The Guanidine-Promoted Direct Synthesis of Open-Chained Carbonates

Article abstract of DOI:10.1071/CH18623

In order to reduce CO₂ accumulation in the atmosphere, chemical fixation methodologies were developed and proved to be promising. In general, CO₂ was turned into cyclic carbonates by cycloaddition with epoxides. However, the cyclic carbonates need to be converted into open-chained carbonates by transesterification for industrial usage, which results in wasted energy and materials. Herein, we report a process catalyzed by tetramethylguanidine (TMG) to afford linear carbonates directly. This process is greener and shows potential for industrial applications.

Full text of DOI:10.1071/CH18623

CSIRO PUBLISHING
Aust. J. Chem. 2019, 72, 933–938
Full Paper
https://doi.org/10.1071/CH18623
The Guanidine-Promoted Direct Synthesis of
Open-Chained Carbonates
A
A
A B C
, ,
A B C
, ,
Yuhan Shang, Mai Zheng, Haibo Zhang,
and Xiaohai Zhou
A
College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072,
Hubei, China.
B
Engineering Research Center of Organosilicon Compounds and Materials,
Ministry of Education, Wuhan University, Wuhan 430072, Hubei, China.
C
Corresponding authors. Email: haibozhang1980@gmail.com; zxh7954@whu.edu.cn
In order to reduce CO2 accumulation in the atmosphere, chemical fixation methodologies were developed and proved to be
promising. In general, CO2 was turned into cyclic carbonates by cycloaddition with epoxides. However, the cyclic
carbonates need to be converted into open-chained carbonates by transesterification for industrial usage, which results in
wasted energy and materials. Herein, we report a process catalyzed by tetramethylguanidine (TMG) to afford linear
carbonates directly. This process is greener and shows potential for industrial applications.
Manuscript received: 19 December 2018.
Manuscript accepted: 18 August 2019.
Published online: 17 September 2019.
Introduction
similar reaction catalyzed by the complex of 1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU) and CO2,[7] and further experiments
were carried out to determine the crystal structure and chemical
properties of the reaction species.[8] In 2005, Heldebrant and
coworkers thoroughly investigated the reaction of DBU with
CO2.[9] A hemicarbonate salt intermediate was reported in their
work, thus refreshing the understanding of the interaction
between CO2 and organic superbases.
Guanidine and its derivatives have been widely used in CO2
activation.[10] The resultant guanidine-CO2 additives have been
applied in the preparation of cyclic carbamates,[11] the carbox-
ylation reaction with CO2,[12] and the direct or indirect synthesis
of cyclic carbonates.[13] Besides, guanidine derivatives showed
potential for CO2 capture and sequestration.[14] For example,
several fixation systems based on diols were reported recently,
and the advantages of reversible capture[15] and availability for
further reactions[16] are shown in the related works.
Despite many efforts, there are still some disadvantages in the
linear-carbonate direct synthesis system, such as the usage of
volatile organic solvents or metal salts, high CO2 pressure,
toxicity and expensiveness of catalysts, and so on. Herein, we
report a green and convenient synthesis of dialkyl carbonate
promotedbytetramethylguanidine(TMG).Thissynthesisprocess
can be carried out under atmospheric CO2 pressure and mild
temperature with no solvent occupation, and 99 % conversion is
achieved with .99 % carbonate selectivity. Besides, the reaction
mechanism was established based on NMR spectroscopic experi-
ments and catalytic screening. This reaction meets the principles
of green chemistry and shows potential for industrial applications.
Although the main reasons are still controversial, global warming
has aroused social concern in recent years. Typically, the green-
house effect caused by the accumulation of greenhouse gases
(GHGs) in the atmosphere is considered to be a possible factor.
The Intergovernmental Panel on Climate Change (IPCC) assess-
ment report[1] has defined global warming potential (GWP) to
measure the ability of gases to trap heat. In this definition, carbon
dioxide is chosen as the standard due to its significant contribution
to the warming process. To reduce the CO2 concentration in the
atmosphere, different methodologies are being developed, among
which chemical fixation has been proven to be promising. In
general, CO2 can be used as a C1 feedstock to produce value-
added chemicals. By the bond formation reaction with different
substrates and proper catalysts, CO2 can overcome its inertness
and be converted into formic acid, formaldehyde, methanol, or
methane by reaction with H2, into organic carbonate monomers,
oligomers, or polymers by reaction with O-containing chemicals,
and into carboxylic acids by reaction with metal alkyls.[2]
The cycloaddition with epoxide is a mature process for CO2
fixation, and many related studies have been reported.[3] The
product cyclic carbonate is an important organic building block
in the polycarbonate industry. However, in the industrial pro-
duction sequence of polycarbonates, cyclic carbonates are
always used after transesterification with methanol or phenol,
which gives open-chained carbonates as the desired substrate
and diols as a by-product.[4] This additional reaction step lowers
the atomic efficiency and overall yield of the whole process, thus
causing wasted materials and energy.[5] In this case, the direct
synthesis of dialkyl carbonate is a very attractive methodology
because it meets the requirements of green chemistry.
Results and Discussion
Since being reported in 1993 by McGhee and coworkers,
organic superbases have been used as catalysts to obtain organic
carbamate or carbonate.[6] Later, E. R. Pe´rez et al. reported a
Taking the ease of separation and analysis into account, the
reaction between 1-bromobutane (nBuBr) and 1-butanol
(nBuOH) was chosen as the model reaction. Different bases
Journal compilation Ó CSIRO 2019
www.publish.csiro.au/journals/ajc

934
Y. Shang et al.
were placed into the reaction system to provide a basic environ-
ment for CO2 fixation, and the activity screenings of the product
dibutyl carbonate (DBC) are listed in Table 1 (see also Table S2,
Supplementary Material). As expected, the substrates did not
react with each other in the absence of bases (Table 1, entry 1).
When alkali hydroxides were used, moderate bromide conver-
sions were given, but the carbonate selectivities were low
(Table 1, entry 2). To improve the selectivity, different cocata-
lysts were chosen to cooperate with potassium hydroxide. The
usage of 18-crown-6 or aprotic ionic liquids (1-butyl-3-methyli-
midazolium bromide (BmimBr)) led to significant growth in
bromide conversion, but the selectivity still remained low
(Table 1, entries 3 and 4). Interestingly, when protic ionic liquid
(PIL) 1,1,3,3-tetramethylguanidinium hydrobromide (TMGHBr)
was used, the conversion went down, but the selectivity greatly
increased (Table 1, entry 5). With an increase in the PIL from
1 mmol to 5 mmol, the selectivity finally reached 94 % in 6 h
(Table 1, entry 6). As a comparison, the PIL alone showed no
activity in this reaction (Table 1, entry 7).
tested under the same conditions. It was found that weak
organic bases led to poor conversion and selectivity
(Table S3, entries 5–11, Supplementary Material). Strong
bases, such as guanidine derivatives (Table 1, Entries 9 and
10) or DBU (Table S3, entry 12, Supplementary Material),
performed much better, and the substitution on the imine N
atom showed little impact on DBC yield. Besides, potassium
bromide (KBr) and water were chosen as additives to test their
effects on the reaction system. The addition of KBr had no
effect on the conversion and selectivity (Table 1, entry 12), but
the conversion decreased significantly when water was intro-
duced into the system (Table 1, entries 11 and 13). These results
indicate that water in the reaction system would compete with
alcohol, thus lowering the overall carbonate yield.
After that, the effect of TMG concentration was tested. In
the concentrated solution (alcohol/bromide/TMG ¼ 1.2 : 1 : 1,
Table 2, entry 1), the conversion was a slightly lower due to the
precipitation of ionic intermediates. When the solution was
diluted to alcohol/bromide/TMG ¼ 4 : 1 : 1 (Table 2, entry 2),
the conversion became even lower, which was caused by the
decrease in intermediate concentration and effective collision
frequency. The effect of reaction temperature was also tested
As a result, TMG was chosen as the base, and 34 % conver-
sion with 99 % selectivity was obtained (Table 1, entry 8).
Furthermore, the activities of various organic bases were also
Table 1. Activity screening of bases and co-componentsA
O
base
O
R1OH + R2X
+
CO2
+
R1
R2
50°C, no solvent
R1O
OR2
R1, R2 = n-butyl, X = Br
Entry
Base [mmol]
Co-component [mmol]
Conversion [%]B
Selectivity [%]B
1
–
–
trace
29
0
0
2
KOH [5]
KOH [5]
KOH [5]
KOH [5]
KOH [5]
TMGHBr [5]
TMG [5]
C2TMGC [5]
C4TMGD [5]
TMG [5]
TMG [5]
TMG [5]
–
3
18-Crown-6 [1]
72
0
4
BmimBr [1]
61
0
5
TMGHBr [1]
8
62
94
0
6
TMGHBr [5]
10
7
–
trace
34
8
–
99
99
99E
99
99
98
9
–
33
34E
10
11
12
13
–
H2O [5]
11
KBr [5]
36
KBr [5] þ H2O [5]
12
AReaction conditions: 1-butanol/1-bromobutane ¼ 2 : 1, 508C, 6 h, CO2 (99.99 %, bubbling, 5 mL minꢀ1).
BDetermined by GC yields.
CC2TMG ¼ 2-ethyl-1,1,3,3-tetramethylguanidine.
DC4TMG ¼ 2-butyl-1,1,3,3-tetramethylguanidine.
EDetermined by isolated yield.
Table 2. Effects of concentration and temperatureA
Entry
n(BuOH)/n(BuBr)
Temperature [8C]
Conversion [%]B
Selectivity [%]B
1
2
3
4
5
6
1.2
4
50
50
25
40
60
80
31
23
15
28
38
20
99
99
99
99
99
88
2
2
2
2
AReaction conditions: 1-bromobutane [5 mmol], TMG [5 mmol], 6 h, CO2 (99.99 %, bubbling, 5 mL minꢀ1).
BCalculated based on GC yields.

Synthesis of Open-Chained Carbonates
935
(Table 2, entries 3–5). As shown, the conversion increased with
the reaction temperature rising from 258C to 608C, while its
growth rate fell continuously. When the temperature reached
808C, a decrease in carbonate yield (20 %) occurred, and the yield
of the side-product dibutyl ether (DBE) increased at the same
time (Table 2, entry 6). According to E. R. Pe´rez’s work, the
complex formed with TMG and CO2 would decompose at
748C.[17] So in this system, the CO2-contained intermediate could
be unstable when the temperature was above 708C, and low
carbonate selectivity was given as a result.
Therefore, the reaction temperature of 508C and the alco-
hol/bromide/TMG ratio of 2 : 1 : 1 were chosen as optimized
conditions. Under these conditions, a time-dependent experi-
ment was carried out. As shown in Fig. 1, the conversion of
substrate reached 99 % at 24 h, and the carbonate selectivity
increased to more than 98 % during the entire reaction period.
Interestingly, the curve plotted with conversion against time
appears to be S-shaped, and an increase in carbonate selectiv-
ity from 0 to 6 h is observed. These phenomena indicate that
the reaction between alcohol, alkyl bromide, TMG, and CO2
had an initial period during which CO2 was captured by the
proton-transfer product of TMG and alcohol, and resulted in
the domination of carbonate product over time. As a result, this
reaction should be a cascade reaction composed of a series of
reaction steps.
Further experiments were carried out to verify these spec-
ulations. First, the 1H NMR spectra of the reaction species and
TMG are shown in Fig. 2. When the alcohol was mixed with
TMG, its hydroxyl proton signal (Hb, 1.88 ppm, s, 1H) moved to
5.14 ppm and broadened. Also, the methylene Ha signal moved
from 3.64 ppm (t, 2H) to 3.56 ppm. These phenomena indicate
that nBuOH was deprotonated and the proton had transferred to
the TMGHþ cation. When CO2 was bubbled into the system, a
0
new Ha peak appeared at 3.86 ppm, and the Hb peak further
moved to 8.45 ppm, which revealed the formation of the hemi-
carbonate species.
The comparison of the 13C NMR spectra of the reaction
species and DBC is shown in Fig. 3. The addition of 1-butanol
had little influence on the signal of the central carbon (Ca)
atom of TMG. However, when CO2 was added, the Ca atom
peak of TMG moved to 162.38 ppm, and a new C signal
appeared at 157.60 ppm. Compared with the Cb signal of
DBC (155.49 ppm), the new peak could be attributed to the
hemicarbonate intermediate.
After NMR spectroscopic analysis, a series of substrate
control experiments were carried out to test the reactivity of
different intermediate species. The composition of the experi-
mental groups is shown in Scheme 1. The two experimental
100
80
n
groups mixing BuOH with TMG in the initial step obtained
32 % and 29 % carbonate yields. Whereas in the nBuBr þ TMG
group, the initial step led to n-butyl tetramethylguanidinium
bromide, which showed no activity in the subsequent reaction
step. This phenomenon indicates that the reactive intermediates
should be tetramethylguanidinium alkoxide and its correspond-
ing hemicarbonate.
60
40
Conv %
20
Sel %
0
According to these results, a probable reaction mechanism
was established and is shown in Scheme 2. At the beginning, the
proton transfer occurred between alcohol R1OH and TMG, and
[TMGH]þ[OR1]ꢀ (Int1) was given. Then, the C atom of CO2
was attacked by the alkoxy anion to form a carbonate anion
0
5
10
15
20
25
Time [h]
Fig. 1. The time-dependent plots of dibutyl carbonate conversion (£) and
selectivity (¢).
HN
HN
N
N
a
a
HaЈ
+
CO2
+
+
OHb
OHb
Ha
N
N
Hb
Ha
Hb
a
Ha
Hb
OHb
HN
N
N
9.0
8.5
8.0 5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
Chemical shift [ppm]
Fig. 2. The 1H NMR spectra of the reaction species.

936
Y. Shang et al.
O
b
Cb
BuO
HN
OBu
Ca
CbЈ
N
CO2
+
a
+
+
OH
OH
N
HN
HN
N
N
a
Ca
N
a
Ca
N
180
175
170
165
160
155
150
145
Chemical shift [ppm]
Fig. 3. The 13C NMR spectra of the reaction species.
O
+ CO2
N
NH
Br
Br
OH
OH
Br
BuO
OBu
N
32 %, 98 %
O
+ CO2
BuO
OBu
29 %, 99 %
+ CO2
OH
No reaction
Scheme 1. The reactant control experiments of the different reaction species. Reaction conditions: 508C, CO2 bubbling (5 mL minꢀ1), 6 h per step.
OH
NH
N
H2O
R1
N
KOH
+
O
X
R1
R2
R2
+
NH2
N
N
NH2
O–
[X–]
R1
N
N
Int1
R1
O
O
+
NH2
O–
CO2
O
N
R
1
O
R2
O
N
X
R2
Int2
Scheme 2. A possible reaction mechanism.
intermediate [TMGH]þ[OCOOR1]ꢀ (Int2) with the assistance
of the protonated imine group. Finally, a substitution reaction
took place between the carbonate anion and an alkyl halide R2X,
and the target material (the open-chained carbonate
R1OCOOR2) was generated with the protonated base TMGHX.
When the substitution reaction took place before the CO2

Synthesis of Open-Chained Carbonates
937
Table 3. Open-chained carbonate synthesis from various substratesA
O
TMG
O
R1OH + R2Br
+
CO2
+
50°C, no solvent
R1
R2
R1O
OR2
Entry
R1
R2
Conversion [%]B
Selectivity [%]B
1
butyl
ethyl
92
96
70
46
45
0
99
99
99
99
99
0
2
butyl
propyl
hexyl
dodecyl
benzyl
phenyl
butyl
3
butyl
4
butyl
5
butyl
6
butyl
7
ethyl
90
91
96
46
97
89
73
77
90
99
80
94
77
72
8
propyl
hexyl
dodecyl
phenyl
benzyl
benzyl
butyl
9
butyl
10
11
12
13
butyl
butyl
butyl
tert-butyl
AReaction conditions: alcohol/organobromide/TMG ¼ 10 : 5 : 5 [mmol], 508C, 24 h, CO2 (99.99 %, bubbling, 5 mL minꢀ1).
BCalculated based on isolated yield.
fixation, the by-product (the dialkyl ether R1OR2) was yielded
as a result. Besides, TMGHX could be recycled by metathesis
with NaOH or KOH in alcohol solution. After filtering out the
metal halide and distilling away the mixture of alcohol and H2O,
TMG could be recovered with the rate .85 %.
Conclusion
In summary, a methodology for synthesizing dialkyl carbonate
promoted by tetramethylguanidine was reported. In this system,
99 % carbonate yield was achieved with .99 % selectivity
under mild temperature (508C) and atmospheric CO2 pressure in
24 h. Further tests were carried out to identify the intermediates
and test the substrate applicability. Based on the experimental
results, a probable mechanism was established, in which CO2
was fixed with the assistance of the guanidinium cation, and the
open-chained carbonate was obtained by nucleophilic substi-
tution between the carbonate anion and alkyl bromide. This
reaction provides a pathway to obtaining open-chained carbo-
nates under mild conditions, and TMG is generally considered to
be cheap and green, and was proved to be recyclable.
Moreover, the activity of TMG towards various substrates
was tested, and the results are listed in Table 3. First, the
reactions between nBuOH and different organic bromides were
carried out. When the number of side chain C atoms was less
than 6, the yields increased slightly with the increase in the
number of C atoms (Table 3, entries 1 and 2). However, when the
number of side chain C atoms was greater than 6, the yields
decreased significantly because the large side chain led to low
reactivity (Table 3, entries 3 and 4). A poor yield of 45 % was
given when benzyl bromide was used (Table 3, entry 5), which
was caused by the hindrance of the benzyl group. Further, no
reaction occurred between bromobenzene and nBuOH due to the
stability of bromobenzene and the abundance of p-electrons on
the phenyl group (Table 3, entry 6).
Supplementary Material
Further experimental details are available on the Journal’s
website.
The reactivity of various alcohols was investigated as well.
The reactant conversion, which was affected by the elongation
of the alcohol alkyl chain, was quite similar to those of the alkyl
bromides (Table 3, entries 7–10). However, when phenol was
used as the reactant, a 97 % yield was achieved with 94 %
selectivity (Table 3, entry 11). Considering that the hydroxyl
group of phenol was easy to deprotonate, this result is reason-
able. Benzyl alcohol, with the hydroxyl acidity weaker than that
of phenol, led to a lesser yield of 89 % (Table 3, entry 12).
The reaction between tert-butyl bromide and benzyl alcohol
(Table 3, entry13)wasalsocarriedoutto comparewiththereaction
of 1-bromobutane. The conversion went down to 73 % due to the
hindrance of the tert-butyl group, which indicated that the SN2
substitution might be dominant during the reaction process.
Interestingly, unlike organic bromides, alcohols with long or
hindered side chains showed lower selectivity of 77–94 %.
Considering that the by-products were derived from the nucleo-
philic attack between alkoxy anion species and organic bro-
mides, the larger hydrocarbon chains were detrimental to the
CO2 fixation process.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgements
This work was financially supported by the Fundamental Research Funds for
the Central Universities (No. 2042016HF1054) and Wuhan University
Experiment Technology Project Funding (No. WHU-2016-SYJS-06). The
authors appreciate the technical support for NMR spectroscopic analysis
from Zuofei Wang and Chunjiang Wang of the College of Chemistry and
Molecular Science, Wuhan University.
References
[1] IPCC, Climate Change 2014: Synthesis Report. Contribution of
Working Groups I, II and III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change 2014 [Core Writing
Team, R. K. Pachauri and L. A. Meyer (eds)], 151 pp. (IPCC: Geneva,
Switzerland).
[2] T. Fujihara, Y. Tsuji, J. Jpn. Petrol. Inst. 2016, 59, 84. doi:10.1627/JPI.
59.84

938
Y. Shang et al.
[3] (a) M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann, F. E. Ku¨hn,
[8] E. R. Pe´rez, R. H. A. Santos, M. T. P. Gambardella, L. G. M. de
Macedo, U. P. Rodrigues-Filho, J.-C. Launay, D. W. Franco, J. Org.
Chem. 2004, 69, 8005. doi:10.1021/JO049243Q
[9] D. J. Heldebrant, P. G. Jessop, C. A. Thomas, C. A. Eckert, C. L. Liotta,
J. Org. Chem. 2005, 70, 5335. doi:10.1021/JO0503759
[10] Q. Song, Z. Zhou, L. He, Green Chem. 2017, 19, 3707. doi:10.1039/
C7GC00199A
[11] T. Niemi, T. Repo, Eur. J. Org. Chem. 2019, 1180. doi:10.1002/EJOC.
201801598
[12] A. Tortajada, F. Julia´-Herna´ndez, M. Bo¨rjesson, T. Moragas, R.
Martin, Angew. Chem. Int. Ed. 2018, 57, 15948. doi:10.1002/ANIE.
201803186
[13] S. Dabral, T. Schaub, Adv. Synth. Catal. 2019, 361, 223. doi:10.1002/
ADSC.201801215
[14] S. Zhang, L. He, Aust. J. Chem. 2014, 67, 980. doi:10.1071/CH14125
[15] R. Steinhardt, S. C. Hiew, H. Mohapatra, Z. O. Du Nguyen, R. Truong,
A. Esser-Kahn, ACS Cent. Sci. 2017, 3, 1271. doi:10.1021/ACS
CENTSCI.7B00418
Angew. Chem. Int. Ed. 2011, 50, 8510. doi:10.1002/ANIE.201102010
(b) Q. He, J. W. O’Brien, K. A. Kitselman, L. E. Tompkins, G. C. T.
Curtis, F. M. Kerton, Catal. Sci. Technol. 2014, 4, 1513. doi:10.1039/
C3CY00998J
(c) D.-H. Lan, N. Fan, Y. Wang, X. Gao, P. Zhang, L. Chen, C.-T. Au,
S.-F. Yin, Chin. J. Catal. 2016, 37, 826. doi:10.1016/S1872-
2067(15)61085-3
[4] (a) S. Fukuoka, M. Tojo, H. Hachiya, M. Aminaka, K. Hasegawa,
Polym. J. 2007, 39, 91. doi:10.1295/POLYMJ.PJ2006140
(b) D. Franco, C. Gu¨rtler, T. Mu¨ller, O. Pieter, P. Angelina, J. Rechner,
F. Risse, A Method for Preparing Diaryl Carbonates from Dialkyl
Carbonates. Patent No. DE102010042937A1 2011
[5] J. Artz, T. E. Mu¨ller, K. Thenert, J. Kleinekorte, R. Meys, A. Sternberg,
A. Bardow, W. Leitner, Chem. Rev. 2018, 118, 434. doi:10.1021/ACS.
CHEMREV.7B00435
[6] (a) W. McGhee, D. Riley, M. Christ, K. Christ, Organometallics 1993,
12, 1429. doi:10.1021/OM00028A069
(b) W. McGhee, D. Riley, K. Christ, Y. Pan, B. Parnas, J. Org. Chem.
1995, 60, 2820. doi:10.1021/JO00114A035
[16] T. Kitamura, Y. Inoue, T. Maeda, J. Oyamada, Synth. Commun. 2016,
46, 39. doi:10.1080/00397911.2015.1116583
[7] E. R. Pe´rez, M. Odnicki da Silva, V. C. Costa, U. P. Rodrigues-Filho,
D. W. Franco, Tetrahedron Lett. 2002, 43, 4091. doi:10.1016/S0040-
4039(02)00697-4
[17] F. S. Pereira, E. R. deAzevedo, E. F. da Silva, T. J. Bonagamba, D. L.
da Silva Agost´ıni, A. Magalha˜es, A. E. Job, E. R. Pe´rez Gonza´lez,
Tetrahedron 2008, 64, 10097. doi:10.1016/J.TET.2008.08.008