Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of glycerol

Article abstract of DOI:10.1016/j.cattod.2016.06.004

The modification effect of P₂O₅ on Cu/SiO₂ catalysts was studied for the hydrogenolysis of glycerol. A series of Cu/P(y)-SiO₂ catalysts with three different loadings of P₂O₅ (2, 3 and 4 wt%) was prepared by successive wet impregnation, and characterized using X-ray Diffraction (XRD), nitrogen sorption, surface acidity technique, UV-Diffuse Reflectance Spectroscopy (UV-DRS) and X-ray Photoelectron Spectroscopy (XPS). The catalysts were evaluated in a batch reactor at 220 °C and 5 MPa of H₂. The addition of phosphorus increased the activity, with the highest activity being achieved over the Cu(x)/P(y)-SiO₂ catalyst with P₂O₅ loading of 2 wt%. The increased activity was attributed to a combination of geometrical and electronic effect induced by P₂O₅ on Cu metallic species. In fact, UV-DRS and TPR measurements showed the existence of a Cu-P interaction, which can increase the distance of the Cu-Cu metal bond, and consequently increase the electron density; this effect enhanced the catalytic activity, compensating for the decrease of Cu dispersion (derived from XPS results). However, this applies only to catalysts with low P₂O₅ loadings because the loss of actives sites by the formation of aggregates at higher P₂O₅ loading cannot be compensated for by Cu-P interaction. The addition of P₂O₅ changes the products distribution because the high electron density of Cu-P favors the adsorption of the C[sbnd]O bond on glycerol.

Full text of DOI:10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Effect of phosphorus on the activity of Cu/SiO2 catalysts in the
hydrogenolysis of glycerol
C. Sepúlvedaa,∗, L. Delgadoa, R. Garcíaa, M. Melendrezb, J.L.G. Fierroc, I.T. Ghampsond,
N. Escalonad,e
a Universidad de Concepción, Facultad de Ciencias Químicas, Edmundo Larenas 129, Casilla 160C, Chile
b Universidad de Concepción, Facultad de Ingeniería, Edmundo Larenas 270 exterior, Casilla 160C, Chile
c Instituto de Catálisis y Petroleoquímica, CSIC, Cantoblanco, 28049 Madrid, Spain
d Departamento de Ingeniería Química y Bioprocesos, Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Avenida Vicu˜na Mackenna 4860,
Macul, Santiago, Chile
e Facultad de Química, Pontificia Universidad Católica de Chile, Chile
a r t i c l e i n f o
a b s t r a c t
Article history:
The modification effect of P2O5 on Cu/SiO2 catalysts was studied for the hydrogenolysis of glycerol. A
series of Cu/P(y)-SiO2 catalysts with three different loadings of P2O5 (2, 3 and 4 wt%) was prepared by
successive wet impregnation, and characterized using X-ray Diffraction (XRD), nitrogen sorption, surface
acidity technique, UV-Diffuse Reflectance Spectroscopy (UV-DRS) and X-ray Photoelectron Spectroscopy
(XPS). The catalysts were evaluated in a batch reactor at 220 ◦C and 5 MPa of H2. The addition of phos-
phorus increased the activity, with the highest activity being achieved over the Cu(x)/P(y)-SiO2 catalyst
with P2O5 loading of 2 wt%. The increased activity was attributed to a combination of geometrical and
electronic effect induced by P2O5 on Cu metallic species. In fact, UV-DRS and TPR measurements showed
the existence of a Cu-P interaction, which can increase the distance of the Cu-Cu metal bond, and con-
sequently increase the electron density; this effect enhanced the catalytic activity, compensating for the
decrease of Cu dispersion (derived from XPS results). However, this applies only to catalysts with low
P2O5 loadings because the loss of actives sites by the formation of aggregates at higher P2O5 loading
cannot be compensated for by Cu-P interaction. The addition of P2O5 changes the products distribution
because the high electron density of Cu-P favors the adsorption of the C O bond on glycerol.
© 2016 Elsevier B.V. All rights reserved.
Received 2 December 2015
Received in revised form 18 April 2016
Accepted 1 June 2016
Available online xxx
Keywords:
Glycerol
Cu
Hydrogenolysis
Octahedral distortion
Electron density
1. Introduction
quantity of glycerol represents an environmental risk [2]. Thus, it
is critical to find ways to utilize the glycerol produced.
The biofuels industry has become very important in the last
decade due to the gradual depletion of crude oil, as well as increased
awareness of the negative environmental impact of crude oil. In
this context, biodiesel produced from transesterification of vegeta-
bles oils has gained a lot of attention because it can be directly
used as biofuels, and its usage can reduce greenhouse gas (GHG)
emissions in comparison to conventional fuels. However, the trans-
esterification process also produces high quantities of glycerol as
a by-product (around 10% in mass) [1] which has contributed to a
sharp decline in the price of this compound; furthermore, the large
Several researchers have made great efforts to find new applica-
tions of glycerol as a low-cost feedstock for functional derivatives
[3]. For example, a series of selective processes for converting glyc-
erol into products with high-added value have been proposed,
including selective oxidation, etherification, hydrogenolysis, dehy-
dration, fermentation, refined glycerol, among others [1,4–7].
Selective hydrogenolysis of glycerol in the presence of hydrogen
and bifunctional catalysts can produce 1.2-propanediol (1.2-PDO),
1.3-propanediol (1.3-PDO) or ethylene glycol (EG) as main prod-
ucts [8–10]. 1,2 PDO is used principally in the chemical industry
to produce polyester resins, liquid detergents, pharmaceuticals,
cosmetics, flavours and fragrances, personal care, paint, pet food,
antifreeze, among others [1]. On the other hand, 1.3-PDO is com-
monly used for the production of polyesters and polyurethane
resins [11,12].
∗
Corresponding author.
E-mail addresses: cathsepulveda@udec.cl,
catherinesepulvedamunoz@gmail.com (C. Sepúlveda).
http://dx.doi.org/10.1016/j.cattod.2016.06.004
0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
2
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
Among the choices of heterogeneous catalysts for the selective
hydrogenolysis of glycerol, copper has been found to be highly
active because it is a good hydrogenating catalyst [13,14], it exhibits
low activity towards breaking C C bonds, and it is highly efficient in
55 ◦C. Finally, the modified support (P (y)-SiO2) was dried for 12 h
at 120 ◦C and calcined at 550 ◦C for 4.5 h under static air [23].
The Cu/SiO2 and Cu/P(y)-SiO2 catalysts were prepared by
wet impregnation of aqueous copper (II) nitrate (Cu(NO3)2·3H2O,
Aldrich) to obtain a 20 wt% of Cu. The mixture was first homoge-
nized for 4 h at 55 ◦C in a rotary evaporator; then, the excess solvent
was removed under vacuum at 55 ◦C. Finally, the samples were
dried at 120 ◦C for 12 h and calcined at 500 ◦C for 2 h under static air.
Before the reaction, the catalysts were reduced ex-situ at 400 ◦C for
4 h in flowing H2 (60 mL min−1). The Cu and P contents were deter-
mined by atomic absorption spectroscopy using a Thermo Scientific
ICE 3000 model series.
C
O bonds hydrogenation and dehydrogenation. Copper catalysts
have been reported to be highly selective towards propanediols,
especially1.2-PDO [8,10,15]. Vila et al. [16] proposed that glycerol
conversion is favored by an optimal Cu0/Cu+ atomic ratio, while
the selectivity towards 1.2-PDO formation was attributed to the
principal presence of Cu0.
In relation to supports, acidic supports with diverse textu-
ral properties such as SiO2-Al2O3, zeolites, Al2O3, ZrO2, activated
carbons, etc. have been used to prepare Cu catalysts for the
hydrogenolysis of glycerol because of the critical functions of
the acidic and textural properties of the supports on this reac-
tion [16–20]. However, SiO2 can also be an effective support
for glycerol hydrogenolysis: its activity and selectivity has been
reported to depend on the interplay between the impregnation
method and the high metal specific surface area of the active
component [8,21].
As previously mentioned, the acidity of the catalysts is fun-
damental in this type of reaction. Phosphorous has been used to
modify the acidic nature of catalysts to change the activity and
selectivity in hydrotreating reactions [22–24]. In hydrodesulfuriza-
tion (HDS) reactions, the addition of P to NiMo/Al2O3 catalysts led to
an optimal combination of acidity and porosity which improved the
dispersion of the active phase on the support, enhancing the activ-
ity and stability of the catalysts [22]. Along the same line, Leiva
et al. [23] studied the effect of the addition of P on Mo/␥-Al2O3
catalysts in hydrodeoxygenation (HDO) reaction, specifically, the
conversion of guaiacol. The authors found that the addition of P
modified the activity which reached a maximum activity at 1.0 wt%
P. This change was correlated with MoS2 dispersion. Also, the prod-
ucts selectivity was modified, attributed to changes in acid strength
by P addition. Huang et. al [24]. compared the activity and selectiv-
ity of Ni2P/SiO2 and Ni/SiO2 catalysts in hydrogenolysis of glycerol
and found that P improves the catalytic activity attributed to syner-
gism of acid and metallic sites from P-OH and Ni, respectively. The
authors proposed that P addition favours C O bond cleavage over
2.2. Characterization of supports and catalysts
The BET specific surface area (SBET) and pore volume of the
supports and catalysts were determined from nitrogen sorption
measurements at −196 ◦C using Micromeritics TriStar II 3020
equipment. Prior to the measurements, the samples were degassed
at 300 ◦C for 2 h. Micropore volume (Vo) was determined by the
Dubinin–Radushkevich method while the total pore volume (Vp)
was recorded by nitrogen adsorption at a relative pressure of 0.99.
Mesopore volume (Vm) was determined by the difference between
Vp and Vo.
Temperature programmed reduction (TPR) was carried out in
a quartz cell in a system equipped with a thermal conductiv-
ity detector. In the experiment, 20 mg of the sample was heated
under 5% H2/Ar flowing at 50 mL min−1. The sample was heated at
10 ◦C min−1 from 25 ◦C to 1050 ◦C.
Powder X-ray Diffraction (XRD) were obtained using a D4
Endeavor Bruker AXS diffractometer equipped with a nickel-
filtered CuK␣1 radiation (␭= 1.5418 Å). The standard scan parame-
ters were 1◦ per min for the 2␪ range from 10 to 90◦. Identification
of the phases was achieved by reference to ICDD files.
The Diffuse Reflectance UV–vis measurements were carried out
using a Lambda 35 spectrophotometer. The Kubelka- Munk’s equa-
tion was used to calculate the intensities (IOH/IOHd) ratio between
Cu2+ octahedral species (IOH) and octahedral distorted Cu2+ species
(IOHd) as a function of P2O5 content.
C
C bond scission observed with Ni/SiO2 catalyst, and attributed
The total acidity of the support and the reduced catalysts
were measured potentiometrically by suspending the catalysts in
acetonitrile and titrating with n-butylamine using an Ag/AgCl elec-
trode [25].
this behaviour principally to geometric and electronic properties.
On the other hand, the effect of phosphorus on Cu catalysts sup-
ported on SiO2 has not been studied. Therefore, the objective of this
work was to study the P effect on the activity and products distribu-
tion of Cu/P(y)-SiO2 catalysts in glycerol conversion. The catalysts
were characterized by different techniques in order to correlate
their activities with the structure of the solids.
The XPS measurements were performed using a VG Escalab
200 R electron spectrometer equipped with a hemispherical elec-
tron analyzer and Mg K␣ (1253.6 eV) X-ray source. The samples
were reduced in-situ at 673 K under H2 flowing and transferred
from the pretreatment chamber to the spectrometer. The inten-
sity of the peaks was estimated by calculating the integral of each
peak after subtracting an S-shaped background and fitting the
experimental curve to a combination of Gaussian/Lorentzian lines.
Atomic ratios were calculated by using peak areas normalized on
the basis of acquisition parameters, sensitivity and transmission
factors provided by the manufacturer, and determined from the
corresponding peak intensities, corrected with tabulated sensitiv-
ity factors, with a precision of 7%.
2. Experimental
2.1. Catalyst preparation
The Cu/SiO2 and Cu/P(y)-SiO2 catalysts were prepared by suc-
cessive impregnation steps. Prior to the metal impregnation, the
SiO2 (Grace 432, Davidson) support was ground and sieved to
obtain particle sizes between 80 and 125 ␮m. Then, the support
was impregnated with an aqueous solution of H3PO4 to obtain final
catalysts with three P loadings (2, 3 and 4 wt.% calculated on P2O5
basis). The range of phosphorous loading was selected according
to previous research, because P loading above 2 wt% increases the
acid strength of the catalysts [23]. Phosphoric acid (85 m/v%) was
also added to SiO2 in addition to 10 mL of deionized water. The pH
was adjusted between 1.5 and 2.0 with HNO3 (65 m/v%). Then, the
mixture was homogenized in a rotary evaporator for 3 h at room
temperature. The excess solvent was removed under vacuum at
2.3. Catalytic reaction test
Hydrogenolysis of glycerol was performed in a 300 mL batch
reactor 4848 Parr model. The concentration of glycerol used was
10 mol L−1. This concentration was selected after rigorous study
of the optimal glycerol amount to avoid the high vapor pressure
of water and the burning of glycerol. Also, the mass of the cata-
lyst used was selected from experimental study of mass of solid
necessary to keep the reaction in kinetic regime. In this context,
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
3
Cu(18.1)/P(4)-SiO
2
0
50000
Cu(18.5)/P(3)-SiO
2
0
Cu(17.5)/P(2)-SiO
2
0
0
Cu(18.4)/SiO
2
100
200
300
400
500
600
Temperature [°C]
Fig. 2. X-ray diffraction patterns of calcined Cu(x)/SiO2 and Cu(x)/P(y)-SiO2 cata-
Fig. 1. Temperature-programmed reduction (TPR) of calcined Cu(x)/SiO2 and
lysts.
Cu(x)/P(y)-SiO2 catalysts.
the liquid reactant feed consisted of an aqueous glycerol solution
(80 wt.%). About 0.5 g of pre-reduced catalyst was added to the reac-
tant. Then, the system was closed and N2 was bubbled through the
solution for 10 min to avoid any air contamination. Still under N2,
the reactor was heated to the reaction temperature of 220 ◦C while
stirring. When the reaction temperature was reached, H2 was intro-
duced into the reactor to 5 MPa which was kept constant during the
course of the experiment. Condensed samples were taken periodi-
cally during the reaction and were identified and quantified using a
Perkin Elmer Autosystem XL gas chromatograph with a flame ion-
ization detector equipped with a column (Nukol, 30 m x 0.53 mm
x0.5 ␮m). The catalytic activity was calculated after 22 h of reaction
according to Eq. (1):
peak at around 300 ◦C, while a low-intense shoulder can also be
observed around 340 ◦C. The peak at 300 ◦C can be attributed to the
reduction of highly dispersed CuO species from Cu2+ to Cu0 [26,27],
while the shoulder at 340 ◦C can be attributed to the reduction Cu+
to Cu0 [28,29], suggesting that this is an intermediate species in the
reduction step. In contrast, the Cu(x)/P(y)-SiO2 catalysts presented
less intense peaks at 300 ◦C attributed to the reduction of highly
dispersed CuO [26,27], while the principal peak observed at 350 ◦C
was assigned to the reduction of Cu2+-like O Cu-P interaction [30].
Finally, the formation of phosphates or Cu polyphosphates was not
detected from TPR analyses. The H2 consumption was calculated
after calibration by the reduction of a reference CuO solid: the
Cu(x)/SiO2 and Cu(x)/P(y)-SiO2 catalysts presented Cu reducibility
of 95% and 98%, respectively, indicating almost complete reduction
of Cu species.
m
r =
(1)
gcat × s
where m is mol of glycerol converted after 22 h calculated from
initial mol of reactant introduced in the reactor, gcat is the mass of
the catalyst (grams), and s is the total time of reaction expressed in
second. The products distribution was calculated when the reaction
was completed.
Fig. 2 presents the XRD results of calcined un-modified and mod-
ified catalysts. All the catalysts showed diffraction peaks at 2␪ = 31◦,
35◦, 38◦, 48◦, 53◦, 58◦, 61◦, 66◦, 67◦, 72◦, 74◦ and 82◦, assigned to
CuO [29]. This result suggests that the addition of small amounts of
P to the support did not modify the Cu crystalline structure. In addi-
tion, phosphates or Cu polyphosphates species were not detected,
in agreement with TPR results.
3. Results and discussion
The UV-DRS spectra of Cu(18.4)/SiO2 and Cu(x)/P(y)-SiO2 cata-
lysts are shown in Fig. 3. The P(2)-SiO2 support displayed a band
between 280 and 400 nm, while the other catalysts displayed two
reflectance bands between 250 and 580 and 800–1000 nm. The
bands between 280 and 580 nm were associated with octahe-
dral geometry (hexacordinated symmetry) of Cu2+ species [31–33].
Meanwhile, the bands around 800–1000 nm were attributed to
d-d electronic transition of Cu2+ according to distorted octahe-
dral geometry (pentacoordinated symmetry) [31–33]. In order to
determine the effect of P2O5 on the symmetry of CuO species,
the intensities (IOH/IOHd) ratio between Cu2+ octahedral species
(IOH) and octahedral distorted Cu2+ species (IOHd) as a function
3.1. Catalysts and support characterization
The textural properties of the SiO2 and P(y)-SiO2 supports,
Cu(18.5)/SiO2 and Cu(x)/P(y)-SiO2 catalysts are presented in
Table 1. The addition of P to SiO2 support slightly modified the
textural properties: the observed changes in surface area can be
attributed to the 10% error associated with these measurements.
Upon Cu impregnation, the textural properties decreased which
can be attributed to pore blockage by Cu species.
Fig. 1 shows the TPR profiles of calcined Cu(x)/SiO2 and
Cu(x)/P(y)-SiO2 catalysts. In this figure, two reduction peaks can be
seen in all the catalysts. The Cu(18.4)/SiO2 catalyst displayed a main
of P2O5 content is shown in Fig. 4. It can be seen that (IOH/IOHd
)
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
4
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
Table 1
Textural and chemical properties of SiO2 and P(y)-SiO2 supports, Cu(x)/SiO2 and Cu(x)/P(y)-SiO2 catalysts.
SBET [m2g−1
]
Vp [cm3g−1
a
]
Vo [cm3g ]
b
Vm [cm3g ]
c
Total acidity (mV)
Supports
SiO2
P(2)-SiO2
P(3)-SiO2
P(4)-SiO2
320
298
321
303
1.13
1.07
1.10
1.07
0.14
0.14
0.15
0.14
0.98
0.93
0.95
0.93
−2
399
–
322
Catalysts
Cu(18.4)/SiO2
Cu(17.5)/P(2)- SiO2
Cu(18.5)/P(3)- SiO2
Cu(18.1)/P(4)- SiO2
238
229
239
251
0.84
0.79
0.83
0.87
0.11
0.11
0.11
0.12
0.73
0.69
0.72
0.75
−50.8
−66.3
−72.3
−58.7
a
Vp: pore volume.
Vo: micropore volume.
Vm: mesopore volume.
b
c
Table 2
120
110
100
90
XPS binding energies (eV) and surface atomic ratio of Cu(x)/SiO2 and Cu(x)/P(y)-SiO2
reduced catalysts.
Catalysts
Cu2p3/2 (eV)
P2p (eV)
Cu/Si at
P/Si at
Cu(18.4)/SiO2
932.7
932.6
932.7
932.8
–
0.218
0.189
0.178
0.172
–
P(2)-SiO2
Cu(17.5)/P(2)-SiO2
Cu(18.5)/P(3)-SiO2
Cu(18.1)/P(4)-SiO2
133.7
133.8
133.8
0.0020
0.0024
0.0029
80
Cu(18.4)/SiO2
Cu(18.5)/P(3)-SiO2
Cu(17.5)/P(2)-SiO2
Cu(18.1)/P(4)-SiO2
70
60
50
40
TPR results. The distortion of the octahedral symmetry of Cu by
P2O5 could increase the distance of the Cu-Cu metallic bonds, sim-
ilar to that observed for Ni catalysts [34], favouring an increase
in the electron density of Cu-P [35]. This effect may increase the
catalytic activity of Cu(x)/P(y)-SiO2 in comparison to Cu(18.4)/SiO2
catalysts.
The acid strength determined by potentiometric titration and
expressed as total acidity (E0) is summarized in Table 1. All the
solids were reduced prior to the measurements. The acid strength
can be determined according to the criterion proposed by Cid and
Pecchi [25]: E0 > 100 mV, very strong acid sites; 0 < E0 < 100 mV,
strong acid sites; −100 < E0 < 0 mV, weak acid sites; E0 < −100 mV,
very weak acid sites. The results indicate that the addition of P
increased the acidity of the SiO2 support to very strong acid sites.
The results suggest that acid strength depends on the nature of P-
SiO2 interaction. Fitz et al. [36] proposed that one hydroxide group
bonded to tetrahedral Al ions on the surface of Al2O3 supports can
react with one of the four hydroxide groups of H3PO4 (aq) to form a
200
400
600
800
1000
Wavelenght (nm)
Fig. 3. UV–vis diffuse reflectance spectra (UV-DRS) of calcined Cu(18.4)/SiO2 and
Cu(x)/P(y)-SiO2 catalysts.
1,6
1,4
1,2
1,0
O
O bond. The authors suggested that there will be an increase in
the surface acidity of the solid if the phosphoric acid molecule binds
to alumina, due to the presence of two available acidic H2 from the
phosphoric acid. Therefore, the observed increase of acidity after
P addition can be attributed to P-Si interaction by one OH-group
of SiO2 formed in the aqueous phase and one OH-group of phos-
phoric acid during impregnation. Notably, the Cu/SiO2 and all the
Cu/P(y)-SiO2 catalysts presented weak acid sites after impregna-
tion of copper: this behaviour can be attributed to the interaction
of Cu2+ with OH-sites of SiO2 and P2O5-SiO2 which decreased the
overall acidity.
Fig. 5 shows the XP spectra of reduced Cu(18.4)/SiO2 (5a) and
Cu(x)/P(y)-SiO2 (5b, c and d) catalysts. Fig. 5(a–d) shows only one
doublet for the Cu 2p3/2 peak for all the catalysts, indicating the
presence of only one phase. Table 2 summarizes the binding ener-
gies (BE) of the most intense Cu 2p3/2 component of each doublet,
the P 2p3/2 contribution, the Cu/Si, and P/Si surface atomic ratios for
all catalysts studied. Table 2 shows that the contribution of binding
energies for Cu in all the catalysts is at 932.7 0.1 eV, assigned to
Cu0 species [16,37], indicating that P addition did not modify the
nature of the active sites. The binding energy at 133.7 0.1 eV for
P 2p3/2 is assigned to P2O5 species [38]. Fig. 6 shows the Cu/Si and
P/Si atomic ratios as a function of P2O5 wt%: a gradual increase of
0,8
0,0
0
1
2
3
4
5
wt.% of P2O5
Fig. 4. Intensities (IOH/IOHd) ratio between Cu2+ octahedral species (IOH) and octa-
hedral distorted Cu2+ species (IOHd)n as a function of P2O5 content.
ratio decreased after P2O5 addition, reaching the minimum value
at 2 wt% of P2O5, and then increased slightly with P content. These
results suggest that the addition of P2O5 distorts the characteris-
tic octahedral symmetry of CuO species formed during calcination,
probably due to an effect of O-Cu-P interaction, as suggested by
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
5
b)
a)
Cu2p
Cu(17.5)/P(2)-SiO2
Cu2p
Cu(18.4)/SiO2
924
936
948
BE (eV)
960
924
936
948
BE (eV)
960
c)
Cu2p
Cu(18.5)/P(3)-SiO2
d)
Cu2p
Cu(18.1)/P(4)-SiO2
924
936
948
BE (eV)
960
924
936
948
BE (eV)
960
Fig. 5. XP spectra of Cu 2p in reduced a) Cu(18.4)/SiO2 b) Cu(17.5)/P(2)-SiO2 and c) Cu(18.5)/P(3)-SiO2 and d) Cu(18.1)/P(4)-SiO2 catalysts.
P/Si atomic ratio with P2O5 content can be observed, suggesting a
homogeneous distribution of phosphorous on the support surface.
Conversely, Cu/Si atomic ratio decreased with P2O5 loading. This
observed trend suggests that Cu-P interaction decreases the homo-
geneous distribution of Cu on SiO2, leading to a gradual loss of Cu
dispersion. In fact, TPR results show the reduction of CuO species
interacting with P2O5, while UV-DRS results (Fig. 4) suggest a dis-
tortion of the octahedral geometry of some CuO species because
of the interaction of Cu-P (P2O5). Taken together, these results are
consistent with the total acidity measurements, where an observed
increase in acid strength after P addition was attributed to P-SiO2
interaction.
glycerol can convert to ethyleneglycol and methanol by hydro-
genation; (ii) dehydration of glycerol via C O bond cleavage,
forming hydroxyacetone (acetol) and 3-hydroxipropanal as inter-
mediates, and subsequent hydrogenation to 1.2-PDO and 1.3-PDO
[15–17]. 3-hydroxipropanal can be further dehydrated to 2-
propenal (acrolein) and propanoic acid [15–17].
Table 3 summarizes the rate of glycerol transformation (mol of
glycerol converted per gram of catalyst per second) on all the cata-
lysts studied, and the distribution of the main products (expressed
in percentage). This table shows that the activity increases with
2 wt% P loading and then decreases. Huang et al. [24] proposed that
the electronic and geometrical properties of catalysts induced by
the addition of P may influence the adsorption of glycerol and, thus,
influence the selectivity. Along the same line, Lee and Oyama [35]
suggested that Ni2P catalysts has a high electron density which
is capable of forming ␲-back-bonded CO species on Ni sites, and
moderate acidity in the form of PO-H form; both of these species
3.2. Catalytic activity of Cu(x)/SiO2 and Cu(x)/P(y)-SiO2 catalysts
The transformation pathway in the hydrogenolysis of glyc-
erol include two principal routes: (i) C C bond breaking, were
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
6
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
Table 3
Catalytic activity of Cu(x)/SiO2 and Cu(x)/P(y)-SiO2 catalysts.
Catalysts
Rate *107 (mol gcat−1 s−1
)
Products distribution (%)
1.2PDO
1.3PDO
Propanic acid
acrolein
1-propanol
Ethyleneglycol
Cu(18.4)/SiO2
9.9
23.5
2.1
5.0
7.9
14.0
30.3
28.9
29.0
8.4
36.8
33.4
22.9
28.2
9.5
Cu(17.5)/P(2)-SiO2
Cu(18.5)/P(3)-SiO2
Cu(18.1)/P(4)-SiO2
22.4
18.6
8.9
10.2
12.7
11.6
19.9
15.2
14.6
13.1
15.4
14.3
2.4
0,004
0,003
0,002
0,001
0,24
0,22
0,20
0,18
Cu/Si atomic ratio. The loss of dispersion due to Cu-P interaction
was compensated for by an increase of Cu-P electron density only
at 2 wt% of P2O5 loading, which favors adsorption on the C O bond
of glycerol. For the catalysts with P content above 2 wt%, the for-
mation of aggregates led to loss of active sites and a consequent
decrease in activity. In relation to products distribution, the addi-
tion of P led to an increase of dehydrated and hydrated compounds
such as propanoic acid and acrolein due to P-OH and Si-OH acid
sites formed in aqueous solution during the reaction.
Acknowledgments
0,000
0
0,00
C. Sepúlveda thanks CONICYT-Chile for FONDECYT N◦
11130376, and FONDEQUIP EQM 120096 grants.
1
2
3
4
wt% of P2O5
References
Fig. 6. Calculated and measured XPS P/Si and Cu/Si atomic ratios as a function of
P2O5 content.
[1] C.-H. Zhou, J.N. Beltramini, Y.-X. Fan, G.Q. Lu, Chem. Soc. Rev. 37 (2008)
527–549.
[2] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[3] Y. Gu, J. Barrault, F. Jérôme, Adv. Synth. Catal. 350 (2008) 2007–2012.
[4] S. Gil, M. Marchena, L. Sánchez-Silva, A. Romero, P. Sánchez, J.L. Valverde,
Chem. Eng. J. 178 (2011) 423–435.
[5] Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catal. Commun. 6
(2005) 645–649.
[6] M.S. Khayoon, B.H. Hameed, Appl. Catal. A 433–434 (2012) 152–161.
[7] F.J. Gutiérrez Ortiz, F.J. Campanario, P.G. Aguilera, P. Ollero, Energy 84 (2015)
634–642.
[8] A. Bienholz, H. Hofmann, P. Claus, Appl. Catal. A 391 (2011) 153–157.
[9] S.S. Priya, V.P. Kumar, M.L. Kantam, S.K. Bhargava, S. Periasamy, K.V.R. Chary,
Appl. Catal. A 498 (2015) 88–98.
[10] Z. Huang, F. Cui, J. Xue, J. Zuo, J. Chen, C. Xia, Catal. Today 183 (2012) 42–51.
[11] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[12] R.M. West, E.L. Kunkes, D.A. Simonetti, J.A. Dumesic, Catal. Today 147 (2009)
115–125.
[13] L. Li, D. Mao, J. Yu, X. Guo, J. Power Sources 279 (2015) 394–404.
[14] E.S. Vasiliadou, T.M. Eggenhuisen, P. Munnik, P.E. de Jongh, K.P. de Jong, A.A.
Lemonidou, Appl. Catal. B 145 (2014) 108–119.
[15] R.V. Sharma, P. Kumar, A.K. Dalai, Appl. Catal. A 477 (2014) 147–156.
[16] F. Vila, M. López Granados, M. Ojeda, J.L.G. Fierro, R. Mariscal, Catal. Today 187
(2012) 122–128.
[17] I. Gandarias, P.L. Arias, J. Requies, M.B. Güemez, J.L.G. Fierro, Appl. Catal. B 97
(2010) 248–256.
[18] L. Ma, D. He, Z. Li, Catal. Commun. 9 (2008) 2489–2495.
[19] N. Hamzah, N.M. Nordin, A.H.A. Nadzri, Y.A. Nik, M.B. Kassim, M.A. Yarmo,
Appl. Catal. A 419–420 (2012) 133–141.
[20] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
[21] E.S. Vasiliadou, A.A. Lemonidou, Appl. Catal. A 396 (2011) 177–185.
[22] P. Rayo, J. Ramírez, P. Torres-Mancera, G. Marroquín, S.K. Maity, J. Ancheyta,
Fuel 100 (2012) 34–42.
[23] K. Leiva, C. Sepúlveda, R. García, J.L.G. Fierro, G. Aguila, P. Baeza, M. Villarroel,
N. Escalona, Appl. Catal. A 467 (2013) 568–574.
enhanced the activity in the catalytic removal of N and S com-
pounds, and probably O compounds too. Therefore, the distortion of
the octahedral geometry of CuO species with P2O5 addition (Fig. 4)
can lead to higher Cu-Cu metallic distance, similar to that observed
for Ni [34], increasing the electron density [35]. This will conse-
quently increase the activity in glycerol conversion. However, the
electronic and geometrical effect of P2O5 on Cu catalysts favors the
catalytic activity only with the lowest P content (2 wt% of P2O5),
were this effect compensates for the loss of dispersion observed by
XPS measurements. For the catalysts with higher P content (3 and
4 wt% of P2O5) a more remarkable decrease of Cu dispersion (XPS
results) lead to loss of active sites by aggregates formation, leading
to decrease in activity despite the presence of Cu-P interaction.
Table 3 presents the products distribution over Cu(18.4)/SiO2
and Cu(x)/P(y)-SiO2 catalysts. It shows that before P2O5 addition,
the principal products were 1.2-PDO and 1-propanol, which are
typical products from supported Cu catalysts. However, all the
Cu(x)/P(y)-SiO2 catalysts presented 1-propanol and propanoic acid
as principal products of the reaction. According to several proposed
reaction schemes [8,16,17], acrolein is formed after two dehydra-
tion steps, followed by hydration to propanoic acid. Therefore, the
increase of the formation of acrolein and propanoic acid can be
attributed to P-OH and Si-OH acid sites formed in the aqueous solu-
tion during the reaction [35], which also improved the catalytic
activity by Cu stabilization on the surface of the support.
[24] J. Huang, J. Chen, Chin. J. Catal. 33 (2012) 790–796.
[25] R. Cid, G. Pecchi, Appl. Catal. 14 (1985) 15–21.
[26] K.V.R. Chary, K.K. Seela, G.V. Sagar, B. Sreedhar, J. Phys. Chem. B 108 (2004)
658–663.
4. Conclusions
The present study showed that the addition of P2O5 modified
the reactivity of supported Cu catalysts and had a positive effect
on the hydrogenolysis activity of glycerol. The addition of P2O5
induced changes in the electronic and geometrical properties of
Cu(x)/P-SiO2 catalysts. The Cu-P interactions inferred from TPR
measurements were supported by observed distortion of the octa-
hedral symmetry of CuO by the presence of P2O5 determined by
UV-DRS analyses. This observed interaction led to a decrease of the
[27] A.J. Marchi, J.L.G. Fierro, J. Santamaría, A. Monzón, Appl. Catal. A 142 (1996)
375–386.
[28] L.-F. Chen, P.-J. Guo, M.-H. Qiao, S.-R. Yan, H.-X. Li, W. Shen, H.-L. Xu, K.-N. Fan,
J. Catal. 257 (2008) 172–180.
[29] M.L. Smith, A. Campos, J.J. Spivey, Catal. Today 182 (2012) 60–66.
[30] G.V. Mamontov, O.V. Magaev, A.S. Knyazev, O.V. Vodyankina, Catal. Today 203
(2013) 122–126.
[31] R. Hierl, H. Knözinger, H.-P. Urbach, J. Catal. 69 (1981) 475–486.
[32] H. Praliaud, S. Mikhailenko, Z. Chajar, M. Primet, Appl. Catal. B 16 (1998)
359–374.
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004

G Model
CATTOD-10238; No. of Pages7
ARTICLE IN PRESS
C. Sepúlveda et al. / Catalysis Today xxx (2016) xxx–xxx
7
[33] B.R. Strohmeier, D.E. Levden, R.S. Field, D.M. Hercules, J. Catal. 94 (1985)
514–530.
[37] J. Agrell, H. Birgersson, M. Boutonnet, I. Melián-Cabrera, R.M. Navarro, J.L.G.
Fierro, J. Catal. 219 (2003) 389–403.
[34] C. Stinner, Z. Tang, M. Haouas, T. Weber, R. Prins, J. Catal. 208 (2002) 456–466.
[35] Y.-K. Lee, S.T. Oyama, J. Catal. 239 (2006) 376–389.
[36] C.W. Fitz, H.F. Rase, Ind. Eng. Chem. Prod. Rd. 22 (1983) 40–44.
[38] T.A. Zepeda, B. Pawelec, J.L.G. Fierro, A. Montesinos, A. Olivas, S. Fuentes, T.
Halachev, Microporous Mesoporous Mat. 111 (2008) 493–506.
Please cite this article in press as: C. Sepúlveda, et al., Effect of phosphorus on the activity of Cu/SiO2 catalysts in the hydrogenolysis of
glycerol, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.06.004