Seung Ju Han a, Yongju BangaSchool of Chemical and Biological Engineerin

151-744, South Korea

ering an

because it possesses high energy density (ca. 120.7 kJ/g) and

burns without emitting any environmental pollutants [1e4].

There are various resources for hydrogen production such as

coal,natural gas, propane,methanol, andethanol.Amongthese

by bio-chemical processes [5]. Thus, ethanol has served an

environmentally friendly resource for hydrogen production.

Reforming technologies for hydrogen production from

ethanol can be classified into steam reforming, auto-thermal

* Corresponding author. Tel.: 82 2 880 9227; fax: 82 2 889 7415.

Available online at www.sciencedirect.com

w.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 3E-mail address: inksong@snu.ac.kr (I.K. Song).1. Introduction

With increasing demand for new and clean energy, hydrogen

has attracted much attention as a promising energy carrier,

resources, ethanol has been considered as a practical resource,

because it is biodegradable, low in toxicity, and free from cata-

lyst poisons suchas sulfur [4].Moreover, ethanol becomesmore

abundantly available due to the increasing ethanol productionMesoporous NieAl2O3eZrO2 xerogel

catalyst

Zr/Al molar ratio

Acidity

increasing Zr/Al molar ratio due to the loss of acid sites of Al2O3 by the addition of ZrO2.

Acidity of Ni-AZ-X catalysts served as a crucial factor determining the catalytic perfor-

mance in the steam reforming of ethanol; an optimal acidity was required for maximum

production of hydrogen. Among the catalysts tested, Ni-AZ-0.2 (Zr/Al 0.2) catalyst withan intermediate acidity exhibited the best catalytic performance in the steam reforming of

ethanol.

Copyright 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rightsreserved.bDepartment of Environmental Engine

Korea

a r t i c l e i n f o

Article history:

Received 18 October 2012

Received in revised form

7 November 2012

Accepted 10 November 2012

Available online 6 December 2012

Keywords:

Hydrogen production

Steam reforming of ethanol0360-3199/$ e see front matter Copyright http://dx.doi.org/10.1016/j.ijhydene.2012.11.0a, Jeong Gil Seo b, Jaekyeong Yoo a, In Kyu Song a,*

g, Institute of Chemical Processes, Seoul National University, Shinlim-dong, Kwanak-ku, Seoul

d Energy, Myongji University, Nam-dong, Cheoin-gu, Yongin-si, Gyeonggi-do 449-728, South

a b s t r a c t

A series of mesoporous NieAl2O3eZrO2 xerogel catalysts (denoted as Ni-AZ-X ) with

different Zr/Al molar ratio (X) were prepared by a single-step epoxide-driven solegel

method, and they were applied to the hydrogen production by steam reforming of ethanol.

The effect of Zr/Al molar ratio of Ni-AZ-X catalysts on their physicochemical properties

and catalytic activities was investigated. Textural and chemical properties of Ni-AZ-X

catalysts were strongly influenced by Zr/Al molar ratio. Surface area of Ni-AZ-X catalysts

decreased with increasing Zr/Al molar ratio due to the lattice contraction of ZrO2 caused by

the incorporation of Al3 into ZrO2. Interaction between nickel oxide species and support

(Al2O3eZrO2) decreased with increasing Zr/Al molar ratio through the formation of NiO

eAl2O3eZrO2 composite structure. Acidity of reduced Ni-AZ-X catalysts decreased withmolar ratio

mesoporous NieAl2O3eZrO2 xerogel catalysts: Effect of Zr/Al

Hydrogen production by steam

journal homepage: ww2012, Hydrogen Energy P57reforming of ethanol over

elsevier .com/locate/heublications, LLC. Published by Elsevier Ltd. All rights reserved.

lysts were obtained with an ASAP-2010 (Micromeritics)

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 en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 3 1377reforming, and partial oxidation [6]. Among these technologies,

ethanol steamreforminghasbeenextensively investigateddue

to its high hydrogen yield compared to the other reforming

technologies [7]. The target reactionof ethanol steamreforming

is represented in Eq. (1). However, it is known that ethanol

steam reforming reaction is accompanied by several reactions,

including ethanol dehydration (Eq. (2)), ethanol decomposition

(Eq. (3)) followed by methane steam reforming (Eq. (4)), and

ethanol dehydrogenation (Eq. (5)). Among these reactions,

ethanol dehydration reaction is mainly responsible for low

hydrogen yield due to the formation of polymeric carbon

species [8]. On the other hand, water-gas shift reaction (Eq. (6))

is an advantageous reaction in the steam reforming of ethanol

because it enhances hydrogen yield.

C2H5OH 3H2O/ 2CO2 6H2 (DHo298 K 174 kJ/mol) (1)

C2H5OH4 C2H4 H2O (DHo298 K 45 kJ/mol) (2)

C2H5OH4 CH4 CO H2 (DHo298 K 49 kJ/mol) (3)

CH4 H2O4 CO 3H2 (DHo298 K 205 kJ/mol) (4)

C2H5OH4 C2H4O H2 (DHo298 K 68 kJ/mol) (5)

CO H2O4 CO2 H2 (DHo298 K 41.2 kJ/mol) (6)

Therefore, developing an efficient ethanol steam reforming

catalyst which can not only promote hydrogen production but

also suppress undesired ethanol dehydration reaction is of

great importance. Conventionally, nickel catalysts supported

on alumina (Ni/Al2O3) have been widely used in the steam

reforming of ethanol due to their high dehydrogenation

activity and low cost [9,10]. However, it is known that Ni/Al2O3catalysts suffer from significant deactivation caused by

carbon deposition on their acid sites during the ethanol steam

reforming reaction [11]. For this reason, many attempts have

been made to increase both catalytic activity and durability of

Ni/Al2O3 catalysts through modification of supporting mate-

rials [12]. Alkaline oxides such as MgO and CaO have been

widely used as additives or promoters, because they can

suppress the coking reaction by neutralizing acid sites of Al2O3[12]. Addition of lanthanide oxides can also increase the

catalytic activity by promoting gasification reaction of dis-

solved carbon species on the surface of nickel catalysts [13].

Among various metal oxides, ZrO2 is known to be the most

effective additive for Ni/Al2O3 catalysts in the steam reform-

ing of ethanol, because ZrO2 can not only enhance the stability

of the catalysts but also promote adsorption and dissociation

of water on the surface of nickel catalysts [14e16]. On the

other hand, Ni/Al2O3 xerogel catalysts have also been inves-

tigated as efficient catalysts for hydrogen production by

reforming reactions due to their high surface area and well-developed mesoporous structure [16,17]. Therefore, a system-

atic investigation on optimizing ZrO2 addition to NieAl2O3instrument. Chemical compositions of Ni-AZ-X catalysts were

determined by ICP-AES (ICPS-1000IV, Shimadzu) analyses.

Crystalline structures of calcined and reduced catalysts were

investigated by XRD (D-Max2500-PC, Rigaku) measurements

using Cu-Ka radiation (l 1.541 A) operated at 50 kV and100 mA. In order to examine the interaction between nickel

species and support, temperature-programmed reduction

(TPR) measurements were conducted in a conventional flow

system with a moisture trap connected to a thermal conduc-

tivity detector (TCD) at temperatures ranging from roomxerogel catalyst for hydrogen production by steam reforming

of ethanol would be worthwhile.

In this work, a series of mesoporous NieAl2O3eZrO2 xero-

gel catalysts (Ni-AZ-X ) with different Zr/Al molar ratio (X )

were prepared by a single-step epoxide-driven solegel

method, and they were applied to the hydrogen production by

steam reforming of ethanol. The effect of Zr/Al molar ratio on

the physicochemical properties and catalytic activities of

mesoporous Ni-AZ-X catalysts in the steam reforming of

ethanol was investigated.

2. Experimental

2.1. Preparation of mesoporous NieAl2O3eZrO2 xerogelcatalysts

A series of mesoporous NieAl2O3eZrO2 xerogel catalysts with

different Zr/Al molar ratio were prepared by a single-step

epoxide-driven solegel method, according to the similar

methods reported in the literatures [18,19]. Known amounts of

aluminum precursor (aluminum nitrate nonahydrate, Sig-

maeAldrich, 98%) and zirconium precursor (zirconium oxy-

nitrate hydrate, SigmaeAldrich, 99%) were dissolved in

ethanol (58 ml). The total amount of two precursors was

0.04 mol. An appropriate amount of nickel precursor (nickel

nitrate hexahydrate, SigmaeAldrich, 97%) was then dissolved

in the solution containing aluminum and zirconium precur-

sors with vigorous stirring. Propylene oxide (29 ml) as a gela-

tion agent was then added into the resulting solution. Upon

maintaining the solution for a few minutes, deprotonation of

hydrated metal salts and subsequent cross-linking of metal

ion complex occurred for the formation of NieAl2O3eZrO2composite gel. After the gel was aged for 2 days, it was washed

with ethanol two times in order tominimize the destruction of

pore structure of the catalyst by removing foreign substances

in the gel with ethanol. The product was then dried at 80 C ina convection oven for 5 days. The resulting powder was finally

calcined at 550 C for 5 h to obtain a mesoporousNieAl2O3eZrO2 xerogel catalyst. The prepared mesoporous

NieAl2O3eZrO2 xerogel catalysts were denoted as Ni-AZ-X

(X 0, 0.1, 0.2, 0.3, and 0.4), where X represented Zr/Al molarratio. Nickel loading of Ni-AZ-X catalysts was fixed at 10 wt%.

2.2. Characterization

Nitrogen adsorptionedesorption isotherms of Ni-AZ-X cata-temperature to 1000 C with a ramping rate of 5 C/min. Forthe TPR measurements, a mixed stream of H2 (2 ml/min) and

AZ-X catalysts decreased with increasing Zr/Al molar ratio.

This might be due to the increased textural density of Ni-AZ-X

catalysts caused by the addition of ZrO2 into NieAl2O3composite [21]. The lowest surface area of Ni-AZ-0.4 catalyst

could be explained by higher surface energy of Zr4 than Al3;the coalescence of structure occurred for the minimization of

surface energy, resulting in the decrease of surface area

[22,23]. However, pore volume and average pore diameter of

Ni-AZ-X catalysts showed no consistent trend with respect to

Zr/Al molar ratio.

Vol

ume

adso

rbed

(A. U

.)

Ni-AZ-0

Ni-AZ-0.1

Ni-AZ-0.2

Ni-AZ-0.3

Ni-AZ-0.4

Table 1eDetailed physicochemical properties of Ni-AZ-Xcatalysts calcined at 550 C for 5 h.

Catalyst Zr/Almolarratioa

Surfacearea

(m2/g)b

Porevolume(cm3/g)c

Averagepore

diameter (nm)d

Ni-AZ-0 0 339 0.66 5.5

Ni-AZ-0.1 0.07 336 0.62 5.3

Ni-AZ-0.2 0.15 320 0.63 5.7

Ni-AZ-0.3 0.25 292 0.49 5.0

Ni-AZ-0.4 0.39 214 0.28 3.9

a Determined by ICP-AES measurement.

b Calculated by the BET equation.

c BJH desorption pore volume.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 31378N2 (20 ml/min) was used for 100 mg of catalyst sample.

Dispersion of metal species on the reduced catalysts was

examined by TEM-EDX analyses (Tecnai F20, FEI). Prior to the

TEM analyses, each catalyst was preliminarily reduced with

a mixed stream of H2 (3 ml/min) and N2 (30 ml/min) at 550 Cfor 3 h. NH3-TPD experiments were conducted to determine

the acid property of reduced catalysts (BELCAT-B, BEL Japan).

0.07 g of each catalyst charged into the TPD apparatus was

reduced at 550 C for 3 h with a mixed stream of H2 (2.5 ml/min) and Ar (47.5 ml/min). After cooling the catalyst to room

temperature, a mixed stream of NH3 (2.5 ml/min) and He

(47.5 ml/min) was introduced into the reactor at 50 C for50 min to saturate acid sites of the catalyst. Physisorbed

ammonia was removed at 100 C for 1 h under a flow of He(50 ml/min). After cooling the sample, furnace temperature

was increased from 50 C to 900 C at a heating rate of 5 C/minunder a flow of He (30 ml/min). The desorbed ammonia was

detected using a TCD (thermal conductivity detector). The

amount of carbon deposition in the used catalysts was

determined by CHNS elemental analyses (CHNS 932, Leco).

2.3. Hydrogen production by steam reforming of ethanol

Hydrogen production by steam reforming of ethanol was

carried in a continuous flow fixed-bed reactor under atmo-

spheric pressure. Each calcined catalyst (100 mg) was reduced

with a mixed stream of H2 (3 ml/min) and N2 (30 ml/min) at

550 C for 3 h. Ethanol and water were sufficiently vaporizedby passing through a pre-heating zone, and they were

continuously fed into the reactor together with N2 carrier

(30 ml/min). Feed composition was fixed at

C2H5OH:H2O:N2 1:6:24.5. Total feed rate with respect tocatalyst weight was maintained at 23,140 ml/hg. Catalyticreaction was carried out at 500 C. Reaction products wereperiodically sampled and analyzed using an on-line gas

chromatograph (ACME 6000, Younglin) equipped with

a thermal conductivity detector (TCD). Ethanol conversion,

hydrogen yield, and carbon-containing product selectivity

were calculated according to the following equations. Fi,in and

Fi,out are molar flow rates of the specific species (i) at the inlet

and outlet of the reactor, and ni is a stoichiometric factor of

carbon-containing product.

Ethanol conversion % FEtOH;in FEtOH;out

FEtOH;in

100 (7)

Hydrogen yield % FH2 ;out3 FEtOH;in FEtOH;out 100 (8)

Si;carboncontaining product% ni Fi;carbon-containing product2 FEtOH;in FEtOH;out 100

(9)

3. Results and discussion

3.1. Physicochemical properties of Ni-AZ-X catalystsTextural properties of Ni-AZ-X (X 0, 0.1, 0.2, 0.3, and 0.4)catalysts were examined by nitrogen adsorptionedesorptionisotherm measurements as represented in Fig. 1. All the Ni-

AZ-X catalysts exhibited type-IV isotherms with H2-type

hysteresis loops, indicating the presence of well-developed

mesopores [20]. Detailed physicochemical properties of Ni-

AZ-X (X 0, 0.1, 0.2, 0.3, and 0.4) catalysts are summarized inTable 1. The observed Zr/Al molar ratios of Ni-AZ-X catalysts

were quite similar to the designed values. Surface area of Ni-

0 0.2 0.4 0.6 0.8 1.0Relative pressure (P/P0)

Fig. 1 e Nitrogen adsorptionedesorption isotherms of Ni-

AZ-X (X [ 0, 0.1, 0.2, 0.3, and 0.4) catalysts calcined at

550 C.d BJH desorption average pore diameter.

3.2. XRD and TPR results of Ni-AZ-X catalysts

Fig. 2 shows the XRD patterns of Ni-AZ-X (X 0, 0.1, 0.2, 0.3,and 0.4) catalysts calcined at 550 C. In the Ni-AZ-0 catalyst,three diffraction peaks indicative of NiAl2O4 phase (solid lines

in Fig. 2) were observed, as reported in the literatures [16,24].

However, the peak intensity of NiAl2O4 gradually decreased

with increasing Zr/Almolar ratio. Instead, the diffraction peak

of tetragonal ZrO2 (1 1 1) (dashed line in Fig. 2) became strong

with increasing Zr/Al molar ratio. It is noticeable that XRD

peak of tetragonal ZrO2 (1 1 1) in the Ni-AZ-X (X 0.1, 0.2, 0.3,and 0.4) catalysts shifted to the higher diffraction angle than

its original angle (2q 30.2), representing the latticecontraction of ZrO2 caused by the incorporation of Al

3 ionsinto ZrO2. This is because the radius of Zr

4 ion (0.84 A) islarger than that of Al3 ion (0.54 A) [25]. Thus, it can beinferred that NiOeAl2O3eZrO2 composite structure was

developed in the Ni-AZ-X catalysts by the simultaneous sol-

idestate reaction among NiO, Al2O3, and ZrO2 [26]. Interest-

ingly, diffraction peaks corresponding to NiO (closed circles in

Fig. 2) suddenly appeared in the Ni-AZ-0.4 catalyst. This is

because some Ni2 ions hardly interacted with Al2O3 when an

excess amount of ZrO2 was added into the Ni-AZ-X catalysts.

In order to elucidate the metalesupport interaction in the

Ni-AZ-X (X 0, 0.1, 0.2, 0.3, and 0.4) catalysts, TPR measure-

eAl2O3eZrO2 composite structure.

0 200 400 600 800 1000

Hyd

roge

n co

nsum

ptio

n (A

. U.)

Temperature (oC)

638

585

572Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

590

454

464

486 568

Fig. 3 e TPR profiles of Ni-AZ-X (X[ 0, 0.1, 0.2, 0.3, and 0.4)

catalysts calcined at 550 C.

20 30 40 50 60 70 80

Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

2 Theta (degree)

Inte

nsity

(A. U

.)

Ni (1 1 1) Ni (2 0 0) Ni (2 2 0)

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 en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 3 1379ments were carried out as shown in Fig. 3. All the catalysts

exhibited a reduction band at relatively high temperature

within the range of 568e638 C. On the other hand, Ni-AZ-X(X 0.2, 0.3, and 0.4) catalysts showed an additional reductionband at relatively low temperature ranging from 454 C to486 C. It has been reported that the band appeared at hightemperature is related to the reduction of nickel oxide species

strongly interactedwith Al2O3 [27], while the band appeared at

20 30 40 50 60 70 80

Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

2 Theta (degree)

Inte

nsity

(A. U

.)

NiO

NiAl2O4ZrO2 (1 1 1)Fig. 2 e XRD patterns of Ni-AZ-X (X [ 0, 0.1, 0.2, 0.3, and

0.4) catalysts calcined at 550 C.low temperature is associated with the reduction of nickel

oxide species weakly interacted with ZrO2 [28,29]. It is

noticeable that reduction peak temperature of nickel oxide

species interacting with Al2O3 decreasedwith increasing Zr/Al

molar ratio, while that of nickel oxide species interacting with

ZrO2 increased with increasing Zr/Al molar ratio. In other

words, reducibility of nickel oxide species in the Ni-AZ-X

catalysts increased with increasing Zr/Al molar ratio. From

these results, it can be inferred that the addition of ZrO2suppressed the interaction between nickel oxide and support

(Al2O3 or Al2O3eZrO2) through the formation of NiO-Fig. 4 e XRD patterns of Ni-AZ-X (X [ 0, 0.1, 0.2, 0.3, and

0.4) catalysts reduced at 550 C.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 313803.3. Crystalline structure and metal dispersion ofreduced Ni-AZ-X catalysts

Fig. 4 shows the XRD patterns of Ni-AZ-X (X 0, 0.1, 0.2, 0.3,and 0.4) catalysts reduced at 550 C. All the reduced catalystsshowed diffraction peaks corresponding to metallic Ni (solid

lines in Fig. 4). Furthermore, diffraction peaks indicative of

NiO and NiAl2O4 were not observed, indicating that nickel

species in all the Ni-AZ-X catalysts were completely reduced

into metallic nickel during the reduction process employed in

this work.

TEM images of Ni-AZ-0, Ni-AZ-0.2, and Ni-AZ-0.4 reduced

at 550 C, and EDXmaps with distribution of Ni, Al, and Zr arerepresented in Fig. 5. It was observed that density of Al

decreased while that of Zr increased with increasing Zr/Al

molar ratio, in accordance with ICP-AES analyses (Table 1). It

is noteworthy that Al and Zr were homogeneously distributed

throughout the catalysts without any significant aggregation

regardless of the different Zr/Al molar ratio.

3.4. Acidity of reduced Ni-AZ-X catalysts

Fig. 6 shows the NH3-TPD profiles of Ni-AZ-X (X 0, 0.1, 0.2,0.3, and 0.4) catalysts reduced at 550 C. Acidity of reduced Ni-AZ-X catalysts was calculated from peak area as summarized

in Table 2. It was observed that acidity of the catalysts

monotonically decreased with increasing Zr/Al molar ratio.

Fig. 5 e TEM images of (a) Ni-AZ-0, (b) Ni-AZ-0.2, and (c) Ni-AZ-

distribution of (I) Ni, (II) Al, and (III) Zr.This can be explained by the fact that acidity of the catalysts is

mainly related to Al2O3 surface, because large amount of acid

sites exists in Al2O3 rather than ZrO2 [24,30e34]. This result

can also be deduced by the fact that the addition of ZrO2

0.4 catalysts reduced at 550 C, and EDX maps with

100 200 300 400 500 600 700 800 900

Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

NH

3de

sorb

ed (A

. U.)

Temperature (oC)Fig. 6 e NH3-TPD profiles of Ni-AZ-X (X [ 0, 0.1, 0.2, 0.3,

and 0.4) catalysts reduced at 550 C.

shift reaction in the steam reforming of ethanol.

Fig. 8 shows the hydrogen yields over Ni-AZ-X (X 0, 0.1,0.2, 0.3, and 0.4) catalysts in the steam reforming of ethanol

obtained after a 900 min-reaction, plotted as a function of Zr/

Al molar ratio and acidity of the catalyst. Hydrogen yields

showed a volcano-shaped curve with respect to Zr/Al molar

ratio and acidity of the catalyst. Hydrogen yield decreased in

the order of Ni-AZ-0.2 > Ni-AZ-0.3 > Ni-AZ-0.1 > Ni-AZ-

0.4 > Ni-AZ-0. Among the catalysts tested, Ni-AZ-0.2 catalyst

exhibited the best catalytic performance in terms of hydrogen

yield. Although acidity of the catalyst was not the sole factor

determining the catalytic performance in the steam reforming

of ethanol, it greatly affected the reaction path and stability

[6,12]. Relatively more acidic catalysts (Ni-AZ-0 and Ni-AZ-0.1)

exhibited low hydrogen yield because undesired ethanol

Table 2 e Acidity of reduced Ni-AZ-X catalysts.

Catalyst Acidity (mmol-NH3/g)a

Ni-AZ-0 131

Ni-AZ-0.1 124

Ni-AZ-0.2 101

Ni-AZ-0.3 86

Ni-AZ-0.4 79

a Calculated from peak area of NH3-TPD profiles in Fig. 6.

80

120

160

200

Ni-AZ-0.4

ogen

yie

ld (%

)

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 en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 3 1381decreased surface area of the catalysts (Table 1), resulting in

the decrease of acid sites of the catalysts [35].

3.5. Hydrogen production by steam reforming of ethanolover Ni-AZ-X catalysts

Fig. 7 shows the hydrogen yields with time on stream over Ni-

AZ-X (X 0, 0.1, 0.2, 0.3, and 0.4) catalysts in the steamreforming of ethanol at 500 C. Both Ni-AZ-0 and Ni-AZ-0.1catalysts experienced a catalytic deactivation resulting from

the significant carbon deposition on the catalyst surface

(Table 3). However, Ni-AZ-X (X 0.2, 0.3, and 0.4) catalystsshowed a stable performance with time on stream. It is

believed that the addition of ZrO2 improved the catalytic

stability by inhibiting coke formation [12].

Detailed catalytic performance of Ni-AZ-X (X 0, 0.1, 0.2,0.3, and 0.4) catalysts in the steam reforming of ethanol at

500 C after a 900 min-reaction is summarized in Table 3.Complete conversion of ethanol was achieved in all the

catalysts under the conditions of high reaction temperature

and excess amount of steam. This is in good agreement with

the previous reports [6,36]. Selectivity for C2H4 over Ni-AZ-X

catalysts decreased in the order of Ni-AZ-0 (9.8%) > Ni-AZ-0.1

(1.6%)> Ni-AZ-0.2 (0%) Ni-AZ-0.3 (0%) Ni-AZ-0.4 (0%). Thiscan be explained by the fact that large amount of acid sites in

the supports promoted the dehydration of ethanol to ethylene

via one-step elimination mechanism [37]. The amount of

carbon deposition decreased in the order of Ni-AZ-

0 (28.8%) > Ni-AZ-0.1 (21.5%) > Ni-AZ-0.2 (3.5%) > Ni-AZ-0.3

(2.6%)> Ni-AZ-0.4 (1.3%) (Table 3). This might be because ZrO2in the Ni-AZ-X catalysts hindered both ethanol dehydration

reaction (Eq. (2)) and ethylene formation reaction related to

coking [38]. It was also observed that CO/CO2 molar ratio

increased with increasing Zr/Al molar ratio (with decreasing

acidity) in the order of Ni-AZ-0 (0.045)

shift reaction [39].

ethanol: a review. Energy Fuels 2005;19:2098e106.[2] Llorca J, Piscina PR, Sales J, Homs N. Direct production of

0.

atio

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 31382In the aspect of textural properties and reducibility, it was

revealed that the addition of ZrO2 decreased surface area and

increased reducibility of the catalysts. As these two factors

affect the catalytic activity in an opposite way, it is believed

that high surface area of Ni-AZ-0 and Ni-AZ-0.1 compensates

low reducibility, and high reducibility of Ni-AZ-0.3 and Ni-AZ-

0.4 compensates low surface area [40]. Consequently, it is

concluded that catalytic performance of Ni-AZ-X catalysts

was well correlated with acidity regardless of the effect of

textural properties and reducibility. Among the catalysts

tested, Ni-AZ-0.2 catalyst with an intermediate acidity

showed the maximum hydrogen yield.

4. Conclusions

A series of mesoporous NieAl2O3eZrO2 (Ni-AZ-X ) xerogeldehydration reaction occurred on the acid-rich surface,

resulting in a severe coke formation [37,38]. On the other

hand, relatively less acidic catalysts (Ni-AZ-0.3 and Ni-AZ-0.4)

showed low hydrogen yield due to low reactivity for water-gas

Hyd

roge

n yi

eld

(%)

Zr/Al molar ratio

Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

a

0 0.1 0.2 0.3 0.490

100

110

120

130

140

Fig. 8 e Hydrogen yields over Ni-AZ-X (X[ 0, 0.1, 0.2, 0.3, and

a 900 min-reaction, plotted as a function of (a) Zr/Al molar rcatalysts with different Zr/Al molar ratio (X) were prepared by

a single-step epoxide-driven solegel method. The prepared

catalysts were applied to the hydrogen production by steam

reforming of ethanol. The effect of Zr/Al molar ratio on the

catalytic performance was investigated. It was found that all

the Ni-AZ-X catalysts exhibited a mesoporous structure.

Surface area of the catalysts decreased with increasing Zr/Al

molar ratio due to the incorporation of ZrO2. Reducibility of Ni-

AZ-X catalysts increased with increasing Zr/Al molar ratio

through the formation of NiOeAl2O3eZrO2 composite struc-

ture. Acidity of Ni-AZ-X catalysts monotonically decreased

with increasing Zr/Al molar ratio. In the hydrogen production

by steam reforming of ethanol, hydrogen yields over Ni-AZ-X

catalysts showed a volcano-shaped curve with respect to Zr/

Almolar ratio and acidity of the catalysts. Among the catalysts

tested, Ni-AZ-0.2 catalyst with an intermediate acidity

exhibited the best catalytic performance. It is concluded that

an optimal Zr/Al molar ratio of NieAl2O3eZrO2 xerogel cata-

lyst was required for maximum production of hydrogen by

steam reforming of ethanol.hydrogen from ethanolic aqueous solution over oxidecatalysts. Chem Commun 2001:641e2.

[3] Schrope M. Which way to energy utopia? Nature 2001;414:682e4.

[4] Youn MH, Seo JG, Lee H, Bang Y, Chung JS, Song IK. Hydrogenproduction by auto-thermal reforming of ethanol over nickelcatalysts supported on metal oxides: effect of supportacidity. Appl Catal B Environ 2010;98:57e64.Acknowledgments

This work was supported by the Global Frontier R&D Program

on Center for Multiscale Energy System funded by the

National Research Foundation under the Ministry of Educa-

tion, Science and Technology, Korea (20110031575).

r e f e r e n c e s

[1] Haryanto A, Fernando S, Murali N, Adhikari S. Current statusof hydrogen production techniques by steam reforming of

70 80 90 100 110 120 130 14090

100

110

120

130

140

Ni-AZ-0.4

Ni-AZ-0.3

Ni-AZ-0.2

Ni-AZ-0.1

Ni-AZ-0

Hyd

roge

n yi

eld

(%)

Acidity of catalyst ( mol-NH3/g)

b

4) catalysts in the steam reforming of ethanol obtained after

and (b) acidity of the catalyst.[5] Balat M, Balat H. Recent trends in global production andutilization of bio-ethanol fuel. Appl Energy 2009;86:2273e82.

[6] Ni M, Leung DYC, Leung MKH. A review on reforming bio-ethanol for hydrogen production. Int J Hydrogen Energy 2007;32:3238e47.

[7] Rabenstein G, Hacker V. Hydrogen for fuel cells from ethanolby steam-reforming, partial-oxidation and combined auto-thermal reforming: a thermodynamic analysis. J PowerSources 2008;185:1293e304.

[8] Rass-Hansen J, Christensen CH, Sehested J, Helveg S,Rostrup-Nielsen JR, Dahl S. Renewable hydrogen: carbonformation on Ni and Ru catalysts during ethanol steam-reforming. Green Chem 2007;9:1016e21.

[9] Vizcano AJ, Lindo M, Carrero A, Calles JA. Hydrogenproduction by steam reforming of ethanol using Ni catalystsbased on ternary mixed oxides prepared by coprecipitation.Int J Hydrogen Energy 2012;37:1985e92.

[10] Wang W, Wang Y. Steam reforming of ethanol to hydrogenover nickel metal catalysts. Int J Hydrogen Energy 2010;34:1285e90.

[11] Rossetti I, Biffi C, Bianchi CL, Nichele V, Signoretto M,Menegazzo F, et al. Ni/SiO2 andNi/ZrO2 catalysts for the steamreforming of ethanol. Appl Catal B Environ 2012;117:384e96.

[12] Sanchez-Sanchez MC, Navarro RM, Fierro JLG. Ethanol steamreforming over Ni/MxOyeAl2O3 (M Ce, La, Zr and Mg)catalysts: Influence of support on the hydrogen production.Int J Hydrogen Energy 2007;32:1462e71.

[13] Fatsikostas AN, Kondarides DI, Verykios XE. Steam reformingof biomass-derived ethanol for the production of hydrogen forfuel cell applications. Chem Commun 2001:851e2.

[14] Hegarty MES, OConnor AM, Ross JRH. Syngas productionfrom natural gas using ZrO2-supported metals. Catal Today1998;42:225e32.

[15] Therdthianwong S, Therdthianwong A, Siangchin C,Yongprapat S. Synthesis gas production from dry reformingof methane over Ni/Al2O3 stabilized by ZrO2. Int J HydrogenEnergy 2008;33:991e9.

[26] Breval E, Deng Z, Chiou S, Pantano CG. Sol-gel prepared Ni-alumina composite materials. J Mater Sci 1992;27:1464e8.

[27] Kim P, Joo JB, Kim H, Kim W, Kim Y, Song IK, et al.Preparation of mesoporous Ni-alumina catalyst by one-stepsol-gel method: control of textural properties and catalyticapplication to the hydrodechlorination of o-dichlorobenzene.Catal Lett 2005;104:184e9.

[28] Seo JG, Youn MH, Park S, Lee J, Lee SH, Lee H, et al. Hydrogenproduction by steam reforming of LNG over Ni/Al2O3eZrO2catalysts: effect of ZrO2 and preparation method ofAl2O3eZrO2. Korean J Chem Eng 2008;25:95e8.

[29] Youn MH, Seo JG, Cho KM, Jung JC, Kim H, La KW, et al. Effectof support on hydrogen production by auto-thermalreforming of ethanol over supported nickel catalysts. Korean

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 en e r g y 3 8 ( 2 0 1 3 ) 1 3 7 6e1 3 8 3 1383[16] Seo JG, Youn MH, Cho KM, Park S, Lee SH, Lee J, et al. Effect ofAl2O3eZrO2 xerogel support on hydrogen production bysteam reforming of LNG over Ni/Al2O3eZrO2 catalyst. KoreanJ Chem Eng 2008;25:41e5.

[17] Tang S, Ji L, Lin J, Zeng HC, Tan KL, Li K. CO2 reforming ofmethane to synthesis gas over solegel-made Ni/geAl2O3catalysts from organometallic precursors. J Catal 2000;194:424e30.

[18] Gao YP, Sisk CN, Hope-Weeks LJ. A sol-gel route to synthesizemonolithic zinc oxide aerogels. ChemMater 2007;19:6007e11.

[19] Bang Y, Lee J, Han SJ, Seo JG, Youn MH, Song JH, et al.Hydrogen production by steam reforming of liquefied naturalgas (LNG) over mesoporous nickel-alumina xerogel catalystsprepared by a single-step carbon-templating sol-gel method.Int J Hydrogen Energy 2012;37:11208e17.

[20] Kim P, Kim Y, Kim C, Kim H, Park Y, Lee JH, et al. Synthesisand characterization of mesoporous alumina as a catalystsupport for hydrodechlorination of 1,2-dichloropropane:effect of catalyst preparation method. Catal Lett 2003;89:185e92.

[21] Faro AC, Souza KR, Camorim VLDL, Cardoso MB. Zirconia-alumina mixing in alumina-supported zirconia prepared byimpregnation with solutions of zirconium acetylacetonate.Phys Chem Chem Phys 2003;5:1932e40.

[22] Dominguez JM, Hernandez JL, Sandoval G. Surface andcatalytic properties of Al2O3eZrO2 solid solutions preparedby sol-gel methods. Appl Catal A Gen 2000;197:119e30.

[23] Hawa T, Zachariah MR. Coalescence kinetics of unequalsized nanoparticles. J Aerosol Sci 2006;37:1e15.

[24] Bang Y, Seo JG, Youn MH, Song IK. Hydrogen production bysteam reforming of liquefied natural gas (LNG) overmesoporous NieAl2O3 aerogel catalyst prepared by a single-step epoxide-driven sol-gel method. Int J Hydrogen Energy2012;37:1436e43.

[25] Youn MH, Seo JG, Jung JC, Park S, Song IK. Hydrogenproduction by auto-thermal reforming of ethanol over nickelcatalyst supported on mesoporous yttria-stabilized zirconia.Int J Hydrogen Energy 2009;34:5390e7.J Chem Eng 2008;25:236e8.[30] Pines H, Haag WO. Alumina: catalyst and support. I.

Alumina, its intrinsic acidity and catalystic activity. J AmChem Soc 1960;82:2471e83.

[31] Krokidis X, Raybaud P, Gobichon AE, Rebours B, Euzen P,Toulhoat H. Theoretical study of the dehydration process ofboehmite to g-Alumina. J Phys Chem B 2001;105:5121e30.

[32] Kim Y, Kim C, Kim P, Yi J. Effect of preparation conditions onthe phase transformation of mesoporous alumina. J Non-cryst Solids 2005;351:550e6.

[33] Onfroy T, Li W-C, Schuth F, Knozinger H. Surface chemistryof carbon-templated mesoporous aluminas. Phys ChemChem Phys 2009;11:3671e9.

[34] Kwak JH, Hu JZ, Kim DH, Szanyi J, Peden CHF. Penta-coordinated Al3 ions as preferential nucleation sites for BaOon g-Al2O3: an ultra-high-magnetic field

27Al MAS NMRstudy. J Catal 2007;251:189e94.

[35] Cutrufello MG, Ferino I, Monaci R, Rombi E, Solinas V. Acid-base properties of zirconium, cerium and lanthanum oxidesby calorimetric and catalytic investigation. Top Catal 2002;19:225e40.

[36] Wang W, Wang YQ. Thermodynamic analysis of steamreforming of ethanol for hydrogen generation. Int J EnergyRes 2008;32:1432e43.

[37] Mattos LV, Jacobs G, Davis BH, Noronha FB. Production ofhydrogen from ethanol: review of reaction mechanism andcatalyst deactivation. Chem Rev 2012;112:4094e123.

[38] Choong CKS, Huang L, Zhong Z, Lin J, Hong L, Chen L. Effectof calcium addition on catalytic ethanol steam reforming ofNi/Al2O3: II. Acidity/basicity, water adsorption and catalyticactivity. Appl Catal A Gen 2011;407:155e62.

[39] Djinovic P, Levec J, Pintar A. Effect of structural and acidity/basicity changes of CuOeCeO2 catalysts on their activity forwater-gas shift reaction. Catal Today 2008;138:222e7.

[40] Rangel MC, Berrocal GP, Maciel CG, Assaf JM. Comparisonbetween zirconia and alumina supports for nickel-basedcatalysts for hydrogen production. Adv Chem Res 2011;10:341e59.

Hydrogen production by steam reforming of ethanol over mesoporous NiAl2O3ZrO2 xerogel catalysts: Effect of Zr/Al molar ratio1. Introduction2. Experimental2.1. Preparation of mesoporous NiAl2O3ZrO2 xerogel catalysts2.2. Characterization2.3. Hydrogen production by steam reforming of ethanol

3. Results and discussion3.1. Physicochemical properties of Ni-AZ-X catalysts3.2. XRD and TPR results of Ni-AZ-X catalysts3.3. Crystalline structure and metal dispersion of reduced Ni-AZ-X catalysts3.4. Acidity of reduced Ni-AZ-X catalysts3.5. Hydrogen production by steam reforming of ethanol over Ni-AZ-X catalysts

4. ConclusionsAcknowledgmentsReferences