Synthesis of carbonates and related compounds from carbon dioxide via methanesulfonyl carbonates

Article abstract of DOI:10.1021/jo026753g

Carbonate anions resulting from reaction of primary or secondary alcohols with carbon dioxide, when added to methanesulfonic anhydride in cooled acetonitrile solution, yield methanesulfonyl carbonates, a new class of synthetic intermediate. Base-mediated reaction of the methanesulfonyl carbonates with alcohols, thiols, and amines yields carbonates, thiocarbonates, and carbamates, respectively. Overall yields for the three steps vary from 19% to 42%.

Full text of DOI:10.1021/jo026753g

Syn th esis of Ca r bon a tes a n d Rela ted Com p ou n d s fr om Ca r bon
Dioxid e via Meth a n esu lfon yl Ca r bon a tes
Mark O. Bratt and Paul C. Taylor*
Department of Chemistry, University of Warwick, Coventry CV4 7AL, UK
p.c.taylor@warwick.ac.uk
Received November 22, 2002
Carbonate anions resulting from reaction of primary or secondary alcohols with carbon dioxide,
when added to methanesulfonic anhydride in cooled acetonitrile solution, yield methanesulfonyl
carbonates, a new class of synthetic intermediate. Base-mediated reaction of the methanesulfonyl
carbonates with alcohols, thiols, and amines yields carbonates, thiocarbonates, and carbamates,
respectively. Overall yields for the three steps vary from 19% to 42%.
SCHEME 1
In tr od u ction
Carbonates 3 are most commonly prepared from phos-
gene 1 via chloroformates 2 (eq 1).1 While this route has
the benefits of simplicity, generality, and high yields,
there are significant problems with the use of highly toxic
phosgene.2 The use of solid phosgene equivalents such
as bis(trichloromethyl) carbonate (triphosgene) 3a or
carbonyl diimidazole 4 can be recommended for small-
scale preparations, but they are expensive, and ulti-
mately such reagents are derived from phosgene.3 In fact,
industrially phosgene is still the most widely used
carbonyl source (80% of usage), with the remainder
largely accounted for by dimethyl carbonate 3b. Synthe-
sis of dimethyl carbonate does not have to be from
phosgene, but the main alternative process is methanol
carbonylation, which involves another toxic gas, carbon
monoxide.4
a good deal of progress has been made in harnessing
carbon dioxide for carbamate (urethane) synthesis.5-7
Amines react readily with carbon dioxide in the presence
of base to form carbamate anions 5. Of particular
relevance to our work are processes in which the base
permits substantial delocalization in its conjugate acid
(typically amidine, pentaalkyl guanidine, or phosphazene
bases are used), ion pairing is reduced, and the anion is
rendered sufficiently nucleophilic to be alkylated to form
carbamates 6 (Scheme 1).5 Similarly, anion 5 can be
Carbon dioxide is an alternative carbonyl source and,
in contrast to phosgene and carbon monoxide, it is quite
harmless, as well as being abundant and cheap. In fact,
(1) (a) Shaikh, A.-A. G.; Sivaram, S. Chem. Rev. 1996, 96, 951. (b)
Parrish, J . P.; Salvatore, R. N.; J ung, K. W. Tetrahedron 2000, 56,
8207. (c) Adams, P.; Baron, F. A. Chem. Rev. 1965, 65, 567.
(2) “Protocol for the prohibition of the use in war of asphyxiating,
poisonous or other gases, and of bacteriological methods of warfare”;
Geneva 17 J une, 1925.
(3) Babad, H.; Zeiler, A. Chem. Rev. 1973, 73, 1; see also ref 1b.
(4) Kirk Othmer Encyclopaedia of Chemical Technology; J ohn
Wiley: New York, 1996; Vol. 18, p 645; Yasuda, H.; Choi, J .-C.; Lee,
S.-C.; Sakakura, T. Organometallics 2002, 21, 1216 and references
therein.
(5) McGhee, W.; Riley, D.; Christ, K.; Pan, Y.; Parnas, B. J . Org.
Chem. 1995, 60, 2820 and references therein.
(6) McGhee, W. D.; Pan, Y.; Talley, J . J . Tetrahedron Lett. 1994,
35, 6, 839; McGhee, W. D.; Talley, J . J . S Patent 5,380,855, 1995.
(7) Shi, M.; Shen, Y. M. Helv. Chim. Acta 2001, 84, 3357. Abla, M.;
Choi, J .-C.; Sakakura, T. Chem. Commun. 2001, 2238. Yoshida, M.;
Hara, N.; Okuyama, S. Chem. Commun. 2000, 151. See also; Salvatore,
R. N.; Chu, F. X.; Nagle, A. S.; Kapxhiu, E. A.; Cross, R. M.; J ung, K.
W. Tetrahedron 2002, 58, 3329.
10.1021/jo026753g CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/18/2003
J . Org. Chem. 2003, 68, 5439-5444
5439

Bratt and Taylor
intercepted by thionyl chloride (or similar reagents) and
transformed to a carbamoyl chloride 7 (or an isocyanate
from a primary amine).6 Further reaction with an alcohol
leads to the carbamate targets.
conditions and equivalents of anhydride. In all cases,
analysis of the evaporated reaction mixture showed a
large number of products (19F NMR spectroscopy indi-
cated the presence of 10 organofluorine compounds from
one experiment), with no evidence for the presence of the
sulfonyl carbonate 9a . The only encouragement came
from the observation of a small amount of di-n-propyl
dicarbonate. However, isopropyl n-propyl carbonate was
also identified, indicating that isomerization was occur-
ring under the reaction conditions. We concluded that
the trifluoromethanesulfonyl intermediates were simply
too reactive. This is consistent with a report of carbonyl
ditriflate 12 that decomposes above -20 °C.11
Alcohols also react with carbon dioxide to form carbon-
ate anions 8. However, the analogous alkylation to form
carbonates directly is successful only in a limited number
of cases.8 Furthermore, reaction with thionyl chloride
does not yield chloroformates 2. In short, the carbamate
chemistry cannot be directly extended to carbonate
synthesis. The carbonate anions can be activated using
Mitsunobu-type chemistry (eq 2), but the intermediates
are not isolable.9 Therefore, this chemistry can only be
used for the synthesis of symmetrical carbonates.
As high reactivity of the sulfonyl reagent seemed to
be far from essential, we decided to investigate the
analogous methanesulfonyl (mesyl) compounds. Meth-
anesulfonic anhydride 11 (R ) CH3) is a cheap, easily
1
handled reagent. Additionally, we expected the H and
13C NMR signals of the methyl signal to be simple to
interpret yet sensitive to the environment in the proposed
intermediates.
We have been interested in the synthesis of dendritic
carbamates using carbon dioxide.10 One synthetic plan
involved the formation of carbamoyl chlorides as in
Scheme 1. However, we were repeatedly frustrated by
the failure of the 5 to 7 transformation (Scheme 1) with
these more complex substrates, which did not appear to
be stable to thionyl chloride under the reaction condi-
tions. We reasoned that both this problem and, more
importantly, the failure of carbonate anions 8 to yield
chloroformates 2 under the same conditions could be
circumvented by sulfonation of 5 or 8 to give sulfonate
analogues of 7 and 2, respectively. Herein, we describe
the development of this idea in the context of carbonate
synthesis and, more briefly, its extension to thiocarbonate
and carbamate synthesis.
Carbon dioxide was bubbled through a solution of
n-propanol and DBU (1 equiv) in acetonitrile at -42 °C.
Methanesulfonic anhydride (1.1 equiv) was added, and
after the mixture had been stirred for a short time, the
1
solvent was evaporated. The H NMR spectrum of the
mixture showed the presence of four products. Two were
unambiguously identified as di-n-propyl carbonate and
n-propyl mesylate. A third compound had a methyl signal
at 3.39 ppm that integrated 3:2 to one of the downfield
methylene signals from the propyl group and was thus
thought to be the desired sulfonyl carbonate intermediate
10a . The fourth compound contained only propyl signals
but could not be identified immediately.
Resu lts a n d Discu ssion
We started our studies with the synthesis of n-propyl
sulfonyl carbonates 9a and 10a . Propanol was chosen as
the test substrate since we anticipated that it would be
easy to remove excess alcohol from reaction mixtures and
we selected diazabicycloundecene (DBU) as base, since
it is commercially available (pentaalkyl guanidine bases
are not and phosphazene bases are very expensive). The
trifluoromethanesulfonyl carbonate 9a was our initial
target, since we reasoned that reaction of the carbonate
anion with the highly electrophilic anhydride 11 (R )
CF3) would be fast.
The mixture was analyzed by GCMS. In agreement
with the 1H NMR spectrum, four components were
observed in the GC trace. The first two peaks were
identified by their mass spectra as di-n-propyl carbonate
and n-propyl mesylate, but we were surprised to find that
the third and fourth peaks were also identified by their
mass spectra to be di-n-propyl carbonate and n-propyl
mesylate! We realized that the sulfonyl carbonate 10a
is likely to decarboxylate on GC analysis, yielding the
mesylate 13, and this led us to the structure of the fourth
unidentified product which we now know to be di-n-
propyl dicarbonate 14, which will decarboxylate to the
carbonate on heating.13
Having identified all four products of reaction of
n-propanol with carbon dioxide and methanesulfonyl
anhydride, we were able to propose a rather complicated
reaction scheme to explain our observations (Scheme 2).
Both components in the initial equilibrium can react with
A number of experiments were attempted in which
carbon dioxide was bubbled through a solution of n-
propanol and DBU in acetonitrile then triflic anhydride
11 (R ) CF3) was added, with variations in reaction
(8) McGhee, W.; Riley, D. J . Org. Chem. 1995, 60, 6205.
(9) Hoffman, W. A. J . Org. Chem. 1982, 47, 5209. Kadokawa, J .;
Habu, H.; Fukamachi, S.; Karasu, M.; Tagaya, H.; Chiba, K. J . Chem.
Soc., Perkin Trans. 1 1999, 2205.
(10) Clark, A. J .; Echenique, J .; Haddleton, D. M.; Straw, T. A.;
Taylor, P. C. J . Org. Chem. 2001, 66, 8687.
(11) J ohri, K. K.; Katsuhara, Y.; Desmarteau, D. D. J . Fluorine
Chem. 1982, 19, 227.
(12) McGhee, W. D.; Paster, M.; Riley, D.; Ruettimann, K.; Solodar,
J .; Waldman, T. ACS Symp. Ser. 1996, 626, 49-58.
(13) Kovacs, A. S.; Wolf, H. O. Ind. Obst.-Gemueseverwert 1963, 48,
229.
5440 J . Org. Chem., Vol. 68, No. 14, 2003

Synthesis of Carbonates and Related Compounds
TABLE 1. Syn th esis of Ca r bon a tes fr om Ca r bon
Dioxid e via Mesyl Ca r bon a tes
SCHEME 2
entry
R1
Me
n-Pr
n-Pr
n-Pr
PhCH2 Ph
PhCH2 t-Bu
i-Pr
t-Bu
Ph
R2
Ph
n-Pr
Ph
base
product isolated yield (%)
1
2
3
4
5
6
7
8
9
pyridine
pyridine
none
3a
3b
3c
3d
3e
3f
26
40
35
28
PhCH2 pyridine
none
pyridine
25
0
PhCH2 pyridine
n-Pr
n-Pr
3g
(26)a
pyridine
pyridine
no reaction
no reaction
mesyl anhydride to yield the mesylate and mesyl carbon-
ate. However, both these products can react further with
carbonate anion to yield carbonate and dicarbonate,
respectively, the latter being transformed to carbonate
by decarboxylation or reaction with alcohol. These path-
ways, though interesting, are only productive if sym-
metrical carbonates are desired. Otherwise, it is essential
to stop the reaction at the mesyl carbonate 10 to permit
reaction with a different nucleophile in a second step.
We were now able to settle on a procedure that
maximized the yield of the mesyl carbonate 10, namely
slow addition of a cooled suspension of the preformed
carbonate anion in acetonitrile to a cooled solution of
methanesulfonic anhydride, also in acetonitrile. For the
n-propyl analogue 10a , we were able to isolate the
proposed intermediate in ca. 45% yield, though only in
90% purity, as a rather unstable oil. Characterization
a
NMR yield; product could not be separated by chromatography
in our hands.
TABLE 2. Syn th esis of Th ioca r bon a tes fr om Ca r bon
Dioxid e via Mesyl Ca r bon a tes
entry
R1
R2
product isolated yield (%)
1
2
3
n-Pr
PhCH2 Et
PhCH2 3,5-dichlorophenyl
3,5-dichlorophenyl
14a
14b
14c
37
25
19
1
was only possible by H and 13C NMR spectroscopy and
(entries 1, 3, and 5) is reactive, as well as primary
alcohols. In short, this is the first synthesis of carbonates
from carbon dioxide which can lead to unsymmetrical and
to aromatic carbonates.
by mass spectrometry. The carboxyl signal at 147.83 ppm
in the 13C NMR spectrum is distinct from that of the
carbonate (155.48 ppm), being closer in value, as ex-
pected, to that of dipropyl dicarbonate (148.64 ppm).
Attempts to replace the sulfonic anhydride with sul-
fonyl chlorides as sulfonating agents were unsuccessful.
Substitution of methanesulfonic anhydride with meth-
anesulfonyl chloride led mainly to mesylation of the
alcohol, whereas use of tosyl chloride led, as with triflic
anhydride, to symmetrical dicarbonates, suggesting that
the intermediates in this case are too reactive for our
purposes.
Syn th esis of Ca r bon a tes. Having optimized the
synthesis of the unstable mesyl carbonate intermediates,
we proceeded to investigate their base-catalyzed reaction
with a second alcohol to complete our new carbonate
synthesis. The results are summarized in Table 1.
At first sight, the isolated yields for these reactions are
poor. However, it should be noted that they are for three
steps, the first of which (formation of the carbonate anion)
is an equilibrium process. Indeed, we believe this equi-
librium to be the major factor in determining the overall
yield, since the only major impurity arises from mesyla-
tion of unreacted alcohol. This is consistent with our
observation (vide supra) that the intermediate 10a is only
formed in ca. 45% yield.
Syn th esis of Th ioca r bon a tes. An obvious extension
of our new procedure was to synthesis of thiocarbonates.
The results are summarized in Table 2. The particularly
low yield for entry 3 is due to competing oxidation of the
thiol under the reaction conditions. We believe this to be
the first report of thiocarbonate synthesis from carbon
dioxide.
Syn th esis of Ca r ba m a tes. As discussed above, there
are already good methods for the synthesis of carbamates
from carbon dioxide, all of which rely on initial reaction
of the amine with carbon dioxide. It was of interest to us
to investigate the alternative route, namely initial reac-
tion of an alcohol with carbon dioxide, followed by the
reaction of an intermediate with an amine. Our results
are presented in Table 3.
To complete the investigation, we briefly investigated
the reactivity of an N-analogue of mesyl carbonates 9,
the mesyl carbamate 15. Using very similar reaction
conditions, carbon dioxide was bubbled through a cooled
acetonitrile solution of diethylamine in the presence of
DBU (eq 3). Once again, we were unable to isolate pure
15, but were able to obtain 1H and 13C NMR data
consistent with the proposed structure. Again, the car-
bamoyl signal in the 13C NMR spectrum (148.00 ppm)
was distinctive. It should be noted that there is literature
precedent for intermediate 15 since an analogous inter-
mediate derived from a primary amine has been pro-
Synthesis of the mesyl carbamate intermediate only
appears to be possible in our hands with primary and
secondary alcohols; use of tert-butyl alcohol or phenol in
the first step led only to recovery of starting alcohol
(entries 9 and 10). However, in the second step, phenol
J . Org. Chem, Vol. 68, No. 14, 2003 5441

Bratt and Taylor
n-propyl mesylate.14 Integration of the methanesulfonyl signals
of the product and n-propyl mesylate in the 1H NMR spectrum
showed that the product was 90% of the oil, overall yield
45%: 1H NMR (CDCl3) δ 1.00 (3H, t, J ) 7 Hz, CH3CH2CH2O),
1.77 (2H tq, J ) 7, 7 Hz, CH3CH2CH2O), 3.39 (3H, s, CH3-
SO2), 4.27 (2H, t, J ) 7 Hz, CH3CH2CH2O); 13C NMR (CDCl3)
δ 9.94 (CH3CH2CH2O), 21.61 (CH3CH2CH2O), 39.65 (CH3SO2),
72.37 (CH3CH2CH2O), 147.83 (CdO); MS (EI) m/z ) 182.
Addition of 1 drop of Et3N or DBU to the NMR sample gave
no n-propyl mesyl carbonate on reanalysis.
TABLE 3. Syn th esis of Ca r ba m a tes fr om Ca r bon
Dioxid e via Mesyl Ca r bon a tes
entry
R3
R1
Bu
Ph
n-Pr
Ph
R2
product
isolated yield (%)
1
2
3
4
5
n-Pr
n-Pr
PhCH2
PhCH2
n-Pr
H
H
H
H
Et
6a
6b
6c
6d
6e
42
21
Gen er a l P r oced u r e for Syn th esis of Ca r bon a tes 3,
Th ioca r bon a tes 14, a n d Ca r ba m a tes 6 fr om Alcoh ols. An
alcohol (9.9 mmol) and DBU (1.25 equiv) were dissolved in
anhydrous MeCN (12 mL), CO2 was bubbled subsurface, and
the solution was cooled to -42 °C for 45 min. The mixture
was allowed immediately to warm to -30 °C and was trans-
ferred by cannula over 60 min to methanesulfonic anhydride
(1.25 equiv) in MeCN (6 mL) at the same temperature. Once
the reaction reached room temperature, the second nucleophile
(1.3 equiv) and pyridine (1.4 equiv) dissolved in MeCN (1 mL)
were added dropwise. The addition was exothermic and
required cooling in ice-water. The reaction was subsequently
stirred at room temperature overnight. Ethyl acetate (50 mL)
was added, and the solution was washed with H2SO4 (0.2 M,
2 × 50 mL) and brine (50 mL). The organic layer was dried
(K2CO3) and filtered, and the solvent was removed in vacuo.
The purification methods are described below for each product.
Meth yl P h en yl Ca r bon a te 3a .15 The general procedure
was followed using methanol and with the following amend-
ment. After the addition of the carbonate salt was complete,
the reaction was allowed to warm to 7 °C. Phenol (1.3 equiv)
was dissolved in MeCN (1 mL) with pyridine (1.4 equiv) and
added to the mesyl carbonate. A temperature increase to 24
°C was noted, and the reaction was cooled back to 7 °C. A
precipitate had formed within 30 s; no further change was
observed within 48 h. The reaction was diluted with Et2O (40
mL) and washed with HCl (0.2 M, 2 × 40 mL), NaOH (0.125
M, 40 mL), and brine (40 mL). The product was isolated as a
colorless oil in 26% yield. Purity by 1H NMR spectroscopy was
>95%: IR (film) 1763 cm-1 (CdO); 1H NMR (CDCl3) δ 3.91
(3H, s, CH3), 7.16-7.41 (5H, m, Ph); 13C NMR (CDCl3) δ 55.37,
121.01 (Ar), 126.06 (C-4), 129.48 (Ar), 151.07 (C-1), 154.28
(CdO); HRMS (EI) calcd 152.0473, found 152.0429.
(48)a
28
Et
35
a
NMR yield; product could not be separated by chromatography
in our hands.
posed.12 In that case, elimination of sulfonic acid to yield
the corresponding isocyanate was observed.
Reaction of 15 with n-propanol under our standard
conditions led smoothly to the known carbamate 6e (eq
3). The increased yield as compared with the syntheses
via mesyl carbonates is consistent with our proposal that
the yield is mainly determined by the initial equilibrium.
Con clu sion
Carbonate anions resulting from reaction of primary
or secondary alcohols with carbon dioxide, when added
to methanesulfonic anhydride in cooled acetonitrile solu-
tion, yield methanesulfonyl carbonates, a new class of
synthetic intermediate. Base-catalyzed reaction of the
methanesulfonyl carbonates with alcohols, thiols, and
amines yields carbonates, thiocarbonates, and carbam-
ates, respectively. Yields for the three steps vary from
19% to 42%, probably being limited by the initial equi-
librium. To our knowledge, this is the first route from
carbon dioxide to unsymmetrical carbonates, to aromatic
carbonates, and to thiocarbonates. Other advantages of
the procedure are that carbon dioxide can be used at
atmospheric pressure and that the base used, DBU, is
readily available.
Di-n -p r op yl Ca r bon a te 3b.16 The general procedure was
followed with n-propanol as both starting alcohol and second
nucleophile. The product was extracted with Et2O and isolated
by evaporation in 40% yield: 1H NMR (400 MHz, CDCl3) δ
0.97 (6H, t, J ) 7 Hz, CH3), 1.70 (4H, tq, J ) 7 Hz, 7 Hz, CH2-
CH2O), 4.09 (4H, t, J ) 7 Hz, CH2O); 13C NMR (100.6 MHz,
CDCl3) δ 10.21 (CH3), 22.08 (CH2), 69.45 (CH2), 155.48
(CdO); HRMS (EI) calcd 146.0943, found 146.0949.
P h en yl n -P r op yl Ca r bon a te 3c.17 The general procedure
was followed with n-propanol as starting alcohol and phenol
as second nucleophile and with the following amendment. No
pyridine was added, and the reaction was heated with phenol
at 40 °C for 22 h. The crude oil was purified by flash
chromatography (SiO2, cyclohexane/CH2Cl2) and isolated as a
Exp er im en ta l Section
1
clear oil in 35% yield: IR (film) 1761 cm-1 (CdO); H NMR δ
Gen er a l Meth od s. Carbon dioxide was supplied by BOC
gases (99%+). NMR spectra were obtained in CDCl3 or
acetone-d6 with TMS as internal standard.
1.00 (3H, t, J ) 7 Hz, CH3), 1.77 (2H, tq, J ) 7 Hz, 7 Hz, CH2-
CH2O), 4.22 (2H, t, J ) 7 Hz, CH2O), 7.16-7.48 (5H, overlap-
ping m, Ph); 13C NMR δ 10.20 (CH3), 22.00 (CH2), 70.34 (CH2),
121.10 (Ar), 125.97 (C-4), 129.47 (Ar), 151.19 (C-1), 153.81
(CdO); MS (EI) m/z ) 180 (M+). Anal. Calcd for C10H12O3: C,
66.65; H, 6.71. Found: C, 66.31; H, 6.66.
P r op yl Meth a n esu lfon yl Ca r bon a te 10a . n-Propanol (0.4
mL, 5.35 mmol) and DBU (1 mL, 6.70 mmol) were dissolved
in anhydrous MeCN (12 mL). CO2 was bubbled subsurface,
and the solution was cooled to -42 °C for 45 min. The mixture
was allowed to warm to -20 °C and was transferred by
cannula over 30 min to methanesulfonic anhydride (1.86 g,
10.7 mmol; 2 equiv) in MeCN (4 mL) at the same temperature.
The mixture was stirred for 10 min, diluted with Et2O (50 mL),
and washed with H2SO4 (0.5 M, 2 × 50 mL) and brine (60 mL).
The ethereal solution was dried with anhydrous potassium
carbonate and filtered, and the solvent was removed in vacuo
to give an oil (493 mg). The product could not separated from
(14) Williams, H. R.; Mosher, H. S. J . Am. Chem. Soc. 1954, 76,
2984.
(15) Yew, K. H.; Koh, H. J .; Lee, H. W.; Lee, I. J . Chem. Soc., Perkin
Trans. 2 1995, 2263. Couture, P.; El-Saidi, M.; Warkentin, J . Can. J .
Chem. 1997, 75, 3, 326.
(16) Kadokawa, J .; Habu, H.; Fukamachi, S.; Karasu, M.; Tagaya,
H.; Chiba, K. J . Chem. Soc., Perkin. Trans. 1 1999, 2205.
(17) Brown, P.; Djerassi, C. J . Am. Chem. Soc. 1966, 88, 2469.
5442 J . Org. Chem., Vol. 68, No. 14, 2003

Synthesis of Carbonates and Related Compounds
Ben zyl n -P r op yl Ca r bon a te 3d .18 The general procedure
was followed with benzyl alcohol as starting alcohol and
n-propanol as second nucleophile and with the following
amendment. A 5 mL (2.3 mmol) aliquot of the reaction was
removed when at 0 °C, to which was added benzyl alcohol (0.3
mL, 2.90 mmol; 1.3 equiv) and pyridine (0.25 mL, 3.09 mmol;
1.39 equiv). TLC after 1 h showed a new product. There was
no further change in the reaction during 3 days. The reaction
was diluted with Et2O (40 mL), washed with HCl (0.125 M, 2
× 40 mL) and brine (40 mL), dried (K2CO3), and filtered.
Isolation by flash chromatography (SiO2, 1:1 cyclohexane/CH2-
Cl2) yielded a colorless oil in 28% yield: IR (film) 1745 cm-1
tography (SiO2, 24:1 cyclohexane/CH2Cl2) yielded 14a as a
1
clear oil in 37% yield: IR (film) 1731 cm-1 (CdO); H NMR
(250 MHz, CDCl3) δ 0.96 (3H, t, J ) 7 Hz, CH3), 1.72 (2H, tq,
J ) 7 Hz, 7 Hz, CH2), 4.23 (2H, t, J ) 7 Hz, CH2), 7.40 (1H, t,
J ) 2 Hz, H-4), 7.44 (2H, apparent d, J ) 2 Hz, H-2 and H-6);
13C NMR (75.5 MHz, CDCl3) δ 10.21 (CH3), 22.00 (CH2CH2O),
70.15 (CH2CH2O), 129.67 (C-4), 130.92 (C-1), 132.57 (C-2),
135.14 (C-3), 168.05 (CdO); HRMS (EI) calcd 263.9779, found
263.9774. Anal. Calcd for C10H10Cl2O2S: C, 45.30; H, 3.80.
Found: C, 45.25; H, 3.71.
Ben zyl S-Eth yl Th ioca r bon a te 14b. The general proce-
dure was followed with benzyl alcohol as starting alcohol and
ethanethiol as second nucleophile. Purification by flash chro-
matography (SiO2, 24:1 cyclohexane/CH2Cl2) gave the isolated
product as a clear oil in 25% yield: IR (film) 1702 cm-1
1
(CdO); H NMR (300 MHz, CDCl3) δ 0.96 (3H, t, J ) 7 Hz,
CH3CH2CH2O), 1.69 (2H, tq, J ) 7 Hz, 7 Hz, CH3CH2CH2O),
4.10 (2H, t, J ) 7 Hz, CH3CH2CH2O), 5.15 (2H, s, PhCH2O),
7.31-7.41 (5H, m, Ph); 13C NMR (CDCl3) δ 10.18 (CH3CH2-
CH2O), 22.01 (CH3CH2CH2O), 69.81 (PhCH2O), 70.08 (CH3-
CH2CH2O), 128.32 (Ar), 128.49 (C-4), 128.59 (Ar), 135.36
(C-1), 155.27 (CdO); MS (EI) m/z ) 194. Anal. Calcd for
1
(CdO); H NMR (CDCl3) δ 1.32 (3H, t, J ) 7 Hz, CH3CH2S),
2.88 (2H, q, J ) 7 Hz, CH2S), 5.23 (2H, s, CH2O), 7.32-7.37
(5H, m, Ph); 13C NMR (CDCl3) δ 14.96 (CH3), 25.38 (CH2S),
68.71 (CH2O), 128.56 (Bn), 128.52 (Bn), 128.44 (Bn), 135.23
(C-1), 164.31 (CdO); HRMS (EI) calcd 196.05580, found
196.05581. Anal. Calcd for C10H12O2S: C, 61.20; H, 6.16.
Found: C, 61.05; H, 6.11.
C
11H14O3: C, 68.02; H, 7.27. Found: C, 67.68; H, 7.18.
Ben zyl P h en yl Ca r b on a t e 3e.19 The general procedure
was followed with benzyl alcohol as starting alcohol and phenol
as second nucleophile and with the following amendment. The
Ben zyl S-3,5-Dich lor op h en yl Th ioca r bon a te 14c. The
general procedure was followed with n-propanol as starting
alcohol and 3,5-dichlorothiophenol as second nucleophile and
with the following amendment. A NaOH wash (0.1 M, 40 mL)
was also carried out. 1H NMR spectroscopy of the crude
product showed a mixture of disulfide and ca. 33% product.
Flash chromatography (SiO2, 24:1 cyclohexane/CH2Cl2) of the
crude oil gave 19% of product free from disulfide: IR (film)
1731 cm-1 (CdO); 1H NMR (CDCl3) δ 5.27 (2H, s, CH2), 7.35-
7.41 (6H, overlapping m, Ar), 7.44 (2H, apparent d, J ) 2 Hz,
H′-2 and H′-6); 13C NMR (CDCl3) 70.01 (CH2), 128.59 (Bn),
128.71 (Bn), 128.83 (Bn), 129.80 (Ar, C-4), 130.62 (Ar C-1),
132.60 (Ar C-2), 135.19 (Ar C-3, C-Cl), 135.76 (Bn C-1), 168.09
(CdO); HRMS (CI) calcd (M+ + 18) 330.0122, found 330.0127.
n -P r op yl N-n -Bu tyl Ca r ba m a te 6a . The general proce-
dure was followed with n-propanol as starting alcohol and
n-butylamine as second nucleophile. Purification by flash
chromatography (SiO2, cyclohexane/CH2Cl2) yielded 6a as a
clear oil in 42% yield: IR (film) 3336, 1695; 1H NMR (CDCl3)
δ 0.89-0.96 (6H, 2 overlapping t, CH3), 1.27-1.55 (4H, 2
overlapping m, CH2CH2CH2N), 1.62 (2H, tq, J ) 7 Hz, 7 Hz,
CH2CH2O), 3.17 (2H, m, J ) 6 Hz, CH2N), 4.01 (2H, t, J ) 7
Hz, CH2O), 4.69 (1H, br, N-H); 13C NMR (50.6 MHz, CDCl3)
δ 10.29 (CH3CH2CH2O), 13.69 (CH3CH2CH2CH2NH), 19.84
(CH3CH2CH2CH2NH), 22.34 (CH3CH2CH2O), 32.05 (CH3-
CH2CH2CH2NH), 40.62 (CH3CH2CH2CH2NH), 66.25 (CH3-
CH2CH2O), 156.79 (CdO); HRMS (CI) calcd. 160.1337, found
160.1332. Anal. Calcd for C8H17NO2: C, 60.35; H, 10.76; N,
8.80. Found: C, 60.05; H, 10.71; N, 8.68.
1
product was extracted with Et2O. Yield by H NMR was 31%.
The crude oil was purified by flash chromatography (SiO2, 5:4
cyclohexane/ CH2Cl2) and isolated as clear oil: IR (thin film)
1
1762; H NMR (300 MHz, CDCl3) δ 5.27 (2H, s, CH2), 7.16-
7.47 (10H, overlapping m, Ar); 13C NMR (75.5 MHz, CDCl3)
70.32 (CH2), 121.01 (Ar), 126.04 (Ar), 128.54 (Ar), 128.67 (Ar),
128.75 (Ar), 129.46 (Ar), 134.73 (Bn, C-1), 151.08 (Ph, C-1),
153.65 (CdO); MS (EI) m/z ) 228. Anal. Calcd for C14H12O3:
C, 73.67; H, 5.30. Found: C, 73.49; H, 5.23.
Ben zyl Isop r op yl Ca r bon a te 3g.20 Isopropyl alcohol (0.4
mL, 5.22 mmol) was dried over 4 Å molecular sieves and then
stirred with DBU (0.78 mL, 5.22 mmol) in MeCN (7 mL). CO2
was bubbled subsurface, and then the reaction was cooled to
-42 °C. CO2 was bubbled for a further 45 min at this
temperature. The resulting carbonate suspension was trans-
ferred to a solution of methanesulfonic anhydride (1.052 g, 6.04
mmol; 1.16 equiv) in MeCN (3.5 mL) at the same temperature,
over 90 min, by cannula. After being warmed to 5 °C, the
reaction was split into two aliquots. Benzyl alcohol (0.35 mL,
3.38 mmol) and pyridine (0.25 mL, 3.13 mmol) were added to
an aliquot. After 1 min, a precipitate formed and the temper-
ature rose to 23 °C. The reaction was cooled to 5 °C again and
stirred overnight. The reaction was filtered and added to Et2O
(30 mL). The organic layer was washed with HCl (0.2M, 2 ×
30 mL) and brine (30 mL). The solvent was removed in vacuo
to give 282 mg. Analysis indicated the required product, benzyl
alcohol, benzyl mesylate, and a small amount of diisopropyl
carbonate in the ratio 26:39:2:1. Yield by 1H NMR was 108
mg (21%): 1H NMR (CDCl3) δ 1.30 (6H, d, J ) 6 Hz, CH3),
4.89 (1H, septet, J ) 6 Hz, CH), 5.14 (2H, s, CH2), 7.35-7.39
(5H, overlapping m, Ar); 13C NMR (CDCl3) δ 21.73 (CH3), 69.25
(CH2), 72.10 (CH), 128.27 (Ar), 128.40 (C-4), 128.51 (Ar), 135.32
(C-1), 154.59 (CdO); HRMS (CI) calcd for C11H14O3 195.1021,
found 195.1019.
n -P r op yl N-P h en yl Ca r ba m a te 6b.21 The general proce-
dure was followed with n-propanol as starting alcohol and
aniline as second nucleophile and with DBU (1 equiv).
Crystallization from cyclohexane yielded 6b as a white solid
in 28% yield: IR (CH2Cl2) 1709 cm-1 (CdO); 1H NMR (CDCl3)
δ 0.98 (3H, t, J ) 7 Hz, CH3), 1.70 (2H, tq, J ) 7 Hz, 7 Hz,
CH2CH2O), 4.13 (2H, t, J ) 7 Hz, CH2), 6.64 (1H, br, NH),
7.03-7.08 (1H, apparent t, J ) 7 Hz, H-4), 7.27-7.40 (4H, m,
Ar);13C NMR (CDCl3) δ 10.33 (CH3), 22.24 (CH2), 66.83 (CH2),
118.58 (C-2), 123.29 (C-4), 129.01 (C-3), 137.93 (C-1), 153.69
(CdO); MS (EI) m/z ) 179. Anal. Calcd for C10H13NO2: C,
67.02; H, 7.31; N, 7.82. Found: C, 67.05; H, 7.33; N, 7.81.
Ben zyl N-n -P r op yl Ca r ba m a te 6c.22 The general proce-
dure was followed with benzyl alcohol as starting alcohol and
propylamine as second nucleophile. Crystallization from hex-
ane gave 6c. As estimated by 1H NMR spectroscopy, this
product was 70% pure, representing a 48% yield of 6c. No
n -P r op yl S-3,5-Dich lor op h en yl Th ioca r bon a te 14a . The
general procedure was followed with n-propanol as starting
alcohol and 3,5-dichlorothiophenol as second nucleophile and
with the following amendment. Analysis by TLC after 1 h
showed a negligible amount of thiol remaining. The reaction
was stirred overnight at ambient temperature. The crude
mixture was diluted with Et2O (50 mL) and washed with H2-
SO4 (0.2 M, 2 × 50 mL), NaOH (0.1 M, 2 × 50 mL), and brine
(50 mL). The solution was dried (K2CO3) and filtered, and the
solvent was removed in vacuo. Purification by flash chroma-
(18) Toa Agricultural Chemical Co., Ltd. J P 6814692, 1968; Chem.
Abstr. 1970, 72, 43193u.
(19) Wakamori, S. Agr. Biol. Chem. 1969, 33, 1367.
(20) Bakhtiar, C.; Smith, E. H. J . Chem. Soc., Perkin. Trans. 1 1994,
239.
(21) Manitto, P.; Speranza, G.; Fontana, G.; Galli, A. Helv. Chim.
Acta 1998, 81, 2005.
(22) Benalil, A.; Roby, P.; Carboni, B.; Vaultier, M. Synthesis 1991,
9, 787.
J . Org. Chem, Vol. 68, No. 14, 2003 5443

Bratt and Taylor
further purification was possible: 1H NMR (CDCl3) δ 0.91 (3H,
t, J ) 7 Hz, CH3), 1.51 (2H, tq, J ) 7 Hz, 7 Hz, CH2CH2N),
3.15 (2H, br m, CH2N), 5.09 (2H, s, CH2O), 7.28-7.37 (5H, m,
Ph); 13C NMR (CDCl3) δ 11.15 (CH3), 23.13 (CH2), 66.52 (CH2),
126.92 (Ar), 127.93 (Ar), 128.36 (Ar), 136.53 (C-1), 156.37
(CdO); HRMS (CI) calcd 194.1180, found 194.1181.
stirred in anhydrous MeCN (8 mL). The flask was purged with
CO2 before cooling to -20 °C, and then CO2 was bubbled
subsurface for 1 h. Methanesulfonic anhydride (0.96 g, 5.51
mmol) in anhydrous MeCN (3 mL) was then transferred
dropwise to the carbamate solution. After addition was com-
plete, the solution was kept cool for 30 min and then allowed
to warm to rt with stirring. n-Propanol (0.4 mL, 5.35 mmol)
and Et3N (0.7 mL, 5.02 mmol) were added, and after being
stirred for 1 h, the reaction was heated at reflux overnight.
After cooling, the solvent was removed in vacuo and the
residue was dissolved in CH2Cl2 (100 mL), solids were filtered
off, and the organic solution was washed with H2SO4 (0.125
M, 2 × 100 mL) and brine (100 mL). The crude product was
dried (MgSO4), filtered, and concentrated to dryness (510 mg).
Flash chromatography (SiO2, 4:1 hexane/EtOAc) gave the
product as a clear oil (200 mg, 26%): 1H NMR (250 MHz,
CDCl3) δ 0.95 (3H, t, J ) 7 Hz, CH3CH2CH2O), 1.12 (6H, t, J
) 7 Hz, CH3CH2N), 1.65 (2H, tq, J ) 7 Hz, 7 Hz, CH3CH2-
CH2O), 3.29 (4H, br m, CH3CH2N), 4.03 (2H, t, J ) 7 Hz, CH3-
CH2CH2O); 13C NMR (CDCl3) δ 10.54 (CH3CH2CH2O), 14.25
(CH3CH2N), 14.39 (CH3CH2N), 22.49 (CH3CH2CH2O), 41.43
(CH3CH2N), 41.74 (CH3CH2N), 66.59 (CH3CH2CH2O), 156.20
(CdO); MS (CI) m/z ) 160 (MH+). Anal. Calcd for C8H17NO2:
C, 60.35; H, 10.76; N, 8.80. Found: C, 59.90; H, 10.76; N, 8.55.
Ben zyl N-P h en yl Ca r ba m a te 6d .23 The general procedure
was followed with benzyl alcohol as starting alcohol and aniline
as second nucleophile. Purification by flash chromatography
(SiO2, cyclohexane/CH2Cl2) and recrystallization from diethyl
ether/hexane gave 6d as a white solid in 28% yield: IR 1726
cm-1; 1H NMR (CDCl3) δ 5.21 (2H, s, CH2), 6.65 (1H, br, N-H),
7.04-7.09 (1H, m, J ) 7 Hz, para H of PhNH), 7.28-7.40 (9H,
overlapping m, Ar); 13C NMR (CDCl3) δ 66.89 (CH2), 118.70
(Ph C-4), 123.40 (Bn C-2), 128.16 (Ar), 128.21 (Ar), 128.49 (Ar),
128.91 (Ar), 135.92 (Bn C-1), 137.68 (Ph C-1), 153.44 (CdO);
HRMS (EI) calcd 227.0946, found 227.0945 Anal. Calcd for
C
14H13NO2: C, 73.99; H, 5.77; N, 6.16. Found: C, 74.02; H,
5.81; N, 6.17.
n -P r op yl N,N-Dieth yl Ca r ba m a te 6e.24 The general
procedure was followed with n-propanol as starting alcohol and
diethylamine as second nucleophile. Product 6e was isolated
in 35% yield, as below.
n -P r op yl N,N-Dieth yl Ca r ba m a te 6e. Diethylamine (0.5
mL, 4.83 mmol) and DBU (0.9 mL, 6.03 mmol; 1.25 equiv) were
Ack n ow led gm en t. We thank the EPSRC for finan-
cial support.
(23) Ninomiya, K.; Dhioiri, T.; Yamada, S. Tetrahedron 1974, 30,
215; see also ref. (5).
(24) Yoshida, Y.; Ishii, S.; Watanabe, M.; Yamashita, T. Bull. Chem.
Soc. J pn. 1989, 62, 1534.
J O026753G
5444 J . Org. Chem., Vol. 68, No. 14, 2003