COMMUNICATIONS
24 h (Table 1, entry 9). Notably, the yield of 3a respectively, while the electronic properties of the
increased to 88% when 2 equiv. of β-hydroxyl ketone substitutes had no significant effect on the reaction
was added (Table 1, entry 10). In the absence of a outcomes. For indolyl substrates with C5 substituents,
catalyst, no reaction took place (Table1, entry 11). the reactions all went on smoothly to give correspond-
°
°
Conducting the reaction at 40 C instead of 60 C, the ing products 3g–3i in moderate yields. It had to been
yield of 3a was decreased to 60% (Table 1, entry 12). mentioned that for substrate bearing a 5-aldehyde
A solvent screen showed that DCE was the most group, relatively low yield of 3h was obtained
suitable solvent for this reaction (Table 1, entries 13– probably due to that the activation of aldehyde with
17). Decreasing or increasing the loadings of catalyst, borane complex would result in unwanted side reac-
the reactions furnished diminished yields of product tions. Substrates with C6 and C7 functionalities were
3a in both cases (Table 1, entries 18 and 19).
also harmonious with this reaction, and furnished the
With the optimal reaction conditions in hand desired products 3j–3o in moderate yields. Specially,
(Table 1, entry 10), we focused our attention to C3 substituted indole was also found to be a good
evaluating the substrate scope of the reaction by substrate and furnished 2-substituted indole product 3p
varying the structure of indoles (Scheme 3). For in 79% yield. N-Methyl, Bn and Ts protected indoles
reactions of C4 substituted indoles with β-hydroxyl were also suitable for this transformation and provided
ketone 2a under the standard reaction conditions, the expected 3-substituted indoles 3q–3s in 60–70%
products 3b–3f were obtained in 55%-72% yields, yields, respectively. For N,N-dimethylaniline and N,N-
diethylaniline derivatives, the reactions also proceeded
smoothly to give the para-functionalized products 3t
and 3u in 60% and 63% yields, respectively. Notably,
for pyrrole substrate, double functionalization took
place and product 3v was obtained in 62% yield. It
should be noted that when anisole was employed as
substrate, no reaction take place under the standard
reaction conditions, probably due to the strong inter-
action of oxygen atom with the boryl complex will
reduce its catalytic activity.
To determine the generality of β-hydroxyketones,
we next examined a series β-hydroxyketones with N-
methylindole under the standard reaction conditions.
The β-hydroxyketones were synthesized by the aldol
reaction of corresponding ketones and aldehydes. As
can be seen in Scheme 4, for both electron-donating
and electron-withdrawing groups on the phenyl group,
the reactions all proceeded smoothly to afford the
desired products 4a–4i in 56–75% yields. It is worth
noting that R4 could be phenyl groups, and correspond-
ing products 4h and 4i were delivered in moderate to
good yields.
To gain a deeper understanding of the mechanism,
several control experiments were conducted. As shown
in Scheme 5, The reactions of indole with ethyl vinyl
ketone (EVK) and chalcone in the present of B
(C6F5)3 ·H2O gave corresponding ketone products 3w
and 4j in excellent yields, indicating the unsaturated
ketone should be the intermediate of the reaction.[10o]
While treatment of indole with acetophenone or
acetone under the standard reaction conditions, no
reaction took place, representing that the ketone
products 3 is not active to react with indoles[6] in the
presence of B(C6F5)3 ·H2O.
On the basis of the mechanistic studies, NMR
experiments (SI) and previous reports,[10o,11g and j,16] we
propose a putative reaction mechanism for the present
dehydrative Friedel-Crafts alkylation in Scheme 6.
Initially, in the presence of in the presence of B
Scheme 3. Substrate Scope of indole and electron-rich arene
derivatives.a,b
Adv. Synth. Catal. 2021, 363, 2386–2391
2388
© 2020 Wiley-VCH GmbH