Quaternary Alkyl Ammonium Salt-Catalyzed Transformation of Glycidol to Glycidyl Esters by Transesterification of Methyl Esters

Article abstract of DOI:10.1021/acscatal.7b03303

Catalytic transformation of glycidol while maintaining its epoxide moiety intact is challenging because the terminal epoxide that interacts with the hydroxyl group via a hydrogen bond is labile for the ring-opening reaction. We found that a quaternary alkyl ammonium salt catalyzes the selective transformation of glycidol to glycidyl esters by transesterification of methyl esters. The developed method can be applied to the synthesis of multiglycidyl esters, which are valuable epoxy resin monomers. Mechanistic studies revealed the formation of a binding complex of glycidol and quaternary alkyl ammonium salt in a nonpolar solvent and the generation of the alkoxide anion as a catalyst through the ring-opening reaction of the epoxide. Computational studies of the reaction mechanism indicated that the alkoxide anion derived from glycidol tends to abstract the proton of another glycidol rather than work as a nucleophile, initiating the catalytic transesterification. Payne rearrangement of the deprotonated glycidol, which produces a destabilized base that promotes nonselective reactions, is energetically unfavorable due to the double hydrogen bond between the anion and diol. The minimal interaction between the quaternary alkyl ammonium cation and the epoxide moiety inhibited the random ring-opening pathway leading to polymerization.

Full text of DOI:10.1021/acscatal.7b03303

Research Article
pubs.acs.org/acscatalysis
Cite This: ACS Catal. 2018, 8, 1097−1103
Quaternary Alkyl Ammonium Salt-Catalyzed Transformation of
Glycidol to Glycidyl Esters by Transesterification of Methyl Esters
Shinji Tanaka,*,† Takuya Nakashima,† Toshie Maeda,† Manussada Ratanasak,‡ Jun-ya Hasegawa,*,‡
Yoshihiro Kon,† Masanori Tamura,† and Kazuhiko Sato*,†
†Interdisciplinary Research Center for Catalytic Chemistry, National Institute of Advanced Industrial Science and Technology,
Central 5, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
‡Institute for Catalysis, Hokkaido University, Kita 21, Nishi 10, Kita-ku, Sapporo 001-0021, Japan
S
* Supporting Information
ABSTRACT: Catalytic transformation of glycidol while
maintaining its epoxide moiety intact is challenging because
the terminal epoxide that interacts with the hydroxyl group via a
hydrogen bond is labile for the ring-opening reaction. We found
that a quaternary alkyl ammonium salt catalyzes the selective
transformation of glycidol to glycidyl esters by transester-
ification of methyl esters. The developed method can be applied
to the synthesis of multiglycidyl esters, which are valuable epoxy
resin monomers. Mechanistic studies revealed the formation of
a binding complex of glycidol and quaternary alkyl ammonium
salt in a nonpolar solvent and the generation of the alkoxide
anion as a catalyst through the ring-opening reaction of the
epoxide. Computational studies of the reaction mechanism indicated that the alkoxide anion derived from glycidol tends to
abstract the proton of another glycidol rather than work as a nucleophile, initiating the catalytic transesterification. Payne
rearrangement of the deprotonated glycidol, which produces a destabilized base that promotes nonselective reactions, is
energetically unfavorable due to the double hydrogen bond between the anion and diol. The minimal interaction between the
quaternary alkyl ammonium cation and the epoxide moiety inhibited the random ring-opening pathway leading to
polymerization.
KEYWORDS: organocatalyst, quaternary ammonium salt, glycidol, glycidyl ester, transesterification
Scheme 1. Catalytic Transformation of Glycidol
INTRODUCTION
■
Glycidol has attracted heightened interest due to its peculiar
structure and reactivity, derived from the proximal accumu-
lation of an epoxide and a hydroxyl group.1−6 The development
of glycidol as a feedstock for various fine chemicals is
considered an important challenge7 because recent studies
suggested that glycidol can potentially be produced from
glycerol and its derivatives.8 Both acids and bases catalyze the
transformation of glycidol (Scheme 1). Acid catalysts activate
the epoxide moiety and convert it to various functional
alcohols, including chiral alcohols, by the ring-opening
reaction;9 at the same time, base catalysts induce the ring-
opening polymerization that results in the formation of
hyperbranched polymers.10−16 Another application of glycidol
involves a transformation that keeps its epoxide intact, serving
as an alternative method for the production of glycidyl
compounds, which are valuable monomers for functional
epoxy resins (Scheme 1).17,18 The labile terminal epoxide
moiety interacts with the hydroxy group via intramolecuar14
and intermolecular hydrogen bonds,19 however, thus hamper-
ing such selective transformations,20 except for reactions with
highly reactive substrates.6,21,22 A strategy for activating glycidol
by a catalyst that is neither an acid nor a base is therefore
required.
Glycidyl esters are an attractive target compound that can be
prepared from glycidol because they are useful monomers of
epoxy resins with high insulating properties.17 Glycidyl esters
Received: September 26, 2017
Revised: December 3, 2017
© XXXX American Chemical Society
1097
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
are conventionally synthesized by an epichlorohydrin process,
in which a chlorine atom of epichlorohydrin is replaced by
carboxylate and an equimolar amount of salt is generated as
waste.17,23 One issue with this process is the side reactions that
generate organochlorine compounds as byproducts, because
contamination by these byproducts negatively affects the
performance of the epoxy resin. Furthermore, the recent use
of halide compounds as commodity chemicals has been
somewhat restricted due to their potential toxicity.24 Trans-
esterification of methyl esters with glycidol would generate
glycidyl esters in a more environmentally benign way than the
conventional process. Current catalysts, however, are limited to
toxic salts such as Tl salts,25 metal cyanide,26,27 and
selenocyanide,26 presumably because common acid catalysts
are not suitable for transesterification.28−30
benzoate (3a) was obtained in low yields in the presence of
bases such as Et3N, (n-Oct)3N, DMAP, and PPh3 (Figure 1,
entries 2−5). Quaternary onium salts facilitated the reaction
(Figure 1, entries 6, 8−12), and interestingly, the use of
quaternary alkyl ammonium chlorides with longer alkyl chains
gave higher yields of 3a, while that of NH4Cl resulted in 0%
yield (Figure 1, entries 7−12). Screening of counteranions of
(n-Bu)4N+ salts indicated that Cl− was optimal (Figure 1, entry
10), and other halogen anions, nitrate, and hydroxide also acted
as catalysts in this reaction (Figure 1, entries 13−20).
Upon optimizing the catalytic conditions,31 we evaluated
methyl esters (Table 1). 4-Substituted phenyl glycidyl esters
a
Table 1. Scope of Substrates
To develop a method in which glycidol acts as a nucleophile
during transesterification, we assumed that a neutral organo-
catalyst would be reasonable to mildly activate a molecule that
is both acid- and base-sensitive. Organic salts having a
distributed cation charge with a less basic anion would separate
glycidol as a monomer form from the dimer structure in
solution1 and enhance the nucleophilic nature under mild
conditions. Here we found that a simple quaternary alkyl
ammonium salt facilitates the transesterification of methyl
esters with glycidol, affording glycidyl esters. The mechanism of
activation of glycidol by a quaternary alkyl ammonium salt
during the catalytic cycle was elaborated by spectroscopic
analysis as well as computational studies.
RESULTS AND DISCUSSION
■
Transesterification of methyl benzoate (1a) with 1.0 equiv of
glycidol (2) in hexane at reflux temperature (69 °C) did not
proceed at all without the catalyst (Figure 1, entry 1). Glycidyl
a
Reaction conditions: methyl ester (1.0 mmol), 2 (1.0−3.0 mmol),
catalyst (0.0010−0.050 mmol), hexane (0.5 mL), at reflux temperature
b
(69 °C) for 2−3 h. Yields of isolated products are shown. 0.1 mol %
c
d
e
f
of catalyst. 3 h. 1 mol % of catalyst. 2.5 mol % of catalyst. 0.5 mol %
g
of catalyst. Hexane (0.5 mL) + toluene (0.25 mL) was used as
solvent. 2.0 equiv of 2. 3.0 equiv of 2.
h
i
such as glycidyl 4-nitrobenzoate (3b), glycidyl 4-cyanobenzoate
(3c), glycidyl 4-bromobenzoate (3d), and glycidyl 4-
chlorobenzoate (3e) were obtained in 75%−87% yields. Yields
of esters with an electron-donating group in the aryl moiety,
such as glycidyl 4-tolylate (3f) and glycidyl 4-anisylate (3g),
however, were relatively low. Esters bearing heteroaromatic
rings, such as glycidyl 2-pyridylate (3h), glycidyl 2-furylate (3i),
glycidyl 2-thienylate (3j), and an alkenyl part, glycidyl
cinnamate (3k) , were also prepared in moderate to high
yields in the presence of a catalytic amount of (n-
Dodec)3MeNCl (Dodec = C12H25). Moreover, substrates
bearing more than two methyl ester moieties were converted
to multiglycidyl compounds using (n-Dodec)3MeNCl and 2;
diglycidyl terephthalate (3l), diglycidyl 2,7-naphthalenedicar-
boxylate (3m), and triglycidyl 1,3,5-benzenetricarboxylate (3n)
were isolated in 43%−64% yields. Notably, those multiglycidyl
esters are useful monomers for functional epoxy resins, and are
conventionally prepared from carboxylic acids and epichlor-
ohydrin.17,23 Alternatively, we successfully prepared these
Figure 1. Catalytic transesterification of 1a with 2. Reaction
conditions: 1a (1.0 mmol), 2 (1.0 mmol), catalyst (0.050 mmol),
hexane (0.5 mL), at reflux temperature (69 °C) for 2 h. Yields are
1
determined by H NMR using biphenyl as internal standard.
1098
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
multiglycidyl esters using by methyl esters and 2 without
epichlorohydrin.
Studies of the transesterification of 1a with 1 equiv of
aliphatic alcohols in the presence of (n-Oct)3MeNCl revealed
the characteristic trend of this reaction. 1-Hexanol (4) was not
converted to hexyl benzoate (5) under the same conditions as
for 2 (Scheme 2a). In contrast, catalytic transesterification of 1a
Scheme 2. Transesterification with an Aliphatic Alcohol
(a,b) and Methyl Carbonate Salt (c) as a Catalyst
Figure 2. Titration of 2 with (n-Oct)3MeNCl monitored by 1H NMR
in toluene-d8 (400 MHz, 25 °C). The concentration of 2 was 12 mM.
A molar ratio of [2]:[(n-Oct)3MeNCl] was varied from 1:0 to 1:10.
Peaks below asterisks (*) are attributed to impurities of [(n-
Oct)3MeNCl].
with 0.5 equiv of 2 and 0.5 equiv of 4 afforded 8% of 5 together
with 3a in 11% (Scheme 2b), indicating the involvement of 2 as
the catalyst. Quaternary alkyl onium salts with methyl
carbonate catalyze the transesterification of dialkyl carbonate32
and methyl esters with aliphatic alcohols.33 Transesterification
of 1a with 2 in the presence of 5 mol % of (n-Oct)3MeN-
(OCO2Me),32 however, did not afford the desired product 3a
(Scheme 2c). Thus, it is likely that the anion species derived
from alkyl ammonium chloride and 2 acts as the catalyst that
converts 2 to glycidyl esters in a selective manner.
research describing the partial ring-opening of the epoxide by
ammonium salts,39 it is considered that an alkoxide anion is
generated from the complex [(n-Oct)3MeNCl·2] through the
ring-opening of the epoxide by attack of the Cl anion.
Possible mechanisms (Paths A, B, B′, B1, and B2) are
depicted in Scheme 3, and the potential energy profiles for each
mechanism were investigated by density functional theory
computations at ωB97XD/6-311+G(d,p)//ωB97XD/6-
31G(d) (Figure 3). For the calculation, (n-Bu)4NCl was used
as a model catalyst, and n-hexane was incorporated as an
implicit solvent (see SI for more detail). Optimized structures
of the reactant, product, intermediate, and transition states are
displayed in Figure S4 for Path A, Figure S5 for Paths B and B′,
Figure S6 for Path B1, and Figure S7 for Path B2. First, the
ammonium salt binds to glycidol, affording a precursor complex
in reactant state R, which generates a ring-opened anion species
in intermediate state A_Int-1 through a transition state,
A_TSR‑1. This ring-opening event goes through a high-energy
barrier (Ea = 26.5 kcal/mol, A_TSR‑1), and the resulting
intermediate state A_Int-1 is destabilized by 8.6 kcal/mol
compared with R. In Path A, A_Int-1 converts to A_Int-2 via a
low activation energy transition state (Ea = 4.3 kcal/mol,
A_TS1−2) by intramolecular proton transfer, and A_Int-2 could
easily convert back to A_Int-1 (Ea = 2.7 kcal/mol, A_TS1−2),
which means that this step is an equilibrium process. After this
step, methyl ester addition occurs (A_Int-3) and nucleophilic
attack of the alkoxide species in the A_Int-3 state on a carbonyl
group of methyl ester gives the intermediate state A_Int-4. The
energy barrier of the C−O bond formation (Ea = 7.1 kcal/mol,
A_TS3−4) is lower than that of A-TSR‑1 (Ea = 26.5 kcal/mol),
indicating that C−O bond formation is not the rate-
determining step in Path A. After the dissociation of methanol
(MeOH), the ring-closing reaction in the A_Int-5 state
produces the glycidyl ester and regenerates the (n-Bu)4NCl,
and the catalytic cycle returns to the reactant state R. In Path A,
To elucidate the role of quaternary alkyl ammonium salts, we
recorded 1H NMR of a mixture with glycidol in toluene-d8. The
1H NMR spectrum of a mixture of (n-Oct)3MeNCl and 2
displayed a lower magnetic-field shift of signals due to 2 and a
higher magnetic-field shift of signals assignable to the −NCH3
and −NCH2− groups in (n-Oct)3MeNCl. In fact, the binding
constant (Ka) for the association between (n-Oct)3MeNCl and
2 in toluene-d8 at 25 °C was 5.9 0.1 × 102 M−1 based on an
NMR titration experiment with a 1:1 complexation nonlinear
regression model (Figure 2 and S1).35,36 Compared with the Ka
of the dimerization of 2 in toluene-d8 (1.2 0.1 M−1),19 the
large Ka value for the complex [(n-Oct)3MeNCl·2] suggests
that 2 in nonpolar solvent favors complexation with (n-
Oct)3MeNCl rather than dimerization. 1H−1H NOESY analysis
of a mixture of (n-Oct)3MeNCl with 2 in toluene-d8 displayed a
cross-correlation peak between all protons of 2 and − NCH3
and −NCH2− (Figure S2), indicating that 2 associates around
the N center of (n-Oct)3MeNCl. The dual interaction at the
epoxide and OH group forming the dimer is known,19 and
accordingly, 2 may bind to (n-Oct)3MeNCl via the hydrogen
bond between the OH group and the anion, forming a
complex.37 In contrast, the IR spectrum of the mixture of
methyl benzoate with (n-Oct)3MeNCl did not show a peak
shift due to a carbonyl group compared with its original
spectrum in hexane (Figure S3).
Although the mixture of 2 and (n-Oct)3MeNCl in solution is
stable at room temperature, oligomerization of 2 proceeded at a
higher temperature.38 Reaction of 2 with a catalytic amount of
(n-Oct)3MeNCl in hexane at reflux temperature resulted in
79% consumption of 2. On the basis of this observation and
1099
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
a
Scheme 3. Possible Mechanism for Catalytic Transesterification by Glycidol
a
Q represents (n-Bu)4N.
Figure 3. Potential energy profiles for the synthesis of glycidyl esters from methyl ester with glycidol which catalyzed by (n-Bu)4NCl.
the reaction must overcome a high-energy barrier of 26.5 kcal/
mol at the ring-opening step of 2 in every catalytic cycle.
Alternatively, ring-opened intermediate states A_Int-1 and
A_Int-2 can initiate other catalytic cycles. The addition of
glycidol to A_Int-1 and A_Int-2 can generate different
pathways, Paths B and B′, respectively. These two pathways
include proton abstraction from glycidol, and the alkoxide
species in the A_Int-1 and A_Int-2 states are structural isomers
of each other. The B and B′ pathways reach a common
intermediate state, B_Int-2, because of their structural
similarity. The calculated activation energy for the proton
abstraction in Paths B and B′ was 5.5 kcal/mol (B_TS1−2) and
5.6 kcal/mol (B′_TS1−2), respectively.
In Path B1, the rate-determining step is the Payne
rearrangement. The calculated activation energy was 25.5
kcal/mol (B1_TS2−3). Because this rearrangement is a ring-
opening step of the epoxide, the amount of Ea is similar to that
of the ring-opening transition state (A_TSR‑1). The transition
state of the Payne rearrangement (B1_TS2−3) is energetically
unfavorable, due to the double hydrogen bonds between the
alkoxide anion and diol in B1_Int-2.41 The resulting anion
species B1_Int-3 is also less unstable by 10.8 kcal/mol
compared with B1_Int-2, even though the alkoxide anion
interacts with positively charged H atoms in the ammonium
cation.42 Due to this destabilized form, the attack of the
alkoxide anion on the carbonyl group is a low-energy barrier
process (Ea = 4.5 kcal/mol, B1_TS4−5). After methyl ester
The alkoxide species in intermediate state B_Int-2 works as
the active species of the nucleophilic addition to methyl ester.
Here, we considered two pathways; the first one (Path B1)
involves isomerization of the alkoxide species via the Payne
rearrangement,40 followed by nucleophilic addition to methyl
ester. The second pathway (Path B2) directly proceeds to the
nucleophilic addition to methyl ester.
addition, the C-OMe breaking (Ea = 5.2 kcal/mol, B1_TS5−6
)
is followed by proton transfer steps (Ea = 1.3 kcal/mol,
B1_TS6−P). The product state from transition state B1_TS6−P
involves alkoxide species that are the same as in A_Int-2.
Therefore, the next catalytic cycle enters Path B′ directly or
Path B via A_Int-1. We note that the reaction process can enter
1100
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
Path A from A_Int-2. As mentioned above, however, this
process is energetically unfavorable, because Path A has to go
through the ring-opening transition state A_TSR‑1 (Ea = 26.5
kcal/mol).
EXPERIMENTAL SECTION
■
1
General Procedure. H NMR (400 MHz) and 13C{1H}
NMR (100 MHz) spectra were recorded on JEOL 400 MHz
NMR spectrometers. All spectra were recorded at 25 1 °C.
Chemical shifts (δ) are in parts per million relative to
In Path B2, the highest energy barrier process is the
nucleophilic addition of alkoxide species (Ea = 15.9 kcal/mol,
B2_TS3−4). This result is consistent with the fact that
substrates having an electron-withdrawing group were
efficiently converted to the desired products (Table 1).
Compared with the rate-determining step in Path B1 (Payne
rearrangement), the calculated activation energy of Path B2 is
lower by 9.6 kcal/mol. From the intermediates B2_Int-4, the
C-OMe breaking through B2_TS4−5 (Ea = 6.3 kcal/mol) is
followed by a proton transfer (Ea = 1.3 kcal/mol, B2_TS5−P) to
reach the product state. The catalytic process returns to A_Int-
2 and reaches B_Int-2 via Path B or Path B′.
Several intermediate states allow glycidol to enter the
polymerization process. These polymerization pathways are
energetically unfavorable, however, because a ring-opening
process is involved. Intermediate states A_Int-1, B_Int-2, and
B1_Int-3 occur just before nucleophilic addition to methyl
ester, and therefore alkoxide species are available. If these active
species attack another glycidol, dimerization is initiated. As
shown in Figure S8, the calculated activation energies for
dimerization from A_Int-1, B_Int-2, and B1_Int-3 are 22.1,
25.2, and 14.0 kcal/mol, respectively. In the first two cases, the
barrier height is as high as the other ring-opening steps (Ea =
1
tetramethylsilane at 0 ppm for H, and CDCl3 at 77.0 ppm
for 13C unless otherwise noted. IR spectra were recorded on
Mettler Toledo ReactIR-4000. Gas chromatographic (GC)
analyses were performed on a Shimadzu GC-2014 using a TC-1
column (0.25 mm × 30 m, GL Sciences Inc.). All samples were
analyzed and quantified by using biphenyl as an internal
standard. (n-Oct)3MeN(OCO2Me) was synthesized by the
reported procedure.34 Other chemicals were purchased from
chemical suppliers. Glycidol was distilled under an atmospheric
pressure before to use in order to remove oligomers. Other
chemicals were used as received. Catalytic reactions were
performed using ChemiStation (Tokyo Rika Inc.) equipped
with thermostated apparatus.
Procedure for Catalytic Reaction. To a solution of
catalyst (0.050 mmol) in hexane (0.5 mL), methyl ester (1.0
mmol) and glycidol (1.0 mmol) were added. After stirring 2 h
at reflux temperature (69 °C), reaction mixture was diluted
with CH3CN (3 mL) and filtered. Extra CH3CN (3 mL × 2)
was used for a collection of residual chemicals in a test tube,
and this solution was also filtered. To a combined solution was
then added measured amount of biphenyl (as an internal
standard for NMR analysis), and an aliquot was used for NMR
measurement. The conversion of substrate and the yield of
1
products were determined by H NMR analysis (400 MHz,
26.5 and 25.5 kcal/mol for A_TSR‑1 and B1_TS2−3
,
CDCl3, 25 °C). GC-FID of same solution was measured for the
determination of conversion of glycidol. For the isolation of
glycidyl esters, silica-gel column chromatography was carried
out using EtOAc and hexane as eluent.
respectively). In the third case, because of the instability of
the B1_Int-3 state, the calculated activation energy was lower
than that of the other ring-opening steps. Once the B1_Int-3
state thermally relaxes into a more stable conformation, such as
the B_Int-2 state, the activation energy becomes much higher.
In any case, access to the B1_Int-3 state is inhibited by the
energy barrier of the Payne rearrangement.
As the energy barrier of ring-opening reactions of epoxide is
relatively high, Path B2 is the most favorable pathway for
transesterification. This result also indicates that the alkyl
ammonium cation has less of an effect to promote the ring-
opening step than the system with Lewis acid,42a which inhibits
the random ring-opening reaction and polymerization.
COMPUTATIONAL DETAILS
■
All the computations were done with Gaussian 09.43 Density
functional theory (DFT) calculations at the ωB97XD44/6-
31G* level were carried out for the structure optimization. In
order to obtain accurate potential energies, single-point
calculations were computed at the ωB97XD/6-311+G**
level. According to a previous assessment45 on the applicability
of the dispersion corrected DFT to the molecular interactions,
ωB97XD performed best among the 18 functionals including
M06-2X and LC-BOP-LRD. The solvation effect was included
with the self-consistent reaction field (SCRF) method, and the
dielectric constant of n-hexane was used.46 Natural population
analysis (NPA) was performed with NBO ver 3.1 in the
Gaussian package.47 The transition-state structures were
verified by frequency calculations.
CONCLUSION
■
In conclusion, we demonstrated the effective transformation of
glycidol to valuable compounds while the epoxide part remains
intact. The unexpected involvement of glycidol as a catalyst
with the alkyl ammonium salt was elucidated experimentally as
well as computationally. The minimal interaction of the alkyl
ammonium cation with epoxide inhibited the undesired ring-
opening polymerization, and enabled the controlled trans-
formation of glycidol. This finding may provide an alternative
strategy for synthesizing functional compounds and materials
containing epoxy moieties, including those derivatized from
epoxy compounds, because a glycidyl group could be installed
not by the electrophilic reaction of epichlorohydrin, but by the
nucleophilic reaction of glycidol. Further studies of the detailed
mechanisms of alkoxide anion generation by onium salt, and
optimization of cation−anion combinations are underway in
our laboratory.
For the consistency of the potential energy profile in Figure
3, energy of glycidol and methyl ester were added to those of
the reactant states, transition states, and intermediate states. For
example, energy of glycidol and methyl ester were added to
those of R, A_TSR‑1, A_Int-1, A_TS1−2, A_Int-2 in Path A. In
calculating the energies of glycidol and methyl ester in their
isolated states, explicit interactions with solvent molecule S (S =
n-hexane) have to be taken into account because the continuum
model cannot describe the explicit hydrogen bonding energy
and CH-π interaction. Otherwise, such interactions were taken
into account only in the complex. We adopted an energy
correction for X (X = glycidol or methyl ester) such as E_int =
E(X+S) − E(X) − E(S). The E_int value for glycidol was 4.0
1101
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
(12) Tsuruta, T.; Inoue, S.; Koenuma, H. Makromol. Chem. 1968,
112, 58−65.
(13) Vandenberg, E. J. J. Polym. Sci., Polym. Chem. Ed. 1985, 23, 915−
949.
kcal/mol, which represents hydrogen bonding. The E_int value
for methyl ester was 8.0 kcal/mol, which represents both
hydrogen bonding and CH−π interactions.
(14) Leibig, D.; Seiwert, J.; Liermann, J. C.; Frey, H. Macromolecules
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.7b03303.
■
2016, 49, 7767−7776.
(15) Goodwin, A.; Baskaran, D. Macromolecules 2012, 45, 9657−
9665.
S
(16) Haag, R.; Sunder, A.; Stumbe, J.-F. J. Am. Chem. Soc. 2000, 122,
2954−2955.
Study on reaction condition of catalytic transesterifica-
tion (Table S1); determination of equilibrium constants
by 1H NMR (Table S2, Figure S1); observation of
glycidol, substrate, and quaternary alkyl ammonium salts
in solution (Figures S2, S3); optimized structures in
computational study (Figure S4−S8); and character-
ization of isolated products (PDF)
(17) Pascault, J. P., Williams, R. J. In Epoxy Polymers: New Materials
and Innovations; Pascault, J. P., Williams, R. J. J., Eds.; Wiley-VCH:
Weinheim, Germany, 2010; pp 1−12.
(18) Auvergne, R.; Caillol, S.; David, G.; Boutevin, B.; Pascault, J.-P.
Chem. Rev. 2014, 114, 1082−1115.
(19) Della Monica, F.; Buonerba, A.; Grassi, A.; Capacchione, C.;
Milione, S. ChemSusChem 2016, 9, 3457−3464.
(20) Reactions of glycidol with an unsaturated bond catalyzed by tin
salts accompany the ring opening reactions. See examples:
(a) Edwards, P. A.; Striemer, G.; Webster, D. C. Prog. Org. Coat.
2006, 57, 128−139. (b) Edwards, P. A.; Striemer, G.; Webster, D. C. J.
Coat. Technol. Res. 2005, 2, 517−527. (c) Chattopadhyay, D. K.;
Zakula, A. D.; Webster, D. C. Prog. Org. Coat. 2009, 64, 128−137.
(d) Pitet, L. M.; Hait, S. B.; Lanyk, T. J.; Knauss, D. M. Macromolecules
2007, 40, 2327−2334.
AUTHOR INFORMATION
Corresponding Authors
*E-mail for S.T.: shinji-tanaka@aist.go.jp.
*E-mail for J.H.: hasegawa@cat.hokudai.ac.jp.
*E-mail for K.S.: k.sato@asit.go.jp.
■
ORCID
(21) (a) McClure, D. E.; Engelhardt, E. L.; Mensler, K.; King, S.;
Saari, W. S.; Huff, J. R.; Baldwin, J. J. J. Org. Chem. 1979, 44, 1826−
1831. (b) Morgan, T. K., Jr.; Lis, R.; Lumma, W. C., Jr.; Wohl, R. A.;
Nickisch, K.; Phillips, G. B.; Lind, J. M.; Lampe, J. W.; Di Meo, S. V. J.
Med. Chem. 1990, 33, 1087−1090.
Shinji Tanaka: 0000-0003-2002-5582
Jun-ya Hasegawa: 0000-0002-9700-3309
Notes
The authors declare no competing financial interest.
(22) Saburi, H.; Tanaka, S.; Kitamura, M. Angew. Chem., Int. Ed.
2005, 44, 1730−1732.
ACKNOWLEDGMENTS
■
(23) Deng, J.; Liu, X.; Li, C.; Jiang, Y.; Zhu, J. RSC Adv. 2015, 5,
15930−15939.
We thank Prof. Dr. Hirose (Osaka University) for his advice on
a NMR titration experiment. S.T. acknowledges financial
support by Grant-in-Aid for Scientific Research on Innovative
Areas (JSPS KAKENHI Grant No. JP16H01044 in Precisely
Designed Catalysts with Customized Scaffolding). J.H. also
appreciates financial support from the same area (No.
JP15H05805). A part of the calculations was performed using
Research Center for Computational Science, Okazaki, Japan.
(24) Coish, P.; Brooks, B. W.; Gallagher, E. P.; Kavanagh, T. J.;
Voutchkova-Kostal, A.; Zimmerman, J. B.; Anastas, P. T. ACS
Sustainable Chem. Eng. 2016, 4, 5900−5906.
(25) Zondler, H.; Trachsler, D.; Lohse, F. Helv. Chim. Acta 1977, 60,
1845−1860.
(26) Degussa Aktiengesellschaft. U.S. Patent No. US4667044A, 1987.
(27) Shah, P. N.; Kim, N.; Huang, Z.; Jayamanna, M.; Kokil, A.; Pine,
A.; Kaltsas, J.; Jahngen, E.; Ryan, D. K.; Yoon, S.; Kovar, R. F.; Lee, Y.
RSC Adv. 2015, 5, 38673−38679.
REFERENCES
■
(28) Otera, J. Chem. Rev. 1993, 93, 1449−1470.
(1) Oki, M.; Murayama, T. Bull. Chem. Soc. Jpn. 1973, 46, 259−263.
(2) (a) Marstokk, K.-M.; Møllendal, H.; Stenstrøm, Y. Acta Chem.
Scand. 1992, 46, 432−441. (b) Marstokk, K.-M.; Møllendal, H.;
Samdal, S.; Stenstrøm, Y. Acta Chem. Scand. 1992, 46, 325−337.
(3) Becucci, M.; Melandri, S. Chem. Rev. 2016, 116, 5014−5037.
(4) Demaison, J.; Craig, N. C.; Conrad, A. R.; Tubergen, M. J.;
Rudolph, H. D. J. Phys. Chem. A 2012, 116, 9116−9122.
(5) Maris, A.; Giuliano, B. M.; Bonazzi, D.; Caminati, W. J. Am.
Chem. Soc. 2008, 130, 13860−13861.
(29) Hatano, M.; Ishihara, K. Chem. Commun. 2013, 49, 1983−1997.
(30) (a) Nakatake, D.; Yazaki, R.; Ohshima, T. Eur. J. Org. Chem.
2016, 2016, 3696−3699. (b) Nakatake, D.; Yokote, Y.; Matsushima,
Y.; Yazaki, R.; Ohshima, T. Green Chem. 2016, 18, 1524−1530.
(c) Horikawa, R.; Fujimoto, C.; Yazaki, R.; Ohshima, T. Chem. - Eur. J.
2016, 22, 12278−12281. (d) Agura, K.; Hayashi, Y.; Wada, M.;
Nakatake, D.; Mashima, K.; Ohshima, T. Chem. - Asian J. 2016, 11,
1548−1554. (e) Mashima, K.; Hayashi, Y.; Agura, K.; Ohshima, T.
Pure Appl. Chem. 2014, 86, 335−343. (f) Hayashi, Y.; Santoro, S.;
Azuma, Y.; Himo, F.; Ohshima, T.; Mashima, K. J. Am. Chem. Soc.
2013, 135, 6192−6199.
(6) Hanson, R. M. Chem. Rev. 1991, 91, 437−478.
(7) (a) Leal-Duaso, A.; Caballero, M.; Urriolabeitia, A.; Mayoral, J.
A.; García, J. I.; Pires, E. Green Chem. 2017, 19, 4176−4185.
(b) Cucciniello, R.; Ricciardi, M.; Vitiello, R.; Di Serio, M.; Proto, A.;
Capacchione, C. ChemSusChem 2016, 9, 3272−3275. (c) Ricciardi,
M.; Passarini, F.; Vassura, I.; Proto, A.; Capacchione, C.; Cucciniello,
R.; Cespi, D. ChemSusChem 2017, 10, 2291−2300.
(31) A study on reaction conditions such as solvent, temperature,
amount of 2 and catalyst is summarized in Table S1 in Supporting
Information.
(32) (a) Selva, M.; Noe,
Chem. 2012, 10, 6569−6578. (b) Selva, M.; Caretto, A.; Noe,
Perosa, A. Org. Biomol. Chem. 2014, 12, 4143−4155.
̀
M.; Perosa, A.; Gottardo, M. Org. Biomol.
̀
M.;
(8) (a) Cespi, D.; Cucciniello, R.; Ricciardi, M.; Capacchione, C.;
Vassura, I.; Passarini, F.; Proto, A. Green Chem. 2016, 18, 4559−4570.
(b) Kondawar, S. E.; Patil, C. R.; Rode, C. V. ACS Sustainable Chem.
Eng. 2017, 5, 1763−1774.
(9) Wang, C.; Luo, L.; Yamamoto, H. Acc. Chem. Res. 2016, 49, 193−
204.
(10) Wilms, D.; Stiriba, S.-E.; Frey, H. Acc. Chem. Res. 2010, 43,
129−141.
(11) Sandler, S. R.; Berg, F. R. J. Polym. Sci., Part A-1: Polym. Chem.
1966, 4, 1253−1259.
(33) Yoshida, Y.; Ogura, Y.; Hatano, M.; Ishihara, K. Presented at the
96th Chemical Society of Japan Annual Meeting, Kyoto, Japan, Mar.
24−27, 2016.
(34) Fabris, M.; Lucchini, V.; Noe,
Eur. J. 2009, 15, 12273−12282.
(35) Hirose, K. J. Inclusion Phenom. Mol. Recognit. Chem. 2001, 39,
193−209.
̀
M.; Perosa, A.; Selva, M. Chem. -
(36) Thordarson, P. Chem. Soc. Rev. 2011, 40, 1305−1323.
1102
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103

ACS Catalysis
Research Article
(37) An interaction of halogen anion of quaternary alkyl onium salts
with OH group was proposed in some literatures. See examples:
(a) Liu, S.; Suematsu, N.; Maruoka, K.; Shirakawa, S. Green Chem.
2016, 18, 4611−4615. (b) Kondo, S.; Okada, N.; Tanaka, R.;
Yamamura, M.; Unno, M. Tetrahedron Lett. 2009, 50, 2754−2757.
(c) Ito, K.; Nishiki, M.; Ohba, Y. Chem. Pharm. Bull. 2005, 53, 1352−
1354.
(38) Attempts to observe the ring-opened products were
unsuccessful. 1H NMR of mixture of (n-Oct)3MeNCl with 2 in
toluene-d8 showed only signals due to oligomers of 2 after heating up
to 50 °C
(39) (a) Buddrus, J. Angew. Chem., Int. Ed. Engl. 1972, 11, 1041−
́
1110. (b) Font, J.; Galan, M. A.; Virgili, A. J. Chem. Soc., Perkin Trans. 2
1986, 75−78.
(40) Payne, G. B. J. Org. Chem. 1962, 27, 3819−3822.
(41) Dua, S.; Taylor, M. S.; Buntine, M. A.; Bowie, J. H. J. Chem. Soc.,
Perkin Trans. 2 1997, 1991−1997.
(42) (a) Ema, T.; Miyazaki, Y.; Shimonishi, J.; Maeda, C.; Hasegawa,
J. J. Am. Chem. Soc. 2014, 136, 15270−15279. (b) Ema, T.; Fukuhara,
K.; Sakai, T.; Ohbo, M.; Bai, F.-Q.; Hasegawa, J. Catal. Sci. Technol.
2015, 5, 2314−2321.
(43) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(44) Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615−6620.
(45) Grimme, S. WIREs Computational Molecular Science 2011, 1,
211−228.
̌
(46) (a) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55,
117−129. (b) Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 106,
5151−5158. (c) Cammi, R.; Mennucci, B.; Tomasi, J. J. Phys. Chem. A
2000, 104, 5631−5638.
(47) (a) Reed, A. E.; Weinhold, F. J. Chem. Phys. 1983, 78, 4066−
4073. (b) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys.
1985, 83, 735−746.
1103
DOI: 10.1021/acscatal.7b03303
ACS Catal. 2018, 8, 1097−1103