Angewandte
Communications
Chemie
to glycolaldehyde followed by amination, in a total yield of
0%. In situ characterization techniques showed that co-
(entry 14), and almost the same ETA yield (94%) was
achieved, demonstrating the utility of the synthesis protocol.
Since glycolaldehyde could be produced from cellulose by
1
existence of Ru and RuO on Ru/ZrO played a crucial role
2
2
[
2c,7d]
on the selectivity of primary amines.
hydrolysis and retro-aldol condensation,
we directly used
Our study began with the reductive amination of glyco-
laldehyde with aqueous ammonia on Ru catalysts with
different supports. Direct hydrogenation of glycolaldehyde
was greatly suppressed on Ru catalysts, while ETA was
obtained as the major product with a small amount of
ethylenediamine. The activity of Ru catalysts depended
greatly on the supports (entries 1–7, Table 1), and Ru/ZrO2
cellulose as feedstock for the production of ETA (for details
see Sections 1.3 and 1.4 in the Supporting Information). In
a first step, cellulose was converted in hot water (1 wt%
cellulose in 10 mL water) using tungstic acid (5 mg H WO ),
2
4
affording glycolaldehyde in a yield of 20.6%. The collected
À1
solution (six runs) was concentrated to 1.2 mL (105 mgmL )
and then submitted to amination with ammonia over the best
catalyst Ru/ZrO -150, producing ETA in 52% yield. Thus, by
2
this two-step approach we produced ETA from cellulose in an
overall yield of about 10% (Scheme 2). To the best of our
knowledge, it is the first example demonstrating the produc-
tion of ETA from lignocellulose. Although the ETA yield is
far from a satisfactory level at present, we believe it will be
greatly enhanced by optimizing the catalyst formulations and
reaction conditions in both steps.
[
a]
Table 1: Catalyst screening for reductive amination of glycolaldehyde.
Entry
Catalyst
Yield [%]
ethylenediamine
ethanolamine
1
2
3
4
5
6
7
Blank
Ru/ZrO -250
Ru/AC
Ru/Al O3
Ru/SiO
Ru/HZSM-5
Ru/Nb O5
Ru/ZrO -250
Ru/ZrO -250
Ru/ZrO -250
Ru/ZrO -350
Ru/ZrO -150
Ru/ZrO -unred
Ru/ZrO -150
n.d.
5
4
7
3
2
n.d.
4
3
3
n.d.
56
16
27
23
26
n.d.
74
46
83
57
93
21
94
2
2
2
2
[
b]
c]
[
8
9
1
1
1
1
1
2
[
Scheme 2. Direct conversion of cellulose to ETA.
2
d]
d]
d]
d]
e]
0
1
2
3
4
2
[
[
[
[
4
2
6
1
2
2
The Ru/ZrO -150 was effective not only for reductive
2
2
amination of glycolaldehyde, but also for a broad range of
biomass-derived aldehydes and ketones, such as hydroxyal-
dehyde, hydroxyacetone, aliphatic and aromatic aldehydes
and ketones (Table 2). For example, glyceraldehyde, which
2
[
a] Reaction conditions: 2 mmol glycolaldehyde, 2.0 mL 25 wt% aque-
ous ammonia, 0.1 g catalyst (metal loading=5 wt%), 858C, 12 h,
.0 MPa H . For Ru/ZrO -x, x is the reduction temperature (8C). Ru/ZrO -
2
2
2
2
[7a]
could be obtained directly from cellulose degradation, or
unred is the unreduced catalyst. n.d.=Not detected. [b] 758C. [c] 758C,
[7e]
from glycerol oxidation,
was aminated smoothly under
1
2
5 h. [d] 758C,12 h, H 3.0 MPa. [e] 10 mmol glycolaldehyde, 10.0 mL
5 wt% aqueous ammonia, 0.5 g catalyst.
2
quite mild conditions (658C, 2.0 MPa H and 12 h) to give 3-
2
amino-1,2-propandiol in 82% yield (Table 2, entry 2). It has
been well known that 3-amino-1,2-propandiol is a primary
intermediate for producing iohexol, a contrast agent widely
stood out giving a promising yield of ETA (56%, entry 2). We
also tested other metals (Pt, Pd, and Ir) supported on ZrO2,
but all of them exhibited much lower activity than Ru/ZrO2
[9]
used as in computed tomography and magnetic resonance.
Clearly, the present approach is more environmentally
benign. In addition to hydroxylaldehydes, aliphatic aldehydes
such as butyraldehyde and pentanal were also readily
aminated to the corresponding primary amines in yields
larger than 70% (entries 3 and 4). The aldehydes containing
a furan ring and aromatic ring such as 5-methylfurfural and
benzaldehyde were also tolerated, producing 5-methylfurfur-
ylamine and benzylamine in yields of 61% and 90.2%,
respectively (Table 2, entries 5 and 6).
(
Table S1). These results indicate that co-existence of Ru
species and ZrO in close proximity is essential for a high yield
2
of ETA. The reaction parameters, such as temperature,
reaction time, hydrogen pressure, and substrate/ammonia
ratio, were also investigated over the Ru/ZrO catalyst (see
2
Figures S1–S4 in the Supporting Information). Under opti-
mized reaction conditions (758C, 3.0 MPa H , 12 h) the ETA
2
yield could reach 83% (Table 1, entry 10).
Our further investigations showed that pre-reduction
In comparison with aldehydes, the amination of ketones is
[
6b,c]
treatment of Ru/ZrO imposed a significant effect on the
more challenging because of their lower reactivity.
How-
2
ETA yield (entries 10–13, Table 1). Without pre-reduction
ever, our Ru/ZrO -150 catalyst was still able to afford
2
treatment, the Ru/ZrO catalyst was poorly active, and the
satisfactory yields for several important biomass-derived
ketones such as methyl isobutyl ketone, isophorone, cyclo-
pentanone and cyclohexanone (entries 7–10). Notably, the
selectivities to the primary amines were rather high (> 90%)
even at elevated temperatures (Table S2, entries 8–10). The
amination of aliphatic ketones such as 2-heptanone and 2-
pentanone was more efficient, producing a yield of amines
larger than 90% (entries 11–12). In contrast, hydroxypropa-
none is more difficult to amination, producing 2-amino-1-
2
ETA yield was only 21%. In contrast, Ru/ZrO reduced at
2
1
508C (denoted as Ru/ZrO -150) gave rise to an ETA yield as
2
high as 93% (entry 12). Further elevating the pre-reduction
temperature resulted in a decline of the ETA yield. These
results suggest that amination of glycolaldehyde is sensitive to
the microstructure and chemical environment of the Ru
species which were formed during the reduction pretreat-
ment. The reaction was also performed on a larger scale
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!