2278
D. Sengupta, B. Basu / Tetrahedron Letters 54 (2013) 2277–2281
Table 1
Optimization of disulfide formation from organic thiocyanates in watera
Entry
Reagentb
Additivec
Temp (°C)/time (h)
Yieldd (%)
1
2
3
4
AmberlystÒ A-26(OH)
AmberlystÒ A-26(OH)
AmberlystÒ A-26(OH)
AmberlystÒ A-26(OH)
AmberlystÒ A-26(OH)
AmberliteÒ IRA-900(Cl)
Nil
Nil
rt/12
60/3
60/20
rt/2
52e
91
76
90
Na2S2O3Á5H2O
L
-Cysteine
5
60/2.5
87
L
-Cystine
6
7
Nil
Nil
60/12
60/6
Nil
81
Amberlite IRA-900(HCOO)
a
b
c
Reactions were performed in 2 mmol scale in water (2 mL).
Different resins were used 150 mg mmolÀ1
Additives were used in stoichiometric quantities.
Yield refers to isolated product after column chromatography.
Starting thiocyanate still remains in the reaction mixture, as indicated on TLC.
.
d
e
polyethylene glycol (PEG 400) following a reported procedure with
minor modifications.10 Optimization of the conversion of organic
thiocyanate to symmetrical disulfide was performed using benzyl
thiocyanate as the model compound (Table 1). In a typical experi-
ment, benzyl thiocyanate (2 mmol) was stirred in water at room
temperature in the presence of commercially available AmberlystÒ
A-26 (OH) (particle size: 0.35 mm, pH 9.4, 300 mg) (entry 1). Since
the conversion was not complete even after 10 h, we increased the
temperature to 60 °C and now a complete conversion was achieved
within 3 h affording dibenzyl disulfide in 91% yield (entry 2). As the
enzymatic (rhodanese) conversion of cyanide to thiocyanate has
occurred in the presence of sulfur donors,11 we tested the same
reaction in the presence of equimolar quantity of Na2S2O3, cystine
or cysteine (entries 3–5). It is observed that while the presence of
Na2S2O3 could have an effect in lowering the rate of the conversion,
the amino acid bearing –SH group (cysteine) did show an apprecia-
ble change in the rate of the conversion (rt/2 h). However, there
was no significant change by using cystine (–S–S– linkage) as the
additive. The same reaction was also tested in the presence of other
anion exchange amberlite resins, like commercially available
AmberliteÒ IRA(Cl) or our laboratory derived Amberlite resin for-
mate (ARF),9c with varying results (entries 6 and 7).
In order to verify any advantage of using poly-ionic resin
hydroxide base, a comparison between the heterogeneous base
AmberlystÒ A-26(OH) and other homogeneous bases (NaOH, NH3,
K2CO3) was performed. Thus, the conversion of benzyl thiocyanate
to dibenzyl disulfide was carried out using homogeneous bases
NaOH, NH3, and K2CO3 that, respectively, afford 83%, 46%, and
25% yields, while AmberlystÒ A-26(OH) can give rise to 91% yield
(Table 3) under similar conditions. The variation of reactivity is
even more prominent in the case of acyl methyl thiocyanates
(Table 4).
Comparative studies using p-methoxy benzoyl methyl thiocya-
nate as a model case revealed that the AmberlystÒ A-26(OH) can
lead to the formation of the corresponding disulfide in 81% yield
with a trace amount of p-methoxy acetophenone, while the base
NaOH affords only p-methoxy acetophenone in 69% yield and no
disulfide (Table 4, entries 1 and 2). On the other hand, use of
K2CO3 afforded a mixture of compounds as seen on tlc (no separa-
tion was attempted), while NH3 solution gave a mixture of the cor-
responding disulfide and acetophenone in 21% and 18% yield,
respectively, the remaining was the starting material (60%), as ana-
lyzed by HPLC (Table 4, entries 3 and 4). These results clearly
establish the advantage of the heterogeneous base AmberlystÒ A-
26(OH) over the existing homogeneous bases.
We had chosen the optimized condition as in entry 2 for the
easy formation of the dibenzyl disulfide from benzyl thiocyanate.
We were interested to establish this mild, efficient, and aqueous
metal-free condition as a general protocol.12 Accordingly, several
substituted benzyl thiocyanates were subjected to a similar reac-
tion. The results are presented in Table 2. It is gratifying that differ-
ent halo-substituted benzyl thiocyanates and naphthyl methyl
thiocyanate underwent smooth conversion to the corresponding
disulfides (entries 2–7). However, the nitro benzyl thiocyanate re-
mained unchanged under the reaction conditions (entry 8). This
could possibly be attributed to the strong electron-withdrawing
nature of the nitro group that reduces nucleophilic character of
the benzyl carbon and electron density on the sulfur atom. Purely
aliphatic thiocyanates, n-octyl thiocyanate, or other compounds
such as 3-phenyl propyl thiocyanate or allyl thiocyanate derivative
(3-phenyl-2-propenyl thiocyanate) also undergo disulfide forma-
tion in good to excellent yields (entries 9–11). In order to further
generalize the reaction, acyl methyl thiocyanates were subjected
to a similar reaction. Organic disulfides flanked by two carbonyl
groups at the C-2 position would be the S-analogue of 1,6-dike-
tones. Such bis(benzoylmethyl) disulfides could be easily prepared
if benzoyl methyl thiocyanate undergoes similar dimerization. We
had prepared different benzoyl methyl thiocyanates and carried
out similar reactions using the resin hydroxide in water. As shown
in Table 2 (entries 12–16), the corresponding disulfides were
formed efficiently and isolated in excellent yields except the ni-
tro-bearing benzoyl methyl thiocyanate (entry 15), possibly due
to the same reason as mentioned in the case of entry 8 of this table.
We then focused our attention to explore whether this mild and
eco-friendly protocol could be employed for the preparation of
unsymmetrical disulfides. In order to get an unsymmetrical disul-
fide, it is required to perform the reaction taking a mixture of two
different organyl thiocyanates. This might also help us to propose a
possible pathway of the reaction. Firstly, we performed one reac-
tion using two different benzyl thiocyanates. For example, carrying
out the reaction using an equimolar mixture of benzyl thiocyanate
and m-chlorobenzyl thiocyanate indeed afforded all three possible
disulfides in varying amounts (two homo-coupled products and
one cross-coupled product) (Table 5, entry 1). After the reaction
and work-up, HPLC analysis of the crude mixture showed three
peaks of which two are from homo-coupled products (confirmed
by co-injection), leaving the third peak assigned for the cross over
product, that is, the unsymmetrical disulfide. Similar results are
observed when a mixture of other alkyl thiocyanates is subjected
to undergo the reaction (Table 5, entries 2 and 3). In all the three
examples, the prominent similarity is in the formation of the
unsymmetrical disulfide in larger quantity than homo-coupled
disulfides. In the case of using acyl methyl thiocyanate as one of
the partners, the cross product could not be distinctly analyzed
by HPLC and appeared as two overlapping peaks of the total area
about 26% only, which is in contrast to the observations in
other cases (entry 4). As regards the mechanism of the reaction,
formation of the cross- over product suggests that the S–S linkage
is made in a step-wise manner and not through a concerted
process. In the presence of heterogeneous base, presumably the
Sengupta, Debasish
