Angewandte
Communications
Chemie
catalyst Ch1 and Ch2, respectively. The addition of 10 mol%
of 2a was obtained. Then a stronger electron donor, that is,
K2CO3 to the Ch2 catalysis system did not inhibit the catalytic
tetrahydrothiopyran (n2), was employed to inhibit the Se···p
bonding interaction. The addition of 2.0 equiv n2 already
resulted in a complete suppression of the Se···p interaction.
Phosphines represent a class of strong electron donors.[16]
Indeed, the addition of 2 mol% triphenylphosphine (n3) or
XantPhos (n4) to the standard reaction resulted in a complete
suppression of the Se···p interaction.
While the cycloaddition reaction using substrate 1e as
a reactant could be promoted by Brønsted acid (i.e. p-
toluenesulfonic acid) or Lewis acid (i.e. scandium triflate),
however, the controlling of stereoselectivity is not productive
(2.0:1 and 2.7:1 2e/2e’) (Table 3, entries 1-2). A detailed
activity, thus excluding possible role of acid. Upon associating
a less coordinating counteranion (i.e. BArF À, tetrakis(3,5-
4
bis(trifluoromethyl)phenyl)borate), catalyst Ch2’ was more
efficient. The product was obtained in 81% yield using
5 mol% of catalyst Ch2’’. In addition, typical binding
constants of catalyst Ch2’’ with representative p systems
were determined by 13C NMR titrations[15] in CD2Cl2, and
values of 0.79 MÀ1 for benzene, 1.43 MÀ1 for 2-vinylnaphtha-
lene (p4) and 1.61 MÀ1 for substrate analogue p10 were
obtained.
In the presence of a bidentate catalyst Ch3, the catalyst
loading could be lowered to 0.5 mol% to give 90% yield of 2a
within 10 min. Furthermore, 1 mol% bidentate catalyst Ch3’
associating a tetrachlorogallate counteranion or Ch4 bearing
a different scaffold was efficient to catalyze the cycloaddition
reaction. The NMR experiments suggest that Se···p bonding
interaction using an alkene-substituted-electron donor is
much stronger than its non-substituted counterpart. To
further support this observation, catalyst Ch5 was used.
Despite the presence of an intramolecular Se···p interaction
between selenium and indole ring in the solid structure of
Ch5, it still showed comparable efficiency in contrast to the
other bidentate catalysts. In line with the observations in
NMR and X-ray studies, a phosphonium catalyst P1 did not
show catalytic activity, thus suggesting that there is no
interferential cation···p interaction. The investigation of the
electron effect of substrates on this chalcogen···p bonding
catalysis approach was carried out (Scheme 1). While both
the electron-withdrawing (i.e. fluoro, 1b) and electron-
donating (i.e. methoxy, 1c) substituents were amendable to
this catalysis approach, methoxy-substituted substrate 1c was
less reactive. Furthermore, an N-H free substrate 1d gave
91% yield in the presence of 1 mol% catalyst Ch3 within
10 min.
Table 3: Controlling of stereoselectivity.[a]
Entry
Catalyst
t [h]
2e [%][b]
2e’ [%][b]
2e/2e’[c]
1
2
TsOH·H2O
Sc(OTf)3
Ch2
Ch2’
Ch2’’
Ch2’
Ch3
Ch4
Ch5
Ch3 + n2
Ch3 + n3
Ch3 + n3
Ch3 + n3
12
12
6
2
8
10
0.6
1.5
1.5
2
1
1.5
6
65
71
74
73
67
73
96
88
89
87
92
88
85
33
23
24
25
23
24
<5
7
6
10
6
9
13
2.0:1
2.7:1
3[d]
4[d]
5[d]
6
2.4:1
2.2:1
2.4:1
2.9:1
>20:1
15.0:1
14.3:1
9.4:1
15.0:1
9.7:1
7
8
9
10[e]
11[f]
12[f]
13[f]
6.4:1
[a] See the Supporting Information for details. [b] Isolated yield.
1
[c] Determined by H NMR analysis of the reaction mixture. [d] 5 mol%
catalyst was used. [e] 1.0 equiv n2 was used. [f] n3 (0.2 mol% for
entry 11; 0.5 mol% for entry 12; 0.8 mol% for entry 13) was used.
The competitive chalcogen bonding between a lone-pair
electron donor and a p-electron donor was investigated
(Table 2). Upon addition of 1.0 equiv tetrahydropyran (n1) to
the standard reaction condition, the reaction worked almost
with the same efficiency. The addition of 5.0 equiv n1 could
inhibit the catalytic activity to a certain extent and 68% yield
investigation reveals that the bidentate catalysts (Ch3–5) can
control the stereoselectivity in an excellent manner (14.3:1 to
> 20:1) while the monodentate catalysts (Ch2, Ch2’, Ch2’’)
resulted in low stereoselectivity (2.2:1–2.9:1) (Table 3,
entries 3–9). These experiments suggest that a dual Se···p
bonding catalysis approach involving the simultaneous inter-
action of a bidentate catalyst with two reactants is in
operation, thus governing the stereoselectivity in the cyclo-
addition process. In this context, it was envisioned that the
stereoselectivity should decrease to some extent upon addi-
tion of an electron donor because it would competitively bind
to the bidentate catalyst to generate a portion of equilibrating
species with mono activating site. In line with this notion, the
addition of 1.0 equiv n2 gave lower stereoselectivity (Table 3,
entry 10). In the presence of a stronger electron-donor n3, the
diastereomer ratio decreased from > 20:1 to 6.4:1 depending
on the amount of n3 (from 0 to 0.8 mol%, Table 3, entries 11–
13).
Table 2: Competition between Se···n and Se···p bonding interaction.[a]
Entry
Competitor
Competitive bonding
Yield [%]
1
2
3
4
5
6
7
no
no
92
84
68
92
<5
<5
<5
n1 (1.0 equiv)
n1 (5.0 equiv)
n2 (1.0 equiv)
n2 (2.0 equiv)
n3 (2.0 mol%)
n4 (2.0 mol%)
Se···O
Se···O
Se···S
Se···S
Se···P
Se···P
Beyond the observation of different performance on the
controlling of stereoselectivity by bidentate and monodentate
catalysts, the dual catalysis mode was further corroborated by
[a] All the experiments were carried out with Ch3 (1 mol%), 1a
(0.2 mmol) and competitor in 1.0 mL CH2Cl2.
Angew. Chem. Int. Ed. 2021, 60, 9395 –9400
ꢀ 2021 Wiley-VCH GmbH