COMMUNICATIONS
products in good yields (Table 3, Entries 7–11). Un-
fortunately, 2-methylbenzonitrile was problematic
under the current conditions, and intramolecular
addition of 1m was observed (see SI). This could be
ascribed to the activation of CÀ H bond of methyl
group by the neighboring CN group under strong basic
condition, (Table 3, Entry 12). Furthermore, nitrile
with ArÀ O bond was tolerated (Table 3, Entry 13). In
addition, the reactions employing heterocyclic nitriles
were explored as well, for example 3-(1H-indol-1-yl)
benzonitrile (1o), 2-cyano quinoline (1p), 6-cyano
quinoline (1q) and 2-cyano pyridine (1r) (Table 3,
Entries 14–17). With slightly modified procedure
(procedure B in the experimental section), the desired
products were obtained in moderate yields apart from
less efficient 1r. Aryl nitriles with electron-withdraw-
ing substituents were also tested in our protocol.
Halides, ester and carboxylic acid were not
tolerated.[19] To our delight, 4-cyano-N,N-diethylbenza-
mide (1s) was a suitable substrate and delivered the
target product in 41% yield under a modified condition
(Table 3, Entry 18).
To further demonstrate the efficiency of this
developed protocol, a scale-up reaction was conducted.
As shown in Scheme 2, 10 mmol of 1a successfully
reacted with 2f under the optimal conditions to deliver
2.25 g of 3af in 82% isolated yields (Scheme 2).
Scheme 3. Mechanistic experiments.
To get the insights into this Ni-catalyzed decyana-
tive amination, mechanistic studies were performed. crossover experiment applying 4 and 2f under the
With the reaction between 1a and 2a as the model, the standard conditions was carried out. Both 3aa and 3af
reaction progress was monitored by GC (see SI). The were produced in a ratio of almost 1:1 (Scheme 3,
results clearly exhibited that 1a was completely Eqn. 2, see SI, Figure S69). In addition, when 4 and 2f
°
consumed within two hours and 71% yield of the were mixed in the presence of KHMDS at 140 C,
desired product was obtained. Simultaneously, a amidine 5 and 2a were detected (Scheme 3, Eqn. 3).
comparable amount of amidine 4 was detected in GC. On the basis of these results, we believed that a
As prolonging the reaction time, the decline of amidine balance between amidine and nitrile existed under
4 and the increase of 3aa were observed. When 1a basic conditions, while the possibility of involving
and 2a were mixed at room temperature in the amidine in the catalytic cycle could not be ruled out
presence of KHMDS, 4 was quickly formed almost in (Scheme 4).
a quantitative yield. With these observations in hands,
In our previous work, complex A was assigned as
we thus reckoned that amidine 4 might be an the key species in nickel-catalyzed decyanation of aryl
intermediated to be involved in the catalytic cycle. To nitriles.[15e] To gain more information about the active
verify this assumption, with 4 as the substrate under nickel species, additional stoichiometric experiments
the standard amination conditions, 3aa was obtained were conducted. Mixing complex A with aniline in the
°
in 23% yield (Scheme 3, Eqn. 1). Surprisingly, 1a was presence of KHMDS at 140 C for 6 h provided the
also detected. Therefore, another possibility might amination product 3ca in 74% yield (Scheme 3,
exist, that is, amidine 4 could decompose to generate Eqn. 4). Besides, with complex A as the catalyst, the
nitrile 1a and amine 2a. To clarify the assumption, a amination of 1a with 2a under the optimal conditions
gave 3aa in 88% yield (Scheme 3, Eqn. 5). Thus,
complex A was believed to be the authentic intermedi-
ate in the catalytic cycle and acted as a key
intermediate of the following CÀ CN cleavage.
With all obtained information, a mechanism for this
Ni-catalyzed amination of nitrile is proposed and
shown in Scheme 4. The Ni(0) species I in situ
generated from nickel precatalyst, firstly coordinated
with the C=C double bond of aromatic ring adjacent to
Scheme 2. Scale-up reaction.
Adv. Synth. Catal. 2021, 363, 1–7
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!