On the gas-phase reforming of glycerol by Pt on carbon black: Effects of metal particle size and pH value of the glycerol stream

Article abstract of DOI:10.1016/j.molcata.2015.12.026

We investigate the effect of the platinum particle size and the pH of the feed solution in the glycerol gas-phase reforming reaction at low-temperature (623?K) over carbon black-supported Pt catalyst with different platinum loadings (1, 3, 5, and 10?wt.%). We also verify the influence of the aqueous glycerol (30?wt.%) feeding solution pH in the reaction at low-temperature (WHSV?=?33?min^?1 and P?=?1?bar) over the 5?wt.% Pt/C catalyst. Glycerol gas-phase reforming demonstrates being a structure-sensitive reaction and it has its activity and selectivity changed with the size of the platinum particle size. The size increasing leads to an increasing of the CO and Glycerol TOF and selectivity towards gaseous products, formed by consecutive C[sbnd]C cleavages. Despite no remarkable change is observed in the selectivity under the reaction condition with different pHs of the glycerol aqueous solution, there is an activity decrease.

Full text of DOI:10.1016/j.molcata.2015.12.026

Accepted Manuscript
Title: On the gas-phase reforming of glycerol by Pt on carbon
black: Effects of metal particle size and pH value of the
glycerol stream
Author: Ricardo R. Soares Douglas F. Martins Douglas E.S.
Pereira Marlon B. Almeida Yiu L. Lam
PII:
S1381-1169(15)30189-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2015.12.026
MOLCAA 9729
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
21-7-2015
30-12-2015
31-12-2015
Please cite this article as: Ricardo R.Soares, Douglas F.Martins, Douglas E.S.Pereira,
Marlon B.Almeida, Yiu L.Lam, On the gas-phase reforming of glycerol by Pt on carbon
black: Effects of metal particle size and pH value of the glycerol stream, Journal of
Molecular Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2015.12.026
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.

Title: On the gas-phase reforming of glycerol by Pt on carbon black: Effects of
metal particle size and pH value of the glycerol stream
Corresponding Author: Prof. Ricardo Reis Soares
First Author: Ricardo Reis Soares
a
b
Order of Authors: Ricardo R. Soares , Douglas F. Martins , Douglas E. S.
a
c
c
Pereira , Marlon B. Almeida , Yiu L. Lam .
Affiliations:
a: School of Chemical Engineering, Federal University of Uberlândia, Av. João
Naves de Ávila, 2121; Bloco 1K, Uberlândia/MG, Brasil, 38408-100;
b: Institute of Chemistry, Federal University of Uberlândia, Av. João Naves de
Ávila, 2121; Bloco 1Q, Uberlândia/MG, Brasil, 38408-100
c: CENPES/PETROBRAS, Av Horácio Macedo, 950 - Cidade Universitária -
Ilha do Fundão; 21941-915 Rio de Janeiro RJ Brasil
Emails:
Ricardo Reis Soares: rrsoares@ufu.br;D.Sc.
Douglas Ferreira Martins: douglas@siamig.com.br; M.Sc.
Douglas Eduardo S. Pereira: gcatufu@yahoo.com.br; M.Sc.
Marlon Brando Almeida: malmeida@petrobras.com.br; M.Sc.
Yiu Lau Lam: lam@petrobras.com.br; D.Sc.

Highlights

Low-temperature glycerol gas-phase reforming demonstrates being a structure-sensitive
reaction over Pt/C catalyst

Activity and selectivity change with the size of the platinum particle size

The pH of the glycerol aqueous solution has not any remarkable effect under the reaction
condition
Abstract
We investigate the effect of the platinum particle size and the pH of the feed
solution in the glycerol gas-phase reforming reaction at low-temperature (623 K)
over carbon black-supported Pt catalyst with different platinum loadings (1, 3, 5,
and 10 wt. %). We also verify the influence of the aqueous glycerol (30 wt. %)
-
1
feeding solution pH in the reaction at low-temperature (WHSV = 33 min and P
1 bar) over the 5 wt. % Pt/C catalyst. Glycerol gas-phase reforming
=
demonstrates being a structure-sensitive reaction and it has its activity and
selectivity changed with the size of the platinum particle size. The size increasing
leads to an increasing of the CO and Glycerol TOF and selectivity towards
gaseous products, formed by consecutive C-C cleavages. Despite no remarkable
change is observed in the selectivity under the reaction condition with different
pHs of the glycerol aqueous solution, there is an activity decrease.
Keywords: glycerol gas-phase reforming; Pt/C-black catalysts; metal particle
size effect; glycerol aqueous solution pH

1
. Introduction
Biodiesel production can be one of the main efforts to obtain “green fuels”.
Its production has grown from 19.6 thousand barrels per day in 2001 to 294.7
thousand barrels in 2010 [1]. Biodiesel is mainly obtained by the trans-
esterification of triglycerides, from vegetable oils and fat with methanol. As a
result, crude glycerol (10 wt. % of total product) is produced as a co-product.
Since crude glycerol is a residue from the biodiesel production, its valorization
represents a way to enhance the biodiesel economy. Moreover, the amount of
glycerol used in commercial applications, sometimes becomes much less than
its production, thereby creating an oversupply crisis [2].
Glycerol of high purity is an important industrial feedstock for applications
in food, cosmetics and pharmaceutical, and others industries; however, it is costly
to refine crude glycerol, especially for medium or small sized plants. Therefore,
many researches focused on innovative processes to convert pure glycerol, as
well as the crude one, into valuable products [3,4].
Glycerol is considered as one of the top twelve building block chemicals
that can be also derived from carbohydrates and converted into to numerous
valuable commodity chemicals (e.g. glyceric acids, propylene glycol, 1,3- and
1
,2-propanediol, branched polyesters and polyols) [5]. These chemicals can be
formed by different routes, including: selective oxidation, hydrogenolysis,
dehydration, transesterification, etherification, oligomerization and
polymerization, as reviewed very well by Zhou et al. [6]. Among possible uses for
glycerol, hydrogen and/or synthesis gas production from glycerol have been
attracting much attention due to their importance in the petrochemical and energy
industries and because they are currently produced from fossil non-renewable

sources. Hydrogen is widely used to power fuel cell systems and syngas is a vital
building block to prepare various chemicals [7]. Moreover, its productions from
biomass process waste or residue has become key to make biorefinery feasible.
Different pathways to produce H
phase reforming [8,9], gasification or thermal reduction [10], steam reforming [11-
3], auto-thermal or partial oxidation reforming [14,15], pyrolysis [16] and
2
or syngas from glycerol: (e.g. aqueous-
1
photocatalysis reforming [17]) have been reported. Recently, a comprehensive
review was published [7], which complemented other important ones [6,8-20].
In this context, Soares et al. [21] have reported a pioneering low-
temperature (e.g., 550 K) gas-phase catalytic process that converts aqueous
glycerol into H /CO gas mixtures over a carbon-supported Pt catalyst.
2
Furthermore, they demonstrated that this glycerol conversion process can be
combined with Fischer-Tropsch synthesis or water-gas-shift (WGS) reaction in a
single reactor process. In this way, the heat generated by the exothermic Fischer-
Tropsch or WGS reaction can supplied to the endothermic glycerol reforming
reaction [21,22]. These combined gas-phase processes produce liquid alkanes
suitable for transportation applications [22] or hydrogen production with low CO
content directly from aqueous glycerol [23,24].
In spite of this advance, only a few studies [25-29] on low-temperature (<
6
73 K) gas-phase glycerol reforming were published after the work by Soares et
al. [21]. Simonetti et al. [26] and Kunkes et al. [27] have shown that Re addition
to 5%Pt/C has a great promotional effect on the rate of glycerol conversion.
Pompeo et al. [28] have shown the influence of the nature of the support on the
catalytic performance of Pt or Ni-based catalysts, and Sutar et al. [29] has
proposed a kinetic model for a Pt/C catalyst.

Since, biodiesel can be synthesized by homogeneous basic or acid
catalysis [2-4,30]; the aqueous glycerol solution derived may be basic or acidic.
Studies on the influence of this solution pH for glycerol or sorbitol liquid-phase
reforming or hydrogenolysis showed that basic aqueous solution enhanced the
hydrogen yield. However, this pH effect on gas-phase glycerol reforming is
hitherto unknown. Therefore, in this work, we investigate in more detail the effect
of the platinum particle size and the pH of the feed solution in the glycerol steam
reforming reaction over carbon-supported Pt catalyst, in various catalytic reaction
conditions with different platinum loadings.
2
. Experimental
2
.1. Catalyst preparation and characterization
Commercial catalysts containing 5 and 10 wt.% Pt supported on carbon
black (Vulcan X-72C) were purchased from BASF FUEL CELL INC. Catalysts
with low platinum loadings, 1 (1.07) and 3 (2.97) wt.%, were prepared by incipient
wetness impregnation of H
2
PtCl
6
.6H O (SIGMA-ALDRICH) aqueous solution on
2
Carbon black (CABOT; Vulcan XC72, Lot-1105528). Then, all the catalysts were
dried at 393 K for 12 h in air. The real Pt contents (in parentheses) were
determined by X-Ray Fluorescence spectrometer (BRUKER – S8 Tiger).
Prior to any CO chemisorption measurements and reaction kinetics
studies, the catalysts were reduced in-situ by heating up in flowing H
2
[35 cm3
-1
-1
(
NTP) min ] with a ramp rate of 0.5 K.min , and then maintained at 623 K for 2
h. In a special case for 5 wt. % Pt the reduction temperature was 473 K giving
the sample with tag 5 wt. %*. Another sample was then generated from this
sample as follows. After remaining at 2h at 473 K under hydrogen flow, the

-
1
temperature was increased at 10 K.min to 623 K and then kept for 2h under He
atmosphere.
CO chemisorption measurements were carried out on all catalysts using a
standard gas adsorption apparatus (MICROMERITICS, model ASAP 2020 C).
The irreversible CO uptake of the catalysts at 300 K was used to determine the
platinum dispersion and crystallite size.
Some samples of spent catalysts (1 and 5 wt. %) were also characterized
by CO chemisorption. Before the analysis, the catalysts were purged with helium
3
-1
(
50 cm .min ) at reaction temperature for 30 min. Next, the material was cooled
down to 273 K and passivated by an air flowing for 30 min.
2
.2. Glycerol Gas-Phase Reforming Tests
Isothermal reaction kinetics studies were conducted on an apparatus
described elsewhere [21-23,26,27]. The reactor consisted of a 26 mm outer
diameter stainless steel tube with a wall thickness of 0.71 mm. A bed consisting
of fresh powder catalyst (20–100 mg) mixed with an equal volume of crushed
fused SiO
2
granules (to decrease pressure drop across the bed) was loaded
granules (−4 + 16 mesh; Sigma
between a quartz wool plug and fused SiO
2
Aldrich). The reactor was heated with a homemade furnace. A J-type
thermocouple was attached to the outside of the reactor to measure temperature,
which was controlled with a temperature controller (THERMA Instruments - TH
2
131P series type). Aqueous glycerol solution (30 wt. % in glycerol, with 99.5%
purity, diluted in deionized water of pH≅ 6.3; VETEC) was introduced into the
reactor in a down-flow configuration. In some experiments, the pH value of the
feeding solution was adjusted to 3 or 10 by adding droplets of an HNO (0.5 M)
3

or NH
4
OH (0.5 M) solutions respectively.
The liquid feed flowing rate was controlled by a pump (ELDEX; Model
440-A-30-S-2 CE), which is connected to a syringe needle (Hamilton; point 5
1
tip) to introduce droplets of the feed into the reactor where vaporization occurs.
Part of the effluent was condensed in a gas–liquid separator at 279 K, and
drained periodically for gas chromatography analysis (Shimadzu equipped with a
flame-ionization detector and a RTx-5 capillary column). Each drain portion was
analyzed for glycerol and trace amounts of other condensable byproducts. The
effluent gas stream was separated by packed column (HAYESEP DB), and
analyzed for H
2
2
, CO, CO , and light hydrocarbons and oxygenates (C
1
–C ) using
4
another Shimadzu gas chromatograph (17AF) equipped with a thermal
conductivity detector followed by with a flame-ionization detector.
The glycerol conversion (X
light hydrocarbons and oxygenates (C
molecules), (X LP (%), were calculated based on the following equation (1) and
2), respectively:
G
)
GP (%) to gaseous products, H
2 2
, CO, CO , and
1
–C ) and to liquid products (condensate
4
G
)
(
F
 F
 F
CH
 2F  3F  4F
CO CO
C
C
C
2
4
2
3
4
(
X ) (%) 
100 (1)
G GP
3.(F
)
glycerol 0
n . F
C (condensate molecule)
(
X ) (%) 
100 (2)
G LP
3.(F
)
glycerol 0
Then, the overall glycerol conversion was calculated by equation (3):
(
X ) (%) (X ) (%)(X ) (%) (3)
G T1 G GP G LP
Alternatively, the overall glycerol conversion was also calculated by
equation (4):

(F
) -.(F
)
glycerol 0
glycerol
)
(
X ) (%) 
100 (4);
G T2
(F
glycerol 0
-
1
Where (F
j
) and (F
j
)
0
are the molar flow rate (mols.min ) of the molecule “j” at any
is the number of carbon atoms in the
molecule. The selectivities towards any molecule were calculated by:
time and at the inlet, respectively, and n
C
(
n ) . F
C j j
(
S) (%) 
100 (5)
j
3
.(F
)
glycerol 0
The carbon balance CB (%) was evaluated by:
(
X ) (%) - (X ) (%)
G T2 G T1
CB (%) 
100 (6);
(
X ) (%)
G T2
-
1
The space velocity, WHSV (min ) was calculated after dividing the feeding
-
1
glycerol mass flow (g.min ) rate by the platinum weight (g) in the catalyst bed.
The values of catalytic activity are expressed by the Turnover Frequencies
-
1
(
TOF/min ), which were calculated for glycerol (G) and carbon monoxide (CO)
by the same way calculated elsewhere for same systems [21-23,27]:
(
F ) - (F )
G 0
G
G - TOF 
(7);
(
mols CO uptaken)
(F
)
CO
mols CO uptaken)
CO - TOF 
(8);
(
This calculation is acceptable for any conversions values, since the
reaction is first-order with respect to glycerol, as demonstrated by Sutar et al.
[29].
3
. Results and Discussion

3
.1. Catalyst characterization: CO uptake, platinum crystallite size and
dispersion
Table 1 shows the irreversible CO uptakes as well as the metal dispersion
and particle size calculated from it, for monometallic Pt/C with different platinum
contents. The average of the platinum crystallite size for the low Pt content
catalysts (1 and 3 wt. % Pt/C) are almost the same, 2.2 nm. When the platinum
content increases to 5 and 10 wt. %, the Pt crystallite sizes increase to 3.56 and
6
.10, respectively. This result indicates a trend of platinum agglomeration with
the increase in metal content, especially above 3 wt. %. The ease in sintering for
the high Pt loading is more evident when comparing the Pt crystallite average
sizes obtained from different reduction procedures using the same Pt catalyst (5
wt.%). The average platinum crystallite size increased from 2.07 to 3.56 when
the sample reduced at 473 K was heated to 623 K in He, and then further reduced
for 2h at this temperature.
3
.2. Glycerol Gas-Phase Reforming Tests: effect of the platinum
crystallite size on activity
Figure 1 exhibits the glycerol conversion as a function of time-on-stream
for all catalysts under atmospheric pressure and at 623 K. The profiles show that
all catalyst presents a good stability, without any significant deactivation even
after 24 h time-on-stream. This observation confirms previous result [21-
23,26,27], which demonstrated the excellent catalytic stability of the 5 wt. % Pt/C
catalysts under low-temperature gas-phase glycerol reforming. The platinum
crystallite sizes of the catalysts (with 1 and 5 wt. % loading of Pt) measured after
reaction, were 2.41 and 3.85 nm, respectively. Hence, no remarkable difference

was detected between the particle sizes of the fresh and spent catalysts.
Table 2 shows the conversions, catalytic activity values, and H /CO ratios
attained after 30 hours of reaction. Good material balance was obtained as shown
by two methods of computing the overall glycerol conversions: ((X
T1). Their differences were less than 10% for all runs. The catalysts exhibited
high conversions to gaseous products ((X GP) and H /CO ratios close to 1.33,
the stoichiometric glycerol decomposition value (C → 3CO + 4H ) (6). This
2
G
)T2 and
(
G
X )
G
)
2
3
H
8
O
3
2
indicates that Pt/C-black catalysts are selective towards synthesis gas under the
test conditions, without significant contribution from water-gas shift (WGS) to the
product mixture. The CO selectivity was very low. This result was also similar to
2
the ones reported in the literature [21-23,26,27].
Two considerations demonstrate that our system is free from transport
limitations. First, theoretical analyses based on dimensionless groups [31]
indicate interphase mass transfer can become limiting for particle sizes higher
than 100 μm. The catalysts particle sizes of the catalysts used in this study were
less than 70 μm. Further, the TOF measured increased with the particle size.
Second, the CO-TOF and Glycerol-TOF values are quite similar for the catalysts
1
% Pt/C and 5 wt. % Pt/C (after reduction at 473 K, with similar dispersion as that
of 1% Pt/C). Thus, the Madon-Boudart criterion for the absence of transport
limitations is satisfied for these catalysts [32]. Therefore, our tests results were
obtained under kinetics regime.
Notably the CO-TOF and Glycerol-TOF values increased with the platinum
-
1
crystallite size (CO-TOF from 107 to 580 min and Glycerol-TOF from 42 to 226
-
1
min for catalysts with 1 wt. % to 10 wt. % Pt). More interestingly is the difference
between the TOF values obtained from the catalysts with the same Pt content,

but different platinum crystallite sizes prepared at different reduction
temperatures. A 4-fold increase in TOF was observed when Pt crystallite size
increases 79%. These results suggest that the glycerol reforming reaction is a
structure-sensitive reaction.
3
.3. Glycerol Gas-Phase Reforming Reaction: effect of the Pt
crystallite size on selectivity
Table 3 summarizes selectivity results under different conversions
(
conversions given in Table 2). The selectivity to CO was the highest among all
products. It increased from 74% for catalyst with higher Pt dispersion to 90% or
above for those with lower Pt dispersion. At the same time the selectivity towards
molecules with 2 and 3 carbon atoms size decreases. Moreover, the ratio of
conversion to gaseous products and to condensed ones [(X
G
)
GP/(X )LP] increased
G
as well (Table 2). It implies that more C-C bond cleavages occurred as compared
to C-O ones. However, the selectivities in Table 3 were recorded under different
glycerol overall conversions. Hence, we carried out the glycerol reforming using
the 5 wt. % Pt/C catalysts by using different space velocities to obtain different
conversions. With these results, we can then discuss and compare with the
selectivity values obtained with the low metal-loading catalysts. As shown in table
4
, the overall glycerol conversion attains a value around 20 and 34% when the
-
1
space velocity is 99 and 66 min , respectively. This latter value in the same range
obtained by the catalysts with 1 wt. % and 3 wt. % Pt under typical conditions. In
these conditions, the selectivity towards gaseous products is still higher for the
platinum catalyst with a larger crystallite size (5 wt. %), by comparing the
selectivity ratio (gaseous to condensed/liquid). The CO selectivity was above

8
0% (higher than that of 1 and 3% Pt) and the selectivity values for two or three
carbons were lower. These results indicate that the gas-phase reforming of
glycerol is also structure sensitive regarding its selectivity. Here, we should point
out that aqueous-phase glycerol hydrogenolysis has been shown to be structure
sensitive concerning to the selectivity [34-36] and activity [36].
3
.4. General consideration on structure sensitivity
Even though we observed remarkable changes in reaction rates and
important changes in selectivity in the reforming reaction comparing Pt catalyst
of not very different in Pt crystallite sizes, the reaction being structure sensitive is
rather reasonable. Simonetti et al [26] have indicated that this reaction was
structure sensitive using a catalyst system similar to this work, in a less severe
reaction condition (T = 548 K). Claus and Lehnert [34] also proposed that the
aqueous phase reforming (APR) reaction is structure sensitive as the selectivity
in hydrogen production increased with crystallite size of Pt even though the rate
of reaction was not affected. For the hydrogenolysis of glycerol on ruthenium
catalysts, Miyazawa et al. [35] and Wang et al. [36] reported that the reaction
might be structure sensitive.
As according to Claus and Lehnert [34], one may assume that the
adsorption of the oxygenated hydrocarbons for subsequent cleavage of a C-C
bond occurs preferably at face positions rather than at the edges and corners.
The former positions are more abundant for larger metal crystallites. Hence,
catalysts containing these type of crystallites are more active per surface atom
exposed.
In the steam reforming reaction of glycerol, in addition to the C-C bond

cleavage, C-O bonds cleavages also occurs. The competition of C-C adsorption
and C-O adsorption on adjacent metal sites and their subsequent cleavage may
result in a complex dependence on the crystallite sizes.
3
.5. Glycerol Gas-Phase Reforming Tests: the pH effect of the
glycerol aqueous feed solution
The catalytic performances of the commercial 5 wt. % Pt/C (BASF)
catalysts at using glycerol aqueous feed solutions with various pH (3, 6, and 10)
are shown in Figure 3 and Tables 6 and 7. The tests were performed under typical
conditions: under atmospheric pressure, at 623 K, with a space velocity (WHSV)
-
1
of 33 min . It is worthy to mention that no activity was observed when that blank
tests using the support as catalyst were made.
Figure 3 illustrates stable conversion profiles for all feed solutions with
different pH values in the first 25 h time-on-stream, which demonstrates that basic
and acidic species not poisoning the catalyst as well as no significant deactivation
occurs. The conversion to gas-phase products was the highest at pH=6, (X
G
)
GP
=
77.4, and was lower in basic and acidic medium. Again, both methods of
calculation of conversions agreed within 10%, as shown in Table 6. The activities
observed, TOF, follow the same trend. They decreased considerably, which
indicates that ionized feed does not favor the main reaction. Despite the gaseous
to liquid products ratio increase with the pH of the feed solution, the selectivities
to all identified molecules, shown in Table 7, did not change significantly.
Chaminand et al. [21] obtained a similar result after adding HCl in the aqueous
feed solution.
In summary, an activity decreased and no remarkable changes in
selectivity were observed in the glycerol steam reforming under the reaction

condition with different pH value of the glycerol aqueous solution. Nevertheless,
catalytic performances of the APR or liquid-phase hydrogenolysis reactions were
changed when the pH of the glycerol solution was varied. In basic medium,
selectivity to break C-C bonds leading to carbon monoxide was higher compared
in neutral medium, whereas in acidic media, the cleavage of the C-O, by a
reaction dehydration became competitive with the former. In acidic medium was
also observed a small inhibition of successive falls of the C-C, providing the
formation of acetol, ethylene glycol and methanol.
4
. Conclusions
Carbon black supported platinum catalysts with low Pt-content can also be
used to produce syngas at low temperature (623 K) with reasonable stability until
3
024 h at least. Moreover, a stable steady state is established even under high
-
1
space velocity (99 min ) under acidic and basic condition. The Pt/C system
provides synthesis gas with a H /CO ratio around 1.33.
2
As the activity and selectivity changed with the size of the platinum particle
size, the current results also indicate that the gas-phase reforming of glycerol is
a structure-sensitive reaction. The size increasing leads to an increasing of the
CO and Glycerol TOF and selectivity towards gaseous products, formed by
consecutive C-C cleavages. The Pt-C bonds are more stable than the Pt-O ones
in the Pt/C system. Then, adsorbed species are probably bonded to the catalyst
surface through Pt-C bonds. This enables further subsequent C-C cleavages,
avoiding C-O breakages, which would lead to hydrogenolysis products. The
activity decreases and no remarkable changes in selectivity are observed in the
glycerol gas-phase reforming under the reaction condition with different pHs of

the glycerol aqueous solution.
Acknowledgments
This work was supported by PETROBRAS (Contract: 0050.0031886.07.2)
and we thank Oscar Chamberlain for his confidence on this team. D.M. Ferreira
acknowledges a M.Sc. grant from CAPES. We also thank Dr. Ronald A. Miranda
Acevedo and Eduardo Piola for their valuable collaborations at the end of this
project.

References
1] REN 21’s Renewables 2014, Global Status Report; www.ren21.net.
Accessed 5 Jan 2015
2] A. Konaka, T. Tago, T. Yoshikawa, H. Shitara, Y. Nakasaka, T. Masuda; Ind.
Eng. Chem. Res. 12 (2013) 15509-15515;
3] H.W. Tan, A.R. Abdul Aziz, M.K. Aroua; Renewable and Sustainable Energy
Reviews 27 (2013) 118-127;
4]
http://www.prweb.com/releases/glycerin_natural/oleo_chemicals/prweb4714434
[
[
&
[
[
.
[
htm
5] T.A. Werpy, J.E. Holladay, J.F. White; Top Value added chemicals from
biomass. Technical report, US DOE, PNNL-14808, 2004.
6] C. Zhou, J. N Beltramini, Y. Fan, G. Q Lu (Max), Chem. Soc. Ver. 37 (2008)
27-549.
[
5
[
[
[
[
7] Yu-Chuan Lin, Int. J. of Hydrogen Energy, 38 (2013) 1678-2700.
8] R. D. Cortright, R. R. Davda, J. A. Dumesic, Nature 48 (2002) 964 – 967.
9] G.W. Huber, J.A. Dumesic, Catal. Today 111 (2006) 119-132.
10] S. Czernik, R. French, C. Feik, E. Chornet; Ind. Eng. Chem. Res. 41 (2002)
4
209–4215.
[
[
[
[
[
[
11] S. Adhikari, S, Fernando, A, Haryanto, , Catal Today 129 (2007) 355–364.
12] C.K. Cheng, S.Y. Foo, A. A. Adesina, Catal. Comm. 12 (2010) 292-298
13] N. Luo, X. Fu, F. Cao, T. Xiao, P.P. Edwards, 87 (2008) 3483-3489.
14] P. J Dauenhaurer, J. R Salge, L. D Schmidt, J. Catal 244 (2006) 238-247.
15] Y. Liu, R. Farrauto, A. Lawal, Chem. Eng. Sci. 89 (2013) 31-39.
16] T. Valliyappan, N.N. Bakhshi, A.K. Dalai, Technol. 99 (2008) 4476-4483.

[
17] P. Panagiotopoulou, E.E. Karamerou, D.I. Kondarides, Catalysis Today 209
2013) 91-98.
(
[
[
18] P.D. Vaidya and A.E. Rodrigues, Chem. Eng. Technol. 32 (2009) 1463-1469.
19] A.C.C. Souza and J.L. Silveira, Renew. Sust. Energy Rev. 15 (2011) 1835-
1
850.
20] S. Adhikari, S.D., Fernando, A, Haryanto, Energy Convers. Manage 50
2009) 2600-2604.
21] R. R. Soares, D. A Simonetti, J. A. Dumesic, Angew. Chem. Int. Ed. 45
2006) 3982-3985.
22] D. A. Simonetti, E. I Kunkes, J. Rass-Hansen, R. R Soares, J. A Dumesic,
Green Chem. 9 (2007) 1073–1083.
23] E. I Kunkes, R. R Soares, D. A. Simonetti, J. A Dumesic, Appl. Catal. B 90
2009) 693-698
[
(
[
(
[
[
(
[
[
24] R.D. Cortright et al.; USPTA 20070225383
25] E.B. Pereira, P. Ramirez de la Piscina, N. Homs, Bioresource Technology
1
02 (2011) 3419-3423.
26] D. A Simonetti, E. L Kunkes, J. A Dumesic, Journal of Catalysis 247 (2007)
98-306.
27] E. L Kunkes, D. A Simonetti, J. A Dumesic, W.D. Pyrz, L.E. Murillo, J.G.
Chen, D.J. Buttrey, Journal of Catalysis 260 (2008) 164-177
28] F. Pompeo, G. Santori, N.N. Nichio, Int. J. of Hydrogen Energy 35 (2010)
912-8920.
29] P.N. Sutar, P.D. Vaidya, A.E. Rodrigues, Chem. Eng. Technol. 33,10 (2010)
645-1649.
30] D.A.G. Aranda, R.T.P. Santos, N.C.O. Tapanes, A.L.D. Ramos, O. A.C.
[
2
[
[
8
[
1
[

Antunes, Catalysis Letters 122,1-2 (2008) 20-25
31] J.W Shabaker, R.R Davda, W. G Huber, J. A Dumesic, R.D. Cortright,
Journal of Catalysis 215 (2003) 344-352,
[
[
[
32] R.J. Madon, M. Boudart; Ind. Eng. Chem. Fund. 21 (1982) 438.
33] B.C. Gates, J.R Katzer, G.C.A. Schuit, Chemistry of catalytic processes.
MacGraw-Hill, 1979.
[
34] K. Lehnert, P. Claus, Catalysis Communications 9 (2008) 2543-2546
35] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Appl, Catal. A: General
[
3
18 (2007) 244
[36] J. Wang, S. Shen, B. Li, H. Lin, Y. Yuan, Chemistry Letters 38,6 (2009)

Figure Captions
Figure 1. Glycerol Steam (30 wt.%) Reforming (P = 1 bar, T = 623 K, and WHSV
-
1
=
33 min ) of 1, 3, 5, and 10 wt.% Pt/C catalysts after reduction at 623 and 473
K (w/ 5 wt.%*)
Figure 2. Glycerol Steam (30 wt. %) Reforming at P = 1 bar, T = 623 K of 5 wt. %
-
1
Pt/C catalyst with different WHSV (33, 66, and 99 min )
Figure 3. Glycerol Steam (30 wt. %) Reforming (P = 1 bar, T = 623 K, WHSV =
-
1
3
6
3 min ) of 5 wt. % Pt/C catalyst with different glycerol aqueous solution pH (3,
, 10).

Figure 1.
80
70
60
50
40
30
20
10
0
1
5
5
3
0%
%*
%
%
1%
0
5
10
15
20
25
TOS (h)
Figure 2.
90
80
70
60
50
40
30
20
10
0
6
5
4
3
2
1
0
3
6
9
3 min -1
6 min -1
9 min -1
33 min -1
6
9
6 min -1
9 min -1
0
5
10
TOS (h)
15
20

Figure 3.
9
8
7
6
5
4
3
0
0
0
0
0
0
0
5.5
5
pH 6.07
pH 10.0
pH 3
4
4
3
3
2
2
.5
.5
.5
pH 6.07
pH 10.0
pH 3
0
5
10
TOS (h)
15
20
25

Table 1. CO chemisorption results
Results
1%
3%
5% *
5%
10%
µmolCO uptaken/gcat 26.00 78.80 97.50 81.30 95.70
Metal Dispersion (%)
Pt particle size (nm)
50.81 51.26 53.01 31.74 18.67
2.22 2.21 2.97 3.56 6.10
(
*) The reduction temperature was 473 K, using the same temperature ramp (0.5
-
1
3
-1
K.min ) and H
2
flow rate (35 cm .min ). After remaining 2h at 473 K under
-
1
hydrogen flowing, the temperature was increased (10 K.min ) to 623 K under He
3
-1
flowing (35 cm .min ), remaining 1h at it.
Table 2. Influence of the metal (Pt) particle size on the conversion to Liquid (X
G LP
)
and gas (X
G
)
GP phases, overall glycerol ((X
G
)
T1 and (X )T2) conversions (%), CO
G
-
1
(
CO-TOF) and Glycerol (G-TOF) Turnover frequencies (min ), and H
2
/CO ratio.
CO-TOF G-TOF
Catalyst
(X
G
)
LP
(X
G
)
GP
(X
G
)
T1
(X
G
)
T2
H /CO
2
-
1
(min-1)
42.28
(
min )
1
3
5
%
%
%
2.89
2.42
4.42
4.47
3.58
26.51
20.56
77.45
27.00
54.07
29.4
22.98
81.87
31.47
57.65
27.3
23.86
81.09
32.71
58.96
107.42
97.53
1.36
1.38
1.36
1.38
1.35
37.77
422.25
101.02
580.26
197.57
43.61
5% *
10%
226.62

Table 3. Influence of the metal (Pt) particle size on the selectivity (%) to gas and
, Ethylene (C = ), Propene(C =
liquid phases products: CO, CH
4
2
), Ethane (C
2
3
),
Propane(C
3
), CO , Acetone, Metanol (MeOH), Acetol, Ethylene glycol (EG),
2
Glyceraldehyde (GA).
Cataly
st
C
O
CH
CO MeO Aceto Acet
E
G
2.
1
G
A
=
=
C
2
C
2
C
3
C
3
4
2
H
ne
ol
75. 3.2 1.8 3.1 1.0 1.3 1.3
1%
3%
5%
2.42
1.39
4.76
2.5
3.1
0.1
5.3
0.3
2
7
6
1
8
8
0
75. 2.8 2.4 2.7 2.2 2.5 1.0
2.
1
2.12
1.47
2.71
1.27
1.28
0.02
0.86
0.12
2.95
1.95
7.56
1.15
6
1
8
0
2
0
9
90. 2.8 0.3 0.4 0.3 0.0 1.6
1.
2
8
1
2
3
0
6
1
74. 3.1 1.3 1.7 0.7 0.1 1.3
1.
8
5% *
5
5
8
8
2
4
0
93. 2.2 0.2 0.3 0.1 0.0 1.3
0.
4
1
0%
4
6
9
4
8
0
8
Table 4. Influence of the WHSV on the conversion to Liquid (X
phases, overall glycerol ((X T2) conversions (%), CO (CO-TOF) and
Glycerol (G-TOF) Turnover frequencies (min ), and H
CO-TOF G-TOF
G
)
LP and gas (X
G GP
)
G
)
T1 and (X )
G
-
1
2
/CO ratio.
-
1
WHSV(min ) (X
G
)
LP (X
G
)
GP (X
G
)
T1 (X
G
)
T2
H /CO
2
-
1
(min-1)
156.53
158.89
126.11
(
min )
33
66
99
4.42
2.14
1.72
77.35 81.77 80.78
29.39 31.53 34.20
16.03 17.75 19.87
404.44
300.96
266.0
1.35
1.38
1.36

Table 5. Influence of the WHSV on the selectivity (%) to gas and condensed
phases products: CO, CH4, Ethylene (C = ), Propene(C =
),
2
), Ethane (C
2
3
Propane(C
Glyceraldehyde (GA).
WHSV
3
), CO , Acetone, Metanol (MeOH), Acetol, Ethylene glycol (EG),
2
=
=
CO CH
4
C
2
C
2
C
3
C
3
CO
2
MeOH Acetone Acetol EG GA
-
1
(
min )
33
66
99
85.32 3.38 1.06 1.41 0.53 0.06 1.46
85.40 3.29 0.94 0.38 0.28 0.00 1.19
80.20 2.68 1.07 0.32 0.32 0.00 1.24
2.01
2.29
2.84
0.02
0.29
0.42
1.49 0.84 1.01
3.93 1.23 2.12
5.29 1.85 5.37
Table 6. Influence of the pH of the feed solution on the conversion to Liquid (X
and gas (X GP phases, overall glycerol ((X T2) conversions (%), CO
CO-TOF) and Glycerol (G-TOF) Turnover frequencies (min ), and H
G
)
LP
G
)
G
)
T1 and (X )
G
-
1
(
2
/CO ratio
.
CO-TOF
G-TOF
pH
3
(X
G
)
LP
(X
G
)
GP
(X
G
)
T1
(X
G
)
T2
H /CO
2
-
1
(min-1)
133.13
156.53
103.11
(
min )
3.04
4.42
4.00
65.70
77.35
45.39
68.74
81.77
49.40
68.35
80.78
53.20
337.98
404.44
248.65
1.46
1.38
1.35
6.07
10

Table 7. Influence of the pH of the feed solution on the selectivity (%) to gas and
, Ethylene (C =
), Ethane (C ),
condensed phases products:
CO, CH
4
2
2
Propene(C
3
=), Propane(C
3
), CO , Acetone, Methanol (MeOH), Acetol, Ethylene
2
glycol (EG), Glyceraldehyde (GA).
=
=
pH CO
88.8
.0 85.3
86.7
CH
4
C
2
C
2
C
3
C
3
CO
2
MeOH Acetone Acetol EG
GA
1
0
2.83
3.38
3.61
0.50 0.38 0.13 0.00 1.09
1.06 1.41 0.53 0.06 1.35
0.67 0.56 0.33 0.00 2.28
2.74
2.01
2.76
0.02
0.02
0.02
0.88
1.49
1.18
1.60 1.92
0.84 1.01
1.60 2.28
6
3