2
Li et al. Sci China Chem
tention of researchers and some achievements have been
obtained in the transformation of mono-alcohols to amines.
However, the direct reductive amination of aliphatic diols
to the corresponding diamines was little studied because it
is more difficult for that the intermediates formed with two
functional groups are more active and easily to transform
to by-products through oligomerization and cyclization
etc. [10,11]. For example, in the reductive amination of
1,3-propanediol over heterogeneous Co-Fe catalyst, the
optimal selectivity of 1,3-diaminopropane was 34% at a
conversion of 95% [12,13], and the selectivity was less
than 20% at a conversion of 59% over the Ni/SiO2 catalyst
[14]. In addition, 2,4-diaminopentane was produced with
only a 9% selectivity at the conversion of 85% in the re-
ductive amination of 2,4-pentanediol over Co-Fe catalyst
[15]. With the same Co-Fe catalyst, 1,4-cyclohexanedi-
amine was produced with a 55% selectivity at a conversion
of 76% in the reductive amination of 1,4-cyclohexanediol
[16], which is much lower than the results obtained with
homogenous Ru complexes that the selectivity of 71% (95%
conversion), 78% (99% conversion), 100% (100% conver-
sion) was obtained with Ru(CO)ClH(PPh3)3/Xantphos [17],
Ru(CO)ClH(DPEphos)(PPh3) [18], and Ru3(CO)12/acridine-
based diphosphine [19], respectively. The reductive amina-
tion of biogenic isohexides with homogenous Ru catalysts
produced diamine derivatives in a high selectivity of 95%
[17,19], but a quite low selectivity of 20% was obtained over
a heterogeneous catalyst of Ru/C [20]. For the amination
of HDO, HMDA was produced with a yield of 81% with
homogenous Ru-acridine catalyst [21], but, in contrast, the
heterogeneous catalysts gave much lower yields as 10%–15%
with Ni [22] and 24.8% with Co-Cu [23]. Therefore, it is still
a great challenge to develop efficient heterogeneous catalyst
in the reductive amination of HDO to HMDA.
(HMDA), hexamethyleneimine (HMI) and 6-amino-1-hex-
anol (AHO) were purchased from Tokyo Chemical Industry
(Japan); acetonitrile, tert-butanol and Nb2O5 were purchased
from Beijing Chemical Works (China); NH3 from Dalian
Special Gases (China). All the chemical reagents were used
as received.
2.2 Catalyst preparation
25 wt% M/Al2O3 (M=Co, Ni) catalysts were prepared by the
co-precipitation method. Nb2O5, SiO2 and MgO supported
Ru (5 wt%) were prepared by the impregnation method. The
support materials of Nb2O5, SiO2 and MgO were calcined in
air at 500ꢀ°C for 3ꢀh and cooled down to room temperature.
The co-precipitation and impregnation samples were dried at
60ꢀ°C overnight and calcinated in air at 500ꢀ°C for 4ꢀh. Before
used in the reductive amination, the calcined catalysts were
reduced in H2 flow at the desired temperature for 2ꢀh accord-
ing to the H2-temperature-programmed reduction (H2-TPR)
results. The reduction temperature determined was 550ꢀ°C for
Ni/Al2O3, 650ꢀ°C for Co/Al2O3, and 200ꢀ°C for Ru/Nb2O5,
Ru/SiO2, Ru/MgO.
2.3 Catalyst characterization
Powder X-ray diffraction (XRD) analysis was made on a
Bruker D8 Advance X-ray diffractometer (Germany) with a
Cu Kα source (λ=0.154ꢀnm) in a 2θ range of 10°–80° with
a scan speed of 2°/min. Transmission electron microscopy
(TEM) images were obtained on a JEOL JEM-2010 in-
strument (Japan) operated at an accelerating voltage of
200ꢀkV. H2-TPR and CO2 temperature-programmed desorp-
tion (CO2-TPD) profiles were obtained on a Micromeritics
Autochem II 2920 chemisorption instrument (USA). The
instrument was calibrated by nano CuO. Before H2-TPR
run, 50ꢀmg catalyst was loaded in an U-quartz tube and
pretreated with ultra-pure argon at 120ꢀ°C for 30ꢀmin. After
cooling to room temperature, the sample cell was heated
from room temperature to 800ꢀ°C with a rate of 10ꢀ°C/min in
H2/Ar mixture gas. For CO2-TPD test, 100ꢀmg catalyst was
pre-reduced in H2/Ar mixture gas at desired temperature
for 1ꢀh. After the catalyst was cooled to room temperature,
the adsorption was conducted by flushing the sample with
CO2/He mixture gas for 30ꢀmin. Hydrogen chemisorption
was estimated based on the H2-TPD results, and H2-TPD
was carried out at a temperature range of 25–800ꢀ°C with a
heating rate of 10ꢀ°C/min under flowing of He.
In the present work, the authors screened several heteroge-
neous catalysts of the supported Ni, Co, Ru, Pt, Pd catalysts
for the reductive amination of HDO in supercritical NH3. The
relationship between the catalytic performance and the sur-
face properties of catalysts such as the metal dispersion, the
acid/base properties was examined. The influence of reaction
temperature, time, H2 and NH3 pressure were examined, and
the reaction processes were discussed in detail.
2 Experimental
2.1 Materials
The total acidity of catalyst was measured by means of po-
tentiometric titration [24,25]. In a typical process, the cat-
alyst (0.05ꢀg) was suspended in acetonitrile and agitated for
3ꢀh. Then, the suspension was titrated with 0.1ꢀmol/L n-buty-
lamine in acetonitrile at 0.05ꢀmL/min. The electrode poten-
tial variation was measured with an automatic potentiometric
titrator containing dual electrodes.
5 wt% M/Al2O3 (M=Ru, Pd, Pt) catalysts were pur-
chased from Wako Pure Chemical Industries (Japan);
Co(NO3)2·6H2O, Ni(NO3)2·6H2O, Al(NO3)3·9H2O and MgO
were purchased from Sinopharm Chemical Reagent (China);
RuCl3·3H2O and SiO2 from Sigma-Aldrich (USA); 1,6-hex-
anediol (HDO) from Aladdin (USA); hexamethylenediamine
Li, Yan
