N. Yamanaka et al.
Molecular Catalysis 505 (2021) 111503
and EtOH produced the highest catalytic activity but had a negative
impact on the selectivity to the imine and secondary amine (Table 1,
entries 1 and 2). Some researchers have proposed that a strong inter-
action between protic polar solvents (e.g., C1-C4 primary alcohols) and
2-butanone by hydrogen bonding dramatically lowers the activation
energy barrier and significantly enhances the catalytic activity [25,26].
The catalytic activity of Ni3Sn2/TiO2 alloy catalyst in aprotic polar DMA
solvent, was next to protic polar solvents (Table 1, entry 3). The use of
other aprotic polar solvents such as THF, NEt3, and 1,4-dioxane led to
low-to- moderate catalytic activity (Table 1, entries 4–6). This is prob-
ably due to the strong adsorption of the solvent molecules on the catalyst
surface blocking the accessibility of the active Ni sites [26–28].
Non-polar solvents such as p-xylene, o-xylene, and mesitylene showed a
lower catalytic activity but provided the imine in higher yields without a
hint of benzyl alcohol formation, when compared with aprotic polar
DMA solvent (Table 1, entries 8, 10, and 12). It can be concluded that the
catalytic activity is higher in a more polar solvent with the following
order: protic polar solvent > aprotic polar solvent > non-polar solvent.
The direct one-pot reaction of benzaldehyde with nitrobenzene in the
non-polar solvents steadily proceeded with increasing the reaction time
(Table 1, entries 9, 11, and 13). Although the complete conversion of
benzaldehyde was achieved using o-xylene after 24 h, the yield of the
imine largely decreased and the secondary amine was obtained in 54%
yield together with the by-product (Table 1, entry 9). In contrast, se-
lective formation of the imine was accomplished by employing mesity-
lene and p-xylene (Table 1, entries 11 and 13). The benzaldehyde
conversions were 90% and 82% less than 100%, respectively. On the
basis of these results, it can be assumed that the formation of the desired
secondary amine N-phenylbenzylamine and the by-product benzyl
alcohol is inhibited before the complete exhaustion of benzaldehyde.
Mesitylene yielded the desired imine compound almost quantitatively
and therefore was shown to be the best solvent for the Ni3Sn2/TiO2
alloy-catalyzed reductive imination of benzaldehyde with nitrobenzene.
Fig. 1 presents the time profile for the Ni3Sn2/TiO2 alloy-catalyzed
one-pot reaction of benzaldehyde with nitrobenzene in mesitylene. At
a reaction time of 6 h, the in situ formed aniline (◆) was condensed with
benzaldehyde, giving rise to only the desired imine product (▴). The
yield of the imine (▴) steeply increased with increasing the reaction time
along with a small amount of aniline (◆), and it reached 90% after 24 h.
The formation of the corresponding secondary amine (●) was observed
at 30 h when the benzaldehyde conversion was almost full, more pre-
cisely, 98%. The yield of the secondary amine (●) dramatically
increased to a remarkably high yield of 80% at only 6 h. The control of
the reaction time in mesitylene can allow an efficient one-pot synthesis
of the imine and secondary amine. Notably, the yield of the secondary
amine (●) was slightly lower than that of the parent imine (▴). This is
because in situ formed H2O resulted in recovery of aniline (◆) and
benzaldehyde as a consequence of the imine hydrolysis [29].
3.2. Effect of the addition of molecular sieve 3 Å
The reverse transformation of the imine benzylideneaniline into
benzaldehyde and aniline in the presence of the in situ produced water
prompted us to add molecular sieve 3 Å into the catalytic system for the
removal. The time profile is illustrated in Fig. 2. In comparison with
Fig. 1, it has been proven that the addition of the molecular sieve plays a
significant role in accelerating the imine formation step during the
present one-pot reaction. However, the reverse reaction into aniline and
benzaldehyde proceeded regardless of the presence of the molecular
sieve. A similar yield of the imine and secondary amine to that of the
catalytic system without the molecular sieve was obtained at shorter
reaction times of 19 h and 30 h, respectively.
3.3. Effect of H2 pressure
Optimization of the reaction conditions was carried out for the one-
pot reaction of benzaldehyde with nitrobenzene using Ni3Sn2/TiO2 alloy
catalyst. First, the effect of H2 pressure was investigated at 423 K in
mesitylene. The results are given in Table 2. A lower H2 pressure of
0.6 MPa decreased the yield of the imine (Table 2, entry 1). This is
because the hydrogenation of nitrobenzene to aniline is sensitive to the
initial H2 pressure. The results presented in Table 2, entries 1 and 2 show
that the kinetic behavior of Ni3Sn2/TiO2 alloy catalyst during this pro-
cess enables the selective formation of the imine before the complete
exhaustion of benzaldehyde. To further improve the catalytic efficiency,
we performed the one-pot reaction under a higher H2 pressure of
2.0 MPa (Table 2, entry 3). In this situation, the imine was converted
into the corresponding secondary amine in a moderate yield, accom-
panied by the unwanted side product because the benzaldehyde con-
version reached 100%.
3.4. Effect of reaction temperature
Next, the effect of reaction temperature was studied under H2
1.0 MPa in mesitylene. The results are listed in Table 3. A relatively high
Fig. 1. Time profile for the one-pot reaction of benzaldehyde with nitroben-
zene by Ni3Sn2/TiO2 alloy catalyst in mesitylene. Reaction conditions: nitro-
benzene, 1.0 mmol; benzaldehyde, 1.0 mmol; Ni3Sn2/TiO2 alloy catalyst,
0.10 g; mesitylene, 5.0 mL; naphthalene, 0.30 mmol; H2, 1.0 MPa; reaction
temperature, 423 K.
Fig. 2. Time profile for the one-pot reaction of benzaldehyde with nitroben-
zene by Ni3Sn2/TiO2 alloy catalyst in mesitylene. Reaction conditions: nitro-
benzene, 1.0 mmol; benzaldehyde, 1.0 mmol; Ni3Sn2/TiO2 alloy catalyst,
0.10 g; mesitylene, 5.0 mL; naphthalene, 0.30 mmol; H2, 1.0 MPa; reaction
temperature, 423 K; molecular sieve 3 Å, 0.10 g.
3