New bulk nickel phosphide catalysts for glycerol hydrogenolysis to 1,2-propanediol

Article abstract of DOI:10.1016/j.catcom.2014.10.020

Transitional metal phosphides were found to have outstanding activity and stability in catalytic hydrotreatments. The bulk trinickel phosphide catalyst with the smallest phosphorus content among the nickel phosphides were synthesized by a hydrothermal method followed by an annealing treatment, and the resulting bulk trinickel phosphide catalysts presented a high purity and morphology of hexagonal prisms. The optimized synthesis conditions include a P:Ni ratio of 3 to 1 and a pH value of 5 in the hydrothermal synthesis stage and a calcination temperature of 773 K in the annealing treatment. The synthesized trinickel phosphides exhibited a low-temperature activity to selective glycerol hydrogenolysis and the high selectivity to 1,2-propanediol.

Full text of DOI:10.1016/j.catcom.2014.10.020

Catalysis Communications 59 (2015) 180–183
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
New bulk nickel phosphide catalysts for glycerol hydrogenolysis
to 1,2-propanediol
⁎
Guojun Shi , Lijun Su, Kai Jin
School of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, Jiangsu Province, People's Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 May 2014
Received in revised form 16 October 2014
Accepted 22 October 2014
Available online 29 October 2014
Transitional metal phosphides were found to have outstanding activity and stability in catalytic hydrotreatments.
The bulk trinickel phosphide catalyst with the smallest phosphorus content among the nickel phosphides were
synthesized by a hydrothermal method followed by an annealing treatment, and the resulting bulk trinickel
phosphide catalysts presented a high purity and morphology of hexagonal prisms. The optimized synthesis
conditions include a P:Ni ratio of 3 to 1 and a pH value of 5 in the hydrothermal synthesis stage and a calcination
temperature of 773 K in the annealing treatment. The synthesized trinickel phosphides exhibited a low-
temperature activity to selective glycerol hydrogenolysis and the high selectivity to 1,2-propanediol.
© 2014 Elsevier B.V. All rights reserved.
Keywords:
Trinickel phosphide
Hydrothermal synthesis
Glycerol
Hydrogenolysis
1,2-Propanediol
1. Introduction
such as Ni2P and Ni12P5. Ni3P can be synthesized by chemical reactions
in “chemical plating” [8] or by electrochemical deposition [4]. Supported
Traditional metal phosphides have attracted considerable attentions
in the recent decades due to their excellent properties and wide
application scopes [1]. Used as catalysts, metal-rich phosphides exhibit
higher catalytic activity than the traditional sulfide catalysts in
hydrotreatments and much better stability than other interstitial
compounds such as carbides and nitrides [2]. Especially, the metal-
rich nickel phosphides like Ni2P was found to possess the highest
catalytic activity among the traditional metal phosphides for the
hydrodesulfurization (HDS) reaction of the FCC naphtha [2].
In our previous work, Ni2P catalysts were synthesized by the solid
phase reaction between the nickel cations and hypophosphites at a
much lower temperature than that used in phosphide synthesis by the
temperature-programmed reduction of nickel phosphates, and the
resulting Ni2P catalysts presented high dispersion degree and thus
high catalytic activity in HDS [3,4]. Trinickel phosphide (Ni3P) is the
nickel-rich phosphide with the smallest content of phosphorus and
has a tetragonal crystal structure. It is often found as a main component
in the nickel-containing electroless coating, which is a mixture of Ni and
Ni3P and prepared by plating without electric current [5]. Supported
Ni3P was also tried as catalysts for the synthesis of carbon nanotubes
[6] and the hydrodechlorination of chlorobenzene [7]. Generally, high
ratio of nickel to phosphorus in Ni3P catalysts means more chance for
the exposure of metal nickel to the reactants, which might lead to
higher catalytic activity than other nickel-rich phosphide catalysts
Ni3P catalysts were synthesized by temperature-programmed reduction
of phosphate at a temperature up to 923 K [7].
In the present work, bulk Ni3P catalysts were synthesized by a hydro-
thermal method followed by an annealing treatment. The resulting Ni3P
samples were tested as catalysts for selective glycerol hydrogenolysis to
1,2-propanediol, which is thought a fundamental reaction in a conver-
sion from biomass to fuel for vehicle in the background of diminishing
fossil resource reserves and increasing environmental concerns [9–16].
To our knowledge, the pure bulk Ni3P is synthesized and used for
glycerol hydrogenolysis to 1,2-propanediol for the first time.
2. Experimental
2.1. Catalyst synthesis
All chemical reagents were of analytical grade and used as received.
Bulk trinickel phosphide catalysts were synthesized by a hydrothermal
method followed by an annealing treatment at N2 atmosphere. In a typ-
ical synthesis, nickel hypophosphite and ammonium hypophosphite
with a P:Ni atomic ratio of 3:1 were dissolved by deionized water, and
the pH value of the resulting green solution was adjusted to 5 by an
aqueous solution of ammonia (25 wt.%). The solution then was trans-
ferred to a 100 mL Teflon-lined stainless steel autoclave. The autoclave
was heated to 423 K and maintained at the temperature for 12 h in a
drying oven, and then cooled to room temperature naturally. The mix-
tures obtained by the hydrothermal method were repeatedly washed
with deionized water and filtered, and the resulting black intermediates
⁎
Corresponding author. Tel./fax: +86 514 87937661.
E-mail address: gjshi@yzu.edu.cn (G. Shi).
http://dx.doi.org/10.1016/j.catcom.2014.10.020
1566-7367/© 2014 Elsevier B.V. All rights reserved.

G. Shi et al. / Catalysis Communications 59 (2015) 180–183
181
composition was determined by atomic emission spectroscopy with
Perkin Elmer ICP-OES spectrometer. Powder XRD tests were carried
out in X-ray diffraction (XRD) patterns and were collected in an ambient
atmosphere by a Bruker D8 Advance system using Cu Kα radiation. The
2θ scans covered a range of 10-80° with a step of 0.02°. The applied
voltage and current were 40 kV and 40 mA, respectively. Transmission
electron microscopy (TEM) analysis was carried out using a Philps
Tecnail 12 transmission electron microscope. The samples were pre-
pared by dispersing the powder material in ethanol and dropping the
solution on a carbon-coated Cu grid. The hydrogen temperature-
programmed desorption (H2-TPD) of the samples were performed by
a chemical adsorption instrument equipped with a thermal conductivi-
ty detector (TCD). The samples were pretreated at 373 K in a flowing Ar
for 2 h, and then 5% H2/Ar mixture was introduced through the reactor
for H2-adsorption of the samples for 40 min when the temperature
were decreased to 303 K. The H2-TPD curves were obtained at a ramp
of 10 K/min with 50 sccm of Ar from 303 K to 1073 K.
2.3. Catalytic test
The glycerol hydrogenolysis reaction was carried out in a 50 mL
Teflon-lined stainless steel autoclave. In a typical test, 30 g of glycerol
was dissolved in 27 g of deionized water, and 300 mg of catalyst was
then added into the autoclave reactor. The reactor was sealed and
repeatedly flushed with hydrogen to eliminate the air, then was heated
to 463 K and pressurized to 5.5 MPa by H2. The product analysis was
performed when the reactor was quenched to room temperature by
an ex-situ gas chromatograph (SP 6890) equipped with a flame ioniza-
tion detector (FID) and a capillary column (SE54, 30 m × 0.32 mm).
Fig. 1. The XRD patterns of the samples synthesized under different pH values at 423 K for
12 h followed by an annealing treatment at 773 K in a flowing N2 of 100 sccm for 3 h. (For
interpretation of the references to colour in this figure, the reader is referred to the web
version of this article.)
were collected. The annealing treatment of the intermediates was per-
formed in a horizontal furnace with a quartz tube at a temperature of
773 K in a flowing N2 of 100 standard cubic centimeter per minute
(sccm) for 3 h. The annealed products was washed for 3 times by the
aqueous solution of ammonia followed by deionized water until the
pH value of the filtrate was up to 7. The obtained catalysts were dried
at 373 K for 12 h in a static air atmosphere before the evaluation and
characterization.
3. Results and discussion
Fig. 1 shows the XRD patterns of the samples synthesized under
different pH values, which then were calcined at 773 K in a flowing N2
atmosphere for 3 h. It was found that the pH values in the synthesis pro-
cess of the precursors remarkably affected the phases of the annealed
samples. The sample synthesized under a pH value of 9 mainly present-
ed the diffraction peaks of the metallic nickel after annealing, and the
sample synthesized at a higher pH value (pH 12) was found to be
pure amorphous metallic nickel proved by XRD. The relatively strong
diffraction peaks of Ni3P were observed when the pH value was
2.2. Catalyst characterization
The Brunauer–Emmett–Teller (BET) surface areas were measured by
the Sorptomatic 1990 instrument (Thermo Finnigan). Experiments
were performed at 77.3 K using N2 as the adsorbate. The elemental
Fig. 2. The XRD patterns of the samples synthesized under a pH value of 5 at 423 K for 12 h
followed by an annealing treatment at different temperatures in a flowing N2 of 100 sccm
for 3 h. (For interpretation of the references to colour in this figure, the reader is referred to
the web version of this article.)
Fig. 3. The XRD patterns of the samples synthesized with different P/Ni ratios under a pH
values of 5 at 423 K for 12 h followed by an annealing treatment at 773 K in a flowing N2 of
100 sccm for 3 h. (For interpretation of the references to colour in this figure, the reader is
referred to the web version of this article.)

182
G. Shi et al. / Catalysis Communications 59 (2015) 180–183
Table 1
Specific surface areas and elemental composition of the annealed samples.
Annealing temperature/K
673
773
873
SBET/m2 g−1
7
8
1
Ni/P (atomic ratio)
2.5
2.6
2.6
decreased to 7. In the case of pH value of 5, all diffraction peaks of the
annealed sample can be attributed to the Ni3P phase. Lower pH value
led to the formation of some species that was difficult to assign (Fig. 1).
The dependence of the formation of Ni3P on the pH values in the
hydrothermal synthesis could be attributed to oxidation/reduction ca-
pability of the precursor solution (Ni2+ and H2PO2−). For the precursor
solution, the higher pH value leads to the stronger reduction capability
and thus the phosphorus of the H2PO−2 can be reduced to phosphine
(P3−), which can explain the formation of amorphous metallic nickel
under a higher pH value. On the contrary, the more acidic precursor
solution, with a lower pH value, leads to stronger oxidation capability
and thus the phosphorus of the H2PO−2 is difficult to be reduced to
Ni3P, in which the valence of phosphorus is close to zero. This can
explain why the formation of Ni3P phase is also difficult under a lower
pH value.
Fig. 2 shows the effect of the annealing temperature of precursors,
which were synthesized under a pH value of 5 at 423 K for 12 h, on
the formation of the Ni3P phase. As can be seen, the diffraction peaks
of the sample annealed at 773 K can be exclusively attributed to the
Ni3P phase. A higher annealing temperature (873 K) led to stronger
diffraction peaks and bigger particle sizes and thus lower dispersion
degree, which is usually not desired for the catalysts. However, too
low annealing temperature (673 K) led the production of other species.
Fig. 3 shows the XRD patterns of the annealed samples, which were
synthesized with different Ni/P ratios under a pH value of 5 at 423 K for
12 h followed by an annealing treatment at 773 K in a flowing N2 for 3 h.
The diffraction peaks of the sample synthesized with a P/Ni ratio of 3
well agreed with Ni3P phase. The XRD patterns of the samples synthe-
sized with either a lower P/Ni ratio (pH 2) or a higher one (pH 5)
were found the presence of other diffraction peaks that cannot be
assigned to Ni3P, indicating the importance of an optimized P/Ni ratio.
Table 1 shows the specific surface areas and elemental compositions
of the synthesized Ni3P samples. It was found that the synthesized bulk
Ni3P catalysts exhibited specific surface areas around 7 m2/g when
they were annealed at a temperature less than 773 K. A higher post-
treatment temperature (873 K) led to a sharp decrease of the specific
Fig. 5. TEM image of the bulk Ni3P sample synthesized under a pH value of 5 at 423 K for
12 h followed by an annealing treatment at 773 K in a flowing N2 of 100 sccm for 3 h.
surface area of the sample, suggesting the formation of the larger Ni3P
crystalline particle. The Ni/P atomic ratio of the synthesized Ni3P cata-
lysts determined by ICP-OES was around 2.6, which was roughly in
line with theoretical ratio of 3. Slightly more P can be attributed to the
presence of oxides or acids of phosphorus because of the incomplete
washing.
Fig. 4 compares the hydrogen desorption of bulk trinickel phosphide
catalysts annealed at 673 K and 773 K, respectively. It was found that
hydrogen desorption happens around 700 K for the catalyst annealed
at 673. Higher annealing temperature (773 K) led to the shift of the
hydrogen desorption peak to higher temperature and less amount of
hydrogen desorption, which can be attributed to the bigger Ni3P parti-
cles at higher annealing temperature and thus lower dispersion degree
of metallic nickel.
Fig. 5 shows the TEM images of the bulk Ni3P sample synthesized
under a pH value of 5 at 423 K for 12 h followed by an annealing treat-
ment at 773 K in a flowing N2 for 3 h. It was found the synthesized Ni3P
exhibited the morphology of hexagonal prisms. The hexagonal prisms
are about 70 nm long and have a length/diameter ratio of 2 to 1,
which is roughly in line with the particle size calculated according
Scherrer equation (Table S1).
The catalytic reactivities of the bulk Ni3P catalysts annealed at differ-
ent temperatures are shown in Table 2. No glycerol conversion was de-
tected in the absence of catalyst under the sample experimental
conditions. It was found the catalysts exhibited a relatively low catalytic
glycerol conversion, which can be attributed to the low reaction tem-
perature (463 K) that was restricted by the PTFE lining of the autoclave.
The raw glycerol was highly selectively converted to 1,2-propanediol,
673 K
773 K
Table 2
Catalytic activities of the Ni3P catalysts synthesized under a pH value of 5 at 423 K for 12 h
followed by an annealing treatment at different temperatures in a flowing N2 for 3 h.a
Annealing
temperature (K)
Conversion (%) Selectivity (%)
1,2-Propanediol 1,3-Propanediol Ethanol
400
500
600
700
800
900
Temperature
(
K
)
673
773
873
3.6
5.0
4.4
81.4
86.4
77.4
5.8
5.7
6.4
9.7
5.5
12.3
Fig. 4. The H2-TPD profiles of the samples synthesized under a pH value of 5 at 423 K for
12 h followed by an annealing treatment at different temperatures in a flowing N2 of
100 sccm for 3 h. (For interpretation of the references to colour in this figure, the reader
is referred to the web version of this article.)
a
Reaction conditions: 463 K, 5.5 MPa, residence time 4 h−1, 10% (wt.) glycerol in de-
ionized water.

G. Shi et al. / Catalysis Communications 59 (2015) 180–183
183
Table 3
prisms and exhibit the low-temperature activity to selective glycerol
hydrogenolysis and the high selectivity to 1,2-propanediol.
Stability in catalytic activity of the Ni3P catalyst synthesized under a pH value of 5 at 423 K
for 12 h followed by an annealing treatment at 773 K in a flowing N2 for 3 h.a
Used time
Conversion (%)
Selectivity (%)
Acknowledgement
1,2-Propanediol
1,3-Propanediol
Ethanol
The authors gratefully acknowledge the financial supports from
the Natural Science Foundation of Jiangsu Province (BK2012681)
and PhD Programs Foundation of Ministry of Education of China
(20123250120008) and A Project Funded by the Priority Academic
Program Development of Jiangsu Higher Education Institutions.
1
2
3
5.0
6.4
5.4
86.4
93.7
95.3
5.7
1.3
0.6
5.5
1.7
3.9
a
Reaction conditions: 463 K, 5.5 MPa, residence time 4 h−1, 10% (wt.) glycerol in de-
ionized water.
Appendix A. Supplementary data
which has a much higher value than glycerol. It was also found that the
Ni3P catalyst annealed at 773 K presented the highest glycerol conver-
sion among the investigated catalysts. The further research was carried
out to investigate whether the catalytic activity was stable under reac-
tion conditions. As is shown in Table 3, the Ni3P catalyst synthesized
under a pH value of 5 at 423 K followed by an annealing treatment at
773 K exhibited a relatively stable activity. It is interesting to note that
the recycled catalysts showed higher selectivities to 1,2-propanediol
in the 2nd and the 3rd runs, which could be attributed to the presence
of the induction period of the bulk Ni3P catalysts in the reaction from
glycerol to 1,2-propanediol. Further, the Ni3P phase of the bulk catalyst
was found unchanged by XRD after it was used to catalyze glycerol
hydrogenolysis to 1,2-propanediol (Fig. S2).
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2014.10.020. These data include MOL files
and InChiKeys of the most important compounds described in this
article.
References
[1] R. Prins, M.E. Bussell, Catal. Lett. 142 (2012) 1413–1436.
[2] S.T. Oyama, J. Catal. 216 (2003) 343–352.
[3] G. Shi, J. Shen, J. Mater. Chem. 19 (2009) 2295–2297.
[4] G. Shi, J. Shen, Catal. Commun. 10 (2009) 1693–1696.
[5] H. Pfeiffer, F. Tancret, T. Brousse, Mater. Chem. Phys. 92 (2005) 534–539.
[6] D.Y. Ding, J.N. Wang, Z.L. Cao, J.H. Dai, F. Fu, Chem. Phys. Lett. 371 (2003) 333–336.
[7] X. Liu, J. Chen, J. Zhang, Catal. Commun. 8 (2007) 1905–1909.
[8] K.C. Hung, Y.C. Chan, J. Mater. Sci. Lett. 19 (2000) 1755–1757.
[9] Z. Huang, F. Cui, J. Xue, J. Zuo, J. Chen, C. Xia, Catal. Today 183 (2012) 42–51.
[10] F. Vila, M. Lόpez Granados, M. Ojeda, J.L.G. Fierro, R. Mariscal, Catal. Today 187
(2012) 122–128.
4. Conclusions
[11] J. Zhou, J. Zhang, X. Guo, J. Mao, S. Zhang, Green Chem. 14 (2012) 156–163.
[12] I. Jiménez-Morales, F. Vila, R. Mariscal, A. Jiménez-López, Appl. Catal. B 117–118
(2012) 253.
[13] S. Lee, D. Moon, Catal. Today 174 (2011) 10.
[14] N. Hamzah, N.M. Nordin, A.H.A. Nadzri, Y.A. Nik, M.B. Kassim, M.A. Yarmo, Appl.
Catal. A 419–420 (2012) 133.
[15] B. Peng, C. Zhao, I. Mejía-Centeno, G.A. Fuentes, A. Jentys, J.A. Lercher, Catal. Today
183 (2012) 3.
[16] M.M. Grazia, S.L. Antonio, M. Francesco, P. Rosario, ChemSusChem 4 (2011) 1143.
In summary, the bulk trinickel phosphide catalyst with the smallest
phosphorus content among the nickel phosphides can be synthesized
by a hydrothermal method followed by an annealing treatment. The
optimized synthesis conditions includes a P:Ni ratio of 3 to 1 and a pH
value of 5 in the hydrothermal synthesis stage, and a calcination
temperature of 773 K in annealing treatment. The resulting bulk
trinickel phosphide catalysts present the morphology of hexagonal