82.8% for
cinnamic acid and 88.1% for 4-methoxy cinnamic acid
were obtained after 24 h. The conversion of 2-methoxy
cinnamic acid was 40% after 24 h. The yields of the ester
products are tabulated with the corresponding acid. In all
cases the yields of monoesters exceeded those of the
diesters which indicated that the reaction rates for the
formation of the monoesters was greater than the rates for
the diesters. Selectivities for the monoesters ranged from
2.1 for the 4-methoxy cinnamic acid esters to 3.2 for the 2-
methoxy cinnamic acid esters after 8 h compared to 1.1
and 2.5, respectively, after 24 h.
Based on these results it appears possible to obtain the
monoester product in nearly 100% yield by performing the
reaction in a continuous flow reactor system. In this way
Table 1 Composition of reaction mixtures after 8 and 24 h at reflux
Compound
8 h (%)
24 h (%)
Cinnamic acid
Monoester
41.9
39.1
19.0
47.5
35.8
16.7
67.5
24.8
7.7
17.2
45.1
37.7
11.9
46.7
41.4
60
Diester
4-Methoxy cinnamic acid
Monoester
Diester
2-Methoxy cinnamic acid
Monoester
28.5
11.5
Fig. 2 Concentration versus time plot for the reaction of 4-methoxy
cinnamic acid and glycerol
Diester
123

224
J Am Oil Chem Soc (2008) 85:221–225
Table 2 Equations used to
model the sequential reaction of
cinnamic acid with glycerol
Reaction 1: monoester
Rate expression 1
Cinnamic acid + glycerol ? cinnamoyl glycerol + water
d/dt [cinnamoyl glycerol] = K1*[cinnamic acid]
Reaction 2: diester
Cinnamic acid + cinnamoyl glycerol ? dicinnamoyl
glycerol + water
Rate expression 2
d/dt [dicinnamoyl glycerol] = K2*[cinnamoyl glycerol]
Combined rate expression
-d/dt [cinnamic acid] = d/dt [cinnamoyl glycerol] + d/dt
[dicinnamoyl glycerol]
Square brackets denote
compound concentrations, K1
and K2 are the first-order rate
constants, and the time
derivative operator is
or
-d/dt [cinnamic acid] = K1* [cinnamic
acid] + K2*[cinnamoyl glycerol]
represented by d/dt
conversion of 4-methoxy cinnamic acid was only slightly
slower while 2-methoxy cinnamic acid exhibited the
slowest rate. The rate of diester formation from cinnamic
acid and 4-methoxy cinnamic acid were similar while 2-
methoxy cinnamic acid again exhibited the slowest rate.
The differences in the reaction rates observed between the
2- methoxy and 4-methoxy cinnamic acids are attributed to
the position of the methoxy group on the aromatic ring of
the cinnamic acid. This was interpreted to be due pre-
dominately to conformational effects. The proximity of the
methoxy group to the side chain containing both the
unsaturated bond and the carboxylic acid function is
reduced when the methoxy group is attached at the para
position compared to the ortho position.
Table 3 Rate constants determined by fitting experimental data to
sequential first-order irreversible model equations
K1 (h-1) K2 (h-1)
r
P
Cinnamic acid
0.1286
0.0499
0.0583
0.0266
0.9760 0.0231
0.9778 0.0221
0.9776 0.0223
4-Methoxy cinnamic acid 0.1219
2-Methoxy cinnamic acid 0.0610
the monoester may be removed from the product stream
before the formation of the diester. The appropriate con-
tinuous flow reactor system for the production of
cinnamoyl glycerol would have a residence time of less
than 3 h. The same strategy could be applied for the
reactions of the 2- methoxy and 4-methoxy cinnamoyl
glycerol esters using residence times of less than 4 and 2 h,
respectively. For efficient operation in this continuous
mode the solvent would be recycled to the reactor after
product recovery.
The UV spectra of cinnamic acid, 2-methoxy cinnamic
acid, and 4-methoxy cinnamic acid are shown in Fig 4. The
spectra of the esters exhibit the same absorbance charac-
teristics as the acids from which they are derived.
However, the spectra of the acids display significant dif-
ferences. The absorbance spectra of 2-methoxy cinnamic
acid and 4-methoxy cinnamic acid extend to 350 nm
compared to 315 nm for cinnamic acid. Materials absorb-
ing at the longer wave lengths are more useful as UV filters
[11]. These methoxy cinnamoyl glycerol esters also pos-
sess more hydrophilic character than cinnamoyl esters
prepared from fatty compounds such as vegetable seed oils
and offer potential for use in personal care products as UV
filters where minimal transdermal properties are required.
Solubility parameters were calculated for the monoester
The kinetics of these sequential esterification reactions
were modeled by a series of first-order irreversible rate
expressions (Table 2). Experimental data were fit to the
model equations by a least squares technique to obtain
values of the rate constants. These results are listed in
Table 3. The rate of monoester formation, K1, was more
than twice the rate of diester formation, K2, for all the
reactions studied. The rate of conversion of cinnamic acid
to the monoester was the most rapid. The rate of
Table 4 Calculated solubility parameters for 4-methoxy cinnamoyl
glycerol and cinnamoyl stearoyl glycerol
Compounda
4mCG
D4mCG
CSG
Log KOWb
Hydrophilic/lipophilic balancec
0.354711
10.8524
0.76598
7.78674
5.9949
4.3652
a
4mCG: 4-methoxy cinnamoyl glycerol; D4mCG: 1, 3-di-4-methoxy
cinnamoyl glycerol; CSG: 1-cinnamoyl, 3-stearoyl glycerol
b
Log octanol/water partition coefficient using modified Hansch
fragments
Fig. 4 UV spectra of cinnamic acid, 2-methoxy cinnamic acid, and
4-methoxy cinnamic acid
c
Calculated by Griffin’s method
123

J Am Oil Chem Soc (2008) 85:221–225
225
5. Freitas ZMF, Goncalves JCS, Santos EP, Verganini A (2005)
Glyceridic esters of p-methoxycinnamic acid. A new sunscreen of
the cinnamate class. Eur J Pharm Sci 25:67–72
6. Compton D, Laszlo JA, Berhow MA (2000) Synthesis of ferulate
esters. J Am Oil Chem Soc 77:513–519
7. Compton DL, Laszlo JA, Isabell TA (2004) Cinnamoyl esters of
lesquerella and castor oil: novel sunscreen active ingredients.
J Am Oil Chem Soc 81:945–951
8. Esen H, Kusefoglu SH (2003) Photolytic and free-radical poly-
merization of cinnamate esters of epoxidized plant oil
triglycerides. J Appl Polym Sci 89:3882–3888
9. Benson HA (2000) Assessment and clinical implications of
absorption of sunscreens across skin. Am J Clin Dermatol 1:217–
224
10. Freitas ZMF, Machado GM, Dellamora-Oritz GM, Santos EP,
Goncalves JCS (2000) Evaluation of phototoxicity of different
sunscreens: 1,2,3-propanetriol-1,3-dipalmitoyl-2-p-methoxycin-
namoyl and 1,2,3-propanetriol-1,3-dioctanoyl-2-p-methoxy-
cinnamoyl. STP Pharma Sci 10:239–242
and diester products of 4-methoxy cinnamic acid with
glycerol using molecular modeling techniques. These
results are presented in Table 4 and compared to values
calculated for the diester compound formed from glycerol
with cinnamic acid and stearic acid. The log of the octanol/
water partition coefficient shows an increasing trend as
glycerol is derivatized with increasingly lipophilic sub-
stituents. A similar shift towards lipophilic behavior is seen
in the decreasing values of the hydrophilic/lipophilic bal-
ance (HLB). This solubility behavior can be significant in
formulations where a UV filter is needed but a strongly
lipophilic compound would either not be compatible with
other ingredients or otherwise problematic.
Acknowledgement The author is grateful to Erin Walter for labo-
ratory analyses.
11. Nohynek GJ, Schaefer H (2001) Benefit and risk of organic
ultraviolet filters. Reg Tox Pharma 33:285–299
12. Chatelain E, Gabard B, Surber C (2003) Skin penetration and sun
protection factor of five UV filters: effect of the vehicle. Skin
Phamacol Appl Skin Physiol 16:28–35
References
13. Janjua NR, Mogensen B, Andersson AM, Petersen JH, Henriksen
M, Skakkebaek NE, Wulf HC (2004) Systemic absorption of the
sunscreens benzophenone-2, octyl-methoxycinnamate, and 3-(4-
methyl-benzylidene) camphor after whole-body topical applica-
tion and reproductive hormone levels in humans. J Invest
Dermatol 123:57–61
1. Mitsunobu O (1981) The use of diethyl azodicarboxylate and
triphenylphosphine in synthesis and transformation of natural
products. Synthesis 1:1–28
2. Hughes DL, Reamer RA, Bergan JJ, Grabowski EJJ (1988) A
mechanistic study of the Mitsunobu esterification reaction. J Am
Chem Soc 110:6487–6491
14. Graca J, Pereira H (2000) Suberin structure in potato periderm:
glycerol, long-chain monomers, and glyceryl and feruloyl dimers.
J Agric Food Chem 48:5476–5483
3. Batovska DI, Kishimoto T, Bankova VS, Kamenarska ZG,
Ubukata M (2005) Synthesis of some phenylpropenoid mono-
glycerides via the Mitsunoba protocol. Molecules 10:552–558
4. Freitas ZMF, Goncalves JCS, Santos EP, Verganini A (2001)
Glyceridic esters of p-methoxycinnamic acid. A new sunscreen of
the cinnamate class. Int J Cosmet Sci 23:147–152
123