Mechanism of formation of organic carbonates from aliphatic alcohols and carbon dioxide under mild conditions promoted by carbodiimides. DFT calculation and experimental study

Article abstract of DOI:10.1021/jo050392y

Dicyclohexylcarbodiimide (CyN=C=NCy, DCC) promotes the facile formation of organic carbonates from aliphatic alcohols and carbon dioxide at temperatures as low as 310 K and moderate pressure of CO₂ (from 0.1 MPa) with an acceptable rate. The conversion yield of DCC is quantitative, and the reaction has a very high selectivity toward carbonates at 330 K; increasing the temperature increases the conversion rate, but lowers the selectivity. A detailed study has allowed us to isolate or identify the intermediates formed in the reaction of an alcohol with DCC in the presence or absence of carbon dioxide. The first step is the addition of alcohol to the cumulene (a known reaction) with formation of an O-alkyl isourea [RHNC(ORO=NR] that may interact with a second alcohol molecule via H-bond (a reaction never described thus far). Such an adduct can be detected by NMR. In alcohol, in absence of CO ₂, it converts into a carbamate and a secondary amine, while in the presence of CO₂, the dialkyl carbonate, (RO)₂CO, is formed together with urea [CyHN-CO-NHCy]. The reaction has been tested with various aliphatic alcohols such as methanol, ethanol, and allyl alcohol. It results in being a convenient route to the synthesis of diallyl carbonate, in particular. O-Methyl-N,N′-dicyclohexyl isourea also reacts with phenol in the presence of CO₂ to directly afford for the very first time a mixed aliphatic-aromatic carbonate, (MeO)(PhO)CO. A DFT study has allowed us to estimate the energy of each intermediate and the relevant kinetic barriers in the described reactions, providing reasonable mechanistic details. Calculated data match very well the experimental results. The driving force of the reaction is the conversion of carbodiimide into the relevant urea, which is some 35 kcal/mol downhill with respect to the parent compound. The best operative conditions have been defined for achieving a quantitative yield of carbonate from carbodiimide. The role of temperature, pressure, and catalysts (Lewis acids and bases) has been established. As the urea can be reconverted into DCC, the reaction described in this article may further be developed for application to the synthesis of organic carbonates under selective and mild conditions.

Full text of DOI:10.1021/jo050392y

Mechanism of Formation of Organic Carbonates from Aliphatic
Alcohols and Carbon Dioxide under Mild Conditions Promoted by
Carbodiimides. DFT Calculation and Experimental Study
Michele Aresta,*,† Angela Dibenedetto,† Elisabetta Fracchiolla,† Potenzo Giannoccaro,†
Carlo Pastore,† Imre Pa´pai,‡ and Ga´bor Schubert‡
Department of Chemistry, University of Bari, and CIRCC, via Celso Ulpiani 27, 70126 Bari, Italy, and
Chemical Research Center, Hungarian Academy of Sciences, H-1525 Budapest, P.O.B. 17, Hungary
aresta@metea.uniba.it
Received March 2, 2005
Dicyclohexylcarbodiimide (CyNdCdNCy, DCC) promotes the facile formation of organic carbonates
from aliphatic alcohols and carbon dioxide at temperatures as low as 310 K and moderate pressure
of CO2 (from 0.1 MPa) with an acceptable rate. The conversion yield of DCC is quantitative, and
the reaction has a very high selectivity toward carbonates at 330 K; increasing the temperature
increases the conversion rate, but lowers the selectivity. A detailed study has allowed us to isolate
or identify the intermediates formed in the reaction of an alcohol with DCC in the presence or
absence of carbon dioxide. The first step is the addition of alcohol to the cumulene (a known reaction)
with formation of an O-alkyl isourea [RHNC(OR′)dNR] that may interact with a second alcohol
molecule via H-bond (a reaction never described thus far). Such an adduct can be detected by NMR.
In alcohol, in absence of CO2, it converts into a carbamate and a secondary amine, while in the
presence of CO2, the dialkyl carbonate, (RO)2CO, is formed together with urea [CyHN-CO-NHCy].
The reaction has been tested with various aliphatic alcohols such as methanol, ethanol, and allyl
alcohol. It results in being a convenient route to the synthesis of diallyl carbonate, in particular.
O-Methyl-N,N′-dicyclohexyl isourea also reacts with phenol in the presence of CO2 to directly afford
for the very first time a mixed aliphatic-aromatic carbonate, (MeO)(PhO)CO. A DFT study has
allowed us to estimate the energy of each intermediate and the relevant kinetic barriers in the
described reactions, providing reasonable mechanistic details. Calculated data match very well
the experimental results. The driving force of the reaction is the conversion of carbodiimide into
the relevant urea, which is some 35 kcal/mol downhill with respect to the parent compound. The
best operative conditions have been defined for achieving a quantitative yield of carbonate from
carbodiimide. The role of temperature, pressure, and catalysts (Lewis acids and bases) has been
established. As the urea can be reconverted into DCC, the reaction described in this article may
further be developed for application to the synthesis of organic carbonates under selective and
mild conditions.
Introduction
Molecular organic carbonates (Figure 1) are expected
to experience a considerable expansion of their market
* To whom correspondence should be addressed. Phone: +39-080-
544-2084. Fax: +39-080-544-2429.
† University of Bari and CIRCC.
FIGURE 1. Linear (a) and cyclic (b) molecular organic
carbonates.
‡ Hungarian Academy of Sciences.
10.1021/jo050392y CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/14/2005
J. Org. Chem. 2005, 70, 6177-6186
6177

Aresta et al.
in the coming years, due to their growing use in the
chemical (as solvent and reagent1), pharmaceutical (as
intermediates2), and polymer industries.3 Also, the use
of dimethyl carbonate (DMC) as an additive to gasoline4
will increase its market by nearly 2 orders of magnitude
to an expected 30 Mt per year.5 To satisfy the market
demand, new synthetic technologies need to be developed
that are not based on phosgene,6 a chemical banned in
several countries. The search for newer, safer, environ-
mentally friendly synthetic methodologies has attracted
the attention of a large scientific community. Either CO7,8
or CO2 are used as substitutes of phosgene in the
scientific and patent literature for the synthesis of linear
carbonates, cyclic monomeric carbonates, or polycarbon-
ates.9,10
The direct carboxylation of alcohols (eq 1) responds to
the requisites of sustainable chemistry. Such a reaction
shows acceptable thermodynamics when R is an aliphatic
group, but has such a high positive free energy when R
is phenyl or, in general, an aromatic group, that thus far
the synthesis of aromatic carbonates according to eq 1
was prevented.
by cooling to ambient temperature the reaction mixture
that, after dewatering, can be heated again to the
reaction temperature. This procedure is very energy-
intensive and does not meet the economic and environ-
mental targets. Alternatively, there have been attempts
to run separately the reaction that causes the water
formation from the process of synthesis of the carbonate.
To this end, ortomethylesters, acetals,14 and ketals15 have
been reacted with scCO2 to afford DMC. A single report
describes the use of carbodiimide as “water trap” in the
carboxylation of methanol in scCO2 (323 K, 12 MPa),16
but the reaction mechanism has not been investigated.
In this article, we discuss the reaction of dicyclohexyl-
carbodiimide, CyNdCdNCy, 2 (DCC), with several al-
cohols in the absence or presence of carbon dioxide, the
formation of carbonates from aliphatic alcohols and CO2
mediated by DCC and the role of the latter. The mech-
anism of formation of dialkyl carbonates is also discussed,
highlighting the influence of temperature and pressure.
The details of the reaction mechanism are revealed from
the results of density functional calculations. Several
intermediates have been isolated or identified using
spectroscopic techniques. The effect of working in the
presence of Lewis acids such as transition metal systems
is also presented.
2 ROH + CO2
f
(1)
(RO)2CO + H2O
R ) Me, 1a; Et, 1b; Allyl, 1c
1
A few reports in the recent literature show that the
equilibrium concentration of conversion of methanol into
DMC (eq 1) is as low as 2%.11 This is also true for ethanol
and allyl alcohol.12 A limiting factor, beyond the thermo-
dynamics, is represented by water formed in the reaction
that may push the equilibrium to the left or destroy the
catalyst, when the latter is an alkoxo metal derivative.13
To prevent such a negative effect, water traps are needed,
either chemical- or process-based. Inorganic materials
such as zeolites can be used as water traps, but the
(surface)-hydroxo groups formed upon reaction with
water are acidic enough to protonate the carbonate at
the reaction temperature and convert it back into metha-
nol and CO2. It is possible to prevent such back reaction
Discussion
Carbodiimides 2 have been known for some time to
react with methanol (eq 2, R ) Me) to afford O-methyl
isoureas 3 as addition products. The chemistry of the
latter compounds has been only moderately investigated.
3 has been reported to behave as an alkylating agent
toward acids and alcohols,17 but no detailed studies are
reported that describe its behavior varying the reaction
conditions.
(1) Delledonne, D.; Rivetti, F.; Romano, U. Appl. Catal., A 2001, 221,
241-251.
(2) (a) Shaikh, A. A. G.; Sivaram, S. Chem. Rev. 1996, 96, 951-976
and references therein. (b) Stinson, S. C. Chem. Eng. News 2001, 79,
15-16.
We have carried out an extended study on the behavior
of DCC toward various alcohols and under different
conditions and shown the reaction path and the role of
O-alkyl isoureas in the synthesis of organic carbonates
from alcohols and CO2 under very mild conditions. Such
reaction turns out to be a paradigm case for the formation
of carbonates from alcohols and CO2.
Reaction of DCC with Alcohols. DCC reacts with
aliphatic alcohols to produce O-alkyl isoureas according
to eq 2. The latter is a quite effective agent for the
conversion of organic acids into the relevant alkylesters17
at room temperature and of phenols into their ethers at
(3) Volker, Serini. In Ullmann’s Encyclopedia of Industrial Chem-
istry; VCH Publishers: Weinheim, Germany, 1992; pp 197-201.
(4) Pacheco, M. A.; Marshall, C. L. Energy Fuels 1997, 11, 2-29.
(5) (a) Aresta, M.; Dibenedetto, A. In Carbon Dioxide: Recovery and
Utilization; Aresta, M., Ed.; Kluwer Academic Publishers: Dordrecht,
Boston, London, 2003; pp 211-260. (b) TEXACO study, 2001.
(6) (a) Fr. Patent, 7pp FR 2163884 19730831 SNPE, 1973. (b) Damle,
S. B. Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed.;
Wiley: New York, 1993; p 77.
(7) (a) Romano, U.; Tesei, R.; Massi, M. M.; Rebora, P. Ind. Eng.
Chem. Prod. Res. Dev. 1980, 19, 396-403. (b) Romano, U. Chim. Ind.
1993, 75, 303-306.
(8) Nishihira, K.; Tanaka, S.; Kodama, K.; Kaneko, T. Eur. Pat. Appl.
EP. 501,507, 1992.
(9) (a) Aresta, M.; Dibenedetto, A.; Gianfrate, L.; Pastore, C. J. Mol.
Catal. A: Chem. 2003, 204-205, 245-252 and references therein. (b)
Aresta, M.; Dibenedetto, A.; Gianfrate, L.; Pastore, C. Appl. Catal., A
2003, 255, 5-11.
(14) Sakakura, T.; Saito, Y.; Okano, M.; Choi, J.-C.; Sako, T. J. Org.
Chem. 1998, 63, 7095-7096.
(10) Aresta, M.; Dibenedetto, A. J. Mol. Catal. 2002, 182-183, 399-
409.
(15) (a) Sakakura, T.; Saito, Y.; Choi, J.-C.; Masuda, T.; Sako, T.;
Oriyama, T. J. Org. Chem. 1999, 64, 4506-4508. (b) Sakakura, T.;
Saito, Y.; Choi, J.-C.; Sako, T. Polyhedron 2000, 19, 573-576.
(16) Isaacs, N. S.; O’Sullivan, B.; Verhaelen, C. Tetrahedron 1999,
55, 11949-11956.
(11) Tomishige, K.; Sakaihori, T.; Ikeda, Y.; Fujimoto, K. Catal. Lett.
1999, 58, 225-229.
(12) Aresta, M.; Dibenedetto, A.; Pastore, C. Inorg. Chem. 2003, 42,
3256-3261 and references therein.
(13) Ballivet-Tkatchenko, D.; Douteau, O.; Stutzmann, S. Organo-
metallics 2000, 19, 4563-4567.
(17) Dabritz, E. Angew. Chem., Int. Ed. 1966, 5, 470-477 and
references therein.
6178 J. Org. Chem., Vol. 70, No. 16, 2005

Organic Carbonates from Aliphatic Alcohols and CO2
FIGURE 2. H-bonded adducts formed between isourea 3a and MeOH as obtained from DFT calculations. The predicted binding
energies (in kcal/mol) are given in parentheses.
SCHEME 1. Methanol Interaction with 3a
380 K, with quantitative yield. We have shown that such
alkylating properties can be used for the selective alkyl-
ation of the acid with respect to the alkoxo functionality
in the same molecule. In fact, 3a-c can promptly (>99%)
alkylate 4-OH-benzoic acid to 4-OH-benzoic alkylester at
ambient temperature with high selectivity.
easily undergo a further reaction with alcohols to afford
carbamates (4) and secondary amines (5) (eq 3).
Such alkylation reactions are relevant to the mecha-
nism of formation of DMC, or dialkyl carbonates in
general, from alcohols and CO2. Therefore, we have
investigated in detail the factors that influence the yield
and selectivity of eq 2. The rate of the process of
alkylation of 2 to afford 3 (eq 2) is dependent on the
nature of the alcohol used: at 330 K using a stoichio-
metric mixture of alcohol and DCC, the trend is CH3 >
C2H5 > CH2-CHdCH2 . Ph. Phenol reacts very little,
and the corresponding O-phenyl isourea could not be
isolated at 330 K, while other alkyl isoureas were isolated
in very good yield (R ) Me, >98%; R ) Et, >96%; R )
allyl, >95%). The reaction rate of the addition of alcohols
to the CdN bond in general increases with the temper-
ature and is influenced by the presence of a catalyst. We
have found that ZnCl2 and CuO or Cu2I2 are catalysts as
good as CuCl, already described.17 Zinc acetate is also
active, but it generates the acetate alkyl ester formed
upon reaction of alcohol with the acetate moiety linked
to Zn, which is converted into Zn(OH)2, isolated at the
end of the reaction as an insoluble white solid. We have
further investigated the reaction of O-alkyl isoureas with
the relevant alcohol, a reaction that was never described
before. We have found that while O-alkyl isoureas are
stable in aromatic aprotic solvents, chlorinated solvents,
and CH3CN, also at temperatures above 373 K, they quite
Density functional calculations show that this reaction
is strongly exothermic, because 4a + 5a is predicted to
lie 29.2 kcal/mol below the energy of 3a + MeOH. A se-
cond alcohol molecule may interact with isourea through
an H-bond to either the aminic or iminic nitrogen or to
the methoxy oxygen atom (see Figure 2), affording [CyH-
NC(OR)dNCy‚‚‚HOCH3], 6 (R ) Me, 6a; Et, 6b; Allyl, 6c).
Of the three possible adducts, 6aN-im represents the
most stable structure (the calculated binding energy is
6.8 kcal/mol), which is due to the enhanced basicity of
the iminic N atom in 3a. The other two H-bonds are
significantly weaker, the corresponding binding energies
being in the range 3-4 kcal/mol. Infrared spectroscopy
(IR) is not able to show such interaction. Conversely, a
1H NMR study on the equimolar mixture isourea alcohol
in CH3CN has clearly shown the formation of an adduct.
In fact, when methanol is added to an acetonitrile
solution of isourea, a net change of the 1H NMR spectrum
is observed. While the methoxo group of the isourea can
still be distinguished from the incoming HOCH3 (CH3
resonances of 3.55 and 3.29 ppm singlet, respectively),
the proton signals of the N-C-H moiety of the two
initially different cyclohexyl rings of 3a (multiplets
centered at 2.92 (1H) and 3.25 ppm (1H)) disappear,
while a single multiplet at 1.72 ppm appears. Also, the
signal due to the single N-H proton of 3a (found at 3.95
J. Org. Chem, Vol. 70, No. 16, 2005 6179

Aresta et al.
SCHEME 2. Exchange Mechanism of OMe to OR in Isoureas
ppm) disappears, and a new broad signal is found at 2.6
ppm. Similarly, the 13C spectrum confirms the equiva-
lence of the two cyclohexyl groups. The equivalence of
both N-H and N‚‚‚H-O protons can be explained with
the initial formation of adduct 6aNim from which the
resonance structures 7a,b can be generated (Scheme 1)
in which the N-H and N‚‚‚H-O hydrogens become
equivalent, while the methoxo groups remain different.
Heating 3 in an excess of alcohol allows eq 3 to occur.
To gain more insight into this reaction, we have reacted
CyHNC(OMe)dNCy with a heteroalcohol ROH (R ) Et
or Allyl) and found that CyHNC(OR)dNCy and methanol
are formed. This exchange can be explained on the basis
of the equilibria depicted in Scheme 2. The involvement
of more than one alcohol molecule (two, as shown in
Scheme 2) recalls the pathway proposed for the hydration
of carbodiimides.18 Scheme 2 shows that the conversion
of 3a into 3b(c) passes through the formation of inter-
mediate 8, which can be considered the monosolvated
form of the product of addition of a molecule of alcohol
to the CdN bond of isourea 3a, as shown in eq 4. Our
calculations predict that 8a lies only 4.8 kcal/mol above
3a + MeOH (11.6 kcal/mol less stable than the hydrogen-
bonded complex 6aN-im).
A study carried out using different alcohols (i.e.,
methanol in the synthesis of the alkyl isourea and
ethanol for the following reaction) has shown that both
O-methyl and O-ethyl carbamate and the chiral amines
NHMeCy and NHEtCy are formed and can be isolated
very pure. The same is true if allyl OH is used instead of
EtOH. This trend is consistent with the alkyl-exchange
reaction (Scheme 2). Increasing the temperature favors
the formation of the carbamate that is not formed below
332 K (Figure 3). This is of relevance to the presence of
4 and 5 in the reaction mixture of carboxylation of
alcohols in the presence of DCC at temperature above
340 K. The concentration of the reacting alcohol plays a
key role. In fact, if isourea (50 mM) and alcohol are
reacted in a solvent such as toluene or CH3CN, no
reaction takes place if the concentration of the alcohol is
lower than 500 mM, independent of the solvent used.
Reaction of O-Alkyl Isoureas with Alcohols in the
Presence of CO2. O-Alkyl isourea 3 does not show any
tendency to react directly with CO2 either in organic
solvents at 330 K under 6.0 MPa of CO2 or under
supercritical conditions (350 K and 12 MPa).
Equation 5 is shifted to the left under the above
conditions. In agreement with such an observation, DFT
calculations show that the insertion of CO2 into the C-O
bond of O-methyl isourea to give 9a (R ) Me) is an
endothermic process (9a is predicted to lie 17.7 kcal/mol
above 3a + CO2). Clearly 9 is not an intermediate in the
formation of carbonates, which are easily formed from
alcohols and CO2 in the presence of DCC.
A low energy transition state for the direct addition of
a MeOH molecule to 3a to give 8a was not found, which
might be an indication that eq 3 does not involve a single
alcohol molecule, but rather a network of H-bonded
alcohols as reported in Scheme 2. The alkyl group
exchange is, thus, consistent with the intermediacy of the
metastable ureal 8, which was not isolated although
several reaction conditions were tried.
FIGURE 4. Influence of the solvent on the carboxylation rate
of methanol in the presence of 3a: the reaction is faster in
polar solvents. The molar ratio solvent/CH3OH/3a is 50:15:1,
while PCO2 ) 0.1 MPa and T ) 323 K.
FIGURE 3. Influence of the temperature on the yield of
carbamate. Molar ratio MeOH/DCC ) 15:1, PCO2 ) 5.0 MPa.
The carboxylation of alcohols shows an induction time
during the first 2 h (Figure 4), and then the rate
(18) Tordini, F.; Bencini, A.; Bruschi, M.; De Gioia, L.; Zampilla,
G.; Fantucci, P. J. Phys. Chem. A 2003, 107, 1188-1196.
6180 J. Org. Chem., Vol. 70, No. 16, 2005

Organic Carbonates from Aliphatic Alcohols and CO2
TABLE 2. Influence of the Reaction Conditions on the
Yield of Formation of Carbonates from Alcohols or
Isoureas
carbonate yield (%)
entry
reagent
catalyst
6 h
12 h
24 h
In Methanol
no
1
2
3
4
5
6
7
DCCa
DCCa
11.0
30.2
32.6
CuCl
53.0
52.0
51.8
46.8
36.4
83.3
76.4
80.3
76.0
methyl isoureaa no
methyl isoureaa MeOC(O)ONa 33.4
methyl isoureaa CuCl
29.4
22.8
62.2
FIGURE 5. Effect of temperature and pressure on the yield
of DMC formed by reacting methanol and CO2 in the presence
of DCC (methanol as solvent). The temperature effect is
studied at a pressure of carbon dioxide of 5.0 MPa, while the
influence of pressure is investigated at 323 K.
DCCb
DCCc
no
no
In Ethanol
no
CuCl
no
CuCl
8
9
10
11
DCCd
1.0
14.0
15.3
14.1
DCCd
37.1
37.5
37.8
46.1
45.8
44.8
ethyl isouread
ethyl isouread
TABLE 1. Influence of the Alkyl Group on the Yield of
Formation of Dialkyl Carbonates from the Relevant
Alcohols and CO2 in Presence of DCCa
a 338 K, 5.0 MPa of CO2, r ) 100 (alcohol/DCC molar ratio).
b 338 K, 5.0 MPa of CO2, r ) 5 (alcohol/DCC molar ratio). c 353 K,
5.0 MPa of CO2, r ) 5 (alcohol/DCC molar ratio). d 338 K, 5.0 MPa
of CO2, r ) 70 (alcohol/DCC molar ratio).
yield (selectivity) %
starting from DCC
yield (selectivity) %
from the relevant isourea
alcohol
MeOH
EtOH
allyl OH
11 (84)
1 (93)
14.8 (94)
32.6 (92)
15.3 (96)
90.1 (95)
CO2 in CH3CN (eq 6).
a Reaction condition: 323 K, 5.0 MPa, 6 h.
An increase in the CO2 pressure above 1 MPa does not
produce an increase of the conversion yield. As a matter
of fact, the yield and selectivity are optimized working
in alcohol under 1.0 MPa of CO2 at 330 K. Under these
conditions, the rate of conversion of DCC into the
carbonate increases in the order: AllylOH > MeOH >
EtOH, which is not the same trend observed for the
addition of the alcohol to DCC (MeOH > EtOH > Allyl
OH).
When O-alkyl isoureas are used as starting materials,
the same trend is observed as that for DCC and alcohol,
which demonstrates that the conversion of isourea and
alcohol + CO2 into the carbonate is the rate-determining
step. We have also found that when allyl alcohol is used,
a new reaction product (diallyl ether) is observed. The
ether is preferentially formed at temperature above 330
K, in both the presence and absence of CO2.
The allyl inversion, a possible process in both the
formation of ether and carbonate, has been excluded
using crotyl alcohol, which affords only dicrotyl carbonate
without any trace of the inversion product (the (crotyl)-
(1-methyl-1-propenyl)-carbonate). The rate and selectivity
of the ether-forming reaction are positively affected by
an increase of temperature: below 330 K the ether yield
is less than 1%, while it is 50% at 350 K.
Use of Catalysts. To ascertain if a Lewis acid, such
as CuCl, CuCl2, Cu2I2, CuO, ZnCl2, or Zn(acetate)2,
improves the rate of formation of the carbonate, we have
compared the conversion yield and selectivity of the
carboxylation of alcohols using either DCC or preformed
alkyl isoureas. Both reactions were carried out in the
presence or absence of catalysts (Table 2). The experi-
mental results (Table 2) suggest that the addition of the
first methanol unit to DCC is catalyzed by Lewis acids
(entries 1 and 2), while the following steps are not.
(Compare entries 1 and 2 (effect of catalyst), 2 and 3,
and 3-5 in Table 2.)
FIGURE 6. Influence of the PCO2 on the rate of formation of
DMC by reacting methanol with CO2 in the presence of
O-methyl-N,N′-dicyclohexyl isourea in CH3CN (solvent/CH3-
OH/isourea ) 50:15:1, T ) 323 K).
increases. The rate of the reaction is sensitive to the na-
ture of the solvent, the reaction being faster in polar
solvents such as CH3CN or the alcohol itself. Moreover,
while in polar solvents the carboxylation readily occurs
also under 0.1 MPa of CO2, and in apolar solvents the
reaction rate is sensitive to the CO2 pressure, being close
to zero at 0.1 MPa.
An increase of pressure from 0.1 to 6.0 MPa does not
greatly influence the yield using the reagent alcohol as
solvent. Conversely, the temperature plays a much more
important role (Figure 5).
The use of scCO2 at moderate temperature does not
give better results with respect to working in alcohol as
solvent. Table 1 shows that the alkyl group may influence
the yield of the reaction.
Using the preformed O-alkyl isourea, we improved both
yield and selectivity. The explanation is that when O-
alkyl isourea is used, a shorter reaction time is necessary
for the carboxylation to occur, which prevents the conver-
sion of isourea into the relevant carbamate (see above).
To observe the best selectivity toward the carbonate, the
carboxylation reaction should be run at the lowest
possible temperature to prevent eq 3 from occurring.
Figure 6 shows the influence of pressure on the rate of
formation of DMC from methyl isourea, methanol, and
J. Org. Chem, Vol. 70, No. 16, 2005 6181

Aresta et al.
TABLE 3. Reactivity Trend in the Case of Allyl Alcohola
reaction mixture composition (%)
entry
r′
r′′
T (K)
t (h)
DAC
DAE
diallyl ether
DCC unreacted
1
2
60
/
338
338
338
338
338
313
338
323
338
338
338
338
6
2
6
6
3
8
2
5
5
4
4
4
14.8
91
2
7
trace
trace
trace
trace
15
82
0
60 (isourea)
/
3
56
31
15
15
15
15
15
15
15
15
2
40
3.3
56.4
61
>90
trace
trace
3
4
40
21
3
7
5
5
215
412
412
500
400b
670c
206d
500d
6
45
40
9
7
81
8
2
8
49
35
9
9
16
1.06
42
38
3
10
11
12
75
7.05
15
12
71
9
/
4
0.4
93
6
a r′ is the molar ratio “alcohol/DCC” (with the exception of entry 2, where it represents the “alcohol/isourea” ratio). r′′ is the ratio
“DCC/catalyst (CuCl)”. b CuCl2 is used as catalyst. c CuO is used. d Cu2I2 is used, DAC ) diallyl carbonate, DAE ) diallyl ether.
SCHEME 3. Reaction Mechanism for the
Formation of Organic Carbonates from DCC and
Alcohols in the Presence of CO2
Entries 6 and 7 in Table 2 show the influence of
temperature on the conversion yield. The comparison of
entries 1 and 2 with 8 and 9, and 3-5 with 10 and 11 of
Table 2 demonstrates that methanol is more reactive
than ethanol.
Table 3 shows the figures relevant to a detailed study
performed with allyl alcohol. Comparing entries 1 and 2
in Table 3 to entries 1 and 3 in Table 2 it is evident that
allyl alcohol is more reactive than methanol. The best
yield and selectivity are those relevant to entries 2 and
12. In other cases, the formation of diallyl ether is
significant (entries 5-11), or the yield in carbonate is low
(entries 3 and 4).
Basic catalysts such as NaOCH3 (obtained from the
reaction of methanol with sodium metal) or the product
of its reaction with CO2, NaOC(O)OCH3,19 10, does not
cause an increase in the conversion yield with respect to
DCC alone (see entries 3 and 4 in Table 2).
The Mechanism of Formation of Dialkyl Carbon-
ates and the Role of DCC. The information gathered
with the systems described above helps to draw the
reaction mechanism of formation of dialkyl carbonates
from alcohols and CO2 in the presence of DCC. Two
pathways can be envisaged:
the 6aN-im complex takes place in two steps. The first
step corresponds to the proton transfer from the alcohol
to the iminic nitrogen of isourea, promoted by CO2, which
interacts with the methoxy anion giving rise to the “ion
pair” intermediate 12a (eq 8). In the second step, the
positively charged methyl group of isourea migrates to
the hemicarbonate anion -OC(O)OCH3 to afford the DMC
molecule and urea (eq 9).
(i) A preliminary interaction of CO2 with methanol (eq
7) to afford the monomethyl ester of carbonic acid (11),
which is then methylated by the alkyl isourea.
CO2
CyHNC(OCH3)dNCy + HOCH3 y
z
3a
-OC(O)OCH3 (8)
CyHN-C+(OCH3)-NHCy
CH OH + CO h
(7)
HOC(O)OCH3
3
2
12a
11
-OC(O)OCH3 f
CyHC+(OCH3)-NHCy
(ii) The direct pseudo “three-center” interaction of alkyl
isourea with alcohol and CO2.
The first route is ruled out because the CO2 insertion
into the O-H bond of CH3OH does not take place.19 We
propose the reaction mechanism depicted in Scheme 3
for the conversion of DCC and methanol under CO2 into
DMC and urea.
To elucidate the mechanistic details of the CO2 activa-
tion part of the reaction, we have examined possible
reaction intermediates and related transition states from
the H-bonded 6aN-im complex toward the final products.
The structures of the stationary points located on the
potential energy surface are collected in Figure 7. The
calculations indicate that the reaction between CO2 and
12a
CyHNC(O)NHCy + (CH3O)2CO (9)
Calculations were used to examine the first step in
detail. The precursor state of this step is a three-
component adduct 6aN-im‚‚‚CO2 involving an initial weak
electrostatic interaction between CO2 and MeOH (the
interaction energy between 6aN-im and CO2 is 2.4 kcal/
mol). The reaction is initiated by the O-H‚‚‚N f O‚‚‚
H-N hydrogen shift that increases the negative charge
of the methoxo oxygen, causing a strengthening of the
O-CO2 bond. When CH3OD, instead of CH3OH, is
reacted with isourea and CO2, an isotopic effect equal to
1.8 is observed that supports that the H (D) transfer is
(19) Aresta, M.; Dibenedetto, A.; Giannoccaro, P.; Pastore, C.; Pa´pai,
I.; Schubert, G. Submitted for publication, 2005.
6182 J. Org. Chem., Vol. 70, No. 16, 2005

Organic Carbonates from Aliphatic Alcohols and CO2
FIGURE 7. Optimized structures of stationary points involved in the reaction of isourea with CH3OH and CO2 (bond lengths in
Å, angles in degrees).
the rate-determining step.20 The concerted mechanism
O-CH3 bond as indicated by the O-C bond distances in
6aN-im‚‚‚CO2 and 12a (1.43 versus 1.46 Å, respectively).
This favors the methyl transfer to the -OC(O)OMe
moiety, a step represented by TSmethyl. The nearly planar
CH3 group is positioned between the two O atoms, and
the N-H‚‚‚O hydrogen bond is much weaker than that
in the ion pair. Since the migrating CH3 group carries a
positive charge, both components of 13a become neutral,
and the final state is characterized as an H-bonded
complex formed between the cis-trans isomers of urea
and DMC. The relative energies of the stationary points
are summarized in Figure 8.
of the proton shift and the O-C bond formation is clearly
born out by the structure of the transition state TSprot
,
in which the proton is between nitrogen and oxygen,
while the O-C bond distance is 1.79 Å and CO2 is bent.
Therefore, the presence of CO2 in the reaction medium
converts the alkoxo- into the hemicarbonate form, which
is more stable than the former due to charge delocaliza-
tion over the two oxygen atoms of the COO- group.
Calculations carried out for the interaction of a meth-
oxy anion with carbon dioxide show that CO2 addition
to CH3O- anion occurs spontaneously (with no energy
barrier), and the formation of the CH3OCOO- anion is
exothermic by 40.4 kcal/mol, underlying the importance
of stabilization introduced by the presence of CO2. The
protonation of the iminic-N of isourea weakens its
Because of the ionic nature of 12a, we found it
important to incorporate the solvent effects into the
energetics, and therefore we estimated the solvation
energies for the five stationary points in terms of the
polarizable continuum model (PCM) model and corrected
(20) O’Ferral R. A. J. Chem. Soc. B 1970, 785-790.
J. Org. Chem, Vol. 70, No. 16, 2005 6183

Aresta et al.
FIGURE 8. Energy profile of the 6aN-im f 13a reaction
pathway as obtained by the gas phase (dashed) and solvated
(dotted) models. The dielectric constant of the polarizable
medium was set to ꢀ ) 32.6 to model methanol as a solvent.
FIGURE 9. Optimized structure of the bidentate ion pair
formed between the protonated isourea and the hemicarbonate
anion (bond lengths in Å).
the relative energies by solvent effects. As Figure 8
shows, the two models provide rather different energy
profiles. While the gas-phase calculations predict the ion
pair to be well above the 6aN-im‚‚‚CO2 adduct, this step
clearly becomes energetically permitted in the solvated
model; therefore, the energy of TSmethyl shifts to a lower
value. The energy barriers for the two steps of the
reaction are predicted to be 9.8 and 14.5 kcal/mol with
the solvated model, which are reasonably low, and we
may conclude that the two-step reaction mechanism
represents a conceivable low energy route between
6aN-im‚‚‚CO2 and 13a. Furthermore, these results are
consistent with the fact that the reaction takes place
faster in polar solvents, since the stability of the ion-pair
intermediate and the gap of the proton shift appear to
be sensitive to the polarity of the solvent.
We have attempted to find experimental evidence for
such a mechanism with NMR and IR studies. When CO2
is added to a CH3CN solution containing 3a and metha-
nol in equimolar amounts (see above for the NMR data
of the system 3a + MeOH), a new signal located at 3.41
ppm appears due to the proton of the methyl group of
the hemicarbonate moiety. At room temperature, no
further change is observed with time, but upon heating
this solution to 330 K the typical DMC signal at 3.70 ppm
appears. As reported above, IR cannot prove the forma-
tion of 6aN-im, but when an acetonitrile solution of
methanol and isourea is exposed to CO2, new bands
attributed to the symmetrical and asymmetrical carbonyl
stretching of the methyl carbonate anion appear at 1585
and 1283 cm-1. The position of the bands and the
difference of 302 cm-1 between the sym and asym
carbonyl stretching frequency confirm that the methyl
carbonate is not monodentate and behaves as a µ2-OO
bidentate ligand,12 a more stable form (12a′), as shown
in Figure 9. 12a′ is not able to give rise to the formation
of DMC at 300 K. After heating such a solution above
330 K, we find that a new signal centered at 1755 cm-1
appears (CdO asymmetrical stretching due to the di-
methyl carbonate), which increases with time until
complete disappearing of the CdN stretching of isourea
(1660 cm-1). The GC-MS analysis of this solution
confirms the formation of DMC. Our B3LYP/6-311++G**
calculations reveal that indeed 12a′ is 19.8 kcal/mol more
stable than 12a, which is attributed to the formation of
the double H-bond and also to the internal structural
rearrangement of the protonated isourea moiety (Figure
9). The relative stability of 12a′ is rather similar in
energy to that of 13a, as the bidentate complex is
predicted to lie 3.5 kcal/mol above 13a in the gas-phase
model, and it even becomes 0.8 kcal/mol more stable
when solvent effects are included. We point out, however,
that the two components in 13a (urea and DMC) can also
rearrange to their global minima (i.e., w-shape struc-
tures), and as a result, the final products become about
7 kcal/mol more stable than the bidentate H-bonded
complex. These results are in line with the observation
reported above that the ion pair identified as a stable
form at 298 K can be converted into urea + DMC when
the temperature is increased up to 330 K.
It is noteworthy that phenol and CO2 can react with
methyl isourea to afford methylphenyl carbonate. This
is the first evidence of formation of a carbonate contain-
ing the phenyl moiety starting from phenol and CO2. As
reported earlier in this article, the formation of diphenyl
carbonate from CO2 and phenol has not been achieved
thus far, most likely because it is not thermodynamically
permitted. What may prevent the formation of diphenyl
carbonate from phenol and CO2 in the presence of DCC
is the fact that phenol does not give a stable addition
product to DCC (see above). The formation of methylphen-
+
yl carbonate occurs because CH3 is transferred to the
-OC(O)OPh anion. DFT calculations reveal that the
formation of methylphenyl carbonate + urea is exother-
mic by 13.5 kcal/mol relative to isourea + PhOH + CO2
(the exothermicity of the reaction with MeOH is 21.2 kcal/
mol), and the energy gap represented by the transition
state of the methyl transfer step is about 6 kcal/mol
higher for PhOH as compared to the reaction with MeOH.
All these data are in agreement with our experimental
observations.
However, we have clearly demonstrated that the role
of DCC in the reaction of alcohols with CO2 to afford
dialkyl carbonates can be identified in the following
specific actions:
(i) DCC reacts with the alcohol to afford the O-alkyl
+
isourea, which is the precursor of the CH3 moiety that
will be used in the methylation step. With respect to this,
it is noteworthy that phenol does not react with DCC
under the same conditions as aliphatic alcohols, despite
being a better acid (pKa ) 10) than methanol (pKa ) 16),
6184 J. Org. Chem., Vol. 70, No. 16, 2005

Organic Carbonates from Aliphatic Alcohols and CO2
ethanol, and allyl alcohol.21 This means that the stability
of the NC(OR)dN moiety is very important for the
isolation of the isourea. This reaction is catalyzed by
metal systems such as CuCl, ZnCl2, CuO, Cu2I2, and Cu-
(OAc)2. The conversion of DCC into the isourea is
quantitative, and the latter can be easily isolated and
used as a reagent. The temperature plays a key role: to
have good selectivity together with high yield, the reac-
tion should be carried out at a temperature close to 330
K. Above this temperature, carbamates are formed from
isourea and alcohols, lowering the selectivity.
(ii) The isourea reacts at 332 K with methanol and CO2
to afford dimethyl carbonate, a reaction not requiring any
catalyst and stepping through the simultaneous interac-
tion of a second alcohol moiety with 3 and CO2 with
subsequent intramolecular methyl transfer to afford the
carbonate. This reaction already occurs at 310 K and 0.1-
6.0 MPa of CO2, with high selectivity and good yield.
Temperatures higher than 330 K increase the reaction
rate but lower the selectivity as they promote the
conversion of 6 into the relevant carbamate and amine.
This reaction does not need a Lewis acid as catalyst. In
fact, the reaction proceeds through a “base activation”
of methanol to afford the methoxo group: the relevant
role is played by the iminic nitrogen of isourea.
GC-MS analyses were carried out with a gas chromato-
graph (capillary column: 30 m; MDN-5S; L 0.25 mm, 0.25-
µm film) coupled to a QP5050 A mass spectrometer. Quanti-
tative determinations on the reaction solutions were recorded
using a GC-FID (capillary column: 30 m; MDN-5S; L 0.25
mm, 0.25-µm film).
Synthesis of O-Alkyl-N,N′-dicyclohexyl Isourea (R )
Me, Et, and Allyl). A quantity of 2 g of DCC (9.71 mmol) was
dissolved in 10 mL of alcohol (methanol or ethanol or allyl
alcohol) in the presence of catalytic amounts of Cu2I2 (10 mg,
0.05 mmol of Cu) (or CuO or ZnCl2). The solution was stirred
at room temperature until the complete conversion (generally
15 h at room temperature) of DCC in the relevant isourea
(GC-MS test) was observed. The residual alcohol was distilled
under vacuum (0.1 mm of Hg) at 300 K, while O-alkyl isourea
was distilled at 410-440 K (depending on the alkyl group)
under vacuum. MS pattern, infrared, and NMR data for
O-methyl, O-ethyl, and O-allyl isoureas are in agreement with
literature data.24
Methylation of 4-Hydroxybenzoic Acid. A quantity of
0.5 g of 4-hydroxybenzoic acid (3.62 mmol) was placed under
stirring in 10 mL of diethyl ether. To the white suspension,
an ether solution of 3a (0.86 g, 3.62 mmol, in 10 mL of diethyl
ether) was added dropwise. The reaction mixture obtained was
analyzed by gas chromatography with a GC-FID (quantitative
analysis using anisole as internal standard) and with GC-
MS (qualitative analysis) showing that 98.7% of 4-hydroxy-
benzoic acid was converted in the relevant methyl ester. Only
traces of MeOC6H4(p-COOMe) were detected.
DCC may, thus, produce the reaction of alcohols with
carbon dioxide to occur under temperature and pressure
conditions that are much milder than those encountered
for metal-catalyzed reactions.11-15 In fact, the latter
require 420-470 K under sc-conditions for reaching a 2%
concentration of carbonate at the equilibrium. However,
there is no sense in using DCC as a water trap in
combination with metal systems as DCC itself promotes
the carboxylation reaction in milder conditions than
metal systems. The identification of the intermediates
has made it possible to fully describe the reaction path.
DFT calculations provided an energy profile for the
reaction. Such a detailed study has no precedent and can
result in being a paradigm for the carboxylation of
alcohols as well as giving information useful for metal-
catalyzed reactions. Because the resulting urea can be
reconverted into DCC,22 our findings may find practical
application.
Reaction of O-Alkyl-N,N′-dicyclohexyl Isourea (R )
Me, Et, and Allyl) with Alcohols. (a) Reaction at 300 K.
A quantity of 0.050 g of O-alkyl isourea was dissolved in 0.7
mL of dry CD3CN in an NMR tube, and the NMR spectrum
was taken. To the solution, the stoichiometric amount of the
relevant alcohol was added, and the proton spectrum was
immediately recorded. The spectral data are reported in the
Discussion section.
(b) Reaction at 350 K. A quantity of 0.5 g of O-alkyl
isourea was placed in 5 mL of the relevant alcohol under
stirring for 24 h at 350 K under a dinitrogen atmosphere. The
formation of the alkylated primary amine and the relevant
carbamate was shown by GC-MS analysis. The amine and
carbamates were fully characterized by distillation of the
alcoholic phase and column (0.5 m, 3 cm i.d.; hexane-diethyl
ether 50/50 v/v) chromatographic separation, as already
reported.25
(c) Alkyl Exchange Reaction. A quantity of 0.11 g of 3a
(0.46 mmol) in toluene (5 mL) was reacted at 333 K with a
5-fold molar excess of either ethanol or allyl alcohol. After a
few minutes, the solution was analyzed by GC-MS, showing
the presence of O-alkyl isourea (ethyl or allyl) and methanol.
The reaction was complete in 15 h.
Experimental Section
All alcohols, starting reagents (DCC, metal catalysts, etc.),
and solvents were commercial products. Alcohols and solvents
were dried, distilled,23 and stored under dinitrogen. Carbon
dioxide was 99.999% pure.
Nuclear magnetic resonance experiments were carried out
with a 300 MHz apparatus using deuterated solvents. Infrared
spectra were recorded with an FTIR apparatus. High-pressure
reactions were carried out in a 70-mL autoclave thermostated
with an electric heating jacket. The autoclave was equipped
with an inner glass reactor for avoiding metal contamination
of the reagents and with a valve for continuous withdrawing
of liquid samples. Reactions in supercritical carbon dioxide
were carried out in equipment with automatic temperature
and pressure control.
Reaction of DCC with Alcohols and Carbon Dioxide.
The following general procedure was used for all reactions. A
quantity of 1.0 g of DCC (4.85 mmol) was placed in a glass
reactor with the reagent alcohol (10 mL). Toluene (20 µL) was
added as internal standard, and the catalyst (Cu2I2, CuCl,
Cu2O, ZnCl2) was eventually added. The reactor was placed
in a stainless steel autoclave that was closed and pressurized
with carbon dioxide at the desired pressure. The system was
heated at different temperatures and for different times. After
reaction, the autoclave was cooled to room temperature and
depressurized, bubbling the gas phase in a cold DMF solution.
The GC analysis of such solution allowed us to determine the
amount of dialkyl carbonate that had possibly left the auto-
clave with the gas phase. The reaction solution was filtered
for recovering the formed urea, and the clear solution was
analyzed by GC or GC-MS to determine the formed carbonate
(21) (a) Ballinger, P.; Long, F. A. J. Am. Chem. Soc. 1960, 82, 795-
798. (b) Hine, J.; Hine, M. J. Am. Chem. Soc. 1952, 74, 5266-5271.
(22) Stevens, C. L.; Singhal, G. H.; Ash, A. B. J. Org. Chem. 1967,
32, 2895.
(23) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals; Pergamon Press: Oxford, England, 1986.
(24) Mathias, L. J. Synthesis 1979, 8, 561-576 and references in
therein.
(25) Aresta, M.; Quaranta, E. Tetrahedron 1991, 47, 9489-9502.
J. Org. Chem, Vol. 70, No. 16, 2005 6185

Aresta et al.
and residual DCC. The recovered urea was washed with
diethyl ether (2 × 5 mL), dried, and weighted for a double
check on the conversion yield of DCC.
Computational Details
DFT calculations were carried out to obtain information on
the structures and relative energies of energy minima and
transition states relevant to possible reaction pathways. First
we located the stationary points using a smaller model, where
the dicyclohexylcarbodiimide was simplified to dimethylcar-
bodiimide. These structures were used to construct the realistic
model compounds by replacing the methyl groups with the
experimentally used cyclohexyls and repeated the geometry
optimizations for these realistic models. The nature of the
stationary points obtained from the geometry optimizations
was verified by subsequent vibrational frequency analysis.
Intrinsic reaction coordinate (IRC) calculations were performed
from the located transition states to check whether these
structures indeed connect the appropriate minima on the
proposed reaction pathways.
Influence of the Solvent on the Carboxylation of
Alcohols. A quantity of 0.2 g of 3a (0.84 mmol) was dissolved
in 5 mL of toluene or acetonitrile in a glass reactor under an
atmospheric pressure of carbon dioxide. Naphthalene (12.0 mg,
0.094 mmol) was added as internal standard. To the solution,
0.25 mL of methanol (6.23 mmol) was added, and the reactor
was closed and heated in an oil bath at 328 K. The reaction
was followed by gas chromatography for the due time. The
results are reported in Figure 4.
Reaction of DCC with Crotyl Alcohol and Carbon
Dioxide. A quantity of 2 g of DCC (9.7 mmol) and 12 mg of
Cu2I2 (0.063 mmol of Cu) was added to 10 mL of crotyl alcohol
(cis-trans isomer mixture) in a glass reactor that was placed
in a stainless steel autoclave under 6.0 MPa at 338 K. After 2
h, the autoclave was depressurized, and the solution was
filtered to afford the relevant urea that was washed with
diethyl ether (2 × 10 mL). The reaction solution was treated
with distilled water (10 mL) to extract the copper salts and
the organic phase was separated, dried over anhydrous Na2-
SO4, and distilled to afford a dark-yellow liquid that was
characterized as dicrotyl carbonate (1.04 g, 6.1 mmol, isolated
yield 63.9%). The GC-MS and NMR analysis confirmed the
total absence of the product originated by allyl inversion.
Mass pattern (m/z): 55, 71, 72, 108, 116. Infrared data
recorded on the neat compound with KBr disks: 3025 (w), 2932
(s), 2856 (m), 1745 (s), 1450 (m), 1392 (m), 1251 (s), 1081 (m),
1042 (m), 967 (m), 924 (m), 793 (m), 590 (m). 1H NMR (second-
order spectrum: all signals are undefined multiplets): 5.79
(1H), 5.62 (1H), 4.51 ppm (2H), 1.70 ppm (3H). 13C NMR: CH2
at 68.6 ppm, CH at 132.2 ppm, CH at 125.1 ppm, CH3 at 17.9
ppm, and OC(O)O signal at 155.3 ppm.
Spectroscopic Evidence of the Reaction of 3a with
Alcohols and CO2. A quantity of 0.1582 g of 3a (0.665 mmol)
was dissolved in 3 mL of CD3CN under a dinitrogen atmo-
sphere. To this solution, 50 µL of methanol (1.25 mmol) was
added. The 1H NMR was recorded, the dinitrogen atmosphere
was replaced with CO2, and the spectrum was again recorded.
The results are reported in the Discussion section. All steps
were also monitored by FTIR and GC-MS.
The calculations were carried out with Becke’s three-
parameter B3LYP hybrid exchange-correlation functional26
using the Gaussian 98 software package.27 For the geometry
optimizations, frequency calculations, and IRC calculations,
we used the standard 6-31G* basis set; however, to obtain
accurate energetics, single-point energy calculations were
performed at each stationary point using the extended
6-311++G** basis set. For some of the structures, in cases
where we expected considerable solvent effects on the energet-
ics, we estimated the solvation energies in terms of the PCM.28
In these calculations, we used methanol as a solvent with a
dielectric constant of ꢀ ) 32.6. The energies given in the article
generally refer to B3LYP/6-311++G** gas-phase values, ex-
cept those where the solvent effects are examined explicitly.
The structures and the total energies of all located stationary
points are given in the Supporting Information.
Acknowledgment. The Italian authors thank the
Ministry of University and Research for financial sup-
port (COFIN Grants: MM03027791 and 2003039774
and EU-IP “TOPCOMBI”). The Hungarian authors
acknowledge the support of the Hungarian Research
Foundation (OTKA Grants: T037345 and T034547).
Study of the Isotopic Effect in the Reaction of CH3OH/
CH3OD with 3a. A quantity of 0.22 g of 3a (0.92 mmol) was
dissolved in 5 mL of toluene under an atmosphere of carbon
dioxide, and 250 µL of CH3OH (or CH3OD) was added. The
composition of the reaction mixture was monitored by GC-
FID, and the kinetic curve of formation of DMC was obtained.
The kH and kD were determined, and an isotopic effect of 1.8
was calculated.
Supporting Information Available: Cartesian coordi-
nates and total energies of located stationary points relevant
to the title reaction. This material is available free of charge
via the Internet at http://pubs.acs.org.
Synthesis of Methylphenyl Carbonate. A quantity of 1.0
g of 3a (4.20 mmol) was placed in the scCO2 reactor with 1.0
g of phenol (10.64 mmol) at 333 K under 20.0 MPa of carbon
dioxide for 4 h. After being cooled and depressurized, 5 mL of
THF was added to dissolve the reaction mixture. A GC-MS
analysis showed the formation of methylphenyl carbonate, for
which retention time and MS pattern fragmentation are
identical with those of an authentic sample. Working in a
solvent (toluene or CH3CN or an excess of phenol) under 6 MPa
of CO2 does not afford any mixed carbonate.
Synthesis of Mixed Carbonates from 3a and Ethanol
or Allyl Alcohol. A quantity of 0.2 g of 3a (0.84 mmol) was
dissolved in 5 mL of CH3CN. Ethanol (250 µL, 4.35 mmol) or
allyl alcohol (3.67 mmol) was added under an atmosphere of
carbon dioxide. The solution was heated to 333 K. The reaction
mixture was analyzed by GC-MS, confirming the formation
of methyl ethyl- and methyl allyl carbonate, respectively.
Continuous monitoring showed that the selectivity of the
reaction decreases with time because of the trans-alkylation
of methyl isourea with subsequent formation of diethyl- and
diallyl carbonates.
JO050392Y
(26) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789. (c) Stephens,
P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem.
1994, 98, 11623-11627.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A.9; Gaussian, Inc.: Pittsburgh, PA,
1998.
(28) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117-
129.
6186 J. Org. Chem., Vol. 70, No. 16, 2005