Synthesis of chlorothioformates from xanthates

Article abstract of DOI:10.1055/s-2006-950356

The Vilsmeier reagent derived from N-formylmorpholine produces chlorothioformates from primary and secondary alkyl xanthates. The major side products are the corresponding alkyl chlorides. Secondary alkyl chlorothioformates give lower yields due to their instability. Treating xanthates with other common chlorinating agents (oxalyl chloride, thionyl chloride) gives only dialkyl thiodicarbonates. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-950356

4118
SHORT PAPER
Synthesis of Chlorothioformates from Xanthates
C
hlorothioformatesfro
e
m
X
anthate
g
s
an A. Fikse, William E. Bylund, Nicolas E. Holubowitch, Christopher J. Abelt*
Department of Chemistry, College of William and Mary, PO Box 8795, Williamsburg, VA 23187, USA
Fax +1(757)2212715; E-mail: cjabel@wm.edu
Received 26 May 2006; revised 21 August 2006
O
N
O
Vilsmeier
adduct
Abstract: The Vilsmeier reagent derived from N-formylmorpho-
line produces chlorothioformates from primary and secondary alkyl
xanthates. The major side products are the corresponding alkyl
chlorides. Secondary alkyl chlorothioformates give lower yields
due to their instability. Treating xanthates with other common chlo-
rinating agents (oxalyl chloride, thionyl chloride) gives only dialkyl
thiodicarbonates.
S
S
N
+
R
R
R
O
O
S
S
Cl
O
O
S
H
H
1
2
3
1
S
S
Y
Cl
R
R
Y
Key words: halogenation, xanthate desulfurization, Vilsmeier
reagent, chlorothioformate synthesis, chlorothionocarbonates
5
1
Y
S
6, Y=Cl
7, Y=OPOCl2
R
O
Cl
4
Chlorothioformates are used as alkoxythiocarbonylating
agents.1 Small alkyl chlorothioformates are prepared by
the reaction of thiophosgene with a potassium alkoxide.1,2
Optimal conversion requires conducting the reaction at
–65 °C and adding the alkoxide to the thiophosgene.
While this method is reported to give reasonable yields
with alkyl groups of four carbons or less, it has two draw-
backs. First, with larger alkyl chains the solubility of the
potassium alkoxide salts at low temperatures becomes
problematic. Second, the O,O-dialkyl thiocarbonate by-
product is produced in significant amounts (~20% when
R = butyl).1 This by-product is formed by reaction of the
chlorothioformate with another alkoxide. Thus, even at
low temperature the alkoxide reacts at similar rates with
both thiophosgene and the chlorothioformate. One possi-
ble method to overcome the disubstitution problem is to
introduce chloride rather than alkoxide in the last step.
This paper describes such a route using xanthates as the
precursor to chlorothioformates. It focuses attention on
the preparation of larger alkyl chlorothioformates to com-
plement the present method.
Xanthates have been known since 1822.3 They are readily
prepared by the nucleophilic addition of an alkoxide to
carbon disulfide. The chlorination of these salts seems
like a trivial transformation. The analogous reaction of
carboxylates with chlorinating agents gives acid chlo-
rides. However, xanthates react with common chlorinat-
ing agents, such as oxalyl chloride and thionyl chloride, to
give dialkyl thiodicarbonates 5 (Scheme 1). We find that
this transformation occurs even when the xanthate is add-
ed slowly to an excess of chlorinating agent at –78 °C.
Thus, the xanthate 1 reacts faster with the chlorothiofor-
mate product 4 than with oxalyl chloride. Sulfur nucleo-
Scheme 1 Reaction of xanthates with the Vilsmeier reagent
philes are known to prefer the softer thiocarbonyl
electrophile to the harder carbonyl electrophile.4
To avoid the production of the thiodicarbonate, the chlo-
rothioformate must not be generated in the presence of the
xanthate. The synthetic route must involve an intermedi-
ate, such as the adduct between a Vilsmeier reagent and a
xanthate. We find that the decomposition of this adduct
can yield the desired chlorothioformate in moderate yields
under the right conditions.
The viability of this method was gauged through an NMR
experiment. Butyl potassium xanthate (1.0 mL, 0.25 M in
CD3CN) was added to a chilled (–10 °C) solution of the
Vilsmeier reagent derived from N-formylmorpholine and
phosphorus oxychloride (1.0 mL, 0.5 M in CD3CN). The
spectrum of the reaction mixture taken within two minutes
of mixing showed complete conversion to the Vilsmeier
adduct 3 (Scheme 1). The formyl hydrogen of the Vils-
meier adduct appeared at 9.9 ppm, while that of the excess
Vilsmeier reagent appeared at 9.6 ppm. No chlorothiofor-
mate was discerned over 15 minutes while the sample was
held below 0 °C. When the sample was warmed to room
temperature, the triplet methylene signal at 4.6 ppm of bu-
tyl chlorothioformate started to grow in slowly at the ex-
pense of the triplet signal of the Vilsmeier adduct at 4.8
ppm.
The undesired thiodicarbonate can be formed from the
chlorothioformate 4 or from the Vilsmeier adduct 3
(Scheme 1). Mixing the Vilsmeier reagent and the xan-
thate at low temperatures prevents premature production
of the chlorothioformate. Using excess Vilsmeier reagent
inhibits the less competitive reaction between the xanthate
and the Vilsmeier adduct. Two equivalents of the Vils-
meier reagent are sufficient to suppress the formation of
the thiodicarbonate.
SYNTHESIS 2006, No. 24, pp 4118–4120
x
x.
x
x
.
2
0
0
6
Advanced online publication: 02.11.2006
DOI: 10.1055/s-2006-950356; Art ID: M03806SS
© Georg Thieme Verlag Stuttgart · New York

SHORT PAPER
Chlorothioformates from Xanthates
4119
While thiodicarbonate formation can be nearly eliminat- Table 1 shows that the efficiency of the chlorothioformate
ed, the by-products resulting from nucleophilic substitu- synthesis is inversely related to the ease of nucleophilic
tion (6 and 7, Scheme 1) are more troublesome. Chlorides substitution of the alkyl group. Simple primary alkyl
6 are the major substitution product with primary sub- groups with branching give the best yields followed by
strates. Substitution can occur with both the Vilsmeier ad- linear alkyl groups, a trend that follows SN2 reactivity.
duct and the chlorothioformate. Analysis of the early With smaller alkyl groups, the alkyl chloride is sufficient-
stages of the butyl Vilsmeier adduct reaction indicates that ly volatile to be removed easily by vacuum distillation.
this ambident electrophile favors formation of the chlo- Secondary alkyl groups give smaller yields due to their in-
rothioformate over chlorobutane by a factor of four. It is herent liability. The benzyl derivative gives only benzyl
known that the subsequent decomposition of the chlo- chloride. With the benzyl and secondary alkyl groups, the
rothioformate is enhanced by heat and nucleophilic catal- product efficiency is inversely related to the correspond-
ysis.5 As a result, the substitution products will dominate ing carbocation stability, indicative of an SN1 decomposi-
with long reaction times. Since the Vilsmeier reagent and tion mechanism. The exception is the cyclohexyl group
the xanthate are both salts, a polar solvent such as aceto- that ionizes more slowly than expected due to an increase
nitrile is necessary for the initial addition. After addition, in strain.8 The half-life of cyclohexyl chlorothioformate in
the reaction is diluted with diethyl ether. The amount of chloroform-d at 8 °C is five hours. In all cases there is lit-
acetonitrile required is minimized by using the formyl- tle loss of chlorothioformate by rapid vacuum distillation.
morpholine/phosphorus oxychloride Vilsmeier reagent,
In conclusion, the reaction of xanthates with Vilsmeier re-
which is typically formed as an oil, and by using tetraalkyl
agents gives moderate yields of primary chlorothiofor-
ammonium xanthates, which are extremely soluble in ace-
mates and lower yields of inherently unstable secondary
tonitrile. Not diluting with a non-polar solvent gives rise
chlorothioformates. This method is useful with large alkyl
to substantially more substitution product. The addition of
groups where the alkoxide/thiophosgene method is less
diethyl ether gives a biphasic mixture that helps shield the
viable.
chlorothioformate product from nucleophiles. Finally, the
reaction time is minimized by using the 4-(chloromethyl-
ene)morpholinium Vilsmeier reagent because the result-
Xanthates
ing Vilsmeier adduct reacts faster than that derived from
Potassium alkyl xanthates were prepared by adding carbon disulfide
N,N-dimethylformamide.6,7
to a solution of KOH in the alkyl alcohol or to a slurry of finely
ground KOH in the alcohol.3,9,10 They were purified by precipitation
from acetone with PE.3 The potassium xanthates were converted to
Table 1 Preparation of Chloroformates from Xanthates
R
Yield (%)
Crude product
composition (4:6:7)
Distillate
composition (4:6:7)
Bp (°C), (mmHg)
74–79, (30)
67–71, (30)
32–36, (0.1)
32–28, (0.1)
32–35, (0.1)
63–66, (0.1)
46–47, (0.1)
1H NMR, CHnO
13C, CHnOC=S
n-butyl
37
54
47
47
20
33
73
62:37:1
93:7:<1
80:19:1
81:18:1
58:16:26
68:17:4a
97:3:<1
95:1:4
4.54 (t)
78.0, 186.7
isobutyl
n-pentyl
isopentyl
2-pentyl
cyclohexyl
2-ethylbutyl
98:<1:2
98:<1:2
97:1:2
4.31 (d)
83.9, 186.7
4.54 (t)
78.3, 186.8
4.59 (t)
76.9, 186.8
69:1:30
80:7:9a
99:1:<1
5.43 (sext)
86.6, 186.1
5.31 (sept)
87.7, 185.8
4.46 (d)
80.5, 187.0
benzyl
0
0:100:0
74:24:1
0:100:0b
99:<1:1
–
–
n-octyl
56
78–79, (0.1)
4.54 (t)
78.4, 186.8
a Remainder is cyclohexene.
b PhCH2Cl yield is 43%.
Synthesis 2006, No. 24, 4118–4120 © Thieme Stuttgart · New York

4120
M. A. Fikse et al.
SHORT PAPER
their tetraethylammonium salts by warming the potassium xanthate
(25 mmol) and tetraethylammonium chloride (4.15 g, 25 mmol) in
acetone (200 mL) with stirring for 1 h, filtering the precipitated
KCl, concentrating the filtrate in vacuo, and drying the residue in
vacuo (0.1 Torr) overnight.
ments for the chlorothioformates and dichlorophosphonates was
made through independent syntheses using thiophosgene and
POCl3, respectively. Pyridine (1 mmol) was added to a cooled
(0 °C) solution of the corresponding alcohol (1.0 mmol) and chlo-
ride (2.0 mmol) in THF (2 mL). After stirring for 15 min, PE (10
mL) was added, and the mixture was washed with aq H2SO4
(2 × 2mL) and H2O (2 × 2 mL), dried over Na2SO4, filtered, and
concentrated in vacuo.
Vilsmeier Reagent
A solution of POCl3 (7.70 g, 50 mmol) in Et2O (20 mL) was added
dropwise to a solution of N-formylmorpholine (5.75 g, 50 mmol) in
Et2O (20 mL) under N2. Stirring was continued overnight, deposit-
ing the Vilsmeier reagent as a lower golden oil layer. The clear up-
per Et2O layer was removed by vacuum aspiration through a small-
diameter plastic tube.
Acknowledgment
Acknowledgment is made to the Donors of The Petroleum Research
Fund, administered by the American Chemical Society, for their
support of this research.
Alkyl Chlorothioformates
The Vilsmeier reagent (50 mmol) was dissolved in MeCN (5 mL),
and the solution was cooled to –15 °C in an ice–acetone bath under
N2. A solution of the tetraethylammonium xanthate in MeCN (5
mL) was added dropwise slowly with rapid stirring. When addition
was complete, Et2O (80 mL) was added dropwise while maintaining
the bath temperature below –5 °C. The cooling bath was removed,
and the reaction was stirred for 3–4 h. The upper Et2O layer was de-
canted from the reaction mixture. The lower layer was extracted
with PE (3 × 100 mL). The Et2O layer was combined with the PE
washings, and the resulting solution was washed with ice-cold H2O
(3 × 200 mL), dried over Na2SO4, and concentrated in vacuo. The
residue was distilled in vacuo. Boiling points and yields are given
in Table 1.
References
(1) Martinez, M. A.; Vega, J. C. Synthesis 1986, 760.
(2) Zaim, O. Tetrahedron Lett. 1999, 40, 8059.
(3) Rao, S. R. Xanthates and Related Compounds; Marcel
Dekker: New York, 1971.
(4) Castro, E. A. Chem. Rev. 1999, 99, 3505.
(5) Barany, G.; Schroll, A. L.; Mott, A. W.; Halsrud, D. A. J.
Org. Chem. 1983, 48, 4750.
(6) Bergman, J.; Stalhandske, C. Tetrahedron 1996, 52, 753.
(7) Katritzky, A. R.; Shcherbakova, I. V. Can. J. Chem. 1992,
70, 2040.
(8) Masson, E.; Leroux, F. Helv. Chim. Acta 2005, 88, 1375.
(9) Larsen, J.; Lenoir, C. Org. Synth. 1995, 72, 265.
(10) Price, C. C.; Stacy, G. W. Org. Synth. 1948, 28, 82.
Product ratios reported in Table 1 were determined through GC/MS
and/or 1H NMR integration. We noted that chlorothioformates will
partially decompose to alkyl chlorides when the GC inlet tempera-
ture exceeds 100 °C. Confirmation of the NMR and GC/MS assign-
Synthesis 2006, No. 24, 4118–4120 © Thieme Stuttgart · New York