i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Hydrogen production through the aqueous phase reforming of

ethylene glycol over supported Pt-based bimetallic catalysts

Ho-Dong Kim a,b, Hyun Ju Park a, Tae-Wan Kim a, Kwang-Eun Jeong a, Ho-Jeong Chae a,
Soon-Yong Jeong a, Chang-Ha Lee b, Chul-Ung Kim a,*
a
Petroleum Displacement Technology Research Center, Green Chemistry Research Division, Korea Research Institute of Chemical Technology,
P.O. Box 107, Sinseongno 19, Yuseong, Daejeon 305-600, Republic of Korea
b
Department of Chemical and Biomolecular Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-749,
Republic of Korea

article info abstract

Article history: The catalytic activities of supported Pt-based bimetallic catalysts (Pt-M) were studied for
Received 14 December 2011 hydrogen production via aqueous phase reforming (APR) using a 10 wt% ethylene glycol
Received in revised form solution. The catalysts and supports used were characterized via X-ray powder diffraction
22 February 2012 (XRD), transmission electron microscopy (TEM), nitrogen adsorption-desorption, CO
Accepted 27 February 2012 chemisorption, and temperature programmed reduction (TPR) techniques. It was found
Available online 23 March 2012 that the PteMn (Pt:Mn ¼ 1:1, molar ratio) bimetallic catalyst significantly enhanced the
catalytic performances such as the hydrogen yield and hydrogen production when
Keywords: compared with monometallic catalysts and other bimetallic catalysts that were examined.
Aqueous phase reforming The XRD and TPR studies confirmed the interaction between the Pt and Mn species, leading
PteMn bimetal to the PteMn alloys supported on CMK-3. Related to the effect of the type of support, the
Carbon support CMK-3 support demonstrated better performance than the commercial activated carbon
Hydrogen and alumina. Accordingly, it can be understood that the better catalytic performance of the
Ethylene glycol APR reaction over PteMn/CMK-3 catalyst is dependent on the alloy effect as well as the
structural properties and nature of support given by the addition of the second metal.
Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.

1. Introduction oxygenated hydrocarbons (e.g. ethylene glycol, glycerol,

sorbitol etc.) with high selectivity and hydrogen yield in
Hydrogen is a potentially desirable alternative energy source a relatively mild reaction conditions (200e250  C, 15e50 bar)
as a nonpolluting, efficient, and cost attractive energy carrier. without toxic by-products. Thus, the biomass-derived
At present, due to the depletion of petroleum and interna- oxygenated hydrocarbons can be converted to hydrogen,
tional environmental regulations, many researchers are carbon dioxide, and light alkanes with a low level of CO via
working to shift the use of petroleum resources to non- various chemical reactions (e.g. CeC cleavage, water-gas shift
petroleum resources such as biomasses as raw material for (WGS), dehydration/hydration, methanation, Fischer-Tropsch
hydrogen production [1e3]. Many researches have reported (F-T) reaction etc.) [4e15]. In general, the active catalysts for
that hydrogen can be produced efficiently from aqueous APR are claimed to have a high catalytic activity for the WGS
phase reforming (APR) of renewable biomass-derived reaction and a sufficiently high catalytic activity for the CeC

* Corresponding author. Tel.: þ82 42 860 7504; fax: þ82 42 860 7508.
E-mail address: [email protected] (C.-U. Kim).
0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2012.02.160
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7 8311

bond cleavage. Group VIII metals, especially Pt, have shown impregnation method with an aqueous solution of
higher activities for the CeC cleavage than other metals. H2PtCl6∙6H2O (Acros, 99.9%), HReO4 (Alfa, 99%), Mn(NO3)2$H2O
However, the high cost of noble metals is a significant draw- (Aldrich, 99%), Fe(NO3)3∙6H2O (Aldrich, 99%), CsNO3 (Aldrich,
back for their use in industrial applications. 99%), Ba(NO3)2 (Kanto, 99%), Ga(NO3)3∙6H2O (Aldrich, 99%),
In previous APR study on the Pt loading in ordered meso- AgNO3 (Aldrich, 99%), and (NH4)6Mo7O24∙4H2O (Aldrich, 99%),
porous carbon CMK-3 support [16], the hydrogen production respectively. The CMK-3 supported bimetallic PteMn catalysts
rate as a function of the Pt loading increased up to 7 wt% Pt with different mole ratios of 1:0.5 to 1:3 (Pt:Mn) were also
loading and then approached a steady-state value of approxi- prepared by impregnating the support with an aqueous solution
mately 37.8 cc/gcat$min. Thus, it is remains a challenge to using the same procedure. After impregnation, the samples
reduce the expensive Pt loading by the introduction of second were dried at 100  C overnight and reduced via heating under
metal (M) to Pt catalyst in the APR. Of the available Pt-M a hydrogen flow rate of 50 cc/min. The reduction temperature
bimetallic catalytic systems, it is widely accepted that the was linearly increased from room temperature to 350  C, and
PteRe bimetallic catalyst has demonstrated good catalytic then maintained at this temperature for 12 h. The samples were
performance for hydrogen productivity and carbon conversion subsequently cooled with a nitrogen flow for 6 h. For compar-
to gas [12,13]. However, it still have high loadings of both ison, PteMn supported on commercial activated carbon
precious metals with 3 wt% of each Pt and Re metals due to the (Aldrich) and commercial alumina (g-Al2O3, Strem Chemical)
expensive Re cost, which is similar to Pt although the loading of were also prepared using the incipient wetness method and are
Pt is reduced by the addition of Re. Considering the high cost denoted as PteMn/AC and PteMn/alumina, respectively.
and overloading of platinum, Pt-based non-noble bimetallic
catalysts (Pt-M catalysts) have received considerable attention 2.2. Catalyst characterization
as promising catalysts for cheaper second metals and to
increase the activity on the Pt in the APR reaction [8,10,13]. The catalysts were characterized using various analytic tech-
Another important factor that affects the catalytic activity niques such as X-ray powder diffraction (XRD), transmission
in the APR is the nature of the catalyst support. The type of electron microscopy (TEM), nitrogen adsorption-desorption,
support can significantly affect the catalytic performance of temperature programmed reduction (TPR) and CO chemisorp-
the platinum-based catalysts in the glycerol APR reaction tion. Before the characterization, all the samples were reduced
[17,18]. Oxide supports such as Al2O3, SiO2 and ZnO are not under a hydrogen flow rate of 50 cc/min at 350  C. The powder
sufficiently stable in the APR conditions due to their weak XRD patterns were recorded on a Rigaku Multiplex instrument
hydrothermal stability under high-temperatures through using Cu Ka radiation (l ¼ 0.1541 nm), operated at 40 kV and
sintering (pore degradation) and phase transformation 40 mA (1.6 kW). The TEM images were taken from the thin
process [6,19]. Carbon-based supports such as activated edges of particles supported on a porous carbon grid, using
carbon (AC) and carbon black were also examined; However a Philips Tecnai G220 device operated at 200 kV.
despite their good hydrothermal stability with high a surface The nitrogen adsorptionedesorption isotherms were
area, the catalytic activity of the APR is not sufficiently high measured at 196  C on using a Micromeritics Tristar 3000
due to the textural nature of the carbon support (e.g. irregular volumetric adsorption analyzer. Before the adsorption
pore arrangement, broad distribution of pore sizes, high measurements, all samples were outgassed at 300  C for 5 h.
microporosity, and pore resistance in liquid-phase reactants). The Brunauer-Emmett-Teller (BET) equation was used to
In this regard, the ordered mesoporous carbon materials can calculate the apparent surface area from adsorption data
be considered as ideal supports for the APR due to their obtained at P/P0 values between 0.05 and 0.2. The total volume
remarkable properties such as hydrothermal stability, of the micropores and mesopores was calculated using the
uniform pore size distribution, and ordered pore structure amount of nitrogen adsorbed at P/P0 ¼ 0.95. The pore size
[16]. distributions were calculated through analysis of the adsorp-
In this work, we have investigated the effect of the bime- tion branch of the nitrogen physisorption isotherm using the
tallic composition, Pt-M (PteRe, PteMn, PteFe, PteCs, PteBa, Barret-Joyner-Halenda (BJH) method.
PteGa, PteAg, and PteMo) and support (CMK-3, commercial The TPR measurements were carried out in a Micromeritics
AC and alumina) on the APR activity in order to develop AutoChem II 2920 analyzer connected to a thermal conduc-
a cheaper and efficient APR catalyst. The molar ratio of Pt:M in tivity detector (TCD) at temperatures ranging from room
the optimum bimetallic composition is also discussed. temperature to 1000  C with a ramping rate of 10  C/min. For
the TPR measurements, a mixed stream of H2 (3 cc/min) and
N2 (27 cc/min) was used for 0.1 g of the catalyst sample. The
2. Experimental CO chemisorption experiments (Micromeritics ASAP 2020
chemi) were conducted in order to measure the metal
2.1. Catalyst preparation dispersion of the catalysts.

The ordered mesoporous carbon, CMK-3, support was synthe- 2.3. Catalytic activity test
sized using SBA-15 as hard template and furfuryl alcohol as
a carbon precursor according to the procedure reported in the An APR reaction of 10 wt% ethylene glycol (EG) was performed
literature [20e23]. The CMK-3 supported monometallic cata- in a fixed bed reactor system with an up-flow liquid feeding
lysts, Pt and Mn, and bimetallic catalysts, Pt-M (M: Re, Mn, Fe, Cs, system that has been described elsewhere [16]. The reactor
Ba, Ga, Ag and Mo), were prepared via an incipient wetness was a stainless steel tube with a height of 360 mm and an
8312 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7

 
inner diameter of 3/8 inch. Before performing the APR reac-
Alkane selectivity %
tion, the catalyst was subjected to a H2 reduction step under 50
cc/min flow at 350  C (heating rate: 1.0  C/min) for 6 h. After C atoms in alkane produced
¼  100
the reduction, the reforming reaction was performed with Total C atoms produced in gas products
a weight hourly space velocity (WHSV) of 2.0 h1 at 250  C
 
under 45 atm. An aqueous feed of 10 wt% EG was continuously
Carbon conversion to gas %
fed into the reactor at a pumping rate of 0.1 cc/min using
a high-performance liquid chromatography (HPLC) pump. The C atoms in the gas phase products
¼  100
effluent gas products were measured using a gas chromato- Total C atoms in the feedstock
graph (6100 GC, Young Lin Instrument Co.) equipped with
a thermal conductivity detector and flame ionization detector.
The H2, CO, CO2, CH4 and hydrocarbons in the effluent gas
were detected using Carboxen 1000 (1/8 in  15 ft) and GS- 3. Results and discussion
GASPRO (0.32 mm  15 m) columns, respectively. The
acquired data were calculated according to following defini- 3.1. Support and catalyst characterizations
tions as described previously in the literature [5].
  The structure of the ordered mesoporous carbon material
H2 produced experimentally CMK-3 and CMK-3 supported catalysts were analyzed using
H2 yield % ¼
H2 calculates according to theoretical quantity a series of different characterization techniques, including
 100 XRD, TEM, and nitrogen adsorptionedesorption isotherms.
Fig. 1(a) shows the low angle XRD patterns of the supports
 
H2 produced experimentally 1 (SBA-15 and CMK-3) and supported catalysts (1 wt% Pt/CMK-3,
H2 selectivity % ¼ 
C atoms produced in gas products R 1 wt% Mn/CMK-3, and 1 wt% PteMn (1:1)/CMK-3) examined
 100 via the XRD measurements. The crystalline phases according
to the wide-angle XRD patterns and typical TEM image of the
where R is the H2/CO2 reforming ratio of 5/2 for ethylene 1 wt% PteMn (1:1)/CMK-3 are also shown in Fig. 1(b) and (c),
glycol. respectively.

Fig. 1 e (a) Low-angle powder XRD patterns of the CMK-3 and CMK-3 supported catalysts, (b) wide-angle XRD patterns of the
CMK-3 and CMK-3 supported catalysts, and (c) TEM image of 1 wt% PteMn/CMK-3.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7 8313

As depicted in Fig. 1(a), the XRD pattern of CMK-3 clearly

confirms the successful replication of the ordered meso-
porous carbon with the two-dimensional (2-D) hexagonal
space group ( p6mm) from the SBA-15 mesoporous silica hard
template. Thus, the SBA-15 and CMK-3 show three distinct
XRD peaks, which can be indexed as (100), (110), and (200),
reflecting a highly ordered mesoporous structure [24]. After
impregnation of the Pt, Mn, and PteMn metals into the CMK-3,
there were no changes in XRD patterns, indicating that all
metals were well dispersed in the supports.
In the wide-angle powder XRD patterns as shown in Fig. 1(b),
it was found that the CMK-3 support and 1wt% PteMn (1:1)/
CMK-3 catalysts retained a diffuse diffraction peak appearing
near 2q ¼ 23 , indicating an amorphous carbon-like nature due
to the CMK-3 support using the furfuryl alcohol as a carbon
precursor. XRD pattern for Pt or Mn metal showed no distinct
diffraction peaks (Fig. 1 (b)). This indicates that Pt and Mn metals
are highly dispersed in the CMK-3 support. The intrinsic XRD
pattern of the Pt displayed three diffraction peaks appearing at
2q ¼ 39.7, 46.2 and 67.5 , which correspond to the (111), (200) and
(220) X-ray diffractions for typical polycrystalline Pt face-
centered cubic phase. On the other hand, PteMn bimetallic
CMK-3 catalyst shows three diffraction peaks appeared at
2q ¼ 40.0, 46.5 and 67.9 , which are not in accordance with those
of the pure Pt metal. The similar result was reported by Zeng and
Bonakdarpour [25,26]. They also analyzed the XRD patterns for
Pt, PteCo and PteMneCo and reported that Pt (111) diffraction
peak is shifted to a higher angle (2q ¼ 40.3 ) as PteCo and
PteMneCo forms alloy. The average particle size of the PteMn
particles calculated using Scherrer’s equation was 5.2 nm. The
TEM images of 1 wt% PteMn (1:1 M ratio)/CMK-3 sample is dis-
played in Fig. 1(c). As shown in Fig. 1(c), the highly ordered
mesoporous structure with a mesostructure comprising parallel
porous channels and carbon rods was observed, and the average
size of the majority of PteMn nanoparticles remains at
approximately 4e5 nm. This TEM result is in good agreement
with the result of the wide-angle XRD patterns.
Fig. 2 shows the nitrogen adsorptionedesorption isotherms
and BJH pore size distributions of the CMK-3 support, 1 wt% Pt/
CMK-3, 1 wt% Mn/CMK-3, and 1 wt% PteMn/CMK-3 catalysts.
As shown in Fig. 2(a) and (b), the ordered meosoporous carbons
Fig. 2 e (a) Nitrogen sorption isotherm and (b) BJH pore size
exhibited the typical Type IV curve with sharp capillary
distribution of CMK-3 and CMK-3 supported catalysts. The
condensation steps and narrow pore size distributions which
isotherms for 1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3, and 1 wt
indicate a well-defined porous structure. After the impregna-
% PteMn/CMK-3 samples are offset vertically by 200, 400,
tion of the Pt, Mn, and PteMn into the CMK-3, there were no
and 600 cm3 STP/g, respectively. The pore size
changes the isotherms, indicating that all metals are well
distributions for 1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3, and
dispersed in the supports. These results are consistent with the
1 wt% PteMn/CMK-3 samples are offset vertically by 1, 2,
XRD results. The detailed structural properties of all catalysts
and 3 cm3 STP/g, respectively.
are summarized in Table 1.
Fig. 3 shows the TPR profiles of the CMK-3 support, mono
(1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3) and bimetallic catalyst
(1 wt% Pt-Mn/CMK-3) used to investigate the effect of the
metal/support interaction. The TPR pattern of the CMK-3 peaks of the CMK-3 support. The TPR profiles of the 1 wt% Pt/
support exhibited a single broad peak over 500e700  C, CMK-3 and 1 wt% PteMn/CMK-3 catalyst exhibited a distinct
which was attributed to the methane emission resulting from reduction peak of the metal itself, whereas the 1 wt% Mn/
the hydrogenation of the surface functional groups on the CMK-3 catalyst was observed with a broader peak over-
carbon material support [27e29]. However, the TPR profiles of lapping the peak of the CMK-3 support. With the addition of
the CMK-3 supported catalysts (1 wt% Pt/CMK-3, 1 wt% Mn/ Mn into Pt on the CMK-3 support (1 wt% PteMn/CMK-3 cata-
CMK-3 and 1 wt% PteMn/CMK-3) are exhibited two reduc- lyst), a single sharp peak was observed at a slightly shifted
tion peaks accompanying the broad hydrogen consumption position compared with the Pt peak. Thus, the reduction
8314 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7

Table 1 e Structural properties of supports and supported catalysts.

Sample Loading (Mn/Pt) a SBET (m2/g) b Vt (cm3/g) c wBJH (nm) d

CMK-3 e 1129 1.20 3.8

AC e 1412 1.06 4.9
Al2O3 e 208 0.46 8.6
1wt% Pt/CMK-3 e 1137 1.16 3.8
1wt% Mn/CMK-3 e 1098 1.11 3.8
1wt% PteMn(1:0.5)/CMK-3 0.5 1117 1.14 3.8
1wt% PteMn(1:1)/CMK-3 1 1099 1.12 3.8
1wt% PteMn(1:2)/CMK-3 2 1088 1.11 3.8
1wt% PteMn(1:3)/CMK-3 3 1086 1.07 3.8
1wt% PteMn(1:1)/AC 1 1319 0.93 4.6
1wt% PteMn(1:1)/alumina 1 162 0.42 7.9
1wt% PteRe(1:1)/CMK-3 1 1089 1.10 3.8

a Molar ratio of PteMn.

b SBET, apparent BET specific surface area deduced from the isotherm analysis in the relative pressure range from 0.05 to 0.20.
c Vt, total pore volume at relative pressure 0.95.
d wBJH, the pore diameter calculated using the BJH method.

profiles of the bimetallic catalysts do not contain peaks in the a CO:M ratio of unity. We note that the low CO chemisorption
same region as the CMK-3 supported monometallic catalysts value with 1 wt% Mn/CMK-3 catalyst, as well as the decrease
(1 wt% Pt/CMK-3 and 1 wt% Mn/CMK-3). This is as an indica- in the CO chemisorption values observed in our work
tion of more formations of the alloy particles. The actual following the Mn addition, was the same results of the PteRe/
amount of hydrogen consumed was estimated through the carbon catalyst reported by King [13].
integration of each reduction signal by considering with only
metal peak, each value of 1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3, 3.2. Effect of second metal addition in Pt-based CMK-3
and 1 wt% PteMn (1:1)/CMK-3 were 0.24, 0.34 and 0.51 mmol/g, catalysts on the APR reaction
respectively.
Table 3 summarizes the chemisorption results for the In order to improve the hydrogen production in the APR
various 1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3, and different reaction of the EG, the effect of the second metal addition in
supported 1 wt% PteMn (1:1) catalysts used in this study. On the 1 wt% Pt-based CMK-3 catalysts was examined and the
the 1 wt% Pt/CMK-3 catalyst, the irreversible CO uptake indi- results are shown in Table 2. Eight different metals (Re, Fe, Cs,
cates a metal dispersion of 49.9%, based on the assumption of Ba, Ga, and Ag) including reducible oxides (Mn and Mo) were
used and the atomic ratio of the Pt to the second metal was
fixed at 1:1 molar ratio. It can be seen from Table 2 that the
hydrogen production rate and hydrogen yield with the addi-
tion of the second metal in the same CMK-3 support decreased
in the following order: PteMn > PteFe > PteRe > Pt > PteCs >
PteBa > PteGa > PteMo > PteAg. Thus, the highest hydrogen
production rate was obtained over the 1 wt% PteMn/CMK-3
catalyst from the catalysts applied, whereas, the PteRe/CMK-3
catalyst obtained both higher conversion to gas and higher
alkane selectivity. This latter results was in good agreement
with the data from our result with the CMK-3 supported PteRe
catalyst. Interestingly, the CMK-3 supported PteMn, PteRe,
and PteFe catalysts were promoted to catalytic activity
through the addition of a second metal in the Pt compared
with the Pt monometallic catalyst. In particular, Mn is well
known as a promoting metal that is favorable for water-gas
shift reaction compared with the other metal oxides for CO
oxidation reactions and the PteMn alloys supported on carbon
produced some enhancement in the kinetics of the oxygen-
ereduction reaction (ORR) relative to the monometallic Pt
supported catalysts [14,15,30]. The addition of Mn into Pt could
induce increases in the reducibility and alloy formation which
is responsible for the enhancement of the catalytic perfor-
Fig. 3 e Temperature programmed reduction results of the mance in the APR accentuating the hydrogen production rate
CMK-3, 1 wt% Pt/CMK-3, 1 wt% Mn/CMK-3, and 1 wt% for the PteMn/CMK-3 catalyst relative to 1 wt% Pt/CMK-3
PteMn/CMK-3 catalysts. catalyst as shown in Fig. 1(b) and Fig. 3.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7 8315

Table 2 e Catalytic performance of 1 wt% Pt-M (1:1, molar ratio)/CMK-3 catalysts in APR of 10 wt% ethylene glycol at 250  C,
45 atm, WHSV of 2.0 hL1.
Catalyst Pt-M (1:1) Conversion (%) H2 production rate (cc/gcat$min) H2 yield (%) Alkanes selectivity (%)

PteRe 44.2 19.9 30.0 19.8

PteMn 39.7 26.8 40.2 2.9
PteFe 30.4 22.1 33.2 2.0
PteCs 20.1 12.5 18.8 1.5
PteBa 15.3 10.6 15.8 2.7
PteGa 5.7 5.5 8.2 3.85
PteAg 0.2 0.3 0.4 0
PteMo 5.8 3.1 4.6 18.9
Pt 25.4 17.7 26.6 4.5

Table 3 e Results of the CO chemsisorption and catalytic performance of CMK-3 supported Pt, Mn, and different supported
PteMn (1:1, molar ratio) catalysts in the APR of 10 wt% ethylene glycol at 250  C, 45 atm, and WHSV of 2.0 hL1.
Catalyst CO uptake Metal Conversion H2 production H2 Alkanes TOF of
(mmol/g)a dispersion (%)b (%) rate (cc/gcat$min) yield (%) selectivity (%) H2 (min1)c

1wt% Pt/CMK-3 25.6 49.9 25.4 17.7 26.6 4.5 103.0

1wt% Mn/CMK-3 1.9 1.1 3.2 2.0 3.0 12.2 e
1wt% PteMn(1:1)/CMK-3 22.1 43.0 39.7 26.8 40.2 2.9 180.9
1wt% PteMn(1:1)/AC 19.3 37.9 34.2 22.7 34.1 5.5 173.4
1wt% PteMn(1:1)/alumina 17.2 34.5 27.7 19.6 29.4 3.8 162.6
a, b
CO uptake and metal dispersion were measured by CO chemisorptions.
c The TOF of catalyst was calculated by using active Pt site from CO chemisorption and H2 production rate.

3.3. Effect of the catalyst support and PteMn molar and hydrogen yield than the commercial AC and alumina
ratio supported catalyst (1 wt% PteMn/AC, 1 wt% PteMn/alumina).
This superior APR activity of the 1 wt% PteMn/CMK-3 catalyst
The CO chemisorption and catalytic performance of CMK-3 may be attributed to the ordered pores and opened mesopores
supported Pt, Mn, and different supported PteMn (1:1, molar interconnected within the CMK-3 support structure, which is
ratio) catalysts in the APR of 10 wt% EG are listed in Table 3.
The CO chemisorption was performed in order to determine
the CO uptake and metal dispersion. As can be seen in Table 3,
the effect of the active component on the catalytic activity in
the same CMK-3 supports is as follows: 1 wt% Mn < 1 wt% Pt <
1 wt% PteMn. The 1 wt% Mn/CMK-3 catalyst showed little
activity in the APR reaction of EG due to the low value of CO
uptake and low metal dispersion. This means that the Mn
metal itself did not uptake CO significantly. However, the
addition of Mn to the Pt catalyst has a higher catalytic activity
than the monometallic Pt catalyst even though it has a low CO
uptake and low metal dispersion. This indicates that the
increase of the reducibility and alloy formation of the Pt and
Mn is responsible for the enhancement of the catalytic
performance.
Fig. 4 shows the rate of hydrogen production for the sup-
ported 1 wt% PteMn catalysts with different supports such as
CMK-3, AC, and alumina. It can be seen that the three different
supported Pt-Mn catalysts exhibit a relatively constant rate of
hydrogen production with time on stream. This implies that
all catalysts are stable during the overall APR reaction without
deactivation, which should lead to a decrease in the hydrogen
production rate. The rate of hydrogen production was greatly
affected by the type of support employed and decreased in the Fig. 4 e Rates of hydrogen production over 1 wt% PteMn/
following order 1 wt% PteMn/CMK-3 > 1 wt% PteMn/AC > CMK-3, 1 wt% PteMn/AC, and 1 wt% PteMn/Al2O3 with
1 wt% PteMn/alumina. Thus, the CMK-3 supported catalyst, time on-stream at 250  C, 45 atm and WHSV of 2.0 hL1 for
1 wt% PteMn/CMK-3 exhibited better hydrogen production APR of 10 wt% aqueous ethylene glycol.
8316 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7

Table 4 e Results of the CO chemisorption and catalytic performance of the CMK-3 supported Pt, Mn, and different molar
ratios of Pt-Mn catalysts in the APR of 10 wt% ethylene glycol at 250  C, 45 atm, and WHSV of 2.0 hL1.
Catalyst CO uptake Metal Conversion H2 production H2 yield (%) Alkanes
(mmol/g)a dispersion (%)b (%) rate (cc/gcat$min) selectivity (%)

1wt% PteMn(1:0.5)/CMK-3 23.8 46.5 27.6 19.1 28.7 4.0

1wt% PteMn(1:1)/CMK-3 22.1 43.0 39.7 26.8 40.2 2.9
1wt% PteMn(1:2)/CMK-3 20.5 40.1 37.1 26.7 40.1 3.1
1wt% PteMn(1:3)/CMK-3 10.1 19.7 19.9 15.0 21.9 3.4
a, b
CO uptake and metal dispersion were measured by CO chemisorptions.

effective for both good accessibility of the liquid-phase reac- hydrogen to carbon dioxide was close to this theoretical value.
tant to catalytic active sites and an easy escape for the gas This indicates that the addition of Mn was facilitated by the
products (e.g. H2, CO2, CO, and alkanes) from catalytic active main reaction as a promoter in the reforming reaction without
site of metal without a diffusion resistance [31], whereas the the H2 consuming reaction. From the above results, it can be
1 wt% PteMn/alumina and 1 wt% PteMn/AC catalysts with suggested that the PteMn catalyst not only facilitated CeC
irregular porous structures were shown in the metal disper- bond cleavage reaction but also inhibited the C-O bond
sions. These trends were also consistent with the result of CO cleavage and dehydration in the APR reaction.
uptake and metal dispersion described in Table 3. Based on
these results, the hydrogen production of the APR reaction
should relate closely to the nature and structural properties of 4. Conclusions
the catalyst supports.
Table 4 shows the chemisorption results and catalytic In the APR of the EG, the effect of bimetallic compositions
activities in the APR reaction of the EG over the 1 wt% PteMn/ (PteRe, PteMn, PteFe, PteCs, PteBa, PteGa, PteAg, and PteMo)
CMK-3 catalysts with four different Pt:Mn molar ratios (1:0.5, and supports (CMK-3, AC, and alumina) were studied. Among
1:1, 1:2, and 1:3). The catalytic activity of the PteMn catalysts these bimetallic compositions, the addition of Mn to the Pt/
increased with the increase of the Pt:Mn ratio from 1:0.5 to 1:2, CMK-3 bimetallic catalyst exhibited good yield and produc-
except the Pt:Mn ¼ 1:3 molar ratio. The reason indicate that tion rates for hydrogen than the monometallic catalysts and
the addition of the Mn to Pt facilitated the CeC bond cleavage other bimetallic catalysts used, whereas the addition of Re to
and transformation of CO to CO2 in the APR. Moreover the Pt/CMK-3 increased carbon to gas conversion. These results
addition in the Mn content to the Pt can be enhance the indicate that Mn itself does not participate in the reforming
activity of the CO oxidation [15]. Thus, Mn itself does not reaction as a CeC bond cleavage unlike Pt, but rather that it acts
participate in the reforming reaction as a CeC bond cleavage as a promoter to improve the catalytic activity. For the effect of
unlike Pt, but rather it acts as a promoter to improve the the support type, the CMK-3 support showed a better activity
catalytic activity. The same trend was found in the results in compared with the activated carbon and alumina. Also, the
this study, a high Mn ratio at Pt:Mn ¼ 1:3 demonstrated ordered mesoporous carbon CMK-3 supported bimetallic cata-
a significantly lower activity than the monometallic 1 wt% Pt/ lyst demonstrated good conversion and selectivity for
CMK-3. From this result, it can be understood that a part of the hydrogen production compared with the other supports due to
Pt active sites is covered through the addition of the excess the easy accessibility of the liquid-phase reactant resulting
Mn, resulting in a decrease in the catalytic activity. This from its unique structure.This suggests that the Pt-Mn/CMK-3
hypothesis is demonstrated by the CO chemisorptions results catalyst is very efficient in developing a much cheaper and
of all the PteMn bimetallic catalysts. Since manganese has the highly efficient APR catalyst due to its alloy formation and
little effect on the CO consumption during the chemisorption mesoporous structure of the support.
(Table 3), the CO uptakes of all the PteMn bimetallic catalysts
come mainly from the Pt metals. Accordingly, it can be
considered that as the amount of manganese increased to
Acknowledgment
3 mol, the gradually decreased metal dispersion from 49.9 to
19.7% due to the partial coverage of the Pt active sites by the
This work was financially supported by a grant from the
some manganese. The similar result was reported by Huber
Industrial Source Technology Development Programs (Grant
et al. [10]. They also conducted the APR of EG over supported Pt
No. 10033099) of the Ministry of Knowledge Economy (MKE) of
and Pd bimetallic catalysts and reported that when large
Korea.
amount of Ni or Co is added to the Pt catalyst, some of the
metals which do not form alloy with the Pt can cause the
catalyst deactivation. references
In addition, the H2 yield and selectivity of the PteMn
catalyst were higher than the PteRe catalyst. In the APR
reaction of the EG, the theoretical reforming ratio of the [1] Navarro RM, Peña MA, Fierro JLG. Hydrogen production
hydrogen to carbon dioxide was 2.5 [32]. The present APR reactions from carbon Feedstocks: Fossil Fuels and biomass.
results of the PteMn catalyst on the reforming ratio of the Chemical Reviews 2007;107:3952e91.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 8 3 1 0 e8 3 1 7 8317

[2] Lubitz W, Tumas W. Hydrogen: an Overview. Chemical [18] Wen G, Xu Y, Xu Z, Tian Z. Characterization and catalytic
Reviews 2007;107:3900e3. properties of the Ni/Al2O3 catalysts for aqueous-phase
[3] Chheda JN, Huber GW, Dumesic JA. Liquid-Phase catalytic reforming of Glucose. Catalysis Letters 2009;129:250e7.
processing of biomass-derived oxygenated hydrocarbons to [19] Johnson MFL. Surface area stability of aluminas. Journal of
Fuels and Chemicals. Angewandte Chemie International Catalysis 1990;123:245e59.
Edition 2007;46:7164e83. [20] Choi M, Heo W, Kleitz F, Ryoo R. Facile synthesis of high
[4] Shabaker JW, Davda RR, Huber GW, Cortright RD, quality mesoporous SBA-15 with enhanced control of the
Dumesic JA. Aqueous-phase reforming of methanol and porous network connectivity and wall thickness. Chemical
ethylene glycol over alumina-supported platinum catalysts. Communications; 2003:1340e1.
Journal of Catalysis 2003;215:344e52. [21] Gierszal KP, Kim T-W, Ryoo R, Jaroniec M. Adsorption and
[5] Cortright RD, Davda RR, Dumesic JA. Hydrogen from catalytic structural properties of ordered mesoporous carbons
reforming of biomass-derived hydrocarbons in liquid water. synthesized by using various carbon precursors and ordered
Nature 2002;418:964e7. Siliceous P6mm and Ia3d mesostructures as Templates. The
[6] Shabaker JW, Huber GW, Davda RR, Cortright RD, Journal of Physical Chemistry B 2005;109:23263e8.
Dumesic JA. Aqueous-Phase reforming of ethylene glycol [22] Kim T-W, Solovyov LA. Synthesis and characterization of
over supported platinum catalysts. Catalysis Letters 2003;88: large-pore ordered mesoporous carbons using gyroidal silica
1e8. template. Journal of Materials Chemistry 2006;16:1445e55.
[7] Meryemoglu B, Hesenov A, Irmak S, Atanur OM, Erbatur O. [23] Che S, Lund K, Tatsumi T, Iijima S, Joo SH, Ryoo R, et al.
Aqueous-phase reforming of biomass using various types of Direct Observation of 3D mesoporous structure by Scanning
supported precious metal and raney-nickel catalysts for electron microscopy (SEM): SBA-15 silica and CMK-5 carbon.
hydrogen production. International Journal of Hydrogen Angewandte Chemie International Edition 2003;42:2182e5.
Energy 2010;35:12580e7. [24] Jun S, Joo SH, Ryoo R, Kruk M, Jaroniec M, Liu Z, et al.
[8] Kunkes EL, Simonetti DA, Dumesic JA, Pyrz WD, Murillo LE, Synthesis of New, Nanoporous carbon with Hexagonally
Chen JG, et al. The role of rhenium in the conversion of ordered mesostructure. Journal of the American Chemical
glycerol to synthesis gas over carbon supported platinum- Society 2000;122:10712e3.
rhenium catalysts. Journal of Catalysis 2008;260:164e77. [25] Zeng J, Yang Lee J. Ruthenium-free, carbon-supported cobalt
[9] Iriondo A, Barrio VL, Cambra JF, Arias PL, Güemez MB, and tungsten containing binary & ternary Pt catalysts for the
Navarro RM, et al. Influence of La2O3 modified support and Ni anodes of direct methanol fuel cells. International Journal of
and Pt active phases on glycerol steam reforming to produce Hydrogen Energy 2007;32:4389e96.
hydrogen. Catalysis Communications 2009;10:1275e8. [26] Bonakdarpour A, Stevens K, Vernstrom GD, Atanasoski R,
[10] Huber GW, Shabaker JW, Evans ST, Dumesic JA. Aqueous- Schmoeckel AK, Debe MK, et al. Oxygen reduction activity of Pt
phase reforming of ethylene glycol over supported Pt and Pd and PteMneCo electrocatalysts sputtered on nano-structured
bimetallic catalysts. Applied Catalysis B: Environmental thin film support. Electrochimica Acta 2007;53:688e94.
2006;62:226e35. [27] Simonetti DA, Kunkes EL, Dumesic JA. Gas-phase conversion
[11] Luo N, Ouyang K, Cao F, Xiao T. Hydrogen generation from of glycerol to synthesis gas over carbon-supported platinum
liquid reforming of glycerin over Ni-Co bimetallic catalyst. and platinum-rhenium catalysts. Journal of Catalysis 2007;
Biomass and Bioenergy 2010;34:489e95. 247:298e306.
[12] Kunkes EL, Soares RR, Simonetti DA, Dumesic JA. An [28] de Miguel SR, Scelza OA, Román-Martı́nez MC, Salinas-
integrated catalytic approach for the production of hydrogen Martı́nez de Lecea C, Cazorla-Amorós D, Linares-Solano A.
by glycerol reforming coupled with water-gas shift. Applied States of Pt in Pt/C catalyst precursors after impregnation,
Catalysis B: Environmental 2009;90:693e8. drying and reduction steps. Applied Catalysis A: General
[13] King DL, Zhang L, Xia G, Karim AM, Heldebrant DJ, Wang X, 1998;170:93e103.
et al. Aqueous phase reforming of glycerol for hydrogen [29] Fraga MA, Jordão E, Mendes MJ, Freitas MMA, Faria JL,
production over Pt-Re supported on carbon. Applied Figueiredo JL. Properties of carbon-supported platinum
Catalysis B: Environmental 2010;99:206e13. catalysts: role of carbon surface sites. Journal of Catalysis
[14] Ayastuy JL, González-Marcos MP, González-Velasco JR, 2002;209:355e64.
Gutiérrez-Ortiz MA. MnOx/Pt/Al2O3 catalysts for CO [30] Stamenkovic VR, Mun BS, Arenz M, Mayrhofer KJJ, Lucas CA,
oxidation in H2-rich streams. Applied Catalysis B: Wang G, et al. Trends in electrocatalysis on extended and
Environmental 2007;70:532e41. nanoscale Pt-bimetallic alloy surfaces. Nature Materials
[15] Grbic B, Radic N, Markovic B, Stefanov P, Stoychev D, 2007;6:241e7.
Marinova T. Influence of manganese oxide on the activity of [31] Wang Y-g, Cheng L, Li F, Xiong H-m, Xia Y-y. High
Pt/Al2O3 catalyst for CO and n-hexane oxidation. Applied Electrocatalytic performance of Mn3O4/Mesoporous carbon
Catalysis B: Environmental 2006;64:51e6. Composite for oxygen reduction in Alkaline Solutions.
[16] Kim T-W, Kim H-D, Jeong K-E, Chae H-J, Jeong S-Y, Lee C-H, Chemistry of Materials 2007;19:2095e101.
et al. Catalytic production of hydrogen through aqueous- [32] Davda RR, Shabaker JW, Huber GW, Cortright RD,
phase reforming over platinum/ordered mesoporous carbon Dumesic JA. A review of catalytic issues and process
catalysts. Green Chemistry 2011;13:1718e28. conditions for renewable hydrogen and alkanes by aqueous-
[17] Wen G, Xu Y, Ma H, Xu Z, Tian Z. Production of hydrogen by phase reforming of oxygenated hydrocarbons over
aqueous-phase reforming of glycerol. International Journal supported metal catalysts. Applied Catalysis B:
of Hydrogen Energy 2008;33:6657e66. Environmental 2005;56:171e86.