80
X.-H. Lu et al. / Catalysis Communications 55 (2014) 78–82
Tables S3 and S4 exhibit the reaction results of imine with benzyl alco-
A
100
80
60
40
20
0
100
80
60
40
20
0
hol in the presence or absence of solid NaOH. Out of question, our cata-
lytic system could effectively promote the occurrence of hydrogen
transfer from PhCH2OH to imine. It means that the hydrogen transfer
from alcohol to imine could be realized via a certain route during the re-
action process, but did not require the participation of noble metal or
non-noble transition metal catalysts totally, notably different from the
results in the literature [10–14,17–19]. As listed in Table S5, along
with decreasing the molar ratio of PhCH2OH to PhNH2 from 10:1 to
1:1, the secondary amine was still formed in high yields. When excess
aniline was added with the molar ratio of PhNH2 to PhCH2OH from
5:1 to 10:1, the conversion of benzyl alcohol and the selectivity of the
secondary amine product showed an appreciable reduction. This com-
parison clearly implied that the source of hydrogen was coming from
benzyl alcohol rather than aniline. Fig. S3 depicted the dependence of
the reactivity on the reaction time in the absence or presence of 1.0%
PhCHO. Apparently, the addition of aldehyde exhibited one advantage
of accelerating the reaction rate under identical conditions. Similarly,
the addition of imine (N-benzylideneaniline) in the initial reaction mix-
ture could as well improve the reaction rate significantly (Table S6).
These experimental observations make us believe that for the
N-alkylation of aniline with benzyl alcohol both PhCHO and imine
are the actual intermediates.
With reference to the above-stated results, we can propose that the
most feasible reaction mechanism be surely involved in three steps, in-
clusive of the alcohol dehydrogenation, the formation of imine and the
subsequent hydrogenation. However, once the reaction is initiated,
these three steps may simultaneously take place through a synergic
route. As disclosed in Fig. S3 and Table S6 for the reaction of aniline
with benzyl alcohol, the secondary amine product with a selectivity of
40.4% had been detected by GC even in the initial stage of the reaction
(e.g. in 15 min). Figs. S4 and S5 illustrate the reaction circulation and
the structure of transition states involved. Firstly, the starting alcohol
1 (R1\CH2OH) is dehydrogenated to give the corresponding aldehyde
via the intermediate of R1\CH2O− anion (step (i)) resulted from the re-
moval of H+ cation of hydroxyl group (step (ii), Fig. S4) in the alkaline
environment. Then, the condensation and dehydration of amine 2
(R2\NH2) with the generated aldehyde afford the intermediate imine
3 (R2\N_CH\R1) (step (iii), Fig. S4), which can easily occur without
any catalyst [15]. After that, the transfer of H− anions from the interme-
diate R1\CH2O− anion to imine 3 takes place through one six-
membered cyclic transition state to form the intermediate anion
(R2\N−\CH2\R1) (step (iv), Fig. S5), similar to the transition metal-
free Meerwein–Pondorf–Verley reduction of aldehydes/ketones by al-
cohols. Finally, the intermediate anion (R2\N−\CH2\R1) will quickly
seize H+ cation from the starting alcohol 1 to form the targeted product
secondary amine 4 (R2\NH\CH2\R1) and to give R1\CH2O− anion
joining in the next reaction circulation (step (v)). Slightly slow reaction
rate in our system without any transition metal and additive becomes
easily understandable, because our present reaction must firstly start
from the dehydrogenation of alcohol to aldehyde promoted by solid
base.
Conv.(mol%)
Sele.(%)
440
450
460
470
480
490
500
Temperature /K
B
100
80
60
40
20
100
80
60
40
20
Conv.(mol%)
Sele.(%)
0
2
4
6
8
10
Time /h
Fig. 1. Effect of reaction temperature (A) and time (B) on the N-alkylation of aniline with
benzyl alcohol: (♦) aniline conversion, (▲) selectivity of the secondary amine. Reaction
conditions: (A) time 6 h, (B) temperature 493 K.
3.3. Understanding of the reaction mechanism
Based on the GC quantitative analysis, the content of benzaldehyde
(PhCHO) was 0.2% in commercial benzyl alcohol, and reduced to 0.1%
by redistillation. Absolute purity grade of benzyl alcohol was obtained
through redistilling commercial benzyl alcohol three times in the
presence of CaH2, degassed and stored in a Schlenk flask under N2,
which has been proven by GC analysis and 1H NMR spectrum (see the
supporting information). As listed in Table S1, with reducing the content
of PhCHO in the reagent both the conversion of aniline and the selectiv-
ity of the secondary amine were obviously improved. Particularly, using
absolute purity of benzyl alcohol achieved the highest conversion of
aniline (99.6 mol%) and selectivity of amine (99.5%), while the use of
benzyl alcohol containing about 1.0% PhCHO led to the lowest conver-
sion of aniline (97.5 mol%) and selectivity of amine (92.8%). This
phenomenon at least disclosed that the external addition of aldehyde
[15] was not beneficial to the titled reaction.
In order to investigate the reaction mechanism, several experi-
ments were conducted and displayed in Tables S2–S6, in which
N-benzylideneaniline (imine) was specially prepared and purified.
First, the reaction between imine and excess PhCHO was carried out
(Table S2). Just as expected, the hydrogen transfer between PhCHO
and imine took place relatively difficult. On the contrary, the external
addition of PhCHO reduced the yield of the secondary amine (Fig. S3).
The experiments proved that the hydrogen transfer via the Cannizzaro
reaction route was not predominant under the present conditions.
4. Conclusions
This work reports that only solid base NaOH can catalyze highly effi-
cient N-alkylation of amines with alcohols in a solvent-free system to af-
ford the targeted secondary amines. For this process, the essentiality of
either the catalysts containing noble metals/non-noble transition
metals or the addition of additive was eliminated. Under optimal condi-
tions, the reaction between aniline and benzyl alcohol could achieve the
highest aniline conversion of 99.6 mol% with a secondary amine selec-
tivity of 99.5%. Various amines could react with alcohols to gain
high conversions and selectivities, implying the universality of this
methodology. We propose a base-catalyzed mechanism for the present