Mechanistic Insight into Weak Base-Catalyzed Generation of Carbon Monoxide from Phenyl Formate and Its Application to Catalytic Carbonylation at Room Temperature without Use of External Carbon Monoxide Gas

Article abstract of DOI:10.1002/adsc.201700751

The mechanisms of the weak base-catalyzed generation of carbon monoxide (CO) and phenol from phenyl formate were investigated by experimental and theoretical methods. Kinetic studies revealed a first-order reaction in both phenyl formate and the base. The reaction was found to proceed by an E2 α-elimination pathway, which involves the abstraction of the formyl proton of phenyl formate, simultaneously generating CO and phenoxide. The reaction rate was affected by the substituents on phenyl formate, the polarity of solvents, and the basicity of bases. The mechanistic insight obtained from these studies permitted the chemical control of the rate of CO generation, which was the key to the development of the external CO-free Pd-catalyzed phenoxycarbonylation of haloarenes at room temperature. Because of the mild reaction conditions and wide substrate scope, this phenoxycarbonylation constitutes a general, safe, and practical method to synthesize arenecarboxylic acid esters. (Figure presented.).

Full text of DOI:10.1002/adsc.201700751

Title: Mechanistic Insight into Weak-Base-Catalyzed Generation of
Carbon Monoxide from Phenyl Formate and Its Application to
Catalytic Carbonylation at Room Temperature without Use of
External Carbon Monoxide Gas
Authors: Hideyuki Konishi, Mika Matsubara, Keisuke Mori, Takaki
Tokiwa, Sundaram Arulmozhiraja, Yuta Yamamoto,
Yoshinobu Ishikawa, Hiroshi Hashimoto, Yasuteru Shigeta,
Hiroaki Tokiwa, and Kei Manabe
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201700751
Link to VoR: http://dx.doi.org/10.1002/adsc.201700751

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Mechanistic Insight into Weak-Base-Catalyzed Generation of
Carbon Monoxide from Phenyl Formate and Its Application to
Catalytic Carbonylation at Room Temperature without Use of
External Carbon Monoxide Gas
Hideyuki Konishi,a Mika Matsubara,a Keisuke Mori,a Takaki Tokiwa,b,c Sundaram
Arulmozhiraja,d Yuta Yamamoto,d Yoshinobu Ishikawa,a Hiroshi Hashimoto,a Yasuteru
Shigeta,b Hiroaki Tokiwa,d and Kei Manabea,*
a
School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan
E-mail: manabe@u-shizuoka-ken.ac.jp
Center for Computational Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8571, Japan
Department of Chemistry, Graduate School of Science, Tohoku University, Aramaki, Aoba-ku, Sendai, Miyagi 980-
b
c
8578, Japan
d
Department of Chemistry, Rikkyo University, 3-34-1 Nishi-Ikebukuro, Toshima-ku, Tokyo 171-8501, Japan
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. The mechanisms of weak-base-catalyzed permitted the chemical control of the rate of CO generation,
generation of carbon monoxide (CO) and phenol from which were key to the development of the external-CO-free
phenyl formate were investigated by experimental and Pd-catalyzed phenoxycarbonylation of haloarenes at room
theoretical methods. Kinetic studies revealed a first-order temperature. Because of the mild reaction conditions and
reaction in both phenyl formate and the base. The reaction wide substrate scope, this phenoxycarbonylation constitutes
was found to proceed by an E2 -elimination pathway, a general, safe, and practical method to synthesize
which involves the abstraction of the formyl proton of arenecarboxylic acid esters.
phenyl formate, simultaneously generating CO and
phenoxide. The reaction rate was affected by the substituents
of phenyl formate, the polarity of solvents, and the basicity
of bases. The mechanistic insight obtained from these studies
Keywords: carbon monoxide; phenyl formate;
carbonylation; room temperature; mechanism; kinetics;
DFT
used as they are more safe to handle and practical for
reactions. Furthermore, the control of the amount of
CO using CO surrogates, which can be easily
Introduction
Carbon monoxide (CO) is a fundamental chemical weighed, is facile. This is advantageous over
feedstock in synthetic organic chemistry.[1] Because conventional reactions under CO gas, because these
of its wide availability and low cost, CO has been reactions are usually conducted in the presence of a
utilized in metal-catalyzed organic reactions that large excess of CO, which often retards transition
involve the introduction of a carbonyl group in metal-catalyzed reactions by excessive ligation of CO
several industrial processes, as well as in laboratory to the metal center.[5]
experiments, for producing synthetically and
We[6] and Tsuji group[7] have independently
pharmaceutically important materials.[2] However, the reported that phenyl formate (1a) generated CO in
gaseous highly toxic nature of CO restricts its situ in the presence of a weak base such as
utilization, especially for reactions involving the triethylamine (NEt3), at 60–80 C, which was
small-scale preparation of various congeners of subsequently applied to the Pd-catalyzed Heck
carbonyl compounds because reactions with CO carbonylation[8] (phenoxycarbonylation, Scheme 1).
require special caution for handling CO and pressure- Later, our group has successfully developed a highly
resistant equipment. For several decades, such reactive CO surrogate, 2,4,6-trichlorophenyl formate,
disadvantages have fostered the concept of CO which
was
utilized
in
the
Pd-catalyzed
surrogates, which can generate CO in situ for their use carbonylation.[9] These CO surrogates have notable
in organic reactions instead of gaseous CO.[3,4] characteristics. First, they can be stored under
Currently, a number of cost-effective CO surrogates ambient conditions. Second, different from several
are commercially available, and these surrogates are other CO surrogates, CO can be generated using a
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
weak base, and neither strong bases nor additional point (217 C) greater than that of NEt3 (89 C).
metal catalysts are required. Therefore, these When the reaction was conducted by treating 1a with
surrogates are particularly suitable for metal- various amount of NBu3 in MeCN at 80 °C, linear
catalyzed carbonylative reactions, which are approximations were obtained from the plots of
conducted under weakly basic conditions. Third, as log([1a]/[1a]0) versus time, regardless of the amount
the aryl ester products of the carbonylation are of NBu3 ([1a]: concentration of phenyl formate, [1a]0:
activated esters compared to alkyl esters, they can be initial concentration of phenyl formate, see
easily used for further transformation to other Supporting Information (SI), Figure S1). Table 1
carbonyl derivatives.[6,9a] Furthermore, the utility of summarizes the observed rate constants (kobs). In the
these CO surrogates is corroborated by their extensive absence of NBu3, no reaction occurred. A catalytic
use in recently reported carbonylation reactions.[10]
amount of NBu3 (0.30 equiv) could also promote the
reaction. The observation of a linear plot for kobs
versus the initial concentration of NBu3 ([NBu3]0)
(Figure S2) reveals a first-order reaction not only in
1a but also in NBu3. Consequently, the reaction on
the whole is second order, with a rate constant of
0.417 h–1 M–1 (see SI for details).
Table 1. Observed rate constants for CO generation.
Scheme 1. (a) Generation of CO from phenyl formate in
the presence of a weak base and (b) its application to Pd-
catalyzed external-CO-free phenoxycarbonylation.[6a,7]
Entry NBu3 (equiv) Conversion (%)a) kobs (h–1)b)
1
2
0.30
1.0
2.8
82 (30 h)
99 (20 h)
86 (4 h)
6.84 × 10–2
0.229
0.575
3
a)
Formic acid esters are known to generate CO in the
presence of a base;[11] surprisingly, only little
information about its reaction mechanism is
available.[12] During the course of our investigation,
we were interested in the generation of CO via the
interaction between phenyl formate and a weak base.
Understanding the mechanism of CO generation
would permit the control of the rate of CO generation
and subsequent applications for safe, practical
reactions with easy-to-handle CO surrogates while
avoiding the use of externally supplied CO gas. In
this study, detailed experimental and theoretical
studies were conducted to understand the mechanism
of CO generation from phenyl formate in the presence
Time required for the conversion of 1a is given in
parenthesis. b) Observed rate constant.
Kinetic Isotope Effect (KIE) Study. To examine
whether the rate-determining step (RDS) for the
reaction involved the cleavage of the C–H bond of the
formyl group of phenyl formate, kinetic experiments
using deuterated phenyl formate-d (1a–d) were
conducted (Scheme 2). Two parallel reactions were
conducted using 1a–d or 1a in CD3CN in the presence
of 1 equiv of NBu3. The observed kH/kD value, 2.45,
indicated that the RDS involves the C–H bond
cleavage.[14]
of
a
weak
base.
Furthermore,
the
phenoxycarbonylation of haloarenes using phenyl
formate at room temperature was conducted, which
formerly requires either high temperature or highly
reactive CO surrogates.
Results and Discussion
Scheme 2. KIE study.
Mechanistic Studies for the Generation of CO
from Phenyl Formate
DFT Calculation. The mechanism of CO
generation was investigated by density functional
theory (DFT). All the geometries were optimized
using the B3LYP functional utilizing the 6-31+G(d)
basis set under vacuum conditions.[15] For simplicity,
trimethylamine (NMe3) was considered for the
calculations instead of NEt3 and NBu3. Optimization
revealed two stable structures for initial 1a; one with
the carbonyl oxygen and phenyl group located cis to
Reaction Profile and Rate Constant. First, the
kinetics study for the reaction of 1a with NEt3 at 80
C was investigated.[13] However, informative results
were not obtained at low NEt3 concentration,
probably because of the partial evaporation of NEt3.
Then, the kinetics study was repeated by replacing
NEt3 with tributylamine (NBu3), which has a boiling
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
the C–O single bond, s-cis 1a, and the other with
those located trans to the C–O bond, s-trans 1a. The
relative potential energies of s-cis 1a and s-trans 1a
under vacuum at 298.15 K are given in Figure 1. Both
initial structures of 1a afforded the same product via
different transition states (TScis and TStrans, Figure 2).
Intrinsic reaction coordinate (IRC) calculations
confirmed good connections between the initial
structures, TSs, and final product.
These calculations indicated that s-cis 1a and TScis
are energetically more favored compared to their
respective s-trans structures (s-trans 1a and TStrans
)
Figure 2. Optimized TScis and TStrans geometries at the
B3LYP/6-31+G(d) level at 298.15 K under vacuum. Bond
lengths and bond angles are expressed in angstroms and
degrees, respectively. Imaginary wave numbers and
transition vectors of C, H, and O atoms are also shown.
(Figure 1). An energy difference of approximately 21
kJ/mol was observed between the TScis and TStrans
structures, favoring the former one, which suggested
that the reaction preferentially occurs from s-cis 1a.
In both TS structures, the H–C–O(carbonyl) angles of
the formyl group become significantly wider (H-C-
O in TScis and TStrans are 163.8° and 157.4°,
respectively, Figure 2). The transition vectors of these
atoms indicated that the abstraction of proton of the
formyl group by NMe3 and elimination of CO and
phenoxide occur simultaneously. In addition, IRC
analysis revealed that the reaction first produces
Natural Bond Orbital (NBO) Analysis. NBO
analysis[16] was carried out to understand the nature of
the TScis structure. Donor–acceptor interactions
existed between the lone pair of the amine nitrogen
and the antibonding orbital of the formyl C–H bond
(Figure 3a, Movie S1). The interaction between the
lone pair of the carbonyl oxygen and the antibonding
orbital of the cleaving C–O bond was also shown
(Figure 3b, Movie S2). The lone pair electrons on
oxygen transferred to the *C–O orbital to assist the
elimination of the phenoxide. The electron densities
shown in Figure 3 indicated that the charge transfer to
the antibonding orbitals occurs to a certain degree (ca.
0.3) at the TS.
trimethylammonium
phenoxide,
which
then
undergoes an internal acid–base reaction to generate
NMe3 and phenol without any activation barrier.
These calculation results strongly suggest that the
reaction proceeds via bimolecular -elimination,
which will be discussed later.
(a)
(b)
nN and *C–H
(1.628) (0.368)
nO and *C–O
(1.749) (0.337)
Figure 3. Overlaid depiction of NBOs at TScis. Calculated
electron density of each orbital is shown in parenthesis. (a)
Donor–acceptor interaction between nN and *C–H. (b)
Donor–acceptor interaction between nO and *C–O. For an
animated version see the Supporting Information.
Figure 1. Energy profiles for the reaction of 1a with NMe3.
Relative Gibbs energies (G) were calculated at the
B3LYP/6-31+G(d) level of theory at 298.15 K under
vacuum. Black: s-cis, blue: s-trans.
The Eyring Plot and Thermodynamic
Parameters. The Eyring plot was constructed to
obtain thermodynamic parameters by conducting the
reaction in MeCN at different temperatures. However,
the reaction was extremely slow to obtain a rate
constant at low temperature. Changing the solvent
from MeCN to more polar DMF led to the successful
determination of the rate constants even at low
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
temperatures and helped in the construction of the Factors Affecting CO Generation
Eyring plot (see SI, Table S3 and Figure S3).
Table 2 shows the experimental thermodynamic Effect of Substituents on the Reaction Rate
parameters obtained from the Eyring plot. In addition, (Hammett plot). With these mechanistic insights, the
these parameters were theoretically calculated at the factors affecting the reaction rate were examined.
B3LYP/6-31+G(d) level with DMF as a solvent using This information would help to realize control of the
the polarized continuum model (PCM) and the results rate of CO generation, which would be highly
are also given in Table 2. The calculated values were beneficial for carbonylative reactions requiring low
in qualitative agreement with the experimental values. concentration of CO throughout the reaction.
Negative S‡ indicated the existence of a highly
Since the reaction mechanism involves the
organized TS, again suggesting that the reaction elimination of phenoxide, the reaction can be
mechanism involves
pathway.
a
bimolecular elimination accelerated by increasing the leaving ability of the
phenoxy group. Thus, the reaction rate is affected by
the electronic features of phenyl formate, namely by
Table 2. Experimental and calculated thermodynamic
parameters.
the introduction of substituents on the benzene ring.
In addition, the substituents may also affect the
acidity of formyl C–H. Our group has previously
reported that the introduction of an electron-
withdrawing group on the benzene ring resulted in
rate acceleration of the CO generation.[9a] However,
this rate acceleration has not been quantitatively
evaluated.[17]
G‡
H‡
S‡
(kJ mol–1)a) (kJ mol–1)b) (J K–1 mol–1)c)
Then, we tried to obtain kinetic data using aryl
formates 1a–e having various substituents para to the
formyloxy group. As anticipated, the greater the
electron-withdrawing nature of the substituents, the
higher the rate constant (Table 3). The Hammett plot
(Figure S5) clearly indicated a linear relationship with
the substituent constant p.[18] The positive value of
the reaction constant  (3.08) indicated that negative
charge increases in the TS of the reaction.
Exp.
109.7
74.7
95.8
–97.1
–26.9
DFTd) 105.3
a)
Gibbs activation energy at 80 °C (353.15 K). The
experimental value was determined from H‡ and S‡.
b)
c)
d)
Activation enthalpy. Activation entropy. Calculated at
the B3LYP/6-31+G(d) level. Solvent (DMF) effects were
considered using PCM.
Proposed Reaction Mechanism. Considering
these results, the proposed mechanism for the
generation of CO from phenyl formate by reacting
with an amine as a weak base is shown in Scheme 3.
In the bimolecular TS, the amine works as a Brønsted
base to abstract the proton of the formyl group,
simultaneously producing CO and phenoxide. The
lone pair electrons on the carbonyl oxygen also assists
in the formation of CO and phenoxide by transferring
electron density into the * orbital of the cleaving C–
O bond. This step is regarded as E2, concerted -
elimination. Then, an acid–base reaction occurs
between the ammonium and phenoxide ions,
affording phenol and regenerating amine. Thus, the
amine can be catalytically used to generate CO.
Table 3. Effect of substituents on the rate constant.
a)
Entry
R
p
k (h1 M1)b)
1
2
3
4
5
OMe (1b) –0.27 0.106
Me (1c)
H (1a)
–0.17 0.130
0
0.458
3.62
24.4
Cl (1d)
CF3 (1e)
0.23
0.54
a) Substituent constant p (values are taken from ref. [18]). b)
Rate constant.
Effect of Solvents on the Reaction Rate. Since
the reaction rate was accelerated by changing the
solvent from MeCN to DMF, the solvent presumably
affected the reaction rate significantly. To
quantitatively understand this effect, reactions were
conducted in different solvents (Table 4 and Figure
S6).[13] In all solvents, the reactions followed the
kinetics shown in Eq. 2, affording k values. Although
the rate of CO generation was not always correlated
to the dielectric constant[19] of the solvent, a highly
polar solvent typically accelerates the reaction. This
trend is reasonable for the assumed E2 mechanism of
the reaction as polar solvents can stabilize a partially
Scheme 3. Proposed mechanism of weak base-catalyzed
CO generation.
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
charged TS. The wide solvent-dependent range of
The observed k values indicated that a strong base
rates (from around 3 × 10–2 to 7 h1 M1) can be typically increases the reaction rate. These results
exploited for adjusting the rate of CO generation for suggest that the rate of CO generation can be
various purposes.
predicted by considering the pKa values[20] of the
corresponding conjugate acids of the bases. However,
several exceptions were observed. Inorganic bases
such as sodium acetate and sodium bicarbonate
afforded somewhat higher k values than expected
(Table 5, entries 2 and 4). In addition, DABCO and
DMAP afforded relatively higher k values (entries 7
and 8), which highlights another reaction mechanism,
in which the reaction proceeds via nucleophilic
catalysis to form activated intermediates, with
possibly higher reactivity toward Brønsted bases to
more easily generate CO (Scheme 4). At this moment,
the involvement of this reaction pathway for highly
nucleophilic bases such as DABCO and DMAP
cannot be excluded.[21] DBU,[22] cesium carbonate,
and potassium phosphate were too basic to obtain
accurate k values because CO was spontaneously
generated immediately with the addition of these
bases (Table 5, entries 9–11). A guanidine-type base
(1,5,7-triazabicyclo[4.4.0]dec-5-ene, TBD) led to
formyl transfer from 1a instead of CO generation
(entry 12). The rapid formation of formylated TBD
and phenol was confirmed by 1H NMR (Scheme 5).
Table 4. Effect of solvents on the rate constant.
Entry Solvent
a)
k (h1 M1)b)
2.21 3.36 × 10–2
1
2
1,4-dioxane
toluene
2.38 4.42 × 10–2
3
propyl acetate 5.69 5.44 × 10–2
4
5
6
7
2-butanone
t-BuOH
DCE
18.1 0.132
12.5 0.182
10.4 0.273
35.9 0.719
MeCN
8
nitromethane 35.9 1.37
9
DMF
DMSO
36.7 1.74
46.5 7.30
10
a) Dielectric constant (values are taken from ref. [19]). b) Rate
constant.
Effect of Bases on the Reaction Rate. The role of
a base in the reaction mechanism indicated that its
basicity affects the reaction rate. Hence, the effect of
bases is investigated in a manner similar to that of
solvents (Table 5 and Figure S7).
Table 5. Effect of bases on the rate constant.
Scheme 4. Alternative mechanism to generate CO via the
formation of activated intermediate.
Entry base
pKaa)
k (h–1 M–1)b)
1
2
3
4
pyridine 5.37 (12.5) 5.6 × 10–3 c)
NaOAc
4.76
7.12
1.90 × 10–2 c)
4.40 × 10–2 c)
0.221
NMId)
Scheme 5. Formyl transfer from 1a to TBD.
NaHCO3 3.58
5
6
7
NBu3
NEt3
10.9
0.458
Application to Room-Temperature Carbonylation
Using Phenyl Formate
10.7 (18.8) 0.719
DABCO 8.72 1.89
8
DMAP
DBU
Cs2CO3
K3PO4
TBDg)
9.87 (18.0) 2.84
9e)
10e)
11e)
11.6 (24.3) N/Af)
Various factors such as temperature, solvent polarity,
basicity, and amount of base were revealed to affect
the rate of CO generation from 1a. These features
enable 1a to serve as a good CO source in
carbonylative reactions, where CO is released from 1a
with a suitable range of reaction rates by tuning the
reaction conditions.
10.3
12.4
N/Af)
N/Af)
12
14.5 (26.0) N/Ah)
a)
pKa value of the conjugate acid in H2O (values are taken
from ref. [20]). pKa in MeCN is shown in parenthesis. b) Rate
constant. Average value.
c)
d)
e)
N-Methylimidazole. Gas
Our next aim is to apply the insights obtained by
both experimental and theoretical studies to CO
generation for organic synthetic reactions. Although
1a as a CO source is known to be applicable to Pd-
catalyzed phenoxycarbonylation, a relatively high
evolution was observed immediately after the addition of a
f)
base at rt. The value was not obtained because of rapid
g)
h)
CO generation.
1,5,7-Triazabicyclo[4.4.0]dec-5-ene.
Not applicable because of the formation of formylated
TBD.
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
a)
b)
reaction temperature (typically 60–80 °C) is
Isolated yield.
Pd(OAc)2 (0.5 mol %), Xantphos (1
required.[6a,7] Ideally, the reactions should be mol %), 1a (1.5 equiv), and DBU (1.5 equiv) were used.
conducted at room temperature for the highest degree
of safety, ease, and functional group compatibility.
Although our group has already developed 2,4,6-
With the optimized reaction conditions in hand, the
trichlorophenyl formate as a highly reactive CO scope of substrates for the Pd-catalyzed
source for the aryloxycarbonylation of iodoarenes at phenoxycarbonylation of haloarenes 2a–q using 1a at
room temperature,[9a] it is not considered to be room temperature was investigated, and the reaction
desirable to use, especially on a large scale because of exhibited broad substrate tolerance (Table 7). Various
its high chlorine content. Instead of 2,4,6- para-substituted iodoarenes 2a–h afforded the desired
trichlorophenyl formate, 1a was applied to the products 3a–h in high yield (entries 1–8), regardless
reaction at room temperature, but a low yield (10%) of the electronic properties of substituents. With
was obtained. However, with our current mechanistic respect to the sterically demanding substrates, 2-
insights into CO generation, phenoxycarbonylation iodotoluene (2i) was converted to phenyl ester
was investigated again using 1a at room temperature, product 3i in moderate yield (entry 9). Using toluene
considering its wide substrate applicability and as a solvent, the product yield increased to 95%,
adaptability to large-scale synthesis.
suggesting that the optimal solvent is apparently
The optimization of the phenoxycarbonylation of dependent on the structure of the substrates (entry 10).
iodobenzene (2a) catalyzed by Pd‒Xantphos[23] was An electron-donating group at the ortho position was
carried out, with focus on solvent polarity and well tolerated (entry 11). Notably, 1-iodo-2-
basicity of bases[24] (Table 6), which highly affect the nitrobenzene (2k), the substrate with a relatively
rate of CO generation from 1a (vide supra). With the bulky group, afforded 3k in high yield (entry 12). In
combined use of toluene and NEt3 (entry 1), almost this case as well, toluene was effective. Naphthyl-type
all 1a remained, indicating that CO was hardly substrate 2l was applicable for the reaction (entry 13),
generated in the reaction. To accelerate CO and double phenoxycarbonylation was also feasible
generation, toluene was replaced by higher polarity (entry 14). Furthermore, heteroaromatic substrates 2n
solvents (entries 2 and 3). Moderate yields were and 2o tolerated the reaction, affording desired
obtained for product 3a, with marginal amounts of 1a products 3n and 3o in high yields (entries 15 and 16).
remained. Hence, a stronger base, DBU, was In addition to iodoarenes, bromoheteroarenes 2q and
investigated
phenoxycarbonylation proceeded well, affording 3a in Although simple bromobenzene failed to give the
85% yield (entry 4). The increase in the reaction time product, activated electron-deficient
in
toluene.
Gratifyingly, 2p were applied to the reaction (entries 17 and 18).
was effective (entry 5). The use of a more polar bromoheteroarenes afforded the corresponding
solvent, MeCN, further improved the reaction, with a products 3o and 3p in good yields.
product yield of 98% in 3 h (entry 6).[25] However, the
Table 7. Scope of substrates for room-temperature
carbonylation.
combination of highly polar solvents, DMF and
DMSO, resulted in lower yields of 3a (entries 7 and
8). Notably, the reduction in the catalyst loading to
0.5 mol% and the amounts of 1a and DBU
maintained the high yield of 3a (entry 9).
Entry Product
Time
(h)
Yield
(%)a)
(R')
1
2
H (3a)
OMe
(3b)
3
3
93
95
Table 6. Optimization of Reaction Conditions.
3
CO2Et
(3c)
3
98
entry solvent base
time (h) yield (%)a)
1
2
3
4
5
6
7
toluene NEt3
MeCN NEt3
DMSO NEt3
toluene DBU
toluene DBU
MeCN DBU
24
16
24
6
12
3
6
6
3
no reaction
4
5
CN (3d) 12
94
70
35
44
85
97
98
54
73
93
CHO
(3e)
24
6
COMe
(3f)
6
87
7
8
Cl (3g)
NO2
6
24
90
85
DMF
DBU
8
DMSO DBU
MeCN DBU
(3h)
9b)
9
Me (3i)
24
57
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
10b)
11
Me (3i)
OMe
(3j)
NO2
(3k)
24
24
95
91
catalyzed carbonylation using phenyl formate is
considered to be attractive for synthesizing
arenecarboxylic acid derivatives.
12
13
24
24
91
87
The
complementary
approach
utilizing
experimental and theoretical methods was successful
in obtaining fundamental insights into the mechanism
of CO generation from phenyl formate. Based on
these insights, the development of other CO
surrogates generating CO under considerably milder
conditions is feasible. Furthermore, the technique
involving the chemical control of the CO generation
can replace the existing conventional carbonylative
reactions conducted under atmospheric or pressurized
CO with external-CO-free protocols by using CO
surrogates as they are more safe and practical. These
changes should contribute to the increased efficiency
of the synthesis and the construction of compound
libraries.
(3l)
14c)
15
(3m)
(3n)
12
6
95
96
16d)
17d)
(3o)
(3p)
6
24
87
74
18e)
48
69
Experimental Section
General
procedure
for
Pd-catalyzed
phenoxycarbonylation of iodoarenes at room
temperature
a)
b)
c)
Isolated yield. Toluene was used as a solvent. 1,4-
Diiodobenzene (2m) was used as a substrate. The amount
of reagents, except 2m, was doubled. 2-Bromopyridine
d)
Pd(OAc)2 (3.4 mg, 0.0151 mmol, 3 mol %) and Xantphos
(2q) was used as a substrate. 1-Bromoisoquinoline (2p) (17.4 mg, 0.0300 mmol, 6 mol %) were added in a 10-mL
e)
test tube containing a magnetic stirring bar. The tube was
evacuated and backfilled with Ar three times. MeCN (1.00
mL), iodoarene 2a–p (0.500 mmol), 1a (82 µL, 0.752
mmol, 1.5 equiv), and DBU (112 µL, 0.749 mmol, 1.5
equiv) were added to the flask and the tube was sealed with
a screw cap under a flow of Ar. The flask was stirred at rt
for 3–48 h. The reaction mixture was diluted with CH2Cl2,
washed with brine, dried over Na2SO4, filtered, and
was used as a substrate.
Conclusion
The weak base-catalyzed generation of CO from
phenyl formate was investigated by experimental and concentrated. The obtained residue was purified by
preparative TLC (SiO2, hexane/AcOEt mixture as an
theoretical methods. Kinetic experiments clarified that
eluent) to afford the desired phenyl ester 3a–p.
the reaction profile is first order in both phenyl
formate and the base. The study on kinetic isotope
effect showed that the rate-determining step of the
Acknowledgements
reaction involves the deprotonation of the formyl H
from phenyl formate. DFT calculations provided the
This study was partly supported by JSPS KAKENHI Grant
necessary mechanistic insights into the reaction,
which is an E2 concerted -elimination, with the
simultaneous release of CO and phenoxide at the TS.
This result was experimentally supported by
obtaining thermodynamic parameters, which were in
agreement with the results obtained from DFT
calculations. Solvent, base, temperature, and
substituents on the benzene ring of phenyl formate
significantly affected the rate of CO generation.
Numbers 15H04634 and 15K18834, the Uehara Memorial
Foundation, and the Platform Project for Supporting in Drug
Discovery and Life Science Research (Platform for Drug
Discovery, Informatics, and Structural Life Science) from the
Ministry of Education, Culture, Sports, Science and Technology
(MEXT) and Japan Agency for Medical Research and
Development (AMED). Computational parts in this study were
supported by the Rikkyo SFR project, 2014−2016, and the MEXT-
Supported Program for the Strategic Research Foundation at
Private Universities, 2013−2018.
The chemical controllability of CO generation from
phenyl formate was applied to the Pd-catalyzed
phenoxycarbonylation of haloarenes under extremely
mild conditions. The appropriate tuning of solvents
and bases led to optimized conditions, which could
promote the reaction at room temperature. The
reaction could be conducted without external CO or
heating, thereby reflecting considerable safety and
practicality. A wide substrate scope was confirmed,
which included electronically and sterically
References
[1] C-1 Building Blocks in Organic Synthesis (Eds.: P. W.
N. M. van Leeuwen), Thieme, Stuttgart, 2014.
[2] For selected reviews, see: a) G. Kiss, Chem. Rev. 2001,
101, 3435–3456; b) A. Zapf, M. Beller, Top. Catal.
2002, 19, 101–109; c) C. Torborg, M. Beller, Adv.
Synth. Catal. 2009, 351, 3027–3043; d) A. Brennführer,
H. Neumann, M. Beller, Angew. Chem. Int. Ed. 2009,
48, 4114–4133; e) J. Magano, J. R. Dunetz, Chem. Rev.
2011, 111, 2177–2250.
demanding
compounds.
Several
activated
bromoheteroarenes were good substrates. Phenyl
formate can be easily handled and stored. Thus, Pd-
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
[3] For selected reviews, see: a) T. Morimoto, K. Kakiuchi,
Angew. Chem. Int. Ed. 2004, 43, 5580–5588; b) L. R.
Odell, F. Russo, M. Larhed, Synlett 2012, 685–698; c)
L. Wu, Q. Liu, R. Jackstell, M. Beller, Angew. Chem.
Int. Ed. 2014, 53, 6310–6320; d) H. Konishi, K.
Manabe, Synlett 2014, 25, 1971–1986; e) P. Gautam, B.
M. Bhanage, Catal. Sci. Technol. 2015, 5, 4663–4702;
f) S. D. Friis, A. T. Lindhardt, T. Skrydstrup, Acc.
Chem. Res. 2016, 49, 594–605.
Beller, A. F. Indolese, J. Mol. Catal. A: Chem. 2000,
156, 213–221.
[6] a) T. Ueda, H. Konishi, K. Manabe, Org. Lett. 2012, 14,
3100–3103; b) T. Ueda, H. Konishi, K. Manabe,
Tetrahedron Lett. 2012, 53, 5171–5175; c) H. Konishi,
H. Nagase, K. Manabe, Chem. Commun. 2015, 51,
1854–1857.
[7] T. Fujihara, T. Hosoki, Y. Katafuchi, T. Iwai, J. Terao,
Y. Tsuji, Chem. Commun. 2012, 48, 8012–8014.
[4] a) J. S. Morgan, J. Chem. Soc. Trans. 1916, 109, 274–
283; b) T. Okano, T. Kobayashi, H. Konishi, J. Kiji,
Tetrahedron Lett. 1982, 23, 4967–4968; c) P. Isnard, B.
Denise, R. P. A. Sneeden, J. Organomet. Chem. 1983,
256, 135–139; d) J.-F. Carpentier, Y. Castanet, J.
Brocard, A. Mortreux, F. Petit, Tetrahedron Lett. 1991,
32, 4705–4708; e) G. Jenner, E. M. Nahmed, H.
Leismann, J. Organomet. Chem. 1990, 387, 315–321; f)
V. V. Grushin, H. Alper, Organometallics 1993, 12,
3846–3850; g) C. M. Beck, S. E. Rathmill, Y. J. Park, J.
Chen, R. H. Crabtree, L. M. Liable-Sands, A. L.
Rheingold, Organometallics 1999, 18, 5311–5317; h) N.
F. K. Kaiser, A. Hallberg, M. Larhed, J. Comb. Chem.
2002, 4, 109–111; i) Y. Wan, M. Alterman, M. Larhed,
A. Hallberg, J. Org. Chem. 2002, 67, 6232–6235; j) S.
Ko, C. Lee, M.-G. Choi, Y. Na, S. Chang, J. Org. Chem.
2003, 68, 1607–1610; k) S. Cacchi, G. Fabrizi, A.
Goggiamani, Org. Lett. 2003, 5, 4269–4272; l) K. Fuji,
T. Morimoto, K. Tsutsumi, K. Kakiuchi, Angew. Chem.
Int. Ed. 2003, 42, 2409–2411; m) Y. Katafuchi, T.
Fujihara, T. Iwai, J. Terao, Y. Tsuji, Adv. Synth. Catal.
2011, 353, 475–482; n) P. Hermange, A. T. Lindhardt,
R. H. Taaning, K. Bjerglund, D. Lupp, T. Skrydstrup, J.
Am. Chem. Soc. 2011, 133, 6061–6071; o) S. D. Friis, R.
H. Taaning, A. T. Lindhardt, T. Skrydstrup, J. Am.
Chem. Soc. 2011, 133, 18114–18117; p) T. Ueda, H.
Konishi, K. Manabe, Angew. Chem. Int. Ed. 2013, 52,
8611–8615; q) T. Ueda, H. Konishi, K. Manabe, Org.
Lett. 2013, 15, 5370–5373; r) D.-S. Kim, W.-J. Park,
C.-H. Lee, C.-H. Jun, J. Org. Chem. 2014, 79, 12191–
12196; s) W. Li, X.-F. Wu, J. Org. Chem. 2014, 79,
10410–10416; t) K. Natte, A. Dumrath, H. Neumann,
M. Beller, Angew. Chem. Int. Ed. 2014, 53, 10090–
10094; u) M. Babjak, O. Caletková, D. Ďurišová, T.
Gracza, Synlett 2014, 25, 2579–2584; v) S. V. F.
Hansen, T. Ulven, Org. Lett. 2015, 17, 2832–2835; w)
A. Barré, M.-L. Ţînţaş, F. Alix, V. Gembus, C.
Papamicaël, V. Levacher, J. Org. Chem. 2015, 80,
6537–6544; x) H. Konishi, T. Muto, T. Ueda, Y.
Yamada, M. Yamaguchi, K. Manabe, ChemCatChem
2015, 7, 836–845; y) S. K. Murphy, J.-W. Park, F. A.
Cruz, V. M. Dong, Science 2015, 347, 56–60; z) C.
Veryser, S. Van Mileghem, B. Egle, P. Gilles, W. M.
De Borggraeve, React. Chem. Eng. 2016, 1, 142–146;
aa) X. Qi, C.-L. Li, L.-B. Jiang, W.-Q. Zhang, X.-F. Wu,
Catal. Sci. Technol. 2016, 6, 3099–3107.
[8] A. Schoenberg, I. Bartoletti, R. F. Heck, J. Org. Chem.
1974, 39, 3318–3326.
[9] a) T. Ueda, H. Konishi, K. Manabe, Org. Lett. 2012, 14,
5370–5373; b) H. Konishi, T. Ueda, K. Manabe, Org.
Synth. 2014, 91, 39–51.
[10] For selected examples of the use of aryl formates as
CO surrogates, see: a) H. Wang, B. Dong, Y. Wang, J.
Li, Y. Shi, Org. Lett. 2014, 16, 186–189; b) Y.
Hoshimoto, T. Ohata, Y. Sasaoka, M. Ohashi, S.
Ogoshi, J. Am. Chem. Soc. 2014, 136, 15877–15880; c)
S. P. Chavan, B. M. Bhanage, Eur. J. Org. Chem. 2015,
2405–2410; d) Y. Zhang, J.-L. Chen, Z.-B. Chen, Y.-M.
Zhu, S.-J. Ji, J. Org. Chem. 2015, 80, 10643–10650; e)
Q. Yuan, Z.-B. Chen, F.-L. Zhang, Y.-M. Zhu, Org.
Biomol. Chem. 2017, 15, 1628–1635. For the use of
phenyl formate in total synthesis, see: f) W.-Y. Zhang,
Q. Che, S. Crawford, M. Ronn, N. Dunwoody, J. Org.
Chem. 2017, 82, 936–943. For the use of 2,4,6-
trichlorophenyl formate in flow synthesis, see: g) J. L.
Vrijdag, F. Delgado, N. Alonso, W. M. De Borggraeve,
N. Pérez-Macias, J. Alcázar, Chem. Commun. 2014, 50,
15094–15097.
[11] Alkyl formates are known to undergo base-catalyzed
decomposition to CO and alcohols. For example, see: a)
T. Ushikubo, H. Hattori, K. Tanabe, Chem. Lett. 1984,
649–652; b) F. Q. Ma, D. S. Lu, Z. Y. Guo, J. Mol.
Catal. 1993, 78, 309–325.
[12] A theoretical study for the thermal decomposition of
methyl formate without a base has been reported; a) J. S.
Francisco, J. Am. Chem. Soc. 2003, 125, 10475–10480.
The gas-phase reaction of methyl formate with anions,
known as Riveros reaction, has been studied; b) P. C.
Isolani, J. M. Riveros, Chem. Phys. Lett. 1975, 33, 362–
364; c) W. L. Jorgensen, J. F. Blake, J. D. Madura, S. D.
Wierschke, Computational Investigations of Organic
Reaction Mechanisms. ACS Symp. Ser. 1987, 353, 200–
217.
[13] The reaction rates of 1a in benzene and THF at 70 C
have been reported. See ref. [10b]
.
[14] N. S. Isaacs, Physical Organic Chemistry, Longman
Scientific & Technical, Essex, 1987.
[15] Molecular calculations were performed using
Gaussian 09. M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.
[5] In carbonylation using CO, several studies have
indicated that low CO pressure affords good results; a)
F. Ozawa, H. Soyama, H. Yanagihara, I. Aoyama, H.
Takino, K. Izawa, T. Yamamoto, A. Yamamoto, J. Am.
Chem. Soc. 1985, 107, 3235–3245; b) W. Mägerlein, M.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Lett. 2006, 8, 4843–4846; b) J. R. Martinelli, D. A.
Watson, D. M. M. Freckmann, T. E. Barder, S. L.
Buchwald, J. Org. Chem. 2008, 73, 7102–7107.
Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J.
E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A.
J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.
Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D.
J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[24] The base not only catalyzes the CO formation from 1a
but also traps HI produced during the Pd-catalyzed
carbonylation. Therefore, stoichiometric amounts of the
base were required.
[25] We observed that the reaction of 1a with DBU (1.0
equiv) in CH3CN (0.50 M) at room temperature
completed within 3 min.
[16] a) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev.
1988, 88, 899–926; b) F. Weinhold, C. R. Landis,
Valency and Bonding. A Natural Bond Orbital Donor–
Acceptor Perspective, Cambridge University Press,
Cambridge, 2005.
[17] Tsuji et al. have reported the effects of substituents on
the Pd-catalyzed decomposition of phenyl formate. See
ref. [4m]
.
[18] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91,
165–195.
[19] a) C. Reichardt, T. Welton, Solvents and Solvent
Effects in Organic Chemistry, 4th Ed., Wiley-VCH,
Weinheim, 2011; b) J. A. Riddick, E. E. Toops,
Technique of Organic Chemistry. Vol. VII. Organic
Solvents. 2nd Ed., Interscience Publishers, New York,
1954.
[20] a) M. B. Smith, J. March, March’s Advanced Organic
Chemistry. Reactions, Mechanisms, and Structure, 5th
Ed., John Wiley & Sons, New York, 2001; b) F. A.
Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann,
Advanced Inorganic Chemistry, 6th Ed., John Wiley &
Sons, New York, 1999; c) L. Cecchi, F. D. Sarlo, F.
Machetti, Eur. J. Org. Chem. 2006, 4852–4860; d) K.
M. K. Yu, I. Curcic, J. Gabriel, H. Morganstewart, S. C.
Tsang, J. Phys. Chem. A 2010, 114, 3863–3872; e) I.
Kaljurand, A. Kütt, L. Sooväli, T. Rodima, V. Mäemets,
I. Leito, I. A. Koppel, J. Org. Chem. 2005, 70, 1019–
1028.
[21] We performed DFT calculation (B3LYP/6-31+G(d),
298.15 K under vacuum) of the reaction of 1a with
DABCO to give a nucleophilic substitution product.
However, the reaction pathway of the formyl transfer
from 1a to DABCO was obtained, with much higher
G‡ value (187.5 kJ/mol) than that of the concerted -
elimination of 1a by DABCO (151.2 kJ/mol). These
results suggest that the nucleophilic substitution
pathway (as shown in Scheme 4) would not be
operative.
[22] DFT calculation (B3LYP/6-31+G(d), 298.15 K under
vacuum) gave G‡ value of 116.5 kJ/mol for the
concerted -elimination of 1a with DBU.
[23] The effectiveness of Pd(OAc)2‒Xantphos in
aminocarbonylation of aryl bromides using atmospheric
pressure of CO has been demonstrated. a) J. R.
Martinelli, D. M. M. Freckmann, S. L. Buchwald, Org.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700751
Advanced Synthesis & Catalysis
FULL PAPER
Mechanistic Insight into Weak-Base-Catalyzed
Generation of Carbon Monoxide from Phenyl
Formate and Its Application to Catalytic
Carbonylation at Room Temperature without
External Carbon Monoxide Gas
Adv. Synth. Catal. Year, Volume, Page – Page
Hideyuki Konishi, Mika Matsubara, Keisuke Mori,
Takaki Tokiwa, Sundaram Arulmozhiraja, Yuta
Yamamoto, Yoshinobu Ishikawa, Hiroshi
Hashimoto, Yasuteru Shigeta, Hiroaki Tokiwa, and
Kei Manabe*
10
This article is protected by copyright. All rights reserved.