Srijit Biswas and Joseph S. M. Samec
were not optimized to generate the corresponding products
in high yields, but to disclose differences in reactivity among
the electrophiles, nucleophiles and catalysts.
of substrate 1b. If the reaction proceeded through a p-allyl
intermediate, such regioisomers would be expected.
For the less reactive alcohols (1d,e) in terms of selectivity
towards formation of the desired substitution product, the
Lewis acids promoted the thioetherification better, while
the redox metals showed low or no conversion to the de-
sired product. BiIII showed the highest degree of efficiency
and generated the desired products in a 50–79% yield. ReI
catalysts, which are known to generate p-propargylic metal
complexes, were unreactive in this transformation.[22] In
analogy to substrates 1a–c, etherification of 1e to generate
the symmetrical ether was observed. Using BiIII or FeIII cata-
lysts, symmetrical ethers were obtained in a 23–26% yield.
In the case of AuIII, an attack of the triple bond of 1e was
observed.[23] Moreover, the reactions involving tert-butyl al-
cohol 1d produced isobutylene as the major side product for
all the catalyst except bismuth, which generated 3d in 79%
yield (see the Supporting Information for details). The cata-
lysts that gave low conversion in direct substitution reactions
leading to thioetherification generally ended up with more
disulfide formation by oxidation of the unreacted thiophe-
nol.
Choice of Solvent
Pre-screening revealed nitromethane to be the best solvent
for all nucleophile–electrophile combinations. Dichloro-
ethane showed similar reactivity for reactions that were effi-
cient at low temperature. However, certain combinations of
electrophile and nucleophile required higher reaction tem-
perature. Running the reactions in toluene gave poor re-
sults.
Catalytic Nucleophilic Substitution by S-Centered
Nucleophile 2a
The results from the thioetherification reactions are given in
Table 2. The reactions were run under three different reac-
tion conditions, depending on the electrophile. Reactions
employing substrates 1a–1c were run at room temperature
for 10 h, 1e was run at 608C for 60 h, and 1d was run at
808C for 48 h. It is worth noting that, of the metals screened
in this study, only LaIII, BiIII, and PdII have been reported to
be active in the thioetherification reaction before.[15,19a]
Catalytic Nucleophilic
Substitution by C-Centered
Table 2. Conversion in the catalytic nucleophilic substitution of alcohols (1) by thiophenol (2a).[a,b]
Nucleophile 2b
À
The results from the C C
bond-forming reactions are
given in Table 3. The reactions
were run under three different
reaction conditions depending
on the electrophile. Reactions
employing substrates 1a and 1b
were run at room temperature
for 10 h, 1c was run at 608C for
10 h, and 1d and 1e were run
at 808C for 10 h. To our knowl-
edge, only FeIII[7c,o] and Pd0[19k]
have been reported to be active
in the nucleophilic substitution
of 2b.
Products
Catalysts
FeCl3
BiBr3
NaAuCl4·2H2O
69%
92%
81%
59%
86%
88%
99%
47% (70%)[c,d]
50% (85%)[c]
16%[d]
trace (82%)[e]
79%
66% (78%)[c]
74%
41% (61%)[c]
10% (23%)[c]
38% (57%)[c]
0%[d]
12% (45%)[e]
30% (89%)[c,e]
0% (21%)[e]
13% (57%)[e]
6% (51%)[e]
0% (23%)[e]
La
[PdCl
[ReBr(CO)5]
T
66% (79%)[c]
62% (73%)[c]
0%
10% (30%)[c]
0%[d]
A
U
61%
N
50%[d]
0%[d]
MeReO3
HCl[f]
80%
79%
61% (71%)[c,d]
53%[c]
48% (78%)[c,d]
32% (54%)[c]
19%[d]
8%[d]
[a] The reactions were run using alcohol 1 (1 mmol), 2a (1 mmol) in nitromethane (2.5 mL), and metal catalyst
(5 mol%). Reactions to form 3a–c were run at room temperature for 10 h. The reaction to form 3e was run at
608C for 60 h. The reaction to form 3d was run at 808C for 48 h. [b] Conversion was determined by 1H NMR
spectroscopy using toluene as an internal standard. Bold entries indicate reactions with higher efficiencies.
[c] In parenthesis: overall conversion of the electrophile where more than 10% formation of symmetrical
ether were observed. [d] The reactions were repeated at least twice to confirm the reproducibility of the ob-
served yield. [e] In parenthesis: overall conversion of tert-butyl alcohol to product and isobutylene determined
by 1H NMR spectroscopy. [f] 20 mol% with respect to the electrophile was used.
For the carbon nucleophile
2b, the allylic substrate 1b
showed the highest selectivity
for forming the desired product.
Except for ReI and ReVII, most
For thiophenol 2a, the benzylic and allylic substrates 1a–
c were the most selective electrophiles in the catalytic thioe-
therification reaction. Lewis acids generally performed well
in the thioetherification reaction, and BiIII showed the high-
est reactivity and also generated the desired products 3a–
c in 86–99% conversions. Other catalysts showed moderate
to good reactivity for these three electrophiles using the
same reaction conditions. For substrates 1a–c, homoetherifi-
cation was observed for PdII, AuIII, LaIII, ReVII and HCl,
which generated the symmetrical ether product in up to
30% yield. No regioisomers were observed in the reaction
catalysts generated the desired product 4b in 49–88% con-
version in less than 10 h. Generally, the Lewis acids gave
higher yields than the redox metals, where BiIII and FeIII
gave 88% and 87% product yield, respectively. AuIII, which
À
has never been reported to catalyze C C bond-forming re-
actions using 2b, generated the desired product in 80%
yield. Interestingly, PdII produced the product in 74% yield.
À
The mineral acid also catalyzed the C C bond-forming reac-
tion, though it generated the product in a lower yield.
The primary benzylic alcohol 1a was the second most se-
À
lective alcohol in the C C bond-forming reaction with 2b as
Chem. Asian J. 2013, 8, 974 – 981
977
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biswas, Srijit
