Conversion of bis(trichloromethyl), carbonate to phosgene and reactivity of triphosgene, diphosgene, and phosgene with methanol

Article abstract of DOI:10.1021/jo000820u

Triphosgene was decomposed quantitatively to phosgene by chloride ion. The reaction course was monitored by IR spectroscopy (React-IR), showing that diphosgene was an intermediate. The methanolysis of triphosgene in deuterated chloroform, monitored by proton NMR spectroscopy, gave methyl chloroformate and methyl 1,1,1-trichloromethyl carbonate in about a 1:1 ratio, as primary products. The reaction carried out in the presence of large excess of methanol (0.3 M, 30 equiv) was a pseudo-first-order process with a k(obs) of 1.0 x 10^-4 s^-1. Under the same conditions, values of k(obs) of 0.9 x 10^-3 s^-1 and 1.7 x 10^-2 s^-1 for the methanolysis of diphosgene and phosgene, respectively, were determined. The experimental data suggest that, under these conditions, the maximum concentration of phosgene during the methanolysis of triphosgene and diphosgene was lower than 1 x 10^-5 M. Methyl 1,1,1-trichloromethyl carbonate was synthesized and characterized also by the APCI-MS technique.

Full text of DOI:10.1021/jo000820u

8224
J . Org. Chem. 2000, 65, 8224-8228
Con ver sion of Bis(tr ich lor om eth yl) Ca r bon a te to P h osgen e a n d
Rea ctivity of Tr ip h osgen e, Dip h osgen e, a n d P h osgen e w ith
Meth a n ol1
Lucia Pasquato,*,† Giorgio Modena,† Livius Cotarca,*,‡,§ Pietro Delogu,‡ and Silvia Mantovani‡
Centro CNR Meccanismi di Reazioni Organiche† and Dipartimento di Chimica Organica Universita` di
Padova, via Marzolo 1, 35131 Padova, Italy, and Industrie Chimiche Caffaro SpA,‡ Centro Ricerche
Torviscosa, P-le F. Marinotti 1, 33050 Torviscosa Udine, Italy
pasquato@chor.unipd.it
Received May 30, 2000
Triphosgene was decomposed quantitatively to phosgene by chloride ion. The reaction course was
monitored by IR spectroscopy (React-IR), showing that diphosgene was an intermediate. The
methanolysis of triphosgene in deuterated chloroform, monitored by proton NMR spectroscopy,
gave methyl chloroformate and methyl 1,1,1-trichloromethyl carbonate in about a 1:1 ratio, as
primary products. The reaction carried out in the presence of large excess of methanol (0.3 M, 30
equiv) was a pseudo-first-order process with a kobs of 1.0 × 10-4 s-1. Under the same conditions,
values of kobs of 0.9 × 10-3 s-1 and 1.7 × 10-2 s-1 for the methanolysis of diphosgene and phosgene,
respectively, were determined. The experimental data suggest that, under these conditions, the
maximum concentration of phosgene during the methanolysis of triphosgene and diphosgene was
lower than 1 × 10-5 M. Methyl 1,1,1-trichloromethyl carbonate was synthesized and characterized
also by the APCI-MS technique.
In the past decade there has been a growing interest
aspect there are two key points to examine: the mech-
anism of action of triphosgene toward nucleophiles both
in the presence and in the absence of catalysts and if
during these reactions phosgene is formed in harmless
quantities.
in the use of triphosgene (bis(trichloromethyl) carbonate
or BTC) in organic synthesis. Its properties as well as
the synthetic application have been recently reviewed.2
The very large number of patents with respect to the
cumulative number of papers found in Chemical Ab-
stracts (85% in 1996) shows the industrial relevance of
this reagent.3 The main reason for such intensive ap-
plication of BTC in synthesis may be ascribed to two
different aspects, namely: (i) the use of triphosgene as a
safer solid and easier to handle compared to phosgene4,5
and (ii) the specific reactivity of triphosgene that enables
the preparation of unsymmetrical ureas,6 carbamoyl
chlorides and isocyanates,7 unsymmetrical carbonates,8
and carbamates.9 Regarding the first aspect, a central
point is to understand if, and under what conditions,
triphosgene releases phosgene and consequently how
triphosgene can be used safely. Regarding the second
Our interest in the chemistry of triphosgene (1) has
prompted us to investigate the depolymerization of
triphosgene in the presence of chloride ions.10 Particularly
interesting was monitoring the process continuously
using the instrument React-IR since it revealed the
presence of diphosgene (2) as an intermediate and hence
allows the mechanism of the process to be understood.
Furthermore, we investigated the reaction of triphosgene
(1) with methanol (as a model for a weak nucleophile) in
the presence, as well as in the absence, of chloride ions.
For comparison we studied the same reaction with
diphosgene (2) and phosgene (3). Preliminary data,
obtained by monitoring the reactions with proton NMR
spectra, clearly showed that phosgene is the most reactive
species. In addition, on the basis of the experimental
data, we may propose a mechanism for the methanolysis
† CMRO-CNR, University of Padova.
‡ Industrie Chimiche Caffaro.
§ Current address: Zambon Group SpA, Centro Ricerche, 36045
Almisano di Lonigo (VI), Italy. e-mail: livius.cotarca@zambongroup.com.
(1) Part of this work has been presented at the XIII Congresso
Nazionale della Divisione di Chimica Industiale, Napoli, J une 20-23,
1999.
(7) See for example: (a) Pattenden, G.; Reynolds, S. J . J . Chem. Soc.,
Perkin Trans. 1 1994, 379. (b) Pattenden, G.; Reynolds, S. J . Tetra-
hedron Lett. 1991, 32, 259. (c) Blommaert, A. G. S.; Weng, J .-H.;
Dorville, A.; McCort, I.; Ducos, B.; Durieux C.; Roques, B. P. J . Med.
Chem. 1993, 36, 2868. (d) Nowick, J . S.; Holmes, D. L.; Noronha, G.;
Smith, E. M.; Nguyen, T. M.; Huang, S.-L. J . Org. Chem. 1996, 61,
3929. (e) Weiberth, F. J . Tetrahedron Lett. 1999, 40, 2895.
(2) Cotarca L.; Delogu, P.; Nardelli, A,; Sunjic, V. Synthesis 1996,
553.
(3) In the year 1996 we found 117 records in the Chemical Abstracts
and among them 99 patents.
(4) Eckert, H.; Forster, B. Angew. Chem., Int. Ed. Engl. 1987, 26,
894.
(5) Aldrichimica Acta 1988, 21, 47. ALDRICH Technical Information
Bulletin AL-176.
(6) See for example: (a) Majer, P.; Randad, R. S. J . Org. Chem. 1994,
59, 1937. (b) Cotarca, L.; Bacaloglu, R.; Csunderlik, C.; Marcu, N.;
Tarnaveanu, A. J . Prakt. Chem. 1987, 329, 1052. (c) Andre´, F.;
Marroud, M.; Boussard, G. Tetrahedron Lett. 1996, 37, 183. (d)
Purchase, C. F.; White, A. D.; Anderson, M. K.; Bocan, T. M. A.;
Bousley, R. F.; Hamelehle, K. L.; Homan, R.; Krause, B. R.; Lee, P.;
Mueller, S. B.; Speyer, C.; Stanfield, R. L.; Reindel, J . F. Bioorg. Med.
Chem. Lett. 1996, 6, 1753.
(8) See for example: (a) Rohrbach, P.; Heitz, T.; Keul, H.; Hoecker,
H. Makromol. Chem. 1993, 194, 1627. (b) Kricheldorf, H. R.; Luebbers,
D. Macromolecules 1990, 23, 2656. (c) J ochims, J . C.; Hehl, S.;
Herzberger, S. Synthesis 1990, 1128. (d) Larock, R. C.; Lee, N. H.
Tetrahedron Lett. 1991, 32, 6315. (e) Kang, S.-K.; J eon, J .-H.; Nam,
K.-S.; Park, C.-H.; Lee, H.-W. Synth. Commun. 1994, 24, 305.
(9) See for example: (a) Ghosh, A. K.; Duong, T. T.; McKee, S. P.
Tetrahedron Lett. 1991, 32, 4251. (b) Scialdone, M. A.; Shuey, S. W.;
Soper, P.; Hamuro, Y.; Burns, D. M. J . Org. Chem. 1998, 63, 4802.
(10) Decomposition of triphosegene and diphosgene to phosgene have
been observed: Senet, J .-P. The Recent Advance in Phosgene Chemistry;
SNPE Group: Nanterre, 1997; Vol. 1, p 40.
10.1021/jo000820u CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/08/2000

Conversion of Bis(trichloromethyl) Carbonate to Phosgene
J . Org. Chem., Vol. 65, No. 24, 2000 8225
Ch a r t 1
The spectroscopic data definitely indicate the presence
of three detectable species: the reagent, triphosgene (1);
the product, phosgene (3); and an intermediate. A de-
tailed and careful analysis of the IR spectra allows the
identification of the absorption bands pertaining un-
doubtedly to diphosgene (2). The profiles reported in
Figure 1 represent the variation of absorbance vs time
for triphosgene (ν ) 1177 cm-1), phosgene (ν ) 836 cm-1),
and diphosgene (ν ) 1054 cm-1). In Figure 1 the absor-
bance of diphosgene changes as expected for a reactive
intermediate which builds up at early stages of the
reaction and then decreases.
of triphosgene that is also a rationalization for the
formation of unsymmetrical carbonates or ureas. Finally,
an efficient synthesis of the methyl 1,1,1-trichloromethyl
carbonate (4), which is a primary product in the metha-
nolysis of triphosgene and diphosgene, and complete
characterization are reported.
The phosgene produced during the process was con-
densed into a methanol solution at -12 °C. This solution,
at the end of the reaction, was subjected to quantitative
analysis to determine the dimethyl carbonate formed.
Indeed, the reaction is quantitative, and three equiva-
lents of dimethyl carbonate were formed per mole of
triphosgene.
Resu lts a n d Discu ssion
The above experimental evidence points to the mech-
anism outlined in Scheme 1, centered on the nucleophilic
role of the chloride ion.
Con ver sion of Tr ip h osgen e to P h osgen e by Ch lo-
r id e Ion s. We carried out several experiments of decom-
position of triphosgene (1) in different solvents or undi-
luted using catalytic amounts of tetrabutylammonium
salts (5-10% w/w), to determine the products of the
reaction and the stoichiometry of the process. In addition,
some of these reactions have been monitored by means
of the mobile probe of the instrument React-IR-MP, which
records several IR spectra per minute and allows statisti-
cal analysis of the spectroscopic data. In a typical
experiment a solution of triphosgene in n-hexane (10%
w/w) was treated with Aliquat 336 (3 mol % vs triphos-
gene) added at one time, and the reaction, carried out
at room temperature, was monitored with the system
React-IR.
The chloride ion, under these reaction conditions, is a
strong nucleophile that attacks the carbonyl carbon of
triphosgene giving diphosgene and phosgene. In turn,
diphosgene reacts with chloride ion to give two molecules
of phosgene. The chloride ion is needed in catalytic
amount since it is not consumed in the overall process.
The results outlined above suggest that phosgene can
be used safely by producing it in desired quantities from
triphosgene, using a catalytic amount of chloride ions
(3-5%) in a flask directly connected to the reaction vessel
where the phosgene is needed as a reagent. In this
respect we can confirm that triphosgene is a solid
phosgene equivalent.
F igu r e 1. Absorbance vs time (s) as monitored by React-IR in the Aliquat (3%) catalyzed conversion of triphosgene (11% w/w)
to phosgene carried out in n-hexane at room temperature.
Sch em e 1

8226 J . Org. Chem., Vol. 65, No. 24, 2000
Pasquato et al.
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n ts for th e
Rea ction s of P h osgen e, Dip h osgen e, a n d Tr ip h osgen e
(0.01 M) w ith Meth a n ol (0.3 M) in CDCl3 a t 25 °C,
Ca lcu la ted fr om th e In itia l Ra tes
kobs, s-1
MeOH, 0.3 M,
Cl- 5%a
MeOH, 0.3 M,
Cl- 10%a
substrate
MeOH, 0.3 M
phosgene
diphosgene
triphosgene
1.7 × 10-2
9.1 × 10-4
1.0 × 10-4
b
b
1.0 × 10-3
1.1 × 10-3
2.3 × 10-4
2.3 × 10-4
Added as Bu4N+Cl-. Too fast to be measured by NMR.
a
b
F igu r e 2. Reaction of triphosgene (0.01 M) with methanol
(0.3 M) in CDCl3 at 25 °C.
Rea ction s of Tr ip h osgen e, Dip h osgen e, a n d P h os-
gen e w ith Meth a n ol. We carried out two sets of
experiments, both in the absence of catalyst and in the
presence of chloride ions (added as tetrabutylammonium
chloride). All the reactions were carried out in deuterated
chloroform (freshly distilled before use) in the presence
of an excess of methanol (0.3 M, 30 equiv) at 25 °C in
screw cap NMR tubes, and they were monitored by
proton NMR (400 MHz, 1H selective probe) at appropriate
regular intervals.
The reaction of triphosgene with methanol forms a
mixture, one to one, of methyl 1,1,1-trichloromethyl
carbonate11 (4) and methyl chloroformate (5) as reported
in an earlier mechanistic study.4,12 Only with a very long
reaction time (several days) was dimethyl carbonate
obtained as a final product.13 The formation of products
4 and 5 was followed by integration of the corresponding
singlets at 3.941 and 3.948 ppm, respectively, and the
small difference to the one-to-one ratio is likely due to
the integration error due to imperfect resolution of the
two peaks. A typical profile of concentration vs time is
reported in Figure 2. The pseudo first-order rate constant
for this reaction is reported in Table 1.
The reaction of diphosgene (2) (0.01 M) with 30 equiv
of methanol in chloroform gives a mixture of methyl 1,1,1-
trichloromethyl carbonate (4) and methyl chloroformate
(5) in a 8.8 to 3.2 ratio. A typical profile of this reaction
is reported in Figure 3, and the relative pseudo-first-order
rate constant is reported in Table 1.
Phosgene (0.01 M) reacts with 30 equiv of methanol,
giving methyl chloroformate as the only product. The
reaction is very fast at 25 °C, and the estimate pseudo-
first-order rate constant is reported in Table 1.
F igu r e 3. Reaction of diphosgene (0.01 M) with methanol (0.3
M) in CDCl3 at 25 °C.
M and 1 × 10-3 M), we monitored the reactions of
triphosgene, diphosgene, and phosgene with methanol
(0.3 M, 30 equiv). In the first two cases the formation of
the products shows the same qualitative behavior ob-
served in the processes carried out in the absence of
chloride ions. In addition, we could not detect any
appreciable difference using 5% or 10% of chloride ion.
The pseudo-first-order rate constants for these reactions
are reported in Table 1. As the reaction of phosgene,
under these conditions, is too fast (it goes to completion
in the mixing time), we could not determine the rate
constant for this process.
On the basis of the experimental data presented for
the methanolysis of triphosgene, we can suggest for this
process the reaction sequence shown in Scheme 2. The
rate-determining step is probably the formation of a
tetrahedral intermediate or transition state, which then
gives the methyl 1,1,1-trichloromethyl carbonate and
phosgene. In turn, phosgene reacts with methanol to give
methyl chloroformate. Since this last reaction is much
faster than the rate-determining step, under these condi-
tions phosgene will not accumulate. This is supported by
the fitting14 of the process using the differential equations
relative to the reactions reported in Scheme 2 which
indicate a maximum concentration of phosgene of 1 ×
10-5 M.
Under the same conditions but in the presence of small
quantities of tetrabutylammonium chloride (0.5 × 10-3
The data obtained for the methanolysis of diphosgene
are in agreement with the mechanistic picture reported
in Scheme 3.
(11) Compound
4 is know; however, we could not find in the
literature its spectroscopic data since its preparation is described only
in ref 15. This is the reason we report its complete characterization in
the next section.
Diphosgene reacts with methanol to form a tetrahedral
intermediate, which has two good leaving groups: the
chloride ion and the trichloromethoxy groups. The loss
of the former gives methyl 1,1,1-trichloromethyl carbon-
ate whereas the detachment of the latter leads to methyl
(12) Cotarca, L.; Bacaloglu, R.; Marcu, N.; Tarnaveanu, A. J . Prakt.
Chem. 1985, 327, 881.
(13) Methyl chloroformate and methyl 1,1,1-trichloromethyl carbon-
ate react very slowly under the reaction conditions used for the
methanolysis of triphosgene. Only 10% of compound 5 (0.01M) with
20 equiv of methanol and 18 equiv of 2,6-di-tert-butylpyridine (to speed
up the reaction) is consumed in 14 h, giving 2 equiv of dimethyl
carbonate. Under the same reaction conditions, methyl chloroformate
(5) reacts faster; about 20% is consumed in the same time course, giving
1 equiv of dimethyl carbonate.
(14) The program Scientist, MicroMath Scientific Software, was
used.

Conversion of Bis(trichloromethyl) Carbonate to Phosgene
Sch em e 2
J . Org. Chem., Vol. 65, No. 24, 2000 8227
Sch em e 3
F igu r e 4. APCI-MS spectrum of methyl 1,1,1-trichloromethyl carbonate.
chloroformate and phosgene which in turn again gives
methyl chloroformate. Therefore, the lability of the
groups is reflected in the ratio between product 4 and
half of product 5, which is about 5. In this case, also,
phosgene should be present during the reaction as an
intermediate with a quite low maximum concentration
in solution. Indeed, the fitting of the experimental data
allows estimation of a maximum phosgene concentration
smaller than 3 × 10-6 M.
nolysis. Thus, the presence of catalytic quantities of
chloride ions may influence the initial part of the reac-
tion, but should be a negligible factor during the progress
of the process.
Syn th esis a n d Ch a r a cter iza tion of Meth yl 1,1,1-
Tr ich lor om eth yl Car bon ate (4). Compound 4 is known
since 1887 when Hentschel15 prepared it by reaction of
diphosgene in refluxing methanol. Later, two other
groups16,17 reported the formation in small amount of 4
by chlorination of dimethyl carbonate together with other
polychlorinated carbonates.
As we described above, methyl 1,1,1-trichloromethyl
carbonate is formed in the methanolysis of triphosgene
in chloroform. This suggested a modification of this
reaction for the production of 4. A solution of methanol
(2 mL) and diisopropylethylamine (9.7 mL) in 160 mL of
dry methylene chloride was slowly added to a solution
of triphosgene (5 g) dissolved in 90 mL of dry methylene
chloride at -10 °C. After workup, the product was
separated from the crude mixture by Kulgheror distilla-
tion (75-80 °C at 15 mmHg) with an 80% yield. This
lachrimatory colorless oil shows a singlet at 3.941 ppm
in the proton NMR spectrum. Its 13C NMR spectrum
presents three signals at 56.241 ppm for the methoxy
carbon, 107.702 ppm for the trichloromethoxy carbon,
As indicated in Schemes 2 and 3, 2 equiv of hydrogen
chloride are formed in the methanolysis of triphosgene
and diphosgene. This should suggest that even using 30
equiv of methanol one would observed an inhibition due
to the solvation of HCl by the methanol (each HCl would
sequester no less than three molecules of methanol).
Higher methanol concentration and constant ionic strength
would obviate this complication. We did not take this
precaution mainly because of the limitations dictated by
the NMR technique used to monitor the process. How-
ever, as a first approximation, the concentration vs time
in all the experiments carried out follows pseudo-first-
order behavior up to 80% of the reaction.
The methanolysis reaction carried out in the presence
of small quantities of chloride ions added as tetrabuty-
lammonium chloride are indicative of the weak effect of
chloride ion as base and the lack of any nucleophilic role
played by chloride ions. Indeed, as shown by the data
reported in Table 1, the reactions carried out in the
presence of 5% or 10% of chloride ion are very similar to
those without added salt. This behavior is attributed to
the formation of hydrogen chloride during the metha-
(15) Hentschel, W. J . Prakt. Chem. 1887, 36, 304.
(16) Kling, A.; Florentin, D.; J acob, E. C. r. d l’ Acad. Sci. 1920,
170, 111.
(17) Grignard, V.; Rivat, G.; Urbain, E. Ann, Chim. 1920, [9], tXIII,
229. Kling, A.; Florentin, D.; J acob, E. C. r. d l’ Acad. Sci. 1919, 170,
234.

8228 J . Org. Chem., Vol. 65, No. 24, 2000
Pasquato et al.
Kin etic Mea su r em en ts. The concentrations of tetrabutyl-
ammonium chloride, triphosgene, diphosgene, and phosgene
were determined by weighing. The increases in the concentra-
tion of methyl chloroformate, 1,1,1-trichloromethyl methyl
carbonate, and dimethyl carbonate were followed by measuring
the 1H NMR integrated area of the methyl resonances at 3.948,
3.941, and 3.640 ppm, respectively. These concentrations in
CDCl3 were determined by comparison of the integrated areas
of appropriate resonances with that of the proton impurity of
the solvent. The concentration of the impurity had been
previously measured by comparison with the signal of 1,4-
dinitrobenzene, at a concentration determined by weighing.
The reactions have been monitored at regular time intervals,
using NMR tubes equipped with airtight screw caps. At least
three or four independent experiments were carried out for
each reaction, and the pseudo-first-order kinetic constants
reported in Table 1 are average values.
and 148.431 ppm for the carbonyl carbon. The carbonyl
IR absorbance is at 1794.7 cm-1. Only operating at
atmospheric pressure in air using an APCI source
coupled with a quadrupolar mass analyzer could we
observe the MH+ ion (m/z 193, 195, 197 in 1:1:0.3 ratio)
as the predominant species, Figure 4. The spectrum
shows also an easy Cl3C+ loss (m/z 117, 119, 121).
Con clu sion s. We have reported a detailed study of
the conversion of triphosgene to phosgene catalyzed by
chloride ions. The experimental data indicate that triph-
osgene can be used as a safe synthetic equivalent of
phosgene. The process we have described allows produc-
tion of the desired quantity of phosgene under controlled
conditions in a vessel directly connected with the reactor
where the phosgene will be used.
Decom p osition of Tr ip h osgen e. In a typical experiment
a 100 mL round-bottom flask equipped with Claisen condenser,
dropping funnel, magnetic stirring, and the mobile probe of
React-IR was charged with 2.1 g (7.07 mmol) of triphosgene
dissolved in 20 mL of n-hexane. The Claisen condenser and
the reactor connected with it, containing 30 mL of methanol,
were kept at -12 °C. At room temperature, a solution of 0.087
g of Aliquat 336 (0.03 equiv, 0.215 mmol) dissolved in 5 mL of
n-hexane was added dropwise under vigorous stirring to the
triphosgene solution. The phosgene that developed during the
reaction was collected in methanol, and the methyl chlorofor-
mate formed was quantitatively determined by GC using
diethyl ether as solvent and dimethyl oxalate as standard. The
IR spectra recorded during the experiment were elaborated,
and a statistical analysis was carried out using the statistical
software Conc-IRT.
We have also reported a thorough investigation of the
methanolysis of triphosgene, diphosgene, and phosgene.
The pseudo-first-order kinetic constants of these reactions
carried out in the presence of excess of methanol clearly
indicate that phosgene reacts more than 2 orders of
magnitude faster than does triphosgene. Thus, under the
reaction conditions used, phosgene does not accumulate
during the methanolysis of triphosgene and its maximum
-5
concentration is always very low (<1 × 10
M). It is
worth noting that both 1,1,1-trichloromethyl methyl
carbonate and methyl chloroformate react with methanol
at a much slower rate than triphosgene. This very
different reactivity between triphosgene and the primary
products formed by reaction with a nucleophile allows
production of unsymmetrical carbonates and unsym-
metrical ureas.
Meth yl 1,1,1-Tr ich lor om eth yl Ca r bon a te (4).17 A 250
mL round-bottom flask equipped with a dropping funnel,
magnetic stirring bar, and condenser was charged with a
solution of 5 g of BTC (0.017 mol) in 90 mL of distilled CH2-
Cl2. A solution of 9.7 mL (0.05 mol) of DIEA and 2 mL of
methanol (0.05 mol) in 160 mL of distilled CH2Cl2 was added
dropwise during 20 min, keeping the reaction vessel at -10
°C. The mixture was washed with a 10% solution of KHSO4
(2 × 200 mL), a 5% solution of NaHCO3 (2 × 200 mL), and
finally brine (1 × 100 mL) and dried over sodium sulfate.
Solvent was removed in vacuo, and the residual oil was
purified by Kugelrorh distillation (75 °C, 15 mmHg). The
product (2.6 g) was obtained as a colorless lachrimatory oil in
80% yield. 1H NMR (200 MHz, CDCl3) δ: 3.94 (3H, s). 13C NMR
(62.9 MHz, CDCl3) δ: 56.241, 107.702, 148.431. FT-IR (solution
0.3 M in CHCl3, cell CaF2 0.1 mm) ν (cm-1): 1797 (s), 1712
(w), 1442 (m), 1254 (s), 1194 (w), 1095 (s). MS (APCI) see
Figure 4. Anal. Calcd for C3H3Cl3O3: C, 18.63; H, 1.56.
Found: C, 18.37; H, 1.48.
Exp er im en ta l Section
Gen er a l. 1H NMR spectra were recorded at 200 MHz, or
at 400 MHz, 13C NMR at 62.9 MHz, using CDCl3 as solvent.
Reaction kinetics of phosgene were performed with a 200 MHz
NMR instrument, and reaction kinetics of triphosgene and
diphosgene were monitored using a 400 MHz (1H selective
probe) NMR instrument. Both instruments were equipped with
a variable temperature unit, and all kinetics were performed
at 25 °C. Commercial CDCl3 was carefully dried before use by
refluxing over P2O5 and then stored with A4 molecular sieves.
Commercial tetrabutylammonium chloride was dried with
P2O5 in a drying pistol. FT-IR measurements were recorded
with the ReactIR-MP (Mobile Probe) system using n-hexane
as solvent. Routine IR spectra were recorded with a FT IR
Perkin-Elmer 1720 X instrument.
APCI-MS measurements were carried out on a TRIO 100
II instrument (Fisons Instruments, Manchester, UK) equipped
with a Fisons APCI source. A schematic of this source and of
the experimental setup used for the introduction of vaporized
samples was given in an earlier paper.18
GC analysis of methyl chloroformate was made on a Perkin-
Elmer Autosystem instrument using a 25 m capillary column
of 25QC2/BPX5 and helium as carrier gas with a flux of 0.9
mL/min. Methyl chloroformate has a Rf of 4.88 min using
T(injector) 290 °C, T(detector) 300 °C, 40 °C × 6 min, 20 °C/
min. For quantitative determinations, dimethyl oxalate was
used as a standard.
Ack n ow led gm en t. We thank Prof. Cristina Para-
disi and Dr. Anna Nicoletti for the mass spectra of
compound 4 and Dr. Giuseppina Vassallo for the
statistical analysis of IR spectra recorded with React-
IR.
Su p p or tin g In for m a tion Ava ila ble: The stacked plot of
IR spectra recorded for a model reaction of conversion of
triphosgene to phosgene catalyzed by chloride ions and the
stacked plot of the proton NMR spectra of a model reaction of
triphosgene (0.01 M) with methanol (0.3 M) in CDCl3 at 25 °C
are reported in Figure S1 and Figure S2, respectively. This
material is available free of charge on the Internet at
http://pubs.acs.org.
Commercial reagents were purchased from standard chemi-
cal suppliers.
(18) Dono`, A.; Paradisi, C.; Scorrano, G. Rapid Commun. Mass
Spectrom. 1997, 11, 1687.
J O000820U