W. Hu et al. / Tetrahedron Letters 56 (2015) 7056–7058
7057
Table 2
Me CN
CN
Scope of N-cyanation sulfoximinesa
radical process
O
R1
NH
R2
O
R1
N
R2
CN
O
O
Cu2O (10 mol%), DTBP (4.0 equiv)
120 oC, O2, 24h
N
CN
NH
+
S
S
+
CH3CN
S
S
C-CN bond clevage
R1
R1
R2
R2
1
2
Scheme 1. N-cyanation of sulfoximine.
O
S
O
N
CN
Cl
N
CN
CN
O
N
CN
S
Table 1
S
The optimization of reaction conditionsa
Cl
O
O
2a
2b
2c
NH
85%
82%
O
79%
N
N
CN
S
S
+
CH3CN
O
S
N
CN
O
S
N
CN
S
1a
2a
Br
OMe
CN
OMe
Entry
Catalyst
Oxidant
Temp (°C)
Yieldb (%)
2d 69%
2e 71%
2f 59%
O
S
1
2
3
4
5
6
7
8
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
—
CuF2
CuBr
CuI
CuCN
Fe(acac)2
AgF
Cu2O
Cu2O
Cu2O
TBHP
K2S2O8
BPO
DTBP
—
DTBP
DTBP
DTBP
DTBP
DTBP
DTBP
DTBP
DTBP
DTBP
DTBP
120
120
120
120
120
120
120
120
120
120
120
120
100
110
130
26
<1
<1
N
O
S
O
S
N CN
N
CN
Br
85 (80)c (<1)d
Br
Br
<1
8
2g
2h
2i
58%
72%
O
61%
48
75
40
46
<1
19
<1
41
79
O
N
CN
Ph
N
S
CN
S
O
S
N
CN
9
10
11
12
13
14
15
Ph
2j 75%
2k 81%
2l 49%
O
S
N
CN
O
S
N
CN
O
N
CN
S
a
Cl
F
Reaction conditions: 1a (0.1 mmol), oxidant (0.4 mmol), catalyst (10 mol %),
CH3CN (1.5 mL) at 120 °C for 24 h under O2 in a sealed tube. (TBHP = tert-butyl
hydroperoxide, BPO = benzoyl peroxide, DTBP = di-tert-butyl peroxide.)
2m
2n
2o
52%
51%
CN
40%
CN
b
Isolated yield.
Under air.
O
S
N
CN
N
N
c
d
N
N
Under N2.
MeO
proceed via a radical process. Thereafter, reactions were conducted
under air and N2. Sharply in contrast, N-cyanation product 2a was
formed in 80% yield under air while the reaction just did not take
place under N2 (Table 1, entry 4). Based on that, we could make a
conclusion that O2 participated in this procedure. Subsequently,
other copper(I) and copper(II) salts such as CuF2, CuBr, and CuI
were also tested in the procedure and provided the desired product
2a in 48%, 75%, and 40% yields, respectively, (Table 1, entries 7–9).
It is significant and meaningful that when CuCN was tested, pro-
duct 2a was formed only in 46% yield (Table 1, entry 10), which
ruled out the possibility that CN anion is the active compound in
this reaction. No conversion occurred when Fe(acac)2 was used
as the catalyst (Table 1, entry 11), while 2a was formed in 19%
yield in the presence of AgF (Table 1, entry 12). Different temper-
atures had also been tested in this procedure. No reaction took
place at 100 °C (Table 1, entry 13), while 2a was formed in 41%
and 79% yields at 110 °C and 130 °C, respectively, (Table 1, entries
14 and 15).
The substrate scope of this strategy was investigated under the
optimized reaction conditions. Initially, diphenylsulfoximine
derivatives were tested. Generally, various substituted groups
(such as methyl, methoxy, chloro, and bromo groups) were all tol-
erated well and provided the corresponding N-cyanation products
in moderate to excellent yields as shown in Table 2 under the stan-
dard procedure. For instance, 2b, 2c, 2e, and 2k were isolated in
82%, 79%, 71%, and 81% yields, respectively. To our delight,
alkylphenylsulfoximines, arylimine, and guanidine also worked
smoothly under the standard procedure delivering the N-cyanation
products. For example, (S-methylsulfonimidoyl)benzene derivative
provided the 2l in 49% yield and (S-ethylsulfonimidoyl)benzene
2p 54%
2q 73%
2r 31%
aReaction conditions: 1a (0.1 mmol), DTBP (0.4 mmol), Cu2O (10 mol %), CH3CN
(1.5 mL) at 120 °C for 24 h under O2 in a sealed tube.
derivative provided the N-cyanation product 2m in 52% yield.
Meanwhile, we are glad to emphasize that substrates containing
halogen also run smoothly (2n, 2o), which provided an efficient
route for further functionalizations. Interestingly, 1,1,3,3-tetram-
ethylguanidine also worked well in this procedure, and gave the
N-cyanation product in 73% yield (2q). Besides, benzophenone
imine in this reaction led to the desired product in 31% yield
(2r), which further extended the substrate scope.
More experiments were conducted to study the potential mech-
anism. Initially, the radical scavenger TEMPO (2,2,6,6-tetramethyl-
1-piperidinyloxyl) was added to this procedure, no desired product
2a was formed, and the adduct TEMPO-CN was observed by GC–
MS (Figs. S1 and S2 in Supplementary data). Meanwhile, product
2a was not provided either when adding BHT (2,6-di-tert-butyl-
4-methylphenol) to the reaction, and the adduct between BHT
and cyanide was also detected by GC–MS (Figs. S3 and S4 in Sup-
plementary data). These indicate that a radical pathway may be
involved in this transformation (Scheme 2).
Based on the aforementioned experimental results, the possible
mechanism is proposed shown in Scheme 3. Firstly, the catalyst Cu
(I) is oxidized to Cu(II) by O2. Then, the reaction between sulfox-
imines 4 and Cu(II) species 3 produces another Cu(II) species 5.
At the same time, cyano radical is formed via CACN bond cleavage
of acetonitrile promoted by catalyst copper and DTBP. Subse-