CH3COONa as an effective catalyst for methoxycarbonylation of 1,6-hexanediamine by dimethyl carbonate to dimethylhexane-1,6-dicarbamate

Article abstract of DOI:10.1039/b915023b

Methoxycarbonylation of 1,6-hexanediamine (HDA) by dimethyl carbonate (DMC) was carried out, using, for the first time, CH₃COONa as catalyst. The effects of the solvent, reaction temperature, reaction time, and catalyst amount, were investigated. A yield as high as 99.0% of dimethylhexane-1,6- dicarbamate 2 has been obtained at a temperature of 348 K and a reaction time of 6 h. Mechanistic studies revealed that N-substituted acetamide, as the active intermediate product, and NaOH were first formed via the reaction between HDA and CH₃COONa. A further reaction between the N-substituted acetamide and DMC generated carbamates and methyl acetate, via a hexatomic ring intermediate. The CH₃COONa catalyst was finally recovered through the reaction between NaOH and methyl acetate, which thus completed the catalytic cycle.

Full text of DOI:10.1039/b915023b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
CH3COONa as an effective catalyst for methoxycarbonylation of
1,6-hexanediamine by dimethyl carbonate to dimethylhexane-
1,6-dicarbamate†
Da-Lei Sun,a,c Shun-Ji Xie,a Jian-Ru Deng,a Cai-Juan Huang,a Eli Ruckenstein*b and Zi-Sheng Chao*a
Received 17th April 2009, Accepted 6th January 2010
First published as an Advance Article on the web 28th January 2010
DOI: 10.1039/b915023b
Methoxycarbonylation of 1,6-hexanediamine (HDA) by dimethyl carbonate (DMC) was carried
out, using, for the first time, CH3COONa as catalyst. The effects of the solvent, reaction
temperature, reaction time, and catalyst amount, were investigated. A yield as high as 99.0% of
dimethylhexane-1,6-dicarbamate 2 has been obtained at a temperature of 348 K and a reaction
time of 6 h. Mechanistic studies revealed that N-substituted acetamide, as the active intermediate
product, and NaOH were first formed via the reaction between HDA and CH3COONa. A further
reaction between the N-substituted acetamide and DMC generated carbamates and methyl
acetate, via a hexatomic ring intermediate. The CH3COONa catalyst was finally recovered through
the reaction between NaOH and methyl acetate, which thus completed the catalytic cycle.
In the synthesis of carbamates from amines and DMC, Lewis
1. Introduction
acidic catalysts, such as the acetates of Mn, Zr, Sn, Zn and
Organic carbamates, which are very important starting mate-
Pb,4,11-13 the nitrates of Bi and Pb,14,15 the triflates of IIIB metals
rials and/or intermediates, have found versatile applications
(M(OTf)3 with M = Sc, La),16,17 and even CO2,18 were often em-
in the production of fine and commodity chemicals1,2 and in
ployed, providing appreciable yields of carbamates. Because they
organic synthesis.1 Currently, the commercial production of
can be thermally or catalytically converted into diisocyanates,
carbamates is based mainly on the reaction between alcohols
which are widely used as precursors in polymer preparation,19
and isocyanates.3 However, this route is not environmentally
the dicarbamates are of particular industrial interest. Few
friendly, because the industrial production of isocyanates is
papers have dealt with the synthesis of dicarbamate from DMC
associated with a high energy input, toxic phosgene as a raw
and diamines, especially the aliphatic ones, over catalysts such
material, and the corrosive hydrochloric acid as a side product.2
as Bi(NO3)2,14 M(OTf)3 (with M = Sc, La),16,17 NaOCH3
20
Several phosgene-free processes for the production of organic
and ionic liquids MIm(CH2)4SO3HTfO.21 We have previously
carbamates have been explored,4-8 among which the alkoxycar-
reported the synthesis of dimethylhexane-1,6-dicarbamate via
bonylation of amines by alkyl carbonates,4 particularly dimethyl
the methoxycarbonylation of 1,6-hexanediamine (HDA) by
carbonate (DMC),5 have received considerable attention. This
DMC over a ZnAlPO4 berlinite catalyst.22 In the present paper,
route appears to be promising, because on the one hand
sodium acetate was employed, for the first time, as catalyst for
methanol is obtained as the main by-product and on the other
this reaction and a yield as large as 99.0% of 2 (dimethylhexane-
hand DMC is non-toxic and can be prepared on a large scale by
1,6-dicarbamate) under mild conditions has been obtained.
the oxidative carbonylation of methanol.9,10 In addition, DMC
The sodium acetate catalyst is superior to the other catalysts
can be easily handled and is cheap, non-toxic and clean. Thus, if
already reported in the literature, because it is cheaper, non-
the methoxycarbonylation reaction between an amine and DMC
toxic, and much more efficient. In addition, the mechanism of
is combined with the oxidative carbonylation of methanol, a
methoxycarbonylation of HDA by DMC using sodium acetate
green process for the production of organic carbamates, with
as catalyst was examined.
“zero emission”, can be achieved.
2. Experimental
2.1 Reagents
aCollege of Chemistry and Chemical Engineering; Key Laboratory of
Chemometrics & Chemical Biological Sensing Technologies, Ministry of
The reagents involved are as follows: HDA (C.P., Shanghai
Education, Hunan University, Changsha, 410082, China. E-mail:
Lingfeng Chemicals Company); CH3OH (A.R., Shanghai Lu-
dou Chemicals Company); CH3CN (A.R., Tianjin Kemiou
Chemicals Reagents Company); CH3COONa, DMC, DMF,
CH3COOH, (CH3CO)2O, and acetamide (A.R., China Na-
tional Medicines Chemicals Reagents Corporation Ltd.). Before
use, all the agents were pretreated to remove the water they
contained: CH3COONa was dried at 373 K in a vacuum
zschao@yahoo.com; Tel: +86-731-8713257
bDepartment of Chemical and Biological Engineering, State University of
New York at Buffalo, Buffalo, NY, 14260, USA. E-mail:
feaeliru@buffalo.edu; Tel: +1-716-6451179
cDepartment of Chemical Engineering and Light Industry, Guangdong
University of Technology, Guangzhou, 510006, China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/b915023b
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 483–490 | 483
©

View Article Online
oven and the liquid agents were dehydrated using molecular
sieves 4A.
mentioned, the conversion of HDA was calculated on the basis
of the initial amount of HDA, the selectivities to products were
calculated on the basis of the amount of HDA converted, and the
yield of 2 was calculated by multiplying the conversion of HDA
with the selectivity to 2 (ESI: Section 2). The thus-obtained yield
of 2 is denoted as GC yield.
2.2. Methoxycarbonylation of HDA by DMC
The methoxycarbonylation of HDA by DMC was performed
under a N2 atmosphere in a 250 ml three-neck flask equipped
with a condenser and a magnetic stirrer. A solution was
first prepared by mixing selected amounts of HDA, DMC,
CH3COONa and solvent (CH3OH, CH3CN, or DMC) with
stirring at room temperature, until a small amount of white
floccules appeared. Then, a flow of nitrogen was introduced
to drive out the air contained in the three-neck flask, and
the solution was heated to the desired temperature and kept
at that temperature for the reaction to take place. The white
floccules disappeared within the first 10 min of reaction, and
by prolonging the reaction time, the clear solution turned
gradually into a white emulsion. After a certain reaction time,
a white viscous gel was generated. This gel was cooled to room
temperature and then subjected to GC and GC-MS analysis.
To identify the reaction intermediates and products, FT-
1
IR and H NMR spectroscopies were employed. The FT-IR
spectrum was recorded on a Thermo Nicolet Nexus 670 FT-
IR spectrometer equipped with a MCTA detector and a ZnSe
single-bounce attenuated total reflectance (ATR) accessory,
using a 32 scans at 4 cm-1 resolution. The specimen was
prepared via a dip-coating procedure using a KBr disk as
1
support. The H NMR spectrum was recorded on a 400 MHz
Varian INOVA-400 spectrometer, with tetramethylsilane (TMS)
in deuterochloroform (CDCl3) as internal standard.
2.4. Purification and identification of
dimethylhexane-1,6-dicarbamate
After the methoxycarbonylation of HDA by DMC in the
presence of CH3COONa as catalyst, the product mixture was
cooled to room temperature and an excess amount of dilute
aqueous HCl solution was added with stirring. The resulting
mixture was allowed to stand until two layers were formed. A
hydrophilic layer contained mainly the catalyst (CH3COONa),
and a hydrophobic layer consisted mainly of the unconverted
reactants and products. After the separation of the hydrophilic
layer, the hydrophobic layer was vacuum-distilled to remove
the DMC, the methanol, and, if any, solvent, yielding a white
solid, i.e., the crude 2. Then, the crude 2 was subjected to re-
crystallization, involving the steps of dissolution in water at ca.
353 K, aging at room temperature for 2 h, filtration and washing
with cold water. The re-crystallization procedure was repeated
five times, and after drying in vacuo, the purified 2 was finally
obtained. The percentage of the weight of purified 2 relative
to that of 2 predicted theoretically by the stoichiometry of the
reaction was then obtained and denoted as isolated yield of 2.
1H NMR and GC-MS were employed to identify the purified
2 and the results are as follows: 1H NMR (ESI: Fig. S2),
d = 1.33 (m, 4H; -CH2CH2CH2CH2CH2CH2-), 1.48 (m, 4H;
-CH2CH2CH2CH2CH2CH2-), 3.14 (q, 4H; -NHCH2-), 3.66 (s,
6H; -OCH3) and 4.79 (broad, 2H; -NHCO-); GC-MS (ESI: Fig.
S3 and Table S3), m/z = 232, 201, 173, 158, 144, 130, 116, 114,
102, 99, 88, 74, 59, 56 and 44.
2.3. Analysis of intermediates and products
The product mixtures of methoxycarbonylation of HDA by
DMC were sampled, dissolved into methanol and analyzed
using a Varian Saturn2200-CP3800 GC-MS spectrometer
equipped with two 15 m ¥ 0.32 mm CP-Wax 52CB fused
silica capillary columns, which were linked to a mass detector
for qualitative analysis and to a flame ionization detector
(FID) for quantitative analysis. It was reported14,17 that in the
methoxycarbonylation of HDA by DMC, besides the desired
product 2 (dimethylhexane-1,6-dicarbamate), a few by-products,
such as, 1 (methyl-6-amino-hexyl-1-carbamate) and 3–6, might
be generated. In our work, the desired product 2, the main by-
product 1 and the minor by-product 7 have been identified (ESI:
Fig. S1 and Tables S1 and S2) in all the runs. Occasionally,
depending on reaction conditions, trace amounts of the by-
products 3–6 could also be detected. Since the amounts of
by-products 3–7 identified in our experiments were small, their
selectivities are reported together in this paper. Unless otherwise
3. Results and discussion
3.1 Methoxycarbonylation of HDA by DMC
Table 1 presents the results of methoxycarbonylation of HDA by
DMC in the absence and presence of various solvents (methanol,
DMC, or CH3CN), using CH3COONa as catalyst, a reaction
time of 10 h and a reaction temperature of 348 K. All the
batches consisted of 100 mmol HDA, 200 mmol (16.8 ml) DMC,
24.4 mmol (2 g) CH3COONa and 50 ml solvent. Except the
batch that employed DMC as solvent, all other batches had a
stoichiometric molar ratio of DMC/HDA (= 2) required for the
methoxycarbonylation of HDA by DMC. It was found that in
the absence of solvent, the conversion of HDA, the selectivity to
484 | Green Chem., 2010, 12, 483–490
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Effect of solvent on methoxycarbonylation of HDA by DMC using CH3COONa as catalysta
c
e
f
cHDA
Si d(%)
Y2,GC
(%)
Y2,isolated
Solvent
(%)
1
2
3–7
(%)
CH3OH
DMC
100
98
97
2.6
3.9
20.4
29.2
96.1
93.6
80.7
65.8
1.3
2.6
2.0
5.0
96.1
91.7
78.3
48.7
87.2
81.5
62.8
21.9
CH3CN
Noneb
74
a HDA: 100 mmol; DMC: 200 mmol (16.8 ml); CH3COONa: 2.0 g (24.4 mmol); solvent: 50 ml; reaction time: 10 h; reaction temperature: 348 K. b No
solvent was employed. c HDA: conversion of HDA. d Si: selectivity to component i in the products mixture e 2, GC : GC yield of 2. f
c Y Y2, isolated : isolated
yield of 2.
2 and the yield of 2 were only 74.0, 65.8 and 48.7%, respectively.
When a solvent was employed, the conversion of HDA, the
selectivity to 2 and the yield of 2 increased to above 97.0, 80.7 and
78.3%, respectively. This indicates that the solvent stimulated
the methoxycarbonylation of HDA by DMC. The stimulation
by the solvent was in the order CH3OH > DMC > CH3CN >
no solvent. Though the isolated yields of 2 were lower than the
GC ones, due to the loss of 2 during re-crystallization, they
still remained in the order CH3OH > DMC > CH3CN > no
solvent. The above observation is useful, because DMC is, on
the one hand, one of the main raw materials employed in the
synthesis of 2, and on the other hand, is currently manufactured
via the reaction between methanol and CO2.9,10 Consequently,
the product mixture (DMC + CH3OH) obtained during the
DMC synthesis can be directly employed in the synthesis of 2.
The use of DMC or methanol as solvent is consistent with the
“green chemistry rule”, to pursue a synthesis process using as
little solvent and operational steps as possible.23
The above result reveals that both methanol and DMC
are good solvents for the methoxycarbonylation of HDA by
DMC. Since methanol is also a by-product and DMC is one
of the reactants of the methoxycarbonylation of HDA by
DMC, it is kinetically favorable for the amount of methanol
to be reduced and that of DMC increased. Therefore, in
the following experiments, as illustrated by Fig. 1–3, batches
consisting of 50 mmol HDA, 200 mmol (16.8 ml) DMC and
20 ml CH3OH, which have a higher molar ratio of DMC/HDA
and a lower amount of methanol than those used in Table 1,
were employed. One can consider that in these experiments
the methoxycarbonylation of HDA by DMC takes place in the
presence of DMC and methanol as co-solvents. These reactions
were conducted at 348 K for 6 h, using 0.6 g (7.3 mmol) of
CH3COONa as catalyst. By changing the reaction temperature,
reaction time and amount of CH3COONa, the effects of these
factors were investigated.
Fig. 1 Effect of reaction temperature on HDA conversion, yield of
2 and selectivities to reaction products. Reaction conditions: HDA,
50 mmol; CH3COONa, 0.6 g; DMC, 200 mmol; time, 6 h; CH3OH,
20 ml. (Legend: (ꢀ) HDA conversion; (᭺) yield of 2; (ꢁ), (᭹) and (ꢀ),
selectivities to 2, 1 and 3–7, respectively.)
Fig. 1 presents the effect of reaction temperature. One can see
that the selectivity to 3–7 is much lower than those to 1 and 2 at
all temperatures tested. With increasing reaction temperature,
the selectivity to 3–7 increases only slightly. This occurs because
the activation energy for N-methylation is much larger than that
for methoxycarbonylation.17 The selectivity to 2 first increases,
attains a maximum at 348 K, followed by a slight decrease, with
increasing reaction temperature. The change of the selectivity to
1 with temperature exhibits the opposite trend, compared to that
to 2. This indicates that the activation energy for the formation of
2 is higher than that for 1. The decrease in the selectivity to 2 and
Fig. 2 Effect of reaction time on HDA conversion, yield of 2 and
selectivities to reaction products. Reaction conditions: HDA, 50 mmol;
CH3COONa, 0.6 g; DMC, 200 mmol; temperature, 348 K; CH3OH,
20 ml. (Legend: (ꢀ) HDA conversion; (᭺) yield of 2; (ꢁ), (᭹) and (ꢀ),
selectivities to 2, 1 and 3–7, respectively.)
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 483–490 | 485
©

View Article Online
increase rapidly when the reaction time increases up to 6 h. The
further increase of reaction time results in an approach towards
a HDA conversion of 100%, but by a decrease of the yield of
2, caused by the decomposition of 2 to 1. Consequently, the
optimum reaction time is 6 h.
Fig. 3 presents the effect of the amount of CH3COONa. One
can see that in the presence of this catalyst, the conversion of
HDA, the yield of 2 and the selectivity to 2 increase but the
selectivity to 1 decreases, compared to those in the absence of
this catalyst. The selectivity to 3–7 is much lower than those to 1
and 2 and almost unchanged in the absence or in the presence of
various amounts of catalyst. When 0.6 g (7.3 mmol) of catalyst
is employed, a 100% conversion of HDA is achieved. With a
further increase of the amount of catalyst in the range of 0.6 to
1.0 g (12.2 mmol), the conversion of HDA remains 100%, the
selectivity to 2 increases and that to 1 decreases. These results
indicate that the catalyst stimulates the formation of both 1 and
2, with 1 as the primary product and 2 as the secondary one via
the further reaction of 1. After HDA was completely converted,
the increase of the amount of catalyst stimulated the conversion
of 1 to 2, because HDA is no longer available for the formation
of 1. When the amount of catalyst is higher than 1.0 g, the
selectivity to 2 decreases slightly and that to 1 increases. This
occurs most likely because the decomposition of 2 to 1 becomes
predominant, when 1 is no longer available for the formation
of 2. Consequently, the optimum amount of catalyst is 1.0 g
(12.2 mmol).
Table 2 presents the results of methoxycarbonylation of HDA
by DMC in the absence and in the presence of CH3COONa
as catalyst; for comparison, the results obtained with other
catalysts reported in the literature are also included. One can
see that both the conversion of HDA, the selectivity to 2 and
the yield of 2 are small in the absence of the catalyst but large
in its presence. CH3COONa appears to be the most efficient
catalyst for the methoxycarbonylation of HDA by DMC under
comparable reaction conditions, exhibiting a 100.0% conversion
of HDA and a 99% yield of 2 at 348 K in 6 h.
Fig. 3 Effect of amount of catalyst on HDA conversion, yield of
2 and selectivities to reaction products. Reaction conditions: HDA,
50 mmol; DMC, 200 mmol; time, 6 h; temperature, 348 K; CH3OH,
20 ml. (Legend: (ꢀ) HDA conversion; (᭺) yield of 2; (ꢁ), (᭹) and (ꢀ),
selectivities to 2, 1 and 3–7, respectively.)
the increase of that to 1 at temperatures higher than 348 K may
be due to the decomposition of 2 to 1, because 1 is the primary
and 2 the secondary product in the methoxycarbonylation
of HDA by DMC.14 With increasing temperature, both the
conversion of HDA and the yield of 2 increase. At temperatures
≥348 K, the conversion increases to and is maintained at 100%,
however, the yield of 2 decreases, as a result of the decrease of
the selectivity to 2. Consequently, the optimum temperature for
the production of 2 via the methoxycarbonylation of HDA by
DMC is around 348 K.
Fig. 2 outlines the effect of reaction time. One can see that for
all reaction times tested, the selectivity to 3–7, which increases
very slightly with increasing reaction time, was much lower
than those to 1 and 2. By prolonging the reaction time, the
selectivity to 2 increases, attaining a maximum at 6 h, followed
by a decrease. The selectivity to 1 exhibits, however, the inverse
trend, compared to that to 2, with increasing reaction time. This
occurs because 2 is a secondary product formed via the further
reaction of 1. Both the conversion of HDA and the yield of 2
3.2. On the reaction mechanism
It is known that the methoxycarbonylation of amines by DMC is
the result of a SN2 nucleophilic attack by amines on the carbonyl
Table 2 Methoxycarbonylation of 1,6-hexanediamine by DMC with various catalysts
c
Batch composition
Reaction conditions
Sib(%)
Y2
Catalyst and its quantity
HDA/mmol DMC/mmol CH3OH/ml T/K
Time/h cHDAa(%)
1
2
3–7 (%)
None
50.0
50.0
200.0
200.0
20.0
20.0
348
348
6
6
16.0
100.0
51.9
0.9
25.6 22.5 4.1
CH3COONa (1.0 g/
12.2 mmol)
99.0 0.1
99.0
Bi(NO3)3 (0.12 mmol)14
NaOCH3 (28.2 mmol)20
MIm(CH2)4SO3HTfO
(3.26 mmol 1 wt% ionic
liquids)21
3.5
392.0
400.0
21.0
2875.0
820.0
3.0
123.8
0
353
343
353
18
1
4
100.0
99.5
100.0
11.0
N Ad
N A
84.0 5.0
84.0
98.9 N A 98.4
95.0 N A 95.0
Sc(OTf)3 (0.295 mmol)17
ZnAlPO4 (1.0 g)22
11.3
200.0
59.4
1600.0
0
0
293
353
144
8
100.0
99.0
N A
3.9
88.0 N A 88.0
88.6 7.5
87.7
a
c
HDA: conversion of HDA. b Si: selectivity to component i in products mixture. c Y2: yield of 2; all the rows in this column are GC yields, except the
second row from bottom, which is isolated yield. d N. A.: not available.
486 | Green Chem., 2010, 12, 483–490
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
carbon of DMC. Whereas a significantly higher basicity of the
amine than that of the leaving group (methoxyl) is required,4,24-26
methoxycarbonylation depends mostly upon the reactivity of
DMC. The factors that increase the electrophilicity of carbonyl
carbon promote this reaction. An electron-withdrawing effect
exerted on the carbonyl group facilitates its reaction with
a nucleophilic amine.4,24,27 Lewis acidic catalysts were often
employed to promote the methoxycarbonylation of amines by
DMC. For example, over the Lewis acidic catalyst ZnAlPO4,22
dimethylhexane-1,6-dicarbamate is formed via a catalytic cycle,
which involves the activation of DMC due to the coordination
of the carbonyl oxygen with Zn(II) and the nucleophilic attack
by the amino group of HDA on the carbonyl carbon of the
activated DMC. The absence of acidity in the CH3COONa
catalyst suggests that a different mechanism may be involved in
the methoxycarbonylation of HDA by DMC over this catalyst.
Mason and Charles19 found that the organic diisocyanate
is formed via the reaction between diphenyl carbonate and
an organic diformamide at temperatures higher than 473 K.
The organic diformamide can be employed as such or can
be generated in situ by reacting the corresponding diamine
with formic acid. Besides the diisocyanate, byproducts such as
dicarbamate, mixed isocyanate/carbamate and phenyl formiate
are also obtained. While the mechanism was not provided by the
authors, the carbamate was most likely the key intermediate in
the formation of isocyanate from formamide and carbonate,
because the thermolysis of the carbamate yields isocyanate.
For the methoxycarbonylation of HDA by DMC, catalyzed by
CH3COONa, it is suggested that the N-substituted acetamide,
RNHCOCH3, is formed in situ via the reaction between HDA
and CH3COONa and further contributes to the formation of
carbamate by reacting with DMC. The following experiments
provide evidence in this direction.
NaOH in the form of white floccules was removed by centrifugal
separation. The supernatant liquid was characterized by FT-
IR and the results indicated the presence of the N-substituted
acetamide, HDA and acetate (the spectrum was similar to that
of Fig. S4 in ESI and for this reason is not provided). To the
above supernatant liquid, an excess amount of DMC was added.
The resulted mixture was heated to 348 K for further reaction,
and after 6 h, the product mixture was vacuum-distilled. A
white powder was finally obtained and characterized by FT-
IR, which indicated the presence of di-carbamate and mono-
carbamate (the spectrum was similar to that of Fig. S5 in ESI
and therefore is not given). The above results indicate that the
N-substituted acetamide plays the role of an active intermediate
and that NaOH is not a catalyst for the formation of carbamates
during the methoxycarbonylation of HDA by DMC, since the
separation of NaOH from the reaction system did not stop the
formation of carbamate.
One may argue that a part of NaOH was dissolved into the
above supernatant liquid, and after the addition of DMC to the
supernatant liquid, this residual NaOH catalyzed the methoxy-
carbonylation reaction. To clarify this issue, we conducted an
experiment, which can be described as follows. HDA (10 mmol),
DMC (40 mmol) and NaOH (0.4 g/10 mmol) were dispersed
into CH3OH (20 ml) with stirring at room temperature. Then,
the mixture was heated to 348 K and kept at that temperature
for 6 h. After cooling to room temperature, a few white
floccules appeared. The floccules were recovered by centrifugal
separation and identified to be NaOH. The GC-MS analysis of
the supernatant liquid indicated the presence of HDA, DMC
and methanol; no other component could be identified. This
result confirms the conclusion that NaOH is not a catalyst for
the formation of carbamates via the methoxycarbonylation of
HDA by DMC.
Experiment A: Identification of reaction intermediates and
products via a two-step reaction for the methoxycarbonylation
of HAD by DMC using CH3COONa as catalyst. A mixture of
HDA (50 mmol), CH3COONa (1.0 g/12.2 mmol) and CH3OH
(20 ml) was first subjected to reaction at 337 K for 1 h. Chemical
analysis and FT-IR identified that N-substituted acetamide
(ESI: Fig. S4 and Table S4) and NaOH (ESI: Section 5) were
formed via this reaction. Then, 200 mmol (16.8 ml) of DMC
were added to the above products mixture for further reaction.
After heating to 348 K and being kept at this temperature for
reaction for 5 h, the product mixture was analyzed by GC
and GC-MS and the results revealed that the conversion of
HDA was 100% and that the selectivities to 1, 2 and 3–7 were
0.9%, 99% and 0.1%, respectively. The product mixture was also
vacuum-distilled until a white powder was generated. The FT-
IR characterization of the white powder indicated the presence
of mainly 2, much less 1 and very little 3–7 (ESI: Fig. S5 and
Table S5). The above results indicate that both the N-substituted
acetamide and NaOH were formed as intermediates, which were
consumed during the formation of carbamates, by the reaction
between HDA and CH3COONa.
Experiment C: Methoxycarbonylation of acetamide by DMC in
the absence of any catalyst. Acetamide (50 mmol) was mixed
with DMC (200 mmol/16.8 ml) and methanol (20 ml) with
stirring at room temperature, and then the temperature was
raised to 348 K and kept at that temperature for reacting for 6 h.
After that, the reaction mixture was vacuum-distilled, resulting
in a white powder, which was characterized by FT-IR (ESI: Fig.
S6 and Table S6). The results revealed the presence of methyl
carbamate (CH3OCONH2), besides the unconverted acetamide.
It indicates that an amide such as the acetamide can interact
directly with DMC to generate carbamate.
Experiment D: Methoxycarbonylation of HDA by DMC
using either CH3COOH or (CH3CO)2O as catalyst. DMC
(200 mmol/16.8 ml) and methanol (20 ml) were first mixed
and then HDA (50 mmol) was added with stirring at room
temperature, resulting in a clear solution. After that, CH3COOH
(0.8 g/13.3 mmol) was added, and it was found that a white
fume burst out and a few small white crystals appeared in the
solution, releasing a large amount of heat. The crystals were
sampled and identified to be N-substituted ammonium acetate,
which is formed via the neutralization reaction of CH3COOH
with HDA, as shown by eqn (1):
Experiment B: Role of the reaction intermediates N-substituted
acetamide and NaOH. A mixture of HDA (10 mmol),
CH3COONa (0.8 g/9.8 mmol) and CH3OH (20 ml) was first
subjected to reaction at 337 K for 4 h. Then, the formed
CH3COOH + H2N(CH2)6NH2 →
CH3COOH3N(CH2)6NH2
(1)
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 483–490 | 487
©

View Article Online
Scheme 1 Possible mechanism for the methoxycarbonylation of HDA by DMC using CH3COONa as catalyst.
After the batch (HDA + DMC + CH3COOH + CH3OH)
was heated to 348 K and kept at that temperature for 6 h,
neither 1 nor 2 could be identified. When the CH3COOH in the
(HDA + DMC + CH3COOH + CH3OH) batch was replaced
by (CH3CO)2O and heated at 348 K for 6 h, again, neither 1
nor 2 could be identified. This occurred because though the N-
substituted acetamide may be formed via the ammonolysis of
(CH3CO)2O by HDA (eqn (2)), this reaction is competing with
both the alcoholysis of (CH3CO)2O by methanol (eqn (3)) and
the neutralization of HDA by CH3COOH (eqn (1)):
H2N(CH2)6NH2 + (CH3CO)2O →
CH3COHN(CH2)6NH2 + CH3COOH
(2)
CH3OH + (CH3CO)2O → CH3COOCH3 +CH3COOH (3)
Reactions (1) and (3) proceed at much higher rates than
reaction (2), thus transforming HDA mainly into the N-
substituted ammonium acetate. Thus, not enough N-substituted
acetamide remained available for the formation of carbamates
(1 and 2). The above results indicate that, neither CH3COOH
nor (CH3CO)2O play the role of catalyst in the methoxycar-
bonylation of HDA by DMC, since the acidic environment
they provided rapidly transformed HDA into the N-substituted
ammonium acetate instead of N-substituted acetamide.
Scheme 2 Possible mechanism for the formation of N-substituted
acetamide via the reaction between HDA and CH3COONa.
Summing up the results of experiments A to D, one can con-
clude that during the methoxycarbonylation of amines by DMC,
CH3COONa plays the role of a catalyst. The N-substituted
acetamide, RNHCOCH3, and NaOH are first formed via the
reaction between HDA and CH3COONa. NaOH does not play
the role of a catalyst but provides an alkaline environment that
ensures the conversion of HDA to N-substituted acetamide
instead of N-substituted ammonium acetate. The observation
that an alkaline environment is favorable to the methoxycar-
bonylation of amines by DMC has already been reported in
the literature.4,9 The N-substituted acetamide constitutes the
key intermediate product and its further reaction with DMC
results in the formation of the carbamates 1 and 2. A possible
mechanism, shown in Scheme 1, for the methoxycarbonylation
of HDA by DMC, with CH3COONa as catalyst, is proposed,
and more details are provided in Schemes 2–4.
In the methoxycarbonylation of amines by DMC, an acidic
catalyst was usually employed to activate the DMC. However,
the nucleophilicity of the amine is potentially reduced by the
interaction between the basic amine and the acidic catalyst. In
the methoxycarbonylation of amines by DMC, in the presence
of the basic CH3COONa catalyst, N-substituted acetamide is
first formed via the reaction between HDA and CH3COONa.
As shown in Scheme 2, in the absence of a solvent, CH3COONa
may exist as an equilibrium among a few resonant structures,
e.g., a1, a2 and a3. The positively charged carbon of a3 is attacked
by the nucleophilic HDA, generating the intermediate b. The
ring-opening and rearrangement of b results in N-substituted
acetamide and NaOH. Due to the large strain of the tetratomic
ring, the equilibrium among a1, a2 and a3 may be displaced to
a1 and a2, generating less a3 and in turn less b and N-substituted
acetamide. While N-substituted acetamide has been identified as
the active intermediate product for the formation of carbamates,
a relatively low yield of carbamates is expected in the absence
of a solvent. When a polar non-protic solvent, e.g., CH3CN, is
employed, a3 can be stabilized by the dipole–dipole interactions
between solvent and a3. As a result, the yield of carbamates
is increased. When a non-polar non-protic solvent, e.g., DMC,
is employed, a higher yield of carbamates than for the polar
non-protic solvent CH3CN is obtained. This occurs because,
whereas no stabilization of a3 is provided by the solvent, the
increase in the amount of DMC, as one of the reactants of
the methoxycarbonylation, promotes the reaction in the right
direction. When a polar protic solvent is employed, e.g., CH3OH,
a hexatom-ringed intermediate c may be formed between a3 and
CH3OH. The higher stability of c than a3 results in a higher yield
488 | Green Chem., 2010, 12, 483–490
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 3 Possible mechanism for the formation of carbamate via the reaction between N-substituted acetamide and DMC.
generates the intermediate IIIa and the methoxyl group (CH3O-),
both driven by the resonance between IIIa and IIIb. The positive
charge of the carbon in IIIb enables its attack by CH3O-,
resulting in the intermediate IV. Due to the hydrogen bonding
between N, O and H in IV, a tetratomic ring intermediate
V is formed, and the ring-opening and rearrangement of V
generates the carbamate 1 and the methyl acetate. The reaction
between methyl acetate and NaOH, which occurs concomitantly
with the formation of the N-substituted acetamide through
the reaction between HDA and CH3COONa, recovers the
CH3COONa catalyst and produces the byproduct CH3OH.
Similar steps and intermediates are expected to occur for the
reaction between 1 and DMC, in the presence of CH3COONa as
catalyst, to produce the desired product 2. We tried to react N,N¢-
dimethyl formamide (DMF) with DMC, but no carbamates (1
and 2) could be identified. This probably occurred because no
hydrogen atom can be provided by DMF for the formation of
a hexatomic ring intermediate, and for this reason, it cannot
generate carbamates. This indicates that the formation of the
intermediate II in the above steps is essential.
Scheme 4 Possible mechanism for the formation of 3–7 during the
methoxycarbonylation of HDA by DMC using CH3COONa as catalyst.
of carbamates when CH3OH is employed as solvent rather than
CH3CN.
The formation of carbamates is a result of the reaction
between the N-substituted acetamide and DMC. As shown in
Scheme 3, the N-substituted acetamide is expected to exist as
an equilibrium between the resonant structures Ia and Ib. With
the proton of Ib transferred to the carbonyl oxygen of DMC, the
DMC becomes activated and the negatively charged N in the Ib
becomes more active in the nucleophilic attack on the carbonyl
carbon of the DMC. This results in the formation of a hexatomic
ring intermediate II. The ring-opening and rearrangement of II
The formation of the N-alkylated byproducts 3–7, identified
in the present work, may be a result of a nucleophilic attack of the
amino groups of the amines on the methoxyl carbon of DMC,
as shown in Scheme 4. Because of the ambident electrophilicity,
DMC may react with nucleophiles not only at the carbonyl
group but also at the methyl moiety.17 When a Lewis acidic
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 483–490 | 489
©

View Article Online
catalyst was employed in the methoxycarbonylation of amines
by DMC, the carbonyl in DMC was activated by the catalyst,
inducing a positive charge on the carbonyl carbon and further
on the methoxyl carbon.15 The nucleophilic attack of the amines
on the positively charged carbons results in the formation of
both carbamates and N-alkylated amines. High selectivities to
N-alkylated amines were also achieved at relatively high reaction
temperatures.17 In the present work, the basic CH3COONa
catalyst promotes the methoxycarbonylation of amines by DMC
via a mechanism very different from that of the Lewis acidic
catalysts. Carbamates are formed as a result of the reaction
between DMC and the N-substituted acetamides, which are
generated by reacting amines with CH3COONa. It is expected
that the positive charge of the methoxyl carbon of DMC in
the presence of a basic catalyst will be smaller than that for
a Lewis acidic catalyst. This might be another reason for the
extremely low selectivity to 3–7 observed in our work, besides the
relatively low reaction temperature (lower than 360 K) employed
compared with that used in the literature for N-alkylation of
amines.
Acknowledgements
We are grateful for financial support from the Program for
New Century Excellent Talents in University, the Ministry of
Education of P. R. China; and the Program for Lotus Scholar
in Hunan Province, P. R. China.
References
1 T. W. Greene and P. G. M. Wuts, in Protective Groups in Organic
Synthesis, Wiley & Sons, New York, 3rd edn, 1999, p. 503.
2 P. Adams and F. A. Baron, Chem. Rev., 1965, 65, 567.
3 S. L. Shapiro, V. Bandurco and L. Freedman, J. Org. Chem., 1961,
26, 3710.
4 T. Baba, A. Kobayashi, Y. Kawanami, K. Inazu, A. Ishikawa, T.
Echizenn, K. Murai, S. Aso and M. Inomata, Green Chem., 2005, 7,
159.
5 P. Tundo and M. Selva, Acc. Chem. Res., 2002, 35, 706.
6 R. A. Jacobson, J. Am. Chem. Soc., 1938, 60, 1742.
7 N. V. Kaminshkia and N. M. Kostic, Inorg. Chem., 1998, 37, 4302.
8 F. Paul, J. Fischer, P. Ochsenbein and J. A. Osborn, Organometallics,
1998, 17, 2199.
9 K. T. Jung and A. T. Bell, J. Catal., 2001, 204, 339.
10 D. Delledome, F. Rivetti and U. Romano, J. Organomet. Chem., 1995,
488, C15.
11 W. Briu, US Pat., 3 763 217, 1973.
12 E. A. Gurgiolo, US Pat., 4 268 683, 1981.
4. Conclusion
13 C. Calderoni, F. Mizia, F. Riveti, U. Romano and Vimercate, US
Pat., 5 091 556, 1992.
14 R. G. Deleon, A. Kobayashi, T. Yamauchi, J. Ooishi, T. Baba, M.
Sasaki and F. Hiarata, Appl. Catal. A: Gen., 2002, 225, 43.
15 T. Baba, M. Fujiwara, A. Oosaku, A. Kobayashi and R. G. Deleon,
Appl. Catal. A: Gen., 2002, 227, 1.
It is demonstrated, for the first time, that CH3COONa is a
very effective catalyst for the production of dimethylhexane-
1,6-dicarbamate (2) via the methoxycarbonylation of HDA by
DMC, when CH3OH or even DMC itself is used as solvent. At a
temperature of 348 K and reaction timeof 6h, a100%conversion
of HDA with a 99% selectivity to 2 could be achieved. It is shown
that the catalytic cycle of the methoxycarbonylation of HDA by
DMC with CH3COONa as catalyst consists mainly of three
steps: the formation of the N-substituted acetamide and NaOH
via the reaction between HDA and CH3COONa, the generation
of carbamates and methyl acetate via the reaction between the
N-substituted acetamide and DMC, and the regeneration of the
CH3COONa catalyst, releasing methanol as a byproduct, via
the reaction between the NaOH and methyl acetate. The yield
of carbamates is affected largely by the solvent, most likely due
to the different stabilities of the reaction intermediates, which
depend on the nature of the solvent.
16 D. Monica and Q. Eugenio, J. Catal., 2004, 228, 36.
17 M. Distaso and E. Quaranta, Appl. Catal. B: Env., 2006, 66, 72.
18 A. Inesi, V. Mucciante and L. Rossi, J. Org. Chem., 2001, 66, 1035.
19 R. Mason and L. Charles, US Pat., 6 781 010, 2004.
20 T. Yagii, T. Itokazu, and K. Murata, US Pat., 5 773 643, 1994.
21 H. C. Zhou, F. Shi, X. Tian, Q. H. Zhang and Y. Q. Deng, J. Mol.
Catal. A: Chem., 2007, 271, 89.
22 D. L. Sun, J. R. Deng and Z. S. Chao, Chem. Cent. J., 2007, 1, 27,
DOI: 10.1186/1752-153X-1-27.
23 P. T. Anastas and M. M. Kirchhoff, Acc. Chem. Res., 2002, 686, 35.
24 A. B. Shivarkar, S. P. Gupte and R. V. Chaudhari, J. Mol. Catal. A:
Chem., 2004, 223, 85.
25 E. A. Castro, Chem. Rev., 1999, 99, 3505.
26 M. J. Gresser and W. P. Jencks, J. Am. Chem. Soc., 1977, 99, 6963.
27 S. Carloni, D. E. Devos, P. A. Jacobs, R. Maggi, G. Sartori and S.
Raffaella, J. Catal., 2002, 205, 199.
490 | Green Chem., 2010, 12, 483–490
This journal is
The Royal Society of Chemistry 2010
©