A study of glycerol hydrogenolysis over Ru-Cu/Al2O3 and Ru-Cu/ZrO2 catalysts

Article abstract of DOI:10.1016/j.molcata.2016.01.027

This work focuses on the activity, selectivity, and characterization of alumina and zirconia supported Ru(2.5% wt.)-Cu(2.5% wt.) bimetallic catalysts to produce 1,2-propanediol (1,2-PDO) from glycerol hydrogenolysis. Characterization techniques included XRD, BET, TEM, H₂ chemisorption, TPR and DRIFTS. Turnover frequencies were obtained for batch reactions ran at different temperatures. At 473 K and 2.5 MPa of dihydrogen, a conversion of 45% of glycerol with 94% selectivity toward 1,2-PDO was found when the Ru(2.5% wt.)-Cu(2.5% wt.)/Al₂O₃ was used. Recycle experiments were done for the monometallic catalysts, in which Ru/Al₂O₃ showed a better performance.

Full text of DOI:10.1016/j.molcata.2016.01.027

Accepted Manuscript
Title: A study of glycerol hydrogenolysis over Ru-Cu/Al O
2
3
and Ru-Cu/ZrO catalysts
2
Author: Andr e´ V.H. Soares Joyce B. Salazar Derek D. Falcone
Fernanda A. Vasconcellos Robert J. Davis Fabio B. Passos
PII:
S1381-1169(16)30027-9
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.01.027
MOLCAA 9759
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
30-10-2015
12-1-2016
22-1-2016
Please cite this article as: Andr e´ V.H.Soares, Joyce B.Salazar, Derek D.Falcone,
Fernanda A.Vasconcellos, Robert J.Davis, Fabio B.Passos, A study of glycerol
hydrogenolysis over Ru-Cu/Al2O3 and Ru-Cu/ZrO2 catalysts, Journal of Molecular
Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2016.01.027
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

A study of glycerol hydrogenolysis over Ru-Cu/Al
2
O
3
and Ru-Cu/ZrO
2
catalysts
a,b
a
c
a
André V. H. Soares , Joyce B. Salazar , Derek D. Falcone , Fernanda A. Vasconcellos
c
a,
Robert J. Davis and Fabio B. Passos *
aDepartamento de Engenharia Química e de Petróleo, Universidade Federal Fluminense, Rua Passo da Pátria,
1
56, Niterói, RJ CEP 24210-240, Brazil
bInstituto Federal do Rio de Janeiro – IFRJ, Avenida República do Paraguai, 120, Sarapuí - Duque de Caxias
CEP: 25050-100, Brazil
cDepartment of Chemical Engineering, University of Virginia, 102 Engineers Way, Charlottesville, VA 22904-
4
741, USA
*
Corresponding author email addresses: fbpassos@vm.uff.br (F. B. Passos)
Graphical abstract
1

Highlights




Ru‐Cu/Al
Ru‐Cu/ZrO
Ru‐Cu/ZrO
2
O
2
3
and Ru‐Cu/ZrO have been investigated in glycerol hydrogenolysis.
displayed higher turnover frequencies.
was more selective for 1,2 propanediol.
2
2
Optimum Ru/Cu ratio was obtained.
Abstract
This work focuses on the activity, selectivity, and characterization of alumina and zirconia
supported Ru(2.5%wt.)-Cu(2.5%wt.) bimetallic catalysts to produce 1,2-propanediol (1,2-
PDO) from glycerol hydrogenolysis. Characterization techniques included XRD, BET, TEM,
2
H chemisorption, TPR and DRIFTS. Turnover frequencies were obtained for batch reactions
ran at different temperatures. At 473 K and 2.5 MPa of dihydrogen, a conversion of 45% of
glycerol with 94% selectivity towards 1,2-PDO was found when the Ru(2.5%wt.)-
Cu(2.5%wt.)/Al
in which Ru/Al
2
O
3
was used. Recycle experiments were done for the monometallic catalysts,
2
O
3
showed a better performance.
Keywords: Ruthenium; Copper; Glycerol hydrogenolysis; Propylene glycol;
1
. Introduction
The search for alternative energy sources has arisen because of concerns related to the
environmental impact of using fossil resources. Although the discovery of new oil and natural
gas sources can postpone the complete depletion of fossil resources, the problems associated
with a shortage are expected to occur within the next few decades. In this context, the quest
for sustainability has become an important point in the motivation of academic and industrial
2

scientists and government officials in the planning and development of using renewable
biomass as alternative fuels to mitigate the problem. Much of the chemical industry uses oil as
the main feedstock in the production of fine chemicals, polymers, and fuels. Nowadays, there
is considerable interest to generate power from biomass as well as convert it into chemicals
and value-added materials [1].
Biodiesel is a high quality fuel produced by the transesterification of triglycerides,
derived from renewable sources like plant oils and animal fats. This biofuel is a liquid with
similar combustion properties to diesel oil but it is essentially sulfur free, which makes it
more environmentally friendly [2]. Mixing 2 to 30 % biodiesel into fossil derived diesel does
not require changes to engines, and will produce lower amounts of particulate matter [3].
Biodiesel has been produced on an industrial scale in Europe since 1992 encouraged by
institutions and a growing public concern for the environment. In 2012, biodiesel production
in Europe reached 8.6 million tons, which is slightly lower compared to previous years [4]. In
Brazil, federal legislation [5] has driven a compulsory biodiesel demand, by enforcing a
minimum 5% biodiesel content in all diesel fuel consumed in the country. This has led to an
increase in the number of authorized production units and actual biodiesel production. As of
February 2014, the total regulated monthly production of biodiesel in Brazil was 243,670 m³,
which represents only 37 % of the total authorized capacity in the country. In light of this,
production growth is still expected. The Brazilian biodiesel production comes mainly from
soybean oil (70.9%) and bovine fat (24.8%) [6].
For every 9 kg of biodiesel produced, about 1 kg of glycerol is formed as a byproduct.
Although there are many applications for glycerol, its production has exceeded the demand,
which leads to a depreciation in the value of the product [7].
In this context, the search for economic and environmentally useful alternatives for
glycerol becomes imperative, in order to ensure the continuity and viability of biodiesel
3

production [8]. Among the promising techniques that have been developed for glycerol usage
are: hydrogenolysis to produce 1,2- or 1,3-propanediol; aqueous and gas phase reforming to
form CO and H ; etherification reaction to form polyglycerols; and selective oxidation to
2
produce ketones, acids and aldehydes [9-11].
Glycerol conversion by catalytic hydrogenation or hydrogenolysis to glycols, i.e. 1,2-
propanediol (1,2-PDO), 1,3-propanediol (1,3-PDO), and ethylene glycol (EG), is an attractive
process both economically and environmentally. 1,2-PDO is used for polyester resins,
pharmaceuticals, antifreeze, cosmetics, and tobacco humectants. 1,3-PDO is a highly valued
chemical that is mainly used in specialty polyester fibers, films, and coatings [12-13]. The
various products that may be obtained from glycerol hydrogenolysis and other related
reactions are shown in Figure 1.
Different catalysts have been used towards 1,2-PDO production from glycerol with
and without the use of external hydrogen, and a special attention has been given to Pt
catalysts promoted by other metals [14,15] and Pd catalysts [16,17]. Different groups have
studied Ru catalysts yielding different conversions and selectivities towards 1,2-PDO [18-49].
In one of the earliest works on glycerol hydrogenolysis over ruthenium catalysts, Montassier
et al. [19] observed that monometallic Ru catalyst preferentially breaks the C-C bond instead
of the C-O bond, which results in degradation products. These authors utilized Ru/SiO
2
catalysts and found that methane was the main product, along with ethane and propane, as
expected products of complete hydrogenation.
Special attention should be given to acidic promoters for Ru/C, such as Amberlyst 15,
that reach a conversion of about 79% with a selectivity of 75% to 1,2-PD at 393 K, a
relatively mild temperature [18]. Feng et al. [20] used a 5% Ru/Al
2
3
O to catalyze glycerol
hydrogenolysis at 453 K and a hydrogen pressure of 5 MPa to obtain a conversion of 34.3 %
with a 47.3 % selectivity to 1,2-PDO. As it will be shown, these reported values are similar to
4

those found with the home made catalysts used in this work. Lee and Moon [21] have also
used 5% Ru/Al
PDO.
2
3
O , and obtained a 45.6 % conversion with a selectivity of 59.2 % for 1,2-
When investigating the activity of Cu catalysts, Montassier et al. [19] observed that for
the Cu-Raney catalyst, under a hydrogen pressure of 3 MPa and a temperature of 513 K, 86%
of the product selectivity was 1,2-PDO. Other authors have also tested copper for this reaction
and obtained good selectivity for 1,2-PDO. [22-25]
The use of Ru alloys with other noble metals has been also investigated. Maris &
Davis have investigated glycerol hydrogenolysis over Pt and Ru catalysts [26], as well as the
effect of bimetallic Pt-Ru and Au-Ru catalysts [27], and saw levels of conversions near 100%
in alkali reaction environments, with varying selectivities. The authors have seen that
nonetheless Pt is less active than Ru to activate C-C bonds. Roy et al. [28] used commercial
Pt/Al
2
O
3
and Ru/Al
2
O
3
and showed that hydrogenolysis occurred in the absence of H
2
, by
generating hydrogen via heterolytic cleavage of H
2
O on Pt sites.
When studying the effect of Cu in Ru catalysts, Jiang et al. [29] have worked with Ru-
Cu clay supported catalysts for glycerol hydrogenolysis and have seen that a decrease in
Ru/Cu ratio causes a drop in conversion with a relative increase in 1,2-PDO selectivity.
Rouco et al. [30] used Ru-Cu and Ru-Ag catalysts supported on silica for the hydrogenolysis
of ethane and both catalysts displayed a decrease in activity when compared to the
monometallic ruthenium, since Cu and Ag have a lower activity in breaking C-C bonds. Guo
et al. [31] utilized Cu/Al
2
O
3
under mild conditions (493 K and H pressure of 1.5 MPa),
2
yielding a 49.6 % conversion with a selectivity of 96.8 % for 1,2-PDO. Vasiliadou and co-
authors [32] performed the glycerol hydrogenolysis reaction under 513 K and 8 MPa, finding
a conversion of 40.5 % with a 60.5 % selectivity using Ru/ZrO
2
, and a conversion of 26.7 %
with a 39.7 % selectivity when using Ru/Al
2
3
O .
5

In a work with similar interests as this one, Liu et al. [33] investigated the influence of
the support on Ru-Cu catalysts. The catalysts used by those authors had a Ru/Cu atomic ratio
of about 10/1 and were supported on TiO
2
, SiO
2
, ZrO
2
, Al
2
O , HY and NaY. They noted that
3
Ru-Cu/ZrO had the best performance. Since they observed no 1,3-PDO while acetol was
2
present in the products, the authors have proposed a mechanism that differs from the one
envisaged by Montassier et al.: dehydration of glycerol occurs first, generating acetol, then
hydrogenation takes place. The difference between the two reaction pathways, (Figure 2a)
from Montassier et al. [19] and (Figure 2b) from Liu et al. [33], can be seen in Figure 2.
Balaraju et al. [34] have summarized the question stating that solvent medium and catalyst
acidity have a great influence on reaction mechanism, thus pathway (a) is due to alkali
catalysts while acetol in pathway (b) is due to acidic catalysts.
Shimizu et al. [35] have investigated the interaction of hydrogen with bimetallic Ru-
Cu surfaces and have inferred that, given the non-homogeneous alloy, three different kinds of
regions for Cu can be identified on the surface of the system, wherein even small percentages
of Cu on the surface (>4%) may cause H adsorption to drop drastically, compared to Ru
2
adsorption. These authors have also considered that the immiscibility of the metals, and thus
lack of an ensemble effect, allows the presence of various possible new adsorption sites for
2
H , e.g. kink and terrace sites. Sinfelt et al. [36] studied the Ru-Cu system and found the same
trend on the hydrogen chemisorption capacity that the authors above cited.
However, in our previous work [37], we investigated the interaction of bimetallic Ru-
Cu/TiO surfaces by XPS and the binding energies of both Ru and Cu in the bimetallic
2
catalysts were similar compared to the monometallic catalysts and revealed no clear trend
with composition, suggesting that electron transfer between the two metals in the bimetallic
catalysts was minimal. We had also observed that when Cu was mixed with Ru in the given
6

proportion, there was a slight decrease in H/Ru, likely the result of Cu deposition onto the Ru
surface of 1:1 to 1:5 Ru/Cu. In agreement to the fact that hydrogen chemisorption is not
enhanced by increasing the Cu weight loading, Rouco et al. [30] found that as the nominal
Cu/Ru molar ratio of a Cu-Ru/SiO
2
catalyst was increased from 0 to 0.667, the H/Ru ratio
determined by H chemisorption remained approximately constant.
2
More recently, Cu-based glycerol hydrogenolysis studies have been especially focused
on the use of Y-zeolite supported catalysts [38], on the effect of preparation methods in the
0
+
catalyst activity of Cu/SiO
glycerol hydrogenolysis over Pt in tandem with methanol aqueous phase reforming (APR)
over Cu:Zn:Al bulk catalysts [41]. On the other hand, Ru/TiO [42,43], and Ru/zeolite
2
[39], on Cu-Cr bimetallic sites with high Cu /Cu ratios [40], and
2
systems [44,45] have been trending as proposed heterogeneous catalysts towards glycerol
hydrogenolysis.
The present work has the objectives of describing the preparation, characterization,
and evaluation of the performance of supported Ru-Cu bimetallic catalysts, which favor the
production of 1,2-propanediol, 1,3-propanediol and ethylene glycol from glycerol
hydrogenolysis. Moreover, turnover frequencies (TOF) are calculated for the Ru-Cu systems,
which are scarce in the literature. Bimetallic catalysts are known to produce distinct
selectivities in some reactions, especially hydrogenolysis [36]. The following characterization
techniques were employed: temperature programmed reduction (TPR), X-ray diffraction
(
XRD), textural analysis by N
microscopy (TEM), and Diffuse Reflectance Infrared Fourier Transform Spectroscopy
DRIFTS). To evaluate the influence of reaction parameters, such as hydrogen pressure,
2
adsorption, H
2
chemisorption, transmission electronic
(
temperature, and initial concentration of glycerol, catalytic tests were performed in a high-
pressure reactor.
7

2. Experimental
2
.1. Catalyst preparation
The catalysts prepared for this work were monometallic Ru, monometallic Cu, and
Ru-Cu, each supported on Al or ZrO . The loading of each metal was 2.5 wt %. The
catalysts were prepared by successive incipient wetness impregnations of Ru and Cu. The
catalysts were prepared using Ru(NO)(NO .H O (Aldrich), and Cu(NO .3H O (Sigma
Aldrich) as the precursor salts. Al was obtained by calcination of boehmite (Catapal) in air
2
O
3
2
3
)
3
2
3
)
2
2
2
O
3
-1
for 3 h at 823 K with heating rate of 5 K.min . After metal impregnation, the supported
catalysts were dried at 393 K for 12h followed by calcination in air at 623 K for 4h with
-1
heating rate of 2.5 K.min .
2.2. Catalyst characterization
Surface areas and pore volumes were determined by N physisorption at 77 K, using
2
the BET analysis method employing an ASAP 2010 Micromeritics analyzer. Prior to the
measurements, the samples were degassed at 523 K for 6 h.
The X-Ray Diffraction (XRD) patterns were performed in a Rigaku (Miniflex II)
diffractometer with a CuK (1.540 Å) radiation.
The temperature programmed reduction (TPR) experiments were carried out on a
-
1
sample (0.5 g) that was dried under flow of He (30 mL.min ) for 30 min at 423 K and then
cooled to room temperature. Temperature programmed reduction was performed under 30
-1
mL.min flow of a gas mixture of 5% H /Ar, while the samples were heated to 723 K at 10
2
-1
K.min . The measurements were carried out in a multipurpose unit coupled to a quadrupole
mass spectrometer (Pfeiffer Vacuum, Prisma TM).
8

Hydrogen chemisorption was performed in a Micromeritics ASAP 2020 automated
adsorption analyzer. The samples were reduced at 573 K under flowing H for 60 min, and
then outgassed under vacuum at 573 K for 60 min. After this, the catalysts were cooled to 308
K and evacuated again for 60 min followed by analysis at 308 K. The total amount of H
2
2
adsorbed was calculated by extrapolating the hydrogen uptake to zero pressure, assuming a
stoichiometry H/Rusurf equal to unity.
CO chemisorption measurements were carried out in the same multipurpose unit used
-1
for TPR. The catalyst samples (ca. 50 mg) were pretreated under a flow of He (30 mL.min )
for 30 min at 423 K and then cooled to room temperature. The activation step took place at
-1
5
73 K under a flow of 5%H /Ar (30 mL.min ) for 1 h, followed by flushing under a flow of
2
He for 30 min at 573 K and cooling to 300 K. CO uptake was measured by injecting a number
of CO pulses and the exposed Ru was calculated assuming a CO/Ru molar ratio of 1/1.
Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) of adsorbed
carbon monoxide was performed with a Bruker VERTEX 70 spectrometer with a high-
temperature DRIFTS cell (Harrick, HVC-DRP-4), fitted with ZnSe windows, and a diffuse
reflectance accessory with Praying Mantis geometry. Spectra were acquired at a resolution of
-1
4
cm , typically averaging 256 scans. The samples were initially pretreated with a He flow
-
1
(
(
30 mL.min ) for 30 min at 423 K. Then, the temperature was raised to 573 K in pure H
2
flow
-
1
30 mL.min ) for 1h. After the reduction, the sample was flushed in He flow for 30 min,
followed by cooling at 303 K. CO adsorption was then performed under CO/He flow (30
-1
mL.min ) for 30 min at 303 K, followed by flushing of the sample for 15 min with He and
before recording spectra. To determine the thermal desorption of the adsorbents, the sample
temperature was increased stepwise in continuous He flow. Images of the catalysts were
obtained by transmission electron microscopy (TEM) (FEI Titan) at 300 kV and a Gatan 794
Multi-scan Camera to analyze metal particle size. At least 300 particles were analyzed from
9

each sample. The samples were prepared by suspending the catalysts in ethanol, and agitating
them in an ultrasonic bath for 10 min; then drops of the suspension were applied to a Cu mesh
grid.
2.3. Catalytic tests
Glycerol hydrogenolysis reactions were carried out in a 300 mL stainless steel batch
reactor (Parr 4848). The reaction conditions were: temperature of 453 K, initial hydrogen
pressure of 2.5 MPa, mass of catalyst of 1.2 g and 100 mL of glycerol (20 wt% in water). The
reaction time was 24 hours with stirring of 500 rpm.
Prior to the reaction, the catalyst was reduced at 573 K for 1h, in a separate unit, under
-1
a flow of pure H
2
(30 mL.min ), followed by oxidation under a flow of 5% O
min ) for 15 min at 273 K.
The reaction sequence was as follows: load the reactor with the glycerol solution and
2
/He (15 mL
-1
-
1
the appropriate amount of catalyst, purge with a flow of N
2
(30 mL.min ) for 5 min, and treat
(30 mL.min ) at ambient temperature for 1 h in the reaction vessel. The
temperature and H pressure were then increased to the desired values under constant stirring.
-1
in flowing H
2
2
The batch reaction proceeded for a 24 h period. After 24 h, the system was cooled to ambient
temperature and liquid and gaseous samples were collected. The liquid phase products were
analyzed and identified by a GC-MS (Shimadzu, GCMS-QP2010S) equipped with a 95%
polyethylene glycol (PEG) wax RTX column. The gas phase was sampled and analyzed in a
4
90 Micro-GC (Agilent) equipped with three columns: M5A 9 (H
2
, O
2
, N
2
, CH
4
and CO),
5
CB (PoraPLOT U CO and C ) and PPU(CP-Sil 5CB – Hydrocarbons). CH
2
2
H
6
4
was obtained
as a gas product in an amount less than 1%.
The identified products were: 1,2-PDO, 1,3-PDO and 1-propanol as products of the
hydrogenolysis reaction and ethylene glycol (EG), ethanol, methanol and methane as
1
0

degradation products of glycerol. Product and reactant concentrations were calculated using
measured response factors.
3
. Results and discussion
3.1. Catalyst characterization
The surface areas and pore volumes of the supports and prepared catalysts are shown
in Table 1. As it can be seen, for Al
2
3
O supported catalysts, there was an 8 % decrease in area
by the addition of either 2.5 wt % Ru or 2.5 % Ru and 2.5 % Cu. On the other hand, for ZrO
2
supported catalysts, there was the same decrease in area when adding Ru metal, but adding
both Ru and Cu had a stronger effect on the support surface area.
Figure 3 shows the X-ray diffractograms, in which five characteristic peaks for Ru and RuO
x
0
are observed. The broadest peaks are assigned to Ru . There are no detected Cu peaks in the
bimetallic catalyst. The Scherrer formula provides an inverse proportionality relationship
between crystal size and XRD peak width, in a way that broader peaks are related to smaller
particles [46]. If this relation holds for monometallic Ru/Al
O
2 3
and bimetallic Ru-Cu/Al
2
3
O ,
the average particle sizes lie in the range of 3.5-5 nm. Particle size analysis by TEM showed a
slightly larger particle size, but they were on the same order of magnitude. On the other hand,
peaks corresponding to RuO
x
were not detected on the ZrO supported catalyst probably
2
because peaks from the crystalline zirconia support mask any contributions from RuO
x
.
The TPR profiles for the catalysts used in this work are displayed in Figure 4, and the
results of the analyses are shown in Table 2. The reduction efficiency was calculated as the
amount of H
2
consumed over the theoretical amount required to reduce RuO and CuO.
3
According to the literature, precursor salts seem to have a great effect on reduction
efficiency. Many values for RuO reduction temperature can be found, especially for catalysts
prepared with RuCl as a precursor salt [47-52]. Koopman et al. [47] have reported that for
Ru/SiO , reduction of RuO takes place between 450 K and 478 K, and McNicol and Short
48] have reported that there is a single peak at 443 K for bulk unsupported RuO . According
to Hurst et al. [49], the reduction of RuNO(NO may present different profiles, depending
2
3
2
2
[
2
)
3 x
1
1

on the reduction of NO and the x number of NO
ruthenium oxide reduction takes place at 470 K, while a shoulder may also be seen,
representing the reduction of the oxichloride for RuCl . Using Ru(acac) as the precursor,
occurs with a small peak at 393 K,
3
groups. Mazzieri et al. [50] reported that
3
3
Bianchi [51] has reported that reduction of RuO
3
to RuO
2
0
and that RuO
monometallic Ru/Al
73 K, and a broader and larger peak at 558 K. From these peaks it could be inferred that the
first peak represents higher oxidation states, the shoulder would be RuO and the last peak
RuO . This agrees partially with the range of reduction (450-525 K) reported by Lee and
Moon [21]. The TPR profiles for monometallic Ru/ZrO and Ru/Al obtained in this work
differ notably from each other, which might be caused by different interactions of the
precursor salt and the supports. The Ru/ZrO catalyst shows only one broad peak that begins
at 421 K and ends at about 450 K, whereas Ru/Al has a very broad reduction peak around
58 K.
2
to Ru takes place at 483 K with a much larger peak. In this work,
2
3
O showed a small peak of reduction at about 438 K, a shoulder at about
4
3
2
2
2
O
3
2
2
O
3
5
When investigating TPR profiles for Ru-Cu/Al
2
O
3
, Galvagno et al. [52] noted that
using RuCl as the precursor salt, the addition of Cu generated a shoulder, which shifted the
3
Ru reduction peak closer to the lower Cu reduction temperature. This was attributed to the
interactions between support and the metallic salts. This is corroborated by the present work,
in which it has been seen that the Cu addition favored reduction of Ru at lower temperature
for Ru-Cu/Al
2
O
3
. The bimetallic Ru-Cu/ZrO profile displayed a broader range for reduction,
2
which is related to the addition of Cu. Since cluster formation is expected for Ru and Cu
mixtures, instead of homogeneous alloys [46,53], it can be inferred that the shifts in reduction
temperature may be caused by inter-phase hydrogen adsorption, which favors Cu reduction.
1
2

Another evidence of this is the observed simultaneous reduction of Cu and Ru species for the
bimetallic Ru-Cu/Al and Ru-Cu/ZrO , indicating some interaction between the two metals.
Carbon monoxide DRIFTS results are shown in Figure 5. Cu/Al displayed a sharp
showed a very weak peak at 2064 cm , and the bimetallic Ru-
2
O
3
2
2
O
3
-1
-1
peak at 2100 cm , Ru/Al
Cu/Al showed both distinct 2100 cm and a broad peak at 2009 cm . Gottschalk et al.
45] reported the same value of 2064 cm for Ru/Al
2
O
3
-1
-1
2
O
3
-1
[
2
O
3
. Brown and Gonzalez [55] have
-
1
assigned the 2080 cm frequency to a Ru–CO -bond, as the main peak for Ru/SiO
2
-1
supported catalysts, while Unland [56] found that Ru has weak peaks at 2009 cm and 2070
-
1
at 2125 cm-1.
cm , and Kim et al. [57] have associated the CO chemisorption on Cu/SiO
Therefore, it can be said that while Ru had a poor performance in CO adsorption, Cu, on the
other hand, acted as promoter of adsorption in the Ru-Cu/Al bimetallic catalyst, with a
supported catalysts, although Ru/ZrO displayed a
strong peak at 2070 cm , in accordance with Unland [56], Cu addition blue-shifted the peak
2
2
O
3
-1
blue-shift of about 25 cm . For the ZrO
2
2
-
1
-1
+
to about 2060 cm , decreasing its amplitude. The Cu -CO species is known to occur at 2158
-1
-1
cm for zeolites [58], and the obtained value of 2100 cm for the intense absorbance band of
+
2+
Cu/Al
2
O could be attributed to Cu . The formation of Cu sites seems to be disfavored.
3
TEM micrographs of the catalysts and their respective size distributions are presented
in Figure 6. As it can be seen, ruthenium particles were larger for Ru/ZrO than for Ru/Al
particles were somewhat more
2
2
3
O .
Additionally the dimensions of bimetallic Ru-Cu/Al
2
O
3
frequently larger, in the range of 8-10 nm.
The different techniques used to assess the metal particle size must be confronted with
one another to capture a clear description of the dimensions of the active metallic particles.
First, it should be noted that the particle size distribution derived from TEM analysis is
dependent on sampling, and a truly Gaussian distribution was not obtained for Ru/ZrO , since
2
1
3

it can be clearly seen that particles with 22 nm of diameter are relevant, albeit their relative
smaller amount. These large particles have a great impact in decreasing H uptake given the
lower amount of surface Ru, and they were not observed for Ru/Al , which is consistent
with the better dispersion measured from the H chemisorption analysis. Another possible
explanation is the presence of partially reduced ZrO , which would decorate some of the
palladium particles and suppress hydrogen adsorption capacity.
2
2
O
3
2
2
3
.2. Hydrogenolysis of glycerol
Glycerol conversion and the selectivity of Ru based catalysts used for the
hydrogenolysis reaction are presented in Table 3. Selectivity was calculated as the ratio of
product moles produced over the total moles of glycerol converted. The different sets of
catalysts showed various performances towards the hydrogenolysis reaction. For all of the
evaluated catalysts, the products identified were: 1,2-PDO, EG, 1,3-PDO, 1-propanol, ethanol
and methanol. Methane, CO and CO were also found in the gas phase, but the CO quantities
2
were very low in comparison to all other C1 compounds.
Acetol was detected as a final product in these reactions at very low concentrations.
However, Vasiliadou et al. [32] found a significant amount of acetol after 5 h of reaction.
Their experiment was performed under severe conditions, using 8MPa of hydrogen pressure,
at 513K and pure glycerol. The presence of acetol in every experiment was interpreted by
those authors as a possible intermediate product of the dehydration/hydrogenation
mechanism, which finally yields 1,2-PDO after a hydrogenation step, in agreement with the
1
4

literature [59-62]. The absence of acetol in the present work may be possibly explained by the
longer duration of the reaction (24 h).
The turnover frequency was calculated using the dispersion from hydrogen
chemisorption as a measure for the surface Ru active atoms, and since H
for Ru/ZrO , the number of active Ru atoms is also lower. The reaction rate was similar for
Ru/ZrO and Ru/Al
2
uptake was lower
2
-7
-7
2
2
O
3
(7.05
x
10 mol/gcat.s and 7.92 10 mol/gcat.s, respectively), and
x
therefore the turnover frequency must be greater for larger particles, with comparatively fewer
atoms on the surface than well dispersed catalysts.
The main identified product in all catalytic reactions studied was 1,2-PDO, except for
Ru/Al
was obtained with monometallic Ru/Al
products. When compared with Ru/ZrO , it could be inferred that the extra activity of alumina
is due to acidity, since glycerol protonation may favor OH cleavage. The 1,3-PDO selectivity
2
O
3
, which presented about equal quantities of 1,2-PDO and EG. The best conversion
2
O
3
catalyst, with a varying selectivity to possible
2
was not pronounced in practically any of the tested catalysts. The Ru-Cu/ZrO
2
3
3
catalyst
(≈92%).
catalyst
presented the highest selectivity towards 1,2-PDO (100%) followed by Ru-Cu/Al
2
O
O
Under 503 K and 8 MPa, a previous study showed that a 3%Ru-1%Cu/Al
2
converted 100% of glycerol with a selectivity of 85% for 1,2-PDO [29]. Table 4 shows a
comparison between the tested catalysts in this work and those described in the literature. By
comparing the results, although better conversion values have been reported, the selectivity
obtained by this work, 91.9 %, had not been yet reported for that reaction temperature
(
473 K).
As it can be seen from the activity results in Table 3, the addition of Cu to Ru is
responsible for an increment in the selectivity towards the production of 1,2-PDO, mainly by
suppressing C-C cleavage and avoiding the formation of EG and C1 compounds. This effect
does not seem to stem from the ability to chemisorb hydrogen, since our results show no clear
1
5

general trend between Cu increment and H uptake. For Ru/Al
affect the H uptake as considerably as in the case of ZrO -supported catalysts, which is low
for monometallic Ru/ZrO , but almost triples in the case of bimetallic Ru-Cu/ZrO . Since the
2
3
O , the addition of Cu does not
2
2
2
reduction of bimetallic particles is also overall better for the bimetallic catalysts than for
monometallic Ru, as seen from our TPR results, it can be said that Cu stabilizes reduced
metallic particles capable of interacting rather with terminal hydroxyl groups of glycerol than
with the C-C bond.
The effect of temperature on catalyst selectivity for Ru-Cu/Al
2
3
O was also
investigated. Although a more complex model for glycerol hydrogenolysis may be
suitable, available kinetic data for glycerol conversion agree with a first order reaction.
Table 5 shows the apparent specific velocity constants, kapp, the initial rate of
consumption of glycerol per volume, RV,0, the initial rate of consumption per catalyst
gram, Rm,0, the initial turnover frequency, TOF
The squared sample correlation coefficients, r², for all temperatures were between
.992 and 0.996. Although considerable difficulties are encountered in calculating the
actual TOF [63], in this work the TOF was normalized by H chemisorption. Upon
0
, for each of the given temperatures.
0
2
comparing the specific velocities, and using least squares methodology, the apparent
barrier energy can be shown to be 63.5 kJ/mol, which agrees with the value reported
by Lahr and Shanks (62 kJ/mol) for glycerol hydrogenolysis over ruthenium [64]. It
should be noted that the turnover frequencies for Ru/ZrO
2
catalysts are remarkably
adsorption sites
greater than for the other catalysts, due to the smaller number of H
2
that offer relatively the same apparent specific reaction velocity. The conversion
range and kinetic fitting used in this article is in agreement with a prior work done by
1
6

our group with Ru-Cu/TiO
2
catalysts, cited as ref. [37], regarding TOF calculations,
based on first order kinetics.
All catalysts tested in the hydrogenolysis reaction behaved as described in the
corresponding literature. In the present work, Ru-Cu/ZrO displayed the best 1,2-PDO yield,
2
with 100% selectivity at low conversion, as it can be seen in table 3, with slightly less
conversion compared to monometallic ruthenium, corroborating Liu et al. [33]. In the present
work, monometallic Cu/Al
2
3
O catalyst did not show activity for the glycerol hydrogenolysis
reaction, as seen in reference [31]. The discrepancy from the conversion obtained in this work
with other referenced values [29,33] is attributed to the larger glycerol to catalyst mass ratio.
The mass ratio of glycerol/catalyst used in the present work (16.7 gGlycerol/gcat) is much
larger than the ones reported by Jiang et al. (about 5.4 gGlycerol/gcat) [29], and by Liu et al.
(
about 6.0 gGlycerol/gcat) [33]. According to ref. [29] an optimal atomic ratio of Ru/Cu is 3:1,
whereas for ref. [33] optimal ratio was found at 10:1. Since the conditions used in the present
work differ greatly from the conditions used in refs. [29] and [33], regarding glycerol/catalyst
mass ratios, it does not seem adequate to offer accurate comparisons. It is, however, safe to
say that Ru-Cu exhibits interesting catalytic behavior in a range from 10:1 to 10:16 of Ru/Cu
atomic ratios for both ZrO
2
and Al
2
3
O as supports, from which the best activities are found in
the lower Cu content region, since hydrogen chemisorption is not favored by Cu in the active
sites. Besides this, somewhat contradictory to refs. [29] and [33], the greater Ru/Cu fraction
used in this work does not seem to have any negative effect towards 1,2-PDO production,
since 100% selectivity was obtained for Ru-Cu/ZrO , as seen in table 3, which could point to
2
other influencing factors in the reaction, such as solvent action.
The low specific rate per gram of catalyst tested for the reaction can be related to the
low dispersion of the ruthenium particles, as showed in Table 1. The incipient wetness
1
7

impregnation method used for the preparation of the catalysts did not produce a high particle
dispersion. Balaraju et al. [13] evaluated the influence of preparation methods and the metal
content in TiO supported ruthenium catalysts, and observed that when the catalyst was
2
prepared by the incipient wetness impregnation the glycerol conversion (31 %) was lower
than when the method was precipitation (44 %). These authors implied that the low
conversion presented by the catalysts prepared by the conventional method is related to the
low metal particle dispersion over the supporting material.
Recycling experiments were performed for the monometallic catalysts, the results of
which can be seen in Table 6. Both Ru/Al
order of ~ 5% in activity, in the first recycle. The Ru/ZrO
drop in activity after the second recycle (about 58% from the first recycle). Comparatively,
Al appears to offer a more stable support for the catalyst, exhibiting a 20 % drop from the
first recycle.
2
O
3
and Ru/ZrO
2
catalysts showed a drop on the
2
catalyst showed the more dramatic
2
O
3
By interpretation of the obtained characterization and glycerol hydrogenolysis activity
data for the materials, it is possible to propose a model for the bimetallic catalysts’ surfaces,
as follows. The first step is to consider that Ru and Cu are highly immiscible, and given its
atomic size and mass, Cu atoms are more likely to form particles on top of the Ru structure.
The second step is to consider that both metals present different space group structures. While
reduced Ru forms hexagonal close packed (hcp) crystals (P6 /mmc), Cu forms face-centered
3
cubic (fcc) crystals (Fm-3m). For fcc crystals, the basal plane (110) is the one with least
atomic density, while the (111) plane presents the closest packing, from which hexagonal
surfaces should be expected as the most stable cleavage configuration. Therefore, it is
reasonable to suspect that the Cu atoms would organize themselves in hexagonal layers over
already assembled bulk hcp Ru nano-particles. The third step is to consider the actual
1
8

diffractogram of Ru-Cu/Al
2
3
O , in which Cu-related peaks are absent, but Ru-related peaks are
prominent. The influence of Cu coverage over Ru in hydrogenolysis activity is also a function
of the reduction activation temperature prior to the reaction [65]. All reactions of this work
were undertaken using a reduction temperature of 573 K, around the expected temperature
range indicative of Cu mobility for multilayer deposition onto Ru, but far bellow temperatures
of intensive sintering (>780 K). With this in mind, since Ru-Cu homogeneous alloy formation
is unlikely [63,65], given the immiscibility of both metals, multi-layer deposition of Cu
occurs, but not to the extent as to cover entirely the Ru surface at the reduction temperature of
5
73 K, or to separate entirely Cu clusters from the Ru particles.
This is evidence for considering Ru as the bulk particle phase, with thin layers of Cu
on top of it. The bimetallic structure would thus be prone to anisotropy, with parts of the Ru
crystal available for coordination or chemisorption-like phenomena, while the Cu sites would
+
also display enhanced activity, from Cu sites. The monometallic Ru catalyst particles have an
average diameter of 6 nm, and the bimetallic Ru-Cu catalyst particles have an average
diameter of about 11 nm, from which the increased length would be due to the outer Cu
cluster layers.
Finally, the effect these geometric factors have on the electron band structure for the
bimetallic particle should also be considered. Both Ru and Cu are late transition metals, which
is enough to envisage that electronic interactions between metallic particles and molecules
will happen through sp and d bands and molecular orbitals of the adsorbate. While Cu has a
completely filled, narrow d band, very near the Fermi level, Ru atoms have open shell 4d
electrons, and therefore a broader d band in respect to the density of states. Thus, formation of
small Cu clusters on the surface of Ru structures would create an asymmetric d band
distribution along the bimetallic particle. As seen from the obtained activity data, Cu has the
ability to direct selectivity towards the formation of 1,2-PDO in glycerol hydrogenolysis,
1
9

avoiding further C-C bond breaking. This points to the fact that the band distribution is such
that it enables back donation to C-O molecular orbitals upon adsorption, after which the C-C
orbitals can not be further filled by the modified Ru d band, but rather a hydrogenation of the
adsorbed C atom takes place.
4
. Conclusion
The use of ruthenium-copper bimetallic catalysts in glycerol hydrogenolysis presents a
viable alternative for 1,2-PDO production. There is considerable agreement between the
literature and the results obtained in the present work, which shows the highest selectivity for
1
,2-PDO by usage of Ru-Cu/ZrO
greater than for the Al bimetallic catalyst. An optimum proportion of Ru/Cu is situated in
the region of low Cu content for best relation conversion/selectivity towards 1,2-PDO, and it
appears that a convenient atomic ratio is 10:1. However selective Ru-Cu/ZrO is, it should be
noted that Al is a much more stable support for glycerol hydrogenolysis.
2
catalyst. The turnover frequency for the Ru-Cu/ZrO
2
is
2
O
3
2
2
O
3
5
. Acknowledgements
J. B. Salazar, A.V.H. Soares and F. B. Passos acknowledge the financial support from CNPq,
CAPES and FAPERJ. D.D. Falcone and R.J. Davis acknowledge support by the US National
Science Foundation under award number CBET-1157829.
6
. References
[1] L. Prati, P. Spontoni, A. Gaiassi, From Renewable to Fine Chemicals Through Selective
Oxidation: The Case of Glycerol, Top Catal. 52 (2009) 288–296.
2
0

[
2] C.A.G. Quispe, C.J.R. Coronaro, J.A. Carvalho Jr., Glycerol: Production, consumption,
prices, characterization and new trends in combustion, Renew. Sust. Energ. Rev. 27 (2013)
75–493.
3] A. Williams, R.L McCormick, R.R. Hayes, J. Ireland, H.L. Fang, Effect of Biodiesel
4
[
Blends on Diesel Particulate Filter Performance, Powertrain and Fluid Systems Conference
and Exhibition, Society of Automotive Engineers, 2006. Available at:
http://savebiodiesel.com/40015.pdf, accessed on 29 June 2015.
[
4] European Biodiesel Board, available at http://www.ebb-eu.org/stats.php, accessed on 29
June 2015.
5] BRAZIL, Federal Law No. 11097, from January 13 , 2005. Available at
th
[
http://www.planalto.gov.br/ccivil_03/_ato2004-2006/2005/Lei/L11097.htm, accessed on 29
June 2015.
[
6] BRAZIL, Agência Nacional de Petróleo, Gás Natural e Biocombustíveis (ANP). Boletim
Mensal do Biodiesel – February 2014. Available at www.anp.gov.br, accessed on 29 June
015.
7] G. Knothe, J. Van Gerpen, J. Krahl, L.P. Ramos, The Biodiesel Handbook, Edgar Blücher,
São Paulo, 2006.
8] C.J.A. Mota, C.X.A. Silva, V.L.C. Gonçalves, Glycerochemistry: New Products from
Glycerin of Biodiesel Production, Quím. Nova 32 (2009) 639-648.
2
[
[
[
9] K. Lehnert, P. Claus, Influence of Pt particle size and support type on the aqueous-phase
reforming of glycerol, Catal. Commun. 9 (2008) 2543–2546.
10] A. Brandner, K. Lehnert, A. Bienholz, M. Lucas, P. Claus, Production of Biomass-
Derived Chemicals and Energy: Chemocatalytic Conversions of Glycerol, Top. Catal. 52
2009) 278–287.
[
(
2
1

[
11] E. Farnetti, J. Kaspar, C. Crotti, A novel glycerol valorization route: chemoselective
dehydrogenation catalyzed by iridium derivatives, Green Chem. 11(2009)704–709.
12] J.N. Beltramini, C. Zhou, Catalytic Conversion of Glycerol to Valuable Commodity
[
Chemicals, in: Mark Crocker (Ed.) Thermochemical conversion of biomass to liquid fuels and
chemicals, RSC Publishing, Cambridge, 2010, pp. 435-467.
[
13] M. Balaraju, V. Rekha, B.L.A. Prabhavathi Devi, R.B.N. Prasad, P.S. Sai Prasad, N.
Lingaiah, Surface and structural properties of titania-supported Ru catalysts for
hydrogenolysis of glycerol, Appl. Catal. A - Gen. 384 (2010) 107–114.
[14] A.V.-H. Soares, G. Perez, F.B. Passos, Alumina supported bimetallic Pt–Fe catalysts
applied to glycerol hydrogenolysis and aqueous phase reforming, Applied Catalysis B:
Environmental, 185 (2016) 77-87.
[
15] A. Ciftci, S. Eren, D.A.J.M. Ligthart, E.J.M. Hensen, Platinum-Rhenium Synergy on
Reducible Oxide Supports in Aqueous-Phase Glycerol Reforming, Chemcatchem, 6 (2014)
260-1269.
16] F. Mauriello, A. Vinci, C. Espro, B. Gumina, M.G. Musolino, R. Pietropaolo,
1
[
Hydrogenolysis vs. aqueous phase reforming (APR) of glycerol promoted by a heterogeneous
Pd/Fe catalyst, Catalysis Science & Technology, (2015).
[17] J. Ge, Z. Zeng, F. Liao, W. Zheng, X. Hong, S.C.E. Tsang, Palladium on iron oxide
nanoparticles: the morphological effect of the support in glycerol hydrogenolysis, Green
Chemistry, 15 (2013) 2064-2069.
[
18] Y. Nakagawa Y, K. Tomishige, Heterogeneous catalysis of the glycerol hydrogenolysis,
Catal. Sci. Technol. 1 (2011) 179–190.
19] C. Montassier, J.C. Ménézo, L.C. Hoang, C. Renaud, J. Barbier, Aqueous Polyol
Conversions on Ruthenium and on Sulfur-modified Ruthenium, J. Mol. Catal. 70 (1991) 99-
10.
[
1
2
2

[20] J. Feng, H. Fu, J. Wang, R. Li, H. Chen, X. Li, Hydrogenolysis of glycerol to glycols
over ruthenium catalysts: Effect of support and catalyst reduction temperature, Catal.
Commun. 9 (2008) 1458–1464.
[
21] S.H. Lee, D.J. Moon, Studies on the conversion of glycerol to 1,2-propanediol over Ru-
based catalyst under mild conditions, Catal. Today 174 (2011) 10–16.
22] S. Wang, H. Liu, Selective hydrogenolysis of glycerol to propylene glycol on Cu-ZnO
catalysts, Catal. Lett. 117 (2007) 62-67.
23] M. Balaraju, V. Rekha, P.S. Sai Prasad, R.B.N. Prasad, N. Lingaiah, Selective
Hydrogenolysis of Glycerol to 1, 2 Propanediol Over Cu-ZnO Catalysts, Catal. Lett. 126
[
[
(2008)119–124.
[24] A. Bienholtz, F. Schwab, P. Claus, Hydrogenolysis of glycerol over a highly active
CuO/ZnO catalyst prepared by an oxalate gel method: influence of solvent and reaction
temperature on catalyst deactivation, Green Chem. 12 (2010) 290–295.
[25] A. Bienholtz, H. Hoffmann, P. Claus, Selective hydrogenolysis of glycerol over copper
catalysts both in liquid and vapour phase: Correlation between the copper surface area and the
catalyst's activity, Appl. Catal. A – Gen. 391 (2011) 153–157.
[
26] E.P. Maris, R.J. Davis, Hydrogenolysis of glycerol over carbon-supported Ru and Pt
catalysts, J. Catal. 249 (2007) 328–337.
27] E.P. Maris, W.C. Ketchie, M. Murayama, R.J. Davis, Glycerol hydrogenolysis on
carbon-supported PtRu and AuRu bimetallic catalysts, J. Catal. 251 (2007) 281–294.
28] D. Roy, B. Subramaniam, R.V. Chaudhari, Aqueous phase hydrogenolysis of glycerol to
,2-propanediol without external hydrogen addition, Catal. Today 156 (2010) 31–37.
29] T. Jiang, Y. Zhou, S. Liang, H. Liu, B. Han, Hydrogenolysis of glycerol catalyzed by Ru-
Cu bimetallic catalysts supported on clay with the aid of ionic liquids, Green Chem. 11 (2009)
000–1006.
[
[
1
[
1
2
3

[
30] A.J. Rouco, G.L. Haller, J.A. Oliver, C. Kemball, A comparative Investigation of Silica-
supported Ru-Cu and Ru-Ag catalysts, J. Catal. 84 (1983) 297-307.
31] L. Guo, J. Zhou, J. Mao, X. Guo, S. Zhang, Supported Cu catalysts for the selective
hydrogenolysis of glycerol to propanediols, Appl. Catal. A – Gen. 367 (2009) 93–98.
[
[32] E.S. Vasiliadou, E. Heracleous, I.A. Vasalos, A.A. Lemonidou, Ru-based catalysts for
glycerol hydrogenolysis—Effect of support and metal precursor, Appl. Catal. B – Environ. 92
(2009) 90–99.
[33] H. Liu, S. Liang, T. Jiang, B. Han, Y. Zhou, Hydrogenolysis of Glycerol to 1,2-
Propanediol over Ru-Cu Bimetals Supported on Different Supports, Clean–Soil Air Water 40
(2012) 318–324.
[34] M. Balaraju, K. Jagadeeswaraiah, P.S. Sai Prasad, N. Lingaiah, Catalytic hydrogenolysis
of biodiesel derived glycerol to 1,2-propanediol over Cu-MgO catalysts, Catal. Sci. Technol.
(2012) 1967–1976.
35] H. Shimizu, K. Christmann, G. Ertl, Model Studies on Bimetallic Cu-Ru Catalysts. 2.
Adsorption of Hydrogen, J. Catal. 61 (1980) 412-429.
2
[
[
[
36] J.H. Sinfelt, Structure of Bimetallic Clusters, Accounts Chem. Res. 10 (1977) 15-20.
37] J.B. Salazar, D.K. Falcone, H.N. Pham, A.K. Datye, F.B. Passos, R.J. Davis, Selective
production of 1,2-propanediol by hydrogenolysis of glycerol over bimetallic Ru–Cu
nanoparticles supported on TiO , Appl. Catal. A – Gen. 482 (2014) 137-144.
2
[38] L. Niu, R. Wei, F. Jiang, M. Zhou, C. Liu, G. Xiao, Selective hydrogenolysis of glycerol
to 1,2-propanediol on the modified ultrastable Y-type zeolite dispersed copper catalyst, React.
Kinet. Mech. Cat. 113 (2014) 543-556.
2
4

[39] Z. Huang, H. Liu, F. Cui, J. Zuo, J. Chen, C. Xia Effects of the precipitation agents and
rare earth additives on the structure and catalytic performance in glycerol hydrogenolysis of
Cu/SiO2 catalysts prepared by precipitation-gel method, Catal. Today, 234 (2014) 223–232.
[
40] Z. Xiao, X. Wang, J. Xiu, Y. Wang, C.T. Williams, C. Liang Synergetic effect between
0
+
Cu and Cu in the Cu-Cr catalysts for hydrogenolysis of glycerol, Catal. Today, 234 (2014)
2
00–207.
[41] E.S. Vasiliadou, V.L. Yfanti, A.A. Lemonidou, One-pot tandem processing of glycerol
stream to 1,2-propanediol with methanol reforming as hydrogen donor reaction, Appl. Catal.
B - Environ. 163 (2015) 258–266.
[42] J. Feng, B. Xu, D.R. Liu, W. Xiong, J.B. Wang, Hydrogenolysis of Glycerol on
Supported Ru-Co Bimetallic Catalysts, Adv. Mat. Res. 12 (2013) 791-793
[
43] V.P. Kumar, Y. Harikrishna, N. Nagaraju, K.V.R. Chary, Characterization and reactivity
of TiO2 supported nano ruthenium catalysts for vapour phase hydrogenolysis of glycerol,
Indian J. Chem. Sect. A -Inorg. Bio-inorg. Phys. Theor. Anal., 53 (2014) 516-523.
[
44] Y. Li, L. Ma, H. Liu, D. He, Influence of HZSM5 on the activity of Ru catalysts and
product selectivity during the hydrogenolysis of glycerol, Appl. Catal. A - Gen. 469 (2014)
5-51.
4
[
45] S. Jin, Z. Xiao, C. Li, C.T. Williams, C. Liang, Hydrogenolysis of glycerol over HY
zeolite supported Ru catalysts, J. Energy Chem. 23 (2014) 185-192.
46] I. Chorkendorf, J.W. Niemantsverdriet Concepts of Modern Catalysis and Kinetics.
Second, Revised and Enlarged Edition, Wiley-VHC Verlag GmbH & Co. KGaA, Weinheim,
011.
47] P.G.J. Koopmann, A.P.G. Kieboom, H. van Bekkum, Characterization of ruthenium
catalysts as studied by temperature programmed reduction, J. Catal. 69 (1981) 172-179.
[
2
[
2
5

[
48] B.D. McNicol, R.T. Short, Short Communication: The Reducibility of Ru, Pt-Ru and Pt
Oxide Electrocatalysts as Measured by Temperature-Programmed Reduction and Cyclic
Voltammetry, J. Electroanal. Chem. 92 (1978) 115.
[
49] N.W. Hurst, S.J. Gentry, A. Jones, B.D. McNicol, Temperature Programmed Reduction,
Catal. Rev. 24 (1982) 233-309.
50] V. Mazzieri, F. Coloma-Pascual, A. Arcoya, P.C. L’Argentière, N.S. Fígoli, XPS, FTIR
and TPR characterization of Ru/Al catalysts, Appl. Surf. Sci. 210 (2003) 222–230.
51] C.L. Bianchi, TPR and XPS investigations of Co/Al2O3 catalysts promoted with Ru, Ir
and Pt, Catal. Lett. 76 (2001)155-159.
52] S. Galvagno, C. Crisafulli, R. Maggiore, A. Giannetto, J. Sehwank, TPR Investigation of
Bimetallic Ru-Cu samples supported on SiO , Al and MgO, J. Therm. Anal. 32 (1987)
71-483.
53] A. Christensen, A.V. Ruban, P. Stoltze, P.W. Jacobsen, H.L. Skriver, J.K. Nørskov,
Phase diagrams for surface alloys, Phys. Rev. B 56 (1997) 5822-5834.
54] D. Gottschalk, E.A. Hinson, A.S. Baird, H.L. Kitts, K.A. Layman, CO Adsorption on
Hydrated Ru/Al : Influence of Pretreatment, J. Phys. Chem. C 114 (2010) 4950-4960.
55] M.F. Brown, R.D. Gonzalez, An Infrared Study of the Adsorption of Carbon Monoxide
on the Reduced and Oxidized Forms on Silica- Supported Ruthenium, J. Phys. Chem. 80
[
2
O
3
[
[
2
2
O
3
4
[
[
2
O
3
[
(
[
1976) 1731-1735.
56] M.L. Unland, Isocyanate intermediates in the reaction of NO and CO over noble metal
catalysts, J. Catal. 31 (1973) 459-465.
57] M.H. Kim, J.R. Ebner, R.M. Friedman, M.A. Vannice, Determination of Metal
Dispersion and Surface Composition in Supported Cu–Pt Catalysts, J. Catal. 208 (2002) 381–
92.
[
3
2
6

[
58] K. Hadjiivanov, H. Knözinger, FTIR Study of Low-Temperature CO Adsorption on Cu-
2
+
ZSM-5: Evidence of the Formation of Cu (CO)
59] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Development of a Ru/C catalyst for
glycerol hydrogenolysis in combination with an ion-exchange resin, Appl. Catal. A – Gen.
18 (2007) 244–251.
60] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Glycerol hydrogenolysis to 1,2-
2
Species, J. Catal. 191 (2000) 480–485.
[
3
[
propanediol catalyzed by a heat-resistant ion-exchange resin combined with Ru/C, Appl.
Catal. A – Gen. 329 (2007) 30-35.
[
61] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Glycerol conversion in the aqueous
solution under hydrogen over Ru/C + an ion-exchange resin and its reaction mechanism, J.
Catal. 240 (2006) 213–221.
[
62] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Catalytic
performance of Rh/SiO2 in glycerol reaction under hydrogen, Green Chem. 9 (2007) 582–
88.
5
[
63] M. Boudart, Turnover rates in heterogeneous catalysis, Chem. Rev. 95 (1995) 661-666.
64] D.G. Lahr, B.H. Shanks, Effect of sulfur and temperature on ruthenium-catalyzed
[
glycerol hydrogenolysis to glycols, J. Catal. 232 (2005) 386-394.
65] J.H. Sinfelt, Y.H. Lam, J.A. Cusumano, A.E. Barnett, Nature of ruthenium-copper
catalysts, J. Catal. 42(1976)227–237.
66] E. van Ryneveld, A.S. Mahomed, P.S. van Heerden, H.B. Friedrich, Direct
[
[
Hydrogenolysis of Highly Concentrated Glycerol Solutions Over Supported Ru, Pd and Pt
Catalyst Systems, Catal. Lett. 141 (2011) 958–967.
2
7

Figure 1: Possible products of glycerol hydrogenolysis, dehydration and dehydrogenation.
Figure 2: Proposed mechanisms for 1,2-PDO production through glycerol hydrogenolysis.
2
8

Ru
(
a)
RuOx
Ru-Cu/Al O
2
3
Ru/Al O
2
3
Al O
2
3
20
40
60
80
2

Ru-Cu/ZrO2
Ru/ZrO2
(
b)
ZrO2
20
30
40
50
60
70
80
90
2

Figure 3: X-ray diffractograms for (a) Al
2
O
3
, Ru/Al
2
O
3
, Ru-Cu/Al
2
O
3
; (b) ZrO
2
, Ru/ZrO .
2
2
9