Organic Letters
Letter
tBuOK at 130 °C (Scheme 1B),17 both of which require high
temperatures and fall short in terms of substrate generality
studies. Consequently, the development of a milder and more
convenient approach for the direct synthesis of thiophosphanes
remains underdeveloped. Herein, we report the first example of
a general dehydrocoupling of phosphines with thiophenols
enabled by photoredox catalysis under visible-light irradiation
to produce a series of thiophosphanes (Scheme 1C). More
interestingly, we also observed an unexpected dealkylative
coupling process in the reaction of secondary and tertiary
alkylphosphine with thiophenols.
12), while it is worth mentioning that THF, dioxane, and
DMSO gave similar results (Table 1, entries 9−11).
With the optimized reaction conditions in hand (Table 1,
entry 6), we evaluated the generality of our methodology for
the synthesis of various thiophosphanes. First, thiophenols
with different substituents on the phenyl ring were applied in
the reaction with diphenylphosphine (1a) (Scheme 2). In
Scheme 2. Ir(ppy)3-Catalyzed Dehydrogenative Coupling of
Diphenylphosphine with Various Thiophenols under
Visible-Light Irradiation*
We chose diphenylphosphine 1a (0.2 mmol) coupled with
4-methoxythiophenol 2a (0.24 mmol) as the model reaction
and Ir(ppy)3 (1 mol %) as the photoredox catalyst (see Table
S1 for photoredox catalyst screening), under ambient
conditions, with a 22 W white LED plate as the light source.
The results of our initial investigations revealed that, without
any H2-acceptor (HA), only a trace amount of 3a was
produced (Table 1, entry 1). Then, in an effort to promote the
Table 1. Ir(ppy)3-Catalyzed Dehydrogenative Coupling of
Diphenylphosphine (1a) with 4-Methoxythiophenol (2a)
a
under Visible-Light: Condition Optimization
b
entry
HAs
solvent
yield (%)
1
2
3
4
5
6
7
8
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
EtOH
DCM
THF
DMSO
dioxane
toluene
2
29
32
trace
65
96
20
78
88
92
HA-1
HA-2
HA-3
HA-4
HA-5
HA-5
HA-5
HA-5
HA-5
HA-5
HA-5
*
9
Reactions were performed in a 20 mL vial with 1a (0.2 mmol),
thiophenols (0.24 mmol), benzaldehyde (0.2 mmol), Ir(ppy)3 (1 mol
10
11
12
%), and MeCN 1.0 mL under N2 and light irradiation for 12−24 h;
92
76
a
isolated yields are given. Isolated yields after complexation with BH3.
a
Reactions were performed in a 20 mL vial with 1a (0.2 mmol), 2a
general, good to excellent yields were obtained. For example,
thiophenols with −OMe, −Me, −Et, and −iPr groups on the
phenyl ring reacted smoothly with 1a and afforded the
corresponding thiophosphanes 3a−3h in yields ranging from
88% to 94%. We also found that the position of the
substituents on the phenyl ring had no obvious effect on the
results. Furthermore, fluoride- and chloride-substituted thio-
phenols also participated in the dehydrogenative coupling
reaction without side reactions arising from the C−X bonds (X
= F, Cl) (3i−3k). Apart from the monosubstituted
thiophenols, the multisubstituted thiophenols also worked
well in our system; 3l with 2,6-dimethyl groups and 3m with
3,4-dichloro groups were obtained in yields of 82% and 92%,
respectively. Finally, by changing the phosphine to 4-Me-
substituted diphenylphosphines we could obtain the corre-
sponding products in 78% and 75% yields when coupling with
different thiophenols (3n and 3o).
(0.24 mmol), Ir(ppy)3 (1 mol %), HA (0.2 mmol), and solvents (1.0
b
mL) under N2 atmosphere with light irradiation. Yields were
determined by 31P NMR spectroscopy.
dehydrocoupling process, we added 1 equiv of various HAs to
the MeCN solution. To our delight, we obtained improved
results. With benzophenone (HA-1), we obtained 29% of the
dehydrocoupling product 3a (Table 1, entry 2). With an imine
(HA-2), we obtained 32% of 3a (Table 1, entry 3). With
styrene (HA-3), only a trace amount of 3a was observed; this
was due to the competing hydrophosphination of styrene
(Table 1, entry 4). With azobenzene (HA-4), we obtained 65%
of 3a (Table 1, entry 5). Finally, benzaldehyde (HA-5) was
found to be the best HA, which afforded 3a in 96% yield
(Table 1, entry 6, and Figure S1). Then, changing the reaction
solvent from MeCN to other solvents was conducted, and all
the tested solvents gave inferior results (Table 1, entries 7−
B
Org. Lett. XXXX, XXX, XXX−XXX