M. Beller et al.
Table 1. Variation of Lewis acids and conditions for imine hydrogenation:
benchmark reaction.[a]
Entry
Lewis acid
Amount
[mol%]
T
[8C]
Conv.
[%][b]
Yield
[%][b]
1
2
3
4
5
6
7
8
9[c]
10[c]
11
12
13
14
—
–
100
0.01
5
5
5
5
5
5
5
100
25
9
41
–
87
83
13
19
20
30
57
–
–
–
<1
72
–
BF3(C2H5)2O
Fe(OTf)2
Cu(OTf)2
Zn(OTf)2
Zn(OAc)2
ZnBr2
100
100
100
100
100
100
65
100
100
120
120
120
Scheme 1. Hydroamination of alkynes and reductive hydroamination of al-
kynes.
4
4
ZnCl2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
11
42
95
92
47
4
With respect to this work, the use of Zn catalysts by Mꢁller
and Blechert[23] as well as by us[15b] for alkyne hydroaminations
is of special importance.
10
5
2
95
>99
74
Finally, it should be mentioned that Che[24] and Gong[25] have
recently independently described the first examples of inter-
and intramolecular hydroamination reactions of alkynes using
a AuI catalyst and a Hanztsch ester as the reducing agent. Un-
fortunately, on applying their protocols, stoichiometric
amounts of pyridine are produced as a coproduct, and expen-
sive noble metals in combination with phosphine ligands have
to be used.
1
38
[a] Reaction conditions: 1a (0.5 mmol), Lewis acid, toluene (0.5 mL),
80 bar H2, at 1008C for 24 h. [b] Determined by performing GC analysis
using hexadecane as an internal standard. [c] 50 bar H2.
80 bar H2 in the presence of 5 mol% Zn(OTf)2 (Table 1,
entry 12).
Clearly, Zn(OTf)2 is a very unusual catalyst for the activation
of H2. Thus, we initially thought that impurities of precious
metals might be responsible for the observed activity. Howev-
er, elemental analysis and inductively coupled plasma spec-
trometry revealed no detectable amounts of Ru, Rh, Ir, and Pd.
Furthermore, various samples of Zn(OTf)2 from different suppli-
ers behave similarly. Finally, Zn(OTf)2 proved to be inactive
under the optimized conditions for the hydrogenation of ke-
tones, nitro groups, olefins, and alkynes, which proceeded in
the presence of precious metal-based catalysts. To prove the
necessity of H2, we performed one experiment without hydro-
gen gas. As expected, no amine was obtained. Furthermore,
imine 1a was reacted with D2 (Scheme 3). Here, NMR spectros-
copy clearly showed the formation of 5. Deuteration was con-
firmed by using ESI-TOF MS. D/H-exchange at the DÀN posi-
tion was observed, most likely because of the presence of
water in the reaction mixture. As expected, both experiments
showed that hydrogenation clearly originated from H2.[26]
As a preliminary mechanistic proposal, we assume the for-
mation of a Zn-coordinated iminium ion. It is likely that dimeric
or higher oligomeric Zn species are formed on which hetero-
lytic hydrogen cleavage could take place.
Results and Discussion
Based on our work with iron catalysts[14] and because of the in-
dustrial importance of 1-arylethylamines, we used 4-methoxy-
N-(1-phenylethylidene)aniline (1a) as the starting material in
our benchmark reaction (Scheme 2 and Table 1).
Firstly, we investigated the influence of different Lewis acids
as catalysts for the hydrogenation of 1a. As expected, in the
absence of a catalyst, no reduction of the imine took place
(Table 1, entry 1). The strong Lewis acid BF3·(C2H5)2O activated
the imine, but gave no hydrogenation product (Table 1,
entry 2). Similarly, a number of Zn, Cu, and Fe salts showed no
hydrogenation reactivity (Table 1, entries 3, 4, 6–8). However,
to our surprise, the desired amine was formed in 72% yield in
the presence of Zn(OTf)2 (Table 1, entry 5). The differences be-
tween conversion and product yield resulted from aldol con-
densation and decomposition of the imine. Entries 9–14 in
Table 1 represent the optimization of the Zn(OTf)2-catalyzed
hydrogenation. Reducing the H2 pressure from 80 to 50 bar
(1 bar=105 Pa) led to a decreased yield of 42% (Table 1,
entry 10). Decreasing the temperature to 658C resulted in sig-
nificantly lower conversion and produced 2a in only 11% yield
(Table 1, entry 9). Conversely, increasing the catalyst loading to
10 mol% (Table 1, entry 11) at 1008C and 80 bar H2 gave 95%
yield. A similar product yield was obtained at 1208C and
Next, we studied the scope and limitations of this remark-
able catalyst in the hydrogenation of different imines.
As shown in Scheme 4, different ortho-, meta-, and para-sub-
stituted electron-withdrawing as well as electron-donating N-
aromatic imines were hydrogenated to give the corresponding
amines in good isolated yields of 58–91%. In addition, mono-
and disubstituted N-4-methoxyphenylimines with electron-do-
nating and electron-withdrawing groups in meta- and para-po-
sitions were smoothly hydrogenated with isolated yields of
56–81% (Scheme 4, 2 f–j). To our delight, heteroaromatic and
aliphatic imines also furnished the desired products 2k and 2l
Scheme 2. Hydrogenation of 1a.
778
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 777 – 782