Reinvestigation of the conversion of epoxides into halohydrins with elemental halogen catalysed by thiourea

Article abstract of DOI:10.1016/j.tet.2005.02.065

In contrast to a previous literature report, thiourea is not a catalyst in the ring opening reaction of epoxides by means of bromine or iodine. Instead, thiourea reacts with the halogen to give a complex mixture of products, among them hydrogen halogenides, which are in fact the real epoxide ring opening reactants. The presence of water is crucial in this reaction.

Full text of DOI:10.1016/j.tet.2005.02.065

Tetrahedron 61 (2005) 4233–4235
Reinvestigation of the conversion of epoxides into halohydrins
with elemental halogen catalysed by thiourea
Mirosław Soroka and Waldemar Goldeman
*
Politechnika Wrocławska, Instytut Chemii Organicznej, Biochemii i Biotechnologii, Wybrze z˙ e Wyspia n´ skiego 27,
PL-50370 Wrocław, Poland
Received 30 October 2004; revised 11 February 2005; accepted 24 February 2005
Available online 17 March 2005
Abstract—In contrast to a previous literature report, thiourea is not a catalyst in the ring opening reaction of epoxides by means of bromine
or iodine. Instead, thiourea reacts with the halogen to give a complex mixture of products, among them hydrogen halogenides, which are in
fact the real epoxide ring opening reactants. The presence of water is crucial in this reaction.
q 2005 Elsevier Ltd. All rights reserved.
1
In 2003, a paper was published in this journal describing
3
When we repeated the reaction of methyloxirane with
iodine in the presence of 10 mol% of thiourea as described
the ring opening reaction of epoxides to halohydrins by
means of bromine or iodine, in the presence of thiourea as a
catalyst. The authors also discussed the mechanistic details
of this reaction and proposed a four-step mechanism, where
the thiourea plays an essential role as a catalyst. Their
proposal was supported by UV spectra, and by cited
literature precedents. The authors neither isolated nor
confirmed the presence of the catalyst, namely thiourea, in
the reaction mixture after reaction. Moreover, the reaction
of thiourea with bromine or iodine is well-known. It was
1
by Sharghi and Eskandari, we identified the iodohydrins by
1
means of H NMR spectroscopy (see Section 1.1.1), but
with low yield (only about 30%), and a considerable amount
of unreacted iodine. By titration of the residual iodine, we
estimated the stoichiometry of the reaction as about 1 mol of
iodine per 2 mol of thiourea. This means that the thiourea is
not a catalyst but just the reagent.
A similar reaction of methyloxirane with bromine in the
presence of 10 mol% of thiourea (Section 1.1.2) gave us
only 32% yield of distilled bromohydrins and a huge
amount of a dark unidentified residue. When we changed the
molar ratio of bromine to thiourea to 1 to 2 (Section 1.1.3),
we observed almost quantitative yield of crystalline
formamidine disulfide dihydrobromide, and no bromo-
hydrins at all.
2
a
described for the first time by Claus in 1875, and was
subsequently repeated and studied many times.
2
b–g
Generally, the distribution of the products depends strongly
on the conditions of the reaction. Under anhydrous
conditions in organic solvent, the main product is the
formamidine disulfide dihydrohalogenide. For example,
2
f
King and Ryden applied this product as a reactant for the
preparation of thiazoles. However, under stronger con-
ditions (in an excess of halogen or in the presence of water)
the reaction gives a complex mixture of products which,
When we applied 2 mol of thiourea and 1 mol of iodine
(Section 1.1.4), and then we added an excess of water, the
2
c
according to McGovan, contains cyanamide, hydrogen
halogenide, sulfur and even sulfuric acid.
iodine disappeared in a few minutes, and we found in the
reaction mixture more than 70% yield of iodohydrins, and
the yellow precipitate, which was identified by elemental
analysis and IR spectrum as elemental sulfur, in more than
60% yield. The presence of water seems to be essential to
complete the reaction of iodine with thiourea.
Under such circumstances, it is unlikely that thiourea could
be a catalyst in the reaction described by Sharghi and
Eskandari, and could survive in the presence of such strong
1
oxidizing and electrophilic reagents like bromine or iodine.
Therefore, we decided to reinvestigate this reaction.
Our findings are shown by the sequence in Scheme 1.
In conclusion, we could not reproduce the Sharghi and
Keywords: Oxiranes; Ring opening; 2-Halogenoalkanols; Bromohydrins;
Iodohydrins.
1
Eskandari protocol. The only role the thiourea plays in their
protocol is to generate the hydrogen halogenide which
*
Corresponding author. Tel.: C48 71 3202404; fax: C48 71 3284064;
e-mail: miroslaw.soroka@pwr.wroc.pl
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.02.065

4234
M. Soroka, W. Goldeman / Tetrahedron 61 (2005) 4233–4235
Scheme 1.
reacts in situ with the epoxide. When we reinvestigated this
reaction we found that the thiourea reduces molecular
halogen to the hydrogen halogenide, via formamidine
disulfide dihydrohalogenides. It appears that the author’s
extensive discussion and the final conclusions concerning
the catalytic effect of thiourea are in error.
organic phases were collected and dried over Na SO
2 4
(20 g), filtered and evaporated under vacuo from warm
water bath (below 40 8C) to give oily residue (1.3 g, about
35%) identified and assayed by means of NMR as the
mixture of iodohydrins–1-iodopropan-2-ol and 2-iodo-
propanol in 74/26 ratio (calculated from integrals of methyl
groups at 1.30 and 1.89 ppm, respectively), and traces of
unidentified impurities.
Furthermore, we would not advise use of this protocol for
the preparation of halohydrins from epoxides. However, if
elemental halogen must be used as a reactant for epoxide
ring opening, we advise use of our procedure based on
generation of hydrogen halogenide in the halogenation of
active aromatics, developed some time ago, and recently
When an excess of iodine was titrated by 1.00 M of
Na S O in water, as much as about 90% of unreacted
iodine was found remained after reaction.
2
2 3
4
described.
1
.1.2. Experiment 2. The reaction of bromine with
methyloxirane in the presence of 10 mol% amount of
thiourea—a reinvestigation. Methyloxirane (5.8 g,
100 mmol) was added to a stirred suspension of thiourea
1. Experimental
(
0.76 g, 10 mmol) in acetonitrile (50 mL) at 10 8C (ice-
water bath). Next, a solution of bromine (5.1 mL,
00 mmol) in acetonitrile (20 mL) was added dropwise
30 min) to the above mixture at the same temperature. The
reaction mixture was stirred to reach the temperature about
0 8C, and then overnight. The next day, a 0.10 mL of the
reaction mixture was taken, evaporated, and dissolved in a
.50 mL of CDCl , and NMR spectrum was measured, on
1
.1. General
1
(
NMR spectra were recorded by Mr. Rafał Kozicki on a
Bruker Avance 300 MHz spectrometer locked on deuterium
from solvent. Chemical shifts (d [ppm]) were calculated
from chemical shift of deuterium lock and were not
calibrated. FTIR spectra were measured on Perkin Elmer
2
0
3
which only ca 30% of bromohydrins were found. Then, the
whole reaction mixture was evaporated under vacuo from
warm water bath (below 40 8C) to give oily residue which
was distilled under vacuo (bulb to bulb) from water bath to
give 4.4 g of bromohydrins (32% yield), and about 12 g of
unidentified thick oily residue. The distillate was identified
and assayed by means of NMR as a mixture of bromo-
hydrins–1-bromopropan-2-ol and 2-bromopropanol in
2
000 spectrometer in KBr pellets (1/200) by Mrs. Elzbieta
Mr o´ z´ . Mass spectra were measured on HP8542 mass
detector coupled with HP8542 gas chromatograph, by Dr.
Andrzej Nosal. Elemental analyses were done by Mrs.
Czesława Andrzejewska. Melting points were determined
on the Boetius microscope with electrical hot plate and were
corrected. The structures of all compounds were derived
1
from H NMR spectra. The required methyloxirane was
7
1
5/25 ratio (calculated from integrals of methyl groups at
.26 and 1.66 ppm, respectively).
acquired from local manufacturer. All reagents and solvents
were of commercial quality and purchased from local
supplier (POCh Gliwice).
1
.1.3. Experiment 3. The reaction of bromine with
1.1.1. Experiment 1. The reaction of iodine with
methyloxirane in the presence of 10 mol% amount of
thiourea—a reinvestigation. Methyloxirane (1.3 g,
methyloxirane in the presence of stoichiometric amount
of thiourea. Methyloxirane (2.9 g, 50 mmol) was added to a
stirred suspension of thiourea (3.8 g, 50 mmol) in aceto-
nitrile (50 mL) at 10 8C (ice-water bath). Next, a solution of
bromine (1.3 mL, 25 mmol) in acetonitrile (10 mL) was
added dropwise (30 min) to the above mixture at the same
temperature. Immediately white crystals precipitated. The
reaction mixture was stirred to reach the temperature about
20 8C, and additionally half an hour. The precipitate was
filtered off, washed with acetonitrile, and dried on air to give
7.8 g of product (about 100% of yield), identified as
formamidine disulfide dihydrobromide by comparison
2
2 mmol) was added to a stirred suspension of thiourea
0.15 g, 2.0 mmol) in acetonitrile (25 mL) at about 20 8C.
Next, a solution of iodine (5.1 g, 20 mmol) in acetonitrile
25 mL) was added dropwise (30 min) to the above mixture
(
(
at the same temperature. The dark reaction mixture was
stirred 1 h, then water was added in one portion at the same
temperature, and next a solution of Na S O *5H O (10 g,
2
2
3
2
40 mmol) in water (100 mL) was added. Almost immedi-
ately the reaction mixture decolorized. The mixture was
extracted with dichloromethane (1!50, then 2!20 mL),
2
f
with authentic sample by means of IR spectrum, and

M. Soroka, W. Goldeman / Tetrahedron 61 (2005) 4233–4235
4235
elemental analyses. The filtrates after isolation of formami-
dine disulfide dihydrobromide contained no bromohydrins.
sulfur (87% purity, 63% of yield) identified by combustion
to H SO and titration, and by comparison of IR spectrum to
2
4
those from the library as well as from a measurement of pure
element.
1
.1.4. Experiment 4. The reaction of iodine with
methyloxirane in the presence of stoichiometric amount
of thiourea, and then with addition of water. Methyl-
oxirane (2.3 g, 40 mmol) was added to a stirred mixture of
thiourea (1.5 g, 20 mmol) in acetonitrile (50 mL) at about
2
0 8C. Next, a solid iodine (2.5 g, 10 mmol) was added in a
References and notes
few portions to the above mixture at the same temperature.
The reaction mixture was stirred for half an hour more at the
same temperature, then water (50 mL) was introduced in
one portion, and stirring was continued till the reaction
mixture decolorized. The mixture was extracted with
dichloromethane (1!50, then 2!20 mL), organic phases
were collected and dried over Na SO (20 g), filtered and
1
. Sharghi, H.; Eskandari, M. Tetrahedron 2003, 59, 8509–8514.
. (a) Claus, A. Justus Liebigs Ann. Chem. 1875, 179, 128–145. (b)
McGovan, G. J. Prakt. Chem. 1880, 33, 188–193. (c) McGovan,
G. J. Chem. Soc. 1887, 51, 378–382. (d) McGovan, G. J. Prakt.
Chem. 1886, 36, 216–222. (e) Werner, E. A. J. Chem. Soc. 1912,
2
2
4
1
6
01, 2166–2180. (f) King, C.; Ryden, I. J. Am. Chem. Soc. 1947,
9, 1813–1814. (g) Peyronel, G.; Malavasi, W.; Pignedoli, A.
evaporated under vacuo from warm water bath (below
0 8C) to give oily residue (2.8 g, about 76%) identified and
4
Spectrochim. Acta Part A 1983, 39(7), 617–620.
. We chose methyloxirane for this investigation because it gives
products in which all of the methyl groups are separated and
easily visible in the NMR spectra.
assayed by means of NMR as the mixture of iodohydrins–1-
iodopropan-2-ol and 2-iodopropanol in 89/11 ratio (calcu-
lated from integrals of methyl groups at 1.31 and 1.89 ppm,
respectively), and traces of unidentified impurities. From
inhomogenous water phase the yellow precipitate was
isolated by filtration, dried on air to give 0.20 g of elemental
3
4. Soroka, M.; Goldeman, W.; Małysa, P.; Stochaj, M. Synthesis
2003, 2341–2344.