.
Angewandte
Communications
Table 1: Optimization of the reaction conditions.
Table 2: Imidations of 5a and 7a with various 1,4,2-dioxazol-5-ones
(1).[a]
Entry Catalyst
Solv.[a]
T
t [h] hn Yield [%]
6aa 9a
1
2
3
4
[Ru(TPP)CO]
tol[b]
tol[b]
tol
tol
tol
tol
tol
tol
508C 18
À
+
+
+
À
À
À
+
+
+
+
+
+
+
+
+
37
45
99
0
0
42
3
2
9
0
0
61
54
0
0
0
9
11
0
0
0
0
0
0
75
0
Entry
1
R
6
8
[Ru(TPP)CO]
[Ru(TPP)CO]
–
[Ru(TPP)CO]
[Ru(TPP)CO]
[Ru(TPP)CO]
[Ru(TMP)CO]
[Ru(F20-TPP)CO] tol
[Rh2(OAc)4]
[Rh2(OAc)4]
[RuCl2(p-Cym)]2
[RuCl2(p-Cym)]2
[Ru(TPP)CO]
[Ru(TPP)CO]
[Ru(TPP)CO]
RT
RT
RT
RT
6
4
4
18
Yield [%]
Yield [%]
1
2
3
4
5
6
7
1a
1b
1c
1d
Ph
Me
Bn
tBu
CF3
pMeOC6H4
pNO2C6H4
99 (6aa)
95 (6ba)
7 (6ca)[b]
0 (6da)[c]
0 (6ea)[c]
35 (6 fa)
62 (6ga)
99 (8aa)
98 (8ba)
86 (8ca)
0 (8da)[c]
54 (8ea)
99 (8 fa)
72 (8ga)
5[c]
6
508C 18
508C 18
7[c]
8
1e
RT
RT
RT
RT
RT
RT
RT
RT
4
4
4
4
4
4
4
6
6
1 f[d]
1g[e]
9
10
11
12
13
14
15
16
tol
CH2Cl2
[a] Reaction were conducted until full consumption of 1. [b] The
sulfoximine was obtained in a mixture with the corresponding urea.
[c] Decomposition of 1 was observed. [d] Use of 3 mL of toluene instead
of 1 mL. [e] Use of 2 mL of toluene instead of 1 mL.
tol[d]
5
CH2Cl2
CH2Cl2
THF
31
21
0
MeCN[d] RT
0
0
iminations providing the N-acyl sulfoximine 6ca and N-acyl
sulfimide 8ca. Whereas the former was only obtained in
a very low quantity [< 7%, as an inseparable mixture with
N,N’-dibenzylurea (9c)], the latter was isolated in 86% yield
(Table 2, entry 3). The conversions of 5a and 7a with 3-
trifluoromethyl-substituted 1,4,2-dioxazol-5-one (1e; Table 2,
entry 5) were limited by the sensitivity of the iminating agent
towards traces of water, and an undesired deoxygenation of
5a, thus forming sulfide 7a. As a result, 5a could not be
iminated, and the sulfimide 8ea was obtained from 7a in only
moderate yield (54%).
Encouraged by the high reactivity and pronounced
selectivity in the imidations of 7a with various 1,4,2-dioxa-
zol-5-ones (1), respective transformations of diversely sub-
stituted sulfides were investigated next. The results are shown
in Scheme 2. In general, the positive data obtained in the
conversions of 7a were reflected in this study. Thus, sulfides
with an alkyl/alkyl substitution pattern reacted equally well
with 1a, thus providing the corresponding products (8ab, 8ac,
and 8ad) in yields ranging from 82 to 99%. Only the
imidations of diphenylsulfide (7e) and phenylvinyl sulfide
(7 f) with 1a proved more difficult, thus leading to the
corresponding sulfimides (8ae and 8af) in yields of 46 and
63%, respectively. Further improvements were observed
when 1b was used as imidating agent, thus providing
sulfimides 8ba–bi. Now, the yields were generally high (82–
99%) irrespective of the substitution pattern of the sulfide.
Noteworthy also is that sulfides with bromo, nitro, and
methoxy substituents in the para position of the S-aryl group
reacted well, thus providing the corresponding N-acyl sulf-
imides 8bg–bi in high yields. As expected from the initial
study with 7a (Table 2), the other 1,4,2-dioxazol-5-ones (1c,
1e–1g) could be used to deliver the N-acylated sulfimides
8ca–gb in moderate to high yields (43–99%).
[a] Unless noted otherwise, dry solvents were used; tol=toluene.
[b] Non-distilled solvent. [c] The reaction was carried out in the dark.
[d] The catalyst was poorly soluble in this solvent.
ambient temperature (Table 1, entry 3). Additional experi-
ments revealed that the presence of both [Ru(TPP)CO] and
light were essential for promoting the reaction at room
temperature (Table 1, entries 4 and 5). Under thermal con-
ditions using dry toluene as the solvent, 6aa was obtained in
42% yield (Table 1, entry 6). Performing the reaction in the
dark at 508C led to 6aa in only 3% yield (Table 1, entry 7),
thus highlighting the importance of the photochemical
activation in this process. We assume that the light provides
the essential energy for the formation of a reactive rutheno N-
acyl nitrene intermediate without reaching the higher level of
energy required for the Curtius rearrangement which leads to
isocyanates.[16] Other ruthenium porphyrin catalysts such as
[Ru(TMP)CO] and [Ru(F20TPP)CO], as well as [RuCl2(p-
Cym)]2 and [Rh2(OAc)4] were less efficient or ineffective
(Table 1, entries 8–13).[15,17] In terms of solvents, toluene
proved to be superior over CH2Cl2, THF, or acetonitrile
(Table 1, entry 3 versus entries 14–16).
An initial test reaction showed that 1a reacted equally
well with the sulfide 7a, just as it did with the sulfoxide 5a,
under the photocatalytic conditions with [Ru(TPP)CO] to
give 8aa in 99% yield (Table 2, entry 1). Hence, the
subsequent evaluation of the substrate scope involved both
of these starting materials, which were explored in their
reactivity with a range of 1,4,2-dioxazol-5-ones (1). With the
exception of the 3-tert-butyl-substituted derivative 1d
(Table 2, entry 4), which proved to be unstable under the
reaction conditions, all the 1,4,2-dioxazol-5-ones 1 showed
a pronounced N-acyl nitrene transfer ability, thus providing
the corresponding N-acylated sulfoximines 6 and sulfimides 8
in yields of up to 99%. In general, 5a was less reactive than
sulfide 7a, as examplified by the comparison of the respective
Motivated by the positive results of the aforementioned
photochemically-induced ruthenium-catalyzed imidation of
sulfides (Scheme 2), we wondered about the possibility to
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!