X. Jing et al.
Applied Catalysis A, General 541 (2017) 107–111
Table 3
Table 5
Catalyst loading investigations [a].
Test of the generality of this protocol.
entry
catalyst loading/%[b]
2a/%[c]
3a/%[d]
entry
R
T1 (°C) [a]
T2 (°C) [b]
2: yield/% [c]
1
2
5
4
3
2
1
0.5
0.5
94.3
84.5
61.6
58.3
53.5
45.3
49.4
5.8
5.6
5.5
4.7
0
1
2
3
4
5
6
7
4-MeOC
Ph
6
H
5
65
65
65
85
75
65
65
160–162
60–62
65–68
110–113
70–73
163–165
72–74
2a: 91
2b: 84 (78)
2c: 74
2d: 73
2e: 66 (64)
2f: 82
[
[
[
[
[
e]
3
4
5
6
7
4-MeC
4-FC
4-ClC
1-C10
2-C
6
H
5
e]
6
H
5
e]
6
H
5
e]
0
3.8
H
7
e,f]
5
H
4
N
2 g: 69
[
a] 1 mmol of 1a was employed; [b] Catalyst loadings based on 1a; [c] GC yields of 2a
[a] Reaction temperature. [b] Reduced pressure distillation temperature. [c] Gross yields
of the isolated products outside the parentheses; nitrile yields after deduction of the
aldehydes from H NMR inside the parentheses.
based on 1a; [d] GC yields of 3a based on 1a; [e] Reactions not completed; [f] The
reaction time was extended to 48 h.
1
reaction, giving 2a in only 50.3% yield, while the by-product 3a was
generated in 10.3% yield and the reaction did not complete in 24 h, as
monitored by GC analysis (Table 2, entry 2). Electron-deficient 4-
reactions. As shown in Fig. 4, the reaction proceeded quickly and the
product yield reached 80% in the first 1 h reaction (curve 1). The
product yield slowly increased to 93% after 24 h. Using recycled
catalyst, the product yields were not reduced and underwent similar
procedures like the first round of reaction (Fig. 4, curves 2–4). It was
FC
ClC
the reaction (Table 2, entries 3–4 vs. 5–6).
H
6 4
Se(O)OH and 3-FC
H
6 4
Se(O)OH were good catalysts, but 3-
6
H
4
Se(O)OH and 3,5-(CF
3 2 6 3
) C H Se(O)OH were unfavourable for
2 2
notable that addition of catalytic amount of H O was the key for
catalyst recycling and reusing, otherwise the catalyst was deactivated,
resulting in the gradually decreased product yields in the 2nd and 3rd
rounds of reaction (Fig. 4, curves 5–6).
3.3. Catalyst loading investigations
More efforts were tried to reduce the catalyst loading of the reaction
(
Table 3). Unfortunately, the reduced catalyst loadings resulted in
3
.6. Substrate scope extension
decreased product yields (Table 3, entries 1 vs. 2–6). The reaction could
not complete within 24 h when less than 4 mol% of catalyst loadings
were employed (Table 3, entries 3-6). Extending the reaction time to
4
(
Generality of this protocol was also tested (Table 5). A series of
aldoximes 1 were treated under the optimized conditions and the
related products 2 were isolated through reduced pressure distillation
8 h could hardly improve the reaction with reduced catalyst loading
Table 3, entry 7). In summary, heating 1a with 5 mol% of PhSe(O)OH
with the simplified distillation device depicted in Fig. 1. The distilled
as catalyst but without solvent was finally screened out to be the
optimized reaction conditions.
products were directly sent to 1H NMR spectroscopic analysis to
confirm their purities (see supplementary information). Besides anisal-
doxime 1a (Table 5, entry 1), benzaldoxime 1b led to benzonitrile 2b in
7
benzaldehyde were also observed in its H NMR spectrum. Dehydration
of the electron-enriched substrate 4-methylbenzaldoxime 1c afforded
the pure product 4-methylbenzonitrile 2c in 74% yield (Table 5, entry
3
.4. Catalyst recycle & reuse in scaled-up reactions
8% yield (Table 5, entry 2), while traces of the deoximation product
1
A 100 mol-scale-magnified experiment under the optimized reaction
conditions were then performed and the product 2a could be distilled
under reduced pressure by using the simplified distillation device, as
shown in Fig. 1. The residue, which contained a series of low valent
3
). The electron-deficient aldoxime 1d and 1e were also tested, but
because of their high melting points, the reactions should be performed
at higher temperature, affording 2d and 2e in moderate yields (Table 5,
entries 4–5). By using this dehydration protocol, the bulky but electron-
enriched substrate 1-naphthaldehydoxime 1f could also produce the
related product 2f in good yield (Table 5, entry 6). It was notable that
the heterocycle-containing substrate 1 g was also preferable for the
reaction, giving 2 g in 69% yield (Table 5, entry 7).
2 2 2
organoselenium compounds such as (PhSe) , was re-oxidized by H O
to regenerate the catalytic PhSe(O)OH and then directly reused after
adding another portion of 1a. Table 4 shows that the catalyst could be
reused for at least 5 times without deactivation. The distilled product
2
a of the 5th round reaction was directly sent to NMR spectroscopic
1
analysis without any additional purification procedure. H NMR and
1
3
C NMR spectra both indicated that the purity of the product was very
high (Figs. 2 and 3), although fraction cutting was inoperable in the
experimental procedures by using the simplified distillation device
3.7. Reaction mechanism discussion
(
Fig. 1). Moreover, further scaled-up dehydration reaction of 1 kg of 1a
Based on the above experimental results as well as literatures [41], a
successfully produced 0.845 kg of 2a (in 96.0% yield) with carbon mass
balance at 99.3%, showing very good reliability of this method for
possible industrial applications in future.
plausible mechanism was given below. Nucleophilic addition of the pre-
catalyst PhSe(O)OH to anisaldoxime 1a initially generated the inter-
mediate 4 (Scheme 1, eqn. 1), which decomposed into the by-product
3
a and the organoselenium species 5 (Scheme 1, eqn. 2). Further
3
.5. Kinetics studies
decomposition of 5 afforded PhSeOH (Scheme 1, eqn. 3) [41], which
was the catalytic species in anisaldoxime dehydration (Scheme 1).
Dehydration of PhSeOH furnished the highly activated species PhSeO-
SePh (Scheme 1). Condensation of 1a with PhSeOSePh gave the
intermediate 6, which rearranged into the intermediate 7. The selen-
oxide syn-elimination of 7 led to the product 2a and regenerated the
catalytic species PhSeOH [48]. As PhSeOH species was highly activated
To get detailed information of the procedures, kinetics studies were
carried out by comparisons of the results of a series of 10 mmol-scale
Table 4
Catalyst recycle & reuse [a].
Recycle NO.
0 [b]
1
2
3
4
5
2 2 2
and might be reduced into the stable species (PhSe) , addition of H O
was necessary to reactivate the recycled catalyst before next round of
reaction (Fig. 4). Since nucleophilic attack of aldoximes to PhSeOSePh
was the key step during the processes, the electron-enriched anisaldox-
ime 1a as substrate led to higher nitrile yield than other simple or
2
2
weight/g
yield/% [c]
12.1
91.0
12.7
95.5
12.3
92.5
13.2
99.2
13.1
98.5
13.2
99.2
[
a]100 mmol reaction scale; [b] First use of the catalyst; [c] Isolated yields of 2.
110