Organic Letters
Letter
Aliphatic carboxylic acids are readily available, stable, and
low-toxicity reagents. N-Hydroxyphthalimide (NHPI) esters,
which can be easily synthesized from carboxylic acids, have
recently been widely investigated as alkyl radical precursors,
which then couple to radical acceptors or nucleophiles.11,12
Various nucleophiles such as alcohols, amines, and B2Pin2 have
been successfully introduced in the decarboxylative reaction.13
With our ongoing interest in organic phosphor chemistry,14 we
envisioned that P(O)SH might serves as a nucleophile in the
decarboxylative coupling reaction, which could provide a
method for accessing phosphorothioates. Herein we disclose a
photoredox-catalyzed phosphorothiolation of NHPI esters via
the decarboxylative reaction.
Having identified the optimal reaction conditions for this
decarboxylative phosphorothiolation reaction, the scope of
alkyl carboxylic acid was investigated (Scheme 2). First, various
Scheme 2. Substrate Scope for Primary and Secondary
a
Acids
To validate our concept, cyclohexyl NHPI ester (1a) and
(EtO)2P(O)SH (2a) were chosen as model substrates (Table
1). When using 2 mol % Ir(ppy)3 as a photocatalyst and 2
a
Table 1. Optimization of Conditions
b
entry
photocatalyst
Ir(ppy)3
4CzIPN
Ir(ppy)2(dtbbpy)PF6
Eosin Y
Rose Bengal
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
base
solvent
yield (%)
1
2
3
4
5
6
7
8
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
Et3N
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
DMF
17
10
25
0
0
trace
trace
57
77
36
44
34
trace
29
17
a
DBU
Reaction conditions: 1 (0.2 mmol), 2 (2 equiv, 0.4 mmol),
[Ir(ppy)2(dtbbpy)]PF6 (1 mol %, 0.002 mmol), KHF2 (2 equiv, 0.4
mmol), CH3CN (2 mL) at room temperature, 5 W blue LEDs, 12 h
Et2NH
KHF2
NaOH
NaOAc
K3PO4
KHF2
KHF2
KHF2
KHF2
9
b
in N2. 1a (1 mmol), 2 (2 equiv, 2 mmol), [Ir(ppy)2(dtbbpy)]PF6 (1
10
11
12
13
14
15
16
17
mol %, 0.01 mmol), KHF2 (2 equiv, 2 mmol), CH3CN (5 mL) at
room temperature, 5 W blue LEDs, 24 h in N2.
DMSO
toluene
CH3CN
CH3CN
CH3CN
carboxylic acids, for instance, primary, secondary, and tertiary
acids, were all compatible in this decarboxylative reaction, and
the yields were moderate to good. Primary carboxylic acids,
including a benzylic substrate (1b−1f) with different
substituents such as electron-neutral (1b−1d) and electron-
rich substituents (1e) at the para position of the aromatic
rings, all smoothly furnished the desired organothiophosphates
(3b−3e). For disubstituted aromatic rings (1f), the corre-
sponding products 3f were procured in 67% yield. A wide
range of functional groups were also evaluated. For example,
heterocycles such as thiophene, alkanes, terminal alkynes,
esters, heteroatoms, and adamantane with high steric
hindrance all proved to be compatible under the standard
reaction conditions (3g−3m). NHPI ester prepared from
trans-3-hexenoic acid, which contains CC bonds, could not
afforded the desired product. Next, we studied secondary
carboxylic acids. Cyclic acids, including three-membered, four-
membered, five-membered, and six-membered cyclic carbox-
ylic acid, fluorinated alkyl carboxylic acid, and acyclic alkyl,
were all converted to the corresponding S-aryl products in
moderate yields (3n−3s). Interestingly, other alkoxy-substi-
tuted phosphorothioic acids such as dimethyl (2b), diisopropyl
phosphorothioic acid (2c), and O-ethyl S-hydrogen phenyl-
phosphorothioic acid (2d) were also suitable coupling partners
for the decarboxylative reaction, which delivered 3u−3w in 52
NR
NR
NR
Ir(ppy)2(dtbbpy)PF6
Ir(ppy)2(dtbbpy)PF6
c
18
KHF2
a
Reaction conditions: 1a (0.2 mmol), 2a (2 equiv, 0.4 mmol)
photocatalyst (1 mol %, 0.002 mmol), base (2 equiv, 0.4 mmol),
solvent (2 mL) at room temperature, 5 W blue LEDs, 12 h in N2.
b
c
Isolated yield. Without light.
equiv of K2CO3 as base in CH3CN under visible light, 3a was
formed in 17% yield (entry 1). A screening of photocatalysts,
such as 4CzIPN, Eosin Y, and Rose Bengal, indicated that
Ir(ppy)2(dtbbpy)PF6 was the best photocatalyst, yielding 3a in
30% yield (entries 2−5). Subsequently, a variety of bases were
tested, and the results showed that the yield was significantly
affected. Most organic bases except Et2NH resulted in only a
trace amount of 3a (entries 6−8). Among the inorganic bases
investigated, KHF2 provided the best result (entries 9−12).
Several solvents such as dimethylformamide (DMF), dimethyl
sulfoxide (DMSO), and toluene were also screened; however,
they were all less efficient than CH3CN (entries 13−15).
Finally, a series of controlled experiments suggested that visible
light, the photocatalyst, and the base were all equally crucial for
this reaction (entries 16−18).
B
Org. Lett. XXXX, XXX, XXX−XXX