Bioresource Technology 123 (2012) 558–565

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Hydrogen production from steam reforming of acetic acid over Cu–Znsupported calcium aluminate

Pravakar Mohanty a,b,⇑, Madhumita Patel a, Kamal K. Pant a,1

a Department of Chemical Engineering, Indian Institute of Technology, New Delhi 110016, Indiab Department of Chemical Engineering, Catalysis and Chemical Reaction Engineering Laboratories, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5A9

h i g h l i g h t s

" 12CaO�7Al2O3 and Cu–Zn/Ca–Al novel catalysts were tested for steam reforming." Cu–Zn/Ca–Al bimetallic and without support catalyst had higher activity with its H2 yield compared for a self sacrificed catalysts." Detailed Catalyst characterization was done by BET, SEM-EDX, XRD and TEM techniques.

a r t i c l e i n f o

Article history:Received 10 January 2012Received in revised form 21 May 2012Accepted 5 July 2012Available online 20 July 2012

Keywords:H2 productionAcetic acidSteam reformingCa–Al/Cu–Zn catalyst

0960-8524/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.biortech.2012.07.019

⇑ Corresponding author at: Department of Chemicaof Technology, New Delhi 110016, India. Tel.:1126581120.

E-mail addresses: pravakar.mohanty@gmail.comchemical.iitd.ac.in (K.K. Pant).

1 Equally contributed to this work.

a b s t r a c t

Hydrogen can be produced by catalytic steam reforming (CSR) of biomass-derived oil. Typically bio oilcontains 12–14% acetic acid; therefore, this acid was chosen as model compound for reforming of biooilwith the help of a Cu–Zn/Ca–Al catalyst for high yield of H2 with low CH4 and CO content. Calcium alu-minate support was prepared by solid–solid reaction at 1350 �C. X-ray diffraction indicates 12CaO�7Al2O3

as major, CaAl4O7 and Ca5Al6O14 as minor phases. Cu and Zn were loaded onto the support bywet-impregnation at 10 and 1 wt.%, respectively. The catalysts were characterized by Brunauer–Emmett–Teller (BET), Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy TEMand the surface area for both support and Cu–Zn were 10.5 and 5.8 m2/g, respectively. CSR was carriedout in a tubular fixed bed reactor (I.D. = 19 mm) at temperatures between 600 and 800 �C with 3-gloadings and (H2O/acetic acid) wt. ratio of 9:1. Significantly high (80%) yield of hydrogen was obtainedover Cu–Zn/Ca–Al catalyst, as incorporation of Zn enhanced the H2 yield by reducing deactivation ofthe catalyst. The coke formation on the support (Ca-12/Al-7) surface was negligible due to the presenceof excess oxygen in the 12CaO�7Al2O3 phase.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Hydrogen is an ideal energy carrier since it possesses the high-est energy content per unit weight (about 120.7 MJ/kg) of anyknown fuel (Hou et al., 2009). It burns cleanly, without generatingany environmental pollutants and its only by-product is water. H2

is one of the most important energy carriers and is widely used infuel cells, ammonia production, oil refineries, and methanol pro-duction. Over 75% of H2 at room temperature is ortho-hydrogen.At very low temperatures, ortho-hydrogen becomes unstable andchanges to more stable para-hydrogen, releasing heat in the

ll rights reserved.

l Engineering, Indian Institute+91 1126596177; fax: +91

(P. Mohanty), kkpant@

process. This heat can complicate low temperature hydrogen pro-cesses, particularly liquefaction, for which steam reforming at hightemperature is a suitable platform (Mohanty et al., 2011a,b; Patelet al., 2012). Currently, H2 is produced commercially by CSR of nat-ural gas and oil-derived naphtha, partial oxidation of heavy oils,gasification of coal as well as electrolysis of water (Pefia et al.,1996; Takanabe et al., 2004, 2006a,b). Bio-oil produced from bio-mass has been proposed as an alternative source to produce hydro-gen because it does not contribute to a net increase in atmosphericCO2. Bio-oil is a dark brown organic liquid, which exhibits a com-plex composition with more than 200 different compounds,including acids, alcohols, glycerol, aldehydes, ketones as well aslignin derived oligomers emulsified in the presence of water(Dou et al., 2009; Gao et al., 2009; Slinn et al., 2008). Bio-oil is dif-ficult to reform directly, but steam reforming of the main compo-nents in bio-oil such as acetic acid (12–14% by wt.) is feasible(Marquevich et al., 1999; Xun and Gongxuan, 2007; Aristides andXenophon, 2008; Xun and Gongxuan, 2010).

P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565 559

Generally, steam reforming of bio-oil is simplified to the follow-ing reactions. (Chornet and Czernik, 2002; Cortright et al., 2002;Takanabe et al., 2004; Basagiannis and Verykios, 2007):

CnHmOk þ ðn� kÞH2O ¼ nCOþ ðnþm=2� kÞH;2� ðSteam reforming reactionÞ ð1Þ

The above reaction is followed by the water–gas shift reaction:

nCOþ nH2O ¼ nCO2 þ nH2 ð2Þ

Therefore, the overall process can be represented as follows:

CnHmOk þ ð2n� kÞH2O ¼ nCO2 þ ð2nþm=2� kÞH2 ð3Þ

Takanabe et al. (2004) studied the CSR of acetic acid overPt/ZrO2 catalyst and Basagiannis and Verykios (2006), Xun andGongxuan (2007) investigated the influence of Ni and noble metalswith supports (Al2O3, La2O3/Al2O3, MgO/Al2O3) on CSR of aceticacid. Takanabe et al. (2004, 2006a), and Vagia and Lemonidou(2007) studied the thermal decomposition and its reforming ten-dency in the absence and presence of steam for different metalssupported by calcium aluminate. Coke deposition on the catalystsurface is a serious problem and the rate of coke formation is high-er in case of nickel catalyst in comparison to noble metal catalyst,but noble metal catalysts are very expensive. The catalyst prepara-tion and selection of support are the two crucial factors for higherhydrogen yield, low methane and carbon monoxide formation, andlow deactivation of the active sites of the catalyst. The presentstudy focused on H2 generation in presence of Cu supported by cal-cium aluminate with a small amount of Zn. These metals are cheap,easily available, but still showed a high yield of H2 with negligiblecoke formation (Ekaterini and Angeliki, 2007; Xun and Gongxuan,2007; Dave and Pant 2011).

Reforming of acetic acid for H2 production can be summarizedas follows:

(a) Steam reforming of acetic acid reaction as stated below:

Table 1Comparproduct

Cata

Ni–A28%33%Ni–ANi–CNi/ACo/A5%PtMg0.

Mg�CMg�CMgAPt/Zr5% N0.5-R

CH3COOH !H2O2COþ 2H2; DH0

298 K ¼ 213:4 kJ=mol ð4Þ

(b) Followed by water gas shift reaction:

COþH2O !CatalystCO2 þH2; DH0

298 K ¼ �41 kJ=mol ð5Þ

The overall steam reforming reaction of acetic acid is:

CH3COOHþ 2H2O !Catalyst2CO2 þ 4H2; DH0

298 K ¼ 131:4 kJ=mol

ð6Þ

ative analysis of different catalyst, (S/C) ratio, temperature and space velocity usion.

lyst Model compound (MC) T (oC) % H2 yield (g H2/g

l Acetic acid 550–750 66 (0.125)Ni–Al n-Butanol 550–750 76 (0.291)Ni–Al Acetol 550–750 (0.169)l Acetic acid 450–750 (0.136)o Acetic acid 350–550 96.3l2O3 Acetic acid 300–600 38l2O3 Acetic acid 300–600 30/Al2O3 VFAs 300–600 7026Ca Acetol 650 63.5 (0.115)a0 Acetic acid 650 65.61 (0.057)a0 Acetol 650 67.02 (0.166)l0.26 Aq. fraction pyro-liquid 650 67 (0.132)O2 Acetic acid 450–600 87i–CaO2–Al2O3 Acetic acid 550–750 70h% CaO2–Al2O3 Acetone 550–750 80

The maximum stoichiometric yield of H2 from acetic acid reform-ing is 2 mole to mole of carbon. However some undesired reactionstake place inside the reactor. Hence there is large possibilities ofdecomposition of acetic acid to lower molecular weight oxygenates,lighter hydrocarbons (CH4, C2H4) and coke formation (Dave and Pant2011; Patel et al., 2012). The comparative analysis of different cata-lyst used in reforming process towards H2 production with aceticacid and other model compounds, varying different process param-eters like steam to carbon (S/C) ratio, temperature and space velocityby different group of researcher are summarized in Table 1

C2H4O2 ! CxHyOz þ Carbonþ GasesðH2;CO;CO2;CH4;CxHxþ2Þð7Þ

Some amount of coke can also be formed via Boudouardreaction

2CO$ CO2 þ C; DH0298 K ¼ �172:4 kJ=mol ð8Þ

The desired gaseous product like H2 is lost through methanationreaction and reverse water gas shift reaction (Eqs. (9) and (10))

COþ 3H2 $ CH4 þH2O; DH0298 K ¼ �206:1 kJ=mol ð9Þ

CO2 þ 4H2 $ CH4 þ 2H2O; DH0298 K ¼ �165:1 kJ=mol ð10Þ

2CH3COOH$ CH3COCH3 þCO2 þH2O; DH0298 K ¼�165:1 kJ=mol

ð11Þ

2CH3COOH$ ðCH3Þ2COþ CO2 þH2O ð12Þ

Small amount of acetone may be formed by ketonization reac-tion (Eq. (12)) from acetic acid decomposition. Different definitionfor % H2 yield, % acetic acid (AAc.) conversion and % selectivity ofgaseous product distribution are used to describe the catalytic re-sults of CSR using the oxygenates. The definitions are given below(Eqs. (13)–(16)) (Basagiannis and Verykios, 2006; Dave and Pant2011; Patel et al., 2012).

% H2 yield ¼moles of H2 in gaseous product4�moles of acetic acid fed

� 100 ð13Þ

% selectivity of H2¼moles of H2 in gaseous product

C atoms in gaseous phase� 1

RR� 100

ð14Þ

where RR is H2/CO2 reforming ratio and RR = 4/2

% selectivity of ‘i’ ¼ C atom in species ‘i’C atoms in gaseous phase

� 100 ð15Þ

ed for steam reforming of acetic acid and other model compounds of biooil for H2

MC) % MC S/C Opt. cond. ReferenceConv. Ratio (T = oC; h�1)

99.5 5.58 650, >28000 Bimbela et al. (2007)90.01 5.5–14.7 650, >30000 Bimbela et al. (2009)97 5.5–14.7 750, <57000 Bimbela et al. (2009)94.3 5.58 650, >13000 Galdámez et al. (2005)100 2.5–7.5 400, >7.5 Hu and Lu (2007)44 2.5–10 400, >5.0 Hu and Lu (2010)40 2.5–10 400, >5.0 Hu and Lu (2010)99.1 13.7 600, >25000 Jeong et al. (2011)99.8 5.58 650, 6800 Medrano et al. (2009)45.5 5.58 650, 29565 Medrano et al. (2009)101 5.58 650, 8247 Medrano et al. (2009)85 7.64 650, 5411 Medrano et al. (2011)100 5 650, 40000 Takanabe et al. (2006a)100 3 750, 30000 Vagia and Lemonidou (2008)100 3 750, 30000 Vagia and Lemonidou (2008)

560 P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565

% AAc conversion ¼moles of AAc in�moles of AAc outmoles of AAc in

� 100

ð16Þ

Table 2Summary of the catalyst nomenclature and its composition.

Catalyst Composition (%) Method of preparation

Ca Al O Cu Zn Ni

Ca12/Al7

48.5 4.06 46.8 – – – Solid–solid reaction

CuZn 32.6 8.05 48.5 9.86 1.0 – Wet impregnationmethod

2. Methods

2.1. Catalyst preparation

The carrier consisting of xCaO�yc-Al2O3 at a molar ratio x/y = 12/7 was prepared by solid state reaction between CaCO3 and c-Al2O3

at 1300 �C for 20 h. A solution containing the precursor of Cu(NO3)2�3H2O and Zn (NO3)2�6H2O (Merck, Germany) was mixedwith the support particles and dried in a vacuum rotary evaporatorat 90 ± 3 �C for 4 h, where 10%Cu�1%Zn on 12CaO�7Al2O3 was tar-geted. After drying, the sample was kept inside the dryer over nightat 120 ± 5 �C. The catalyst was calcined under air flow at 550 ± 5 �C.The nominal metal composition of the final catalyst was 10 wt.%for Cu and 1 wt.% for Zn. The catalyst support was designated asCa-12/Al-7, and the catalyst as Cu–Zn–Ca12/Al-7.

2.2. Catalyst characterization

The surface areas and pore volumes of the fresh and used cata-lyst were analyzed using the N2 adsorption–desorption isotherm.The BET surface area, total pore volume and pore size distributionof the catalyst were determined from nitrogen adsorption/desorp-tion isotherms measured at �196 �C using a Micromeritics ASAP2010 apparatus. Prior to gas adsorption measurements, the catalystwas degassed at 150 ± 1 �C under vacuum for 6 h. The total porevolume was calculated at a relative pressure of ffi0.989. X-ray dif-fraction (XRD) patterns were obtained using a Philips powder dif-fractometer, with Cu-Ka radiation. The morphology of the catalystsample was investigated using SEM-EDX with a ZEISS EVO SeriesScanning Electron Microscope EVO 50. Scanning Electron Micro-scope EVO 50 having the resolution of 2.0 nm at 30 kV (SE withLaB6 option) with magnification ranges up to 1,000,000� and atan acceleration voltage of 10 kV. TEM was performed using aPHILIPS CM12 microscope operated at an accelerating voltage of100 kV (Patel et al., 2012; Mohanty et al., 2011a).

2.3. Catalytic testing

Testing was performed at atmospheric pressure in a laboratoryunit equipped with a mass flow-controlled system and a fixed-bedtubular reactor designed for an offline gas chromatograph. The exitgas composition from the reactor was determined by a gas chro-matograph (Nucon 5765) equipped with a TCD mode (Nucon Engi-neers, Delhi, India). A peristaltic pump was used for the feeding ofthe mixture of acetic acid and water to the reactor through a pre-heater arrangement. The reactor was made of a cylindrical inconel600 cup of 19 mm diameter and 30 mm length, perforated on itsbottom with orifices of 0.5 mm diameter to enable the flow ofgases. It was assembled in a SS 316 reactor of 780 mmlength � 19 mm ID with a thermo well and feed inlet. Thefurnace consisted of three heating zones and proportional–integral–derivative (PID) controllers with an electronic temperatureindicator installed at each zone for accurate reactor temperaturecontrol. To monitor the temperature periodically a thermocouplewas installed in the thermo well of the reactor assembly. Hot gasesexiting the reactor were cooled to condense the vapour mixture toproduce unconverted liquid reactants and the uncondensed gasesanalyzed using Nucon 5700 gas chromatograph operated on TCDmode. The experiments were performed in the temperature rangeof 600–800 �C at atmospheric pressure for a concentration of10 wt.% of acetic acid in distilled water. Three gram of catalyst

was placed inside the catalyst bed for every run and the flow rateof reactant was maintained at 0.38 mL/min. Nitrogen was used asthe carrier gas at a flow rate of 25 mL/min using a mass flow meter.The catalysts were reduced at 750 ± 5 �C under 50% H2/He flowfor 3 h followed by reforming experiments, as temperature pro-grammed reduction analysis provides the reduction temperatureprofile to be followed (Dave and Pant 2011; Patel et al., 2012).

3. Result and discussion

3.1. Characterization of the support and the catalyst

The details of the catalysts composition as determined by EDXpoint analysis are presented in Table 2. The porosity of the catalystis revealed by their prominent surface area. The N2 adsorption/desorption isotherms are summarized in Table 3. SupplementalFig. 1(A and B) shows the XRD patterns of the support and thecatalyst, after calcination. The XRD peaks at different 2h value forthe support are shown in Table 4. 12CaO�7Al2O3 is the major phasewith some minor phases i.e. CaAl4O7 and Ca5Al6O14 and size of thesupport particle ranges from 8 to 23 nm. Ca5Al6O14, CaAl4O7,12CaO�7Al2O3 and CaCO3 phases were detected based on their dif-ferent 2h values with conformation from JCPDS-720767, relatedto monoclinic and end centered lattice revealed from JCPDS-461475. The lattice parameters (f.c.c) were calculated at (4 4 4)

using the equation: af :c:c ¼ffiffiffiffi2kp

Sinh; where k = is the wave length of radi-ation and h = Bragg’s angle. Calcium and alumina were also found inthe form of Ca5Al6O14 phases, among which some were cubic, body-centered cubic (b.c.c), and some were end-centered with an ortho-rhombic or monoclinic nature (Tien-Thao et al., 2007; Mohantyet al., 2011a; Patel et al., 2012). The surface area of the fresh catalystand the support are provided in Table 3. Supplemental Fig. 2(A andB) shows SEM photographs of fresh catalyst and support aftercalcinations. The metals were uniformly distributed on the catalystsurface. The largest particles and the most heterogeneous size dis-tribution were observed for the calcium support, whereas the alu-mina has a lower, but homogeneous particle size were varied inthe range of�30–50 nm. These findings are in good agreement withother literatures (Tien-Thao et al., 2007; Patel et al., 2012; Vagia andLemonidou, 2008). The boundary between metals and support forCuZnNi and Ca12Al7 became indistinct in comparison with thesupport only. With assumption that all the metals were presenton the surface are spherical in nature; corresponding to sizes anddistribution of weight can be given by TEM, where metal dispersion(DM) was calculated using the following equation given by(Tien-Thao et al., 2007; Mohanty et al., 2011a; Patel et al., 2012).

DM ¼ 6� 107 VMAM

1dðnmÞ

� �; where VM is the bulk atomic volume of

the metals (cm3), AM is the area of an atom (cm2), and d is the metalparticle size in nm. For Zn and Cu (as Cu and Zn having nearer atom-ic weight), their VM is very close to one another, so averageVM ¼ ð1:10� 10�23 cm3 was taken; and their AM ¼ 1� 10�15 cm2

was considered (Tien-Thao et al., 2007). The catalyst with a lowerCu–Zn content had a larger metal dispersion on the face-centeredcubic (f.c.c) crystals, indicating more metal sites are available onthe surface.

Table 3BET surface areas and pore volume of the catalysts.

Catalyst BET S.A.(m2/g)

Pore volume(cm3/g)

Adsorption porewidth (A�)

Ca12/Al7 10.5 ± 1 0.012 ± 0.002 28 ± 4CuZn 5.8 ± 1 0.017 ± 0.002 45 ± 4

Table 4Calculation of the phase and particle size of the support utilising the XRD analysis.

2h Crystal size in (nm) Phases

32.303 16.5350 Ca5Al6O14

34.249 8.3096 CaAl4O7

37.401 20.9595 CaAl4O7

44.341 21.4380 12CaO�7Al2O3

54.129 17.8355 12CaO�7Al2O3

57.502 22.6447 12CaO�7Al2O3

64.284 18.7573 12CaO�7Al2O3

P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565 561

To obtain monophasic calcium aluminates and kinetically fa-vored conditions with a higher (Ca/Al) ratio from the supportmaterial, high temperature calcination is desirable. Whereas after

Fig. 4. (A) Effect of temperature on% H2 yield (pressure, 1 atm; acetic acid concentrationhydrogen yield (Pressure, 1 atm; acetic acid concentration of 10 wt.% at a flow rate of 0

Cu–Zn metal impregnation it desirable to calcine at two stages like550 ± 5 and 850 ± 5 �C, which is most suitable to enhance spill overactivity as well as to cast distinct CuO and ZnO sites. 12CaO�7Al2O3

exhibits as a major phase with additional minor phases like CaA-l4O7 and Ca5Al6O14. Both these phenomena are well supported byother authors (Galdamez et al., 2005; Vagia and Lemonidou,2008, and Patel et al., 2012) for this sort of catalyst to reduce athigher temperature like 600 ± 10 �C, providing two stage of postcalcination for 4 h. The XRD analysis for calcined catalyst providesmany prominently stable and distinct phases. The Cu–Zn catalystsupported on Ca–Al exhibit low surface area as the intermediatehydration with partial dissolution of Ca–Al phases occurred by heattreatment at 900 �C and subsequent formation of the dehydratedphases with addition of Cu–Zn, followed by reduction at 750 �Cmakes the catalyst more stable and prominent. From TPR analysisthe reduction peaks centered at 245 and 295 �C respectively break-ing the CuO to Cu2+ ion whereas the other peak rounded at 490 �Cassociated with the reduction of ZnO to Zn0, which bonded to thesupport (Ca-12/Al-7). Similar TPR results have been presented forthe reduction of CaxAlyOz for promotion of transition metal likeCu and Zn (Patel et al., 2012). For Ca-12/Al-7, a broad peak was ob-tained at slightly higher temperature of 705 �C as it favors both

of 10 wt.% at a flow rate of 0.38 mL/min; 3 g of catalyst) and (B) effect of run time on.38 mL/min; 3 g of catalyst).

Fig. 5. (A) Effect of run time on product distribution (Pressure,1 atm; acetic acid concentration, 10 wt.% at a flow rate of 0.38 mL/min; 3 g of catalyst; Temperature,600 �C) and(B) Effect of run time on product distribution (pressure, 1 atm; acetic acid concentration, 10 wt.% at a flow rate of 0.38 mL/min; 3 g of catalyst; temperature, 650 �C).

562 P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565

WGS and reforming reaction in the reaction zones (650–750 �C) togenerate stable phases.

Supplemental Fig. 3(A and B) shows the TEM image of the freshcatalyst and support, respectively. The support particles are spher-ical in nature and diameter of the support particle varies from 15 to25 nm. The catalyst particle diameter is varied from 28 to 34 nm. Inthe used catalyst most of the metal particles were found above thecarbon filament, although there were some metal particles encap-sulated within the filaments which possibly helped in the simulta-neous water shift gas reaction and carbon formation with an autoconsumption process (Segner et al., 1984). Thus, small metal parti-cles (P10 nm) generate the carbon filaments with a short diameterand a thermodynamically unfavorable growth due to higher (S/C)ratio. Moreover, carbon gasification towards CO2 (a structure-sensitive reaction) is favored in smaller particles (Vagia andLemonidou 2007; Zhang et al., 2008). During CSR of the acetic acid

mixture, some active sites of Cu and Zn on the catalyst surfaceare deactivated by carbon deposition (Chen et al., 2011). Whensubsequent layer of carbon deposition starts to form on differentsites of active metals, the hydrocarbon (like ketons) can beadsorbed on site of catalyst surface and decomposed to carbonatoms giving rise to carbonaceous atomic groups which might stayon catalyst surface for longer duration and blocks the activity ofmetal (Cu) atoms. At that emergency the Zn particle is forced tobreak the inter-cohesive force between catalyst active sites andcarbide, as the carbide starts off to grow.

3.2. Steam reforming of the acetic acid

3.2.1. Effect of temperature on reforming acetic acid reformingThermodynamic analysis suggested that % yield of H2 increased

with temperature and attained the maximum of 96% at 680 �C,

Fig. 6. (A) Effect of run time on product distribution (pressure, 1 atm; acetic acid concentration, 10 wt.% at a flow rate of 0.38 mL/min; 3 g of catalyst; Temperature, 750 �C)and (B) Effect of run time on product distribution (pressure, 1 atm; acetic acid concentration, 10 wt.% at a flow rate of 0.38 mL/min; 3 g of catalyst; Temperature, 800 �C).

P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565 563

1 atm pressure and 10 wt.% acetic acid concentration. Whereas fur-ther increase in temperature favored methanation reaction whichincreased the% H2 yield also devalue the CO selectivity due to therapidity of reverse water gas shift reaction, but CH4 selectivity de-creased with temperature as steam reforming of methanation isinconsistent at high temperature. Thermodynamic findings indi-cates that at 680 �C and 10 wt.% acetic acid concentrations, maxi-mization of H2 can be realized having the selectivity of CO andCH4 at minimum as 9% and 0.1% respectively. Sequentially highertemperature is unsuitable for selectivity of CO2. Therefore basedon thermodynamics result it is found that at 680 �C the percentof H2 yield was maximum. As temperature increased beyond680 �C, percent of H2 yield decreased approximately to 5–6% fromthe maximum value (96%) up to 1000 �C. During reforming processthe carbon species available on catalyst surface are not favouredthermodynamically in the presence of water splitting oxygen. With

higher S/C ratio, higher H2 is always desirable; however the rela-tive energy consumption and unit size increases accordingly. Tem-perature is the one of the dominating parameter for CSR. The yieldof hydrogen was measured as the function of the steam reformingtemperature ranging from 600 to 800 �C for a concentration of 10wt.% acetic acid in distilled water with weight hourly space veloc-ity of 79.2 h�1. Fig. 4(A) shows the variation of the H2 yield withdifferent temperature for the CSR condition. Percentage of H2 yieldis about 23% at 600 �C and it increases to 80% as the temperaturereaches to 800 �C which is according to Le-Chatelier’s principle.Two undesirable products i.e. CO and CH4 are formed during steamreforming via reverse water gas shift and methanation reactions atmoderate temperature (600–650 �C). Some researchers foundmethane formation at this temperature range suits the processby decarboxylation reaction (Patel et al., 2012; Vagia andLemonidou, 2008). As temperature increases above 650 �C the

Fig. 7. Effect of space time (Wt. of catalyst/FA0) on acetic acid conversion at different temperature range (600–800 �C; pressure, 1 atm; acetic acid concentration, 10 wt.%; 3 gof catalyst).

564 P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565

reaction shifts towards ketonization which prevail along withreforming and water gas shift reactions. Sometimes high yield ofketone inevitably spoils the H2 selectivity, so optimization of tem-perature, steam to carbon ratio and space velocity enhances thecarbon conversion and hydrogen yield to desired assessment. Astemperature increased from 650 to 750 �C, acetic acid conversionincreased from 46% to 80%. But at 800 �C, acetic acid conversion re-mains steady�81%. It can be observed that there is a weak effect oftemperature on acetic acid conversion beyond 800 �C.

3.2.2. Effect of run time on hydrogen yield and product distributionFig. 4(A and B) show the moles of H2 obtained at different run

times and temperatures. H2 production increased with tempera-ture. Figs. 5 and 6(A and B) show the product distribution in thegaseous phase with run time at different temperatures. Methaneformation was significant at all temperature, but decreased withan increase in temperature possibly due to steam reforming andless mass transfer limitation. At lower temperatures, a higher rateof carbon monoxide formation took place due to improper watergas shift reaction. CO concentrations were almost negligible athigher temperatures. CO2 decreased and H2 production increasedwith increasing temperature. At 800 �C, maximum hydrogen pro-duction was obtained with negligible amounts of CO and CH4 after7 h. Fig. 7 shows the variation of acetic acid conversion with W/FA0(weight of catalyst to molar flow rate of acetic acid) at four differ-ent temperatures. Maximum acetic acid conversion was obtainedabove 750 �C at a lower acetic acid flow rate (high residence time).As the flow rate increased from 0.38 to 1.3 mL/min at 750 �C, aceticacid conversion decreased to 32%. At 650 �C, acetic acid conversiondecreased from 71% to 36% as the acetic acid flow rate increasedfrom 0.38 to 1.3 mL/min.

4. Conclusions

The representative model compound of bio-oil, acetic acid, wasreformed effectively in a fixed bed in the presence of a Cu–Zn–Ca12/Al-7 catalyst. Effect of process variables such as temper-ature, space time and acetic acid concentration in feed overproduct distribution and acetic acid conversion was investigated.

The yield of H2 was maximum at 800 �C (80%) with negligibleamount of CH4 and CO levels. The effect of coking was negligible.The physical characterization in terms of XRD, SEM, EDX andTEM confirms the presence of (15–34 nm) nano-sized metal-support interaction and its uniform distribution on support(Ca12/Al-7) makes the catalyst more efficient. Phases likeCa5Al6O14, CaAl4O7 and 12CaO�7Al2O3 on support helps the CSRwith negligible carbon formation. Higher temperature, higher spacetime and lower acetic acid feed concentration favor the increase ofhydrogen yield and acetic acid conversion. Consecutively to makethe process feasible attention towards the aqueous phase of biooilseparation with its CSR is need of the hour.

Acknowledgements

The authors wish to thank the financial support provided byCentre for Fire, Explosive and Environment Safety (CFEES) divisionof Defence Research Development Organization (DRDO), Govern-ment of India, for providing financial support for this Project.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biortech.2012.07.019.

References

Aristides, C.B., Xenophon, E.V., 2008. Influence of the carrier on steam reforming ofacetic acid over Ru-based catalysts. Appl. Catal. B: Environ. 82, 77–88.

Basagiannis, A.C., Verykios, X.E., 2007. Catalytic steam reforming of acetic acid forhydrogen production. Int. J. Hydrogen Energy 32, 3343–3355.

Basagiannis, A.C., Verykios, X.E., 2006. Reforming reactions of acetic acid on nickelcatalysts over a wide temperature range. Appl. Catal. A: General 308, 182–193.

Bimbela, F., Oliva, M., Ruiz, J., Garcıa, L., Arauzo, J., 2007. Hydrogen production bycatalytic steam reforming of acetic acid, a model compound of biomasspyrolysis liquids. J. Anal. Appl. Pyrolysis 79, 112–120.

Bimbela, F., Oliva, M., Ruiz, J., Garcıa, L., Arauzo, J., 2009. Catalytic steam reformingof model compounds of biomass pyrolysis liquids in fixed bed: acetol and n-butanol. J. Anal. Appl. Pyrolysis 85, 204–213.

Chen, T., Wu, C., Liu, R., 2011. Steam reforming of bio-oil from rice husks fastpyrolysis for hydrogen production. Bioresour. Technol. 102, 9236–9240.

P. Mohanty et al. / Bioresource Technology 123 (2012) 558–565 565

Chornet, E., Czernik, S., 2002. Renewable fuels: harnessing hydrogen. Nature 418,928–929.

Cortright, R.D., Davda, R.R., Dumesic, J.A., 2002. Hydrogen from catalytic reformingof biomass-derived hydrocarbons in liquid water. Nature 418, 964–967.

Dave, C.D., Pant, K.K., 2011. Renewable hydrogen generation by steam reforming ofglycerol over zirconia promoted ceria supported catalyst. Renew. Energy 36,3195–3202.

Dou, B., Dupont, V., Rickett, G., Blakeman, N., Williams, P.T., Chen, H., Ding, Y.,Ghadiri, M., 2009. Hydrogen production by sorption-enhanced steam reformingof glycerol. Bioresour. Technol. 100, 3540–3547.

Ekaterini, C.V., Angeliki, A.L., 2007. Thermodynamic analysis of hydrogenproduction via steam reforming of selected components of aqueous bio-oilfraction. Int. J. Hydrogen Energy 32, 212–223.

Galdamez, J.R., Garcıa, L., Bilbao, R., 2005. Hydrogen production by steam reformingof bio-oil using coprecipitated Ni–Al catalysts. Acetic acid as a modelcompound. Energy Fuels 19, 1133–1142.

Gao, N., Li, A., Quan, C., 2009. A novel reforming method for hydrogen productionfrom biomass steam gasification. Bioresour. Technol. 100, 4271–4277.

Hou, T., Yuan, L., Ye, T., Gong, L., Tu, J., Yamamoto, M., Torimoto, Y., Li, Q., 2009.Hydrogen production by low-temperature reforming of organic compounds inbio-oil over a CNT-promoting Ni catalyst. Int. J. Hydrogen Energy 34, 9095–9107.

Hu, X., Lu, G., 2007. Investigation of steam reforming of acetic acid to hydrogen overNi–Co metal catalyst. J. Mol. Catal. A: Chem. 261, 43–48.

Hu, X., Lu, G., 2010. Comparative study of alumina-supported transition metalcatalysts for hydrogen generation by steam reforming of acetic acid. Appl. Catal.B: Environ. 99, 289–297.

Jeong, C.M., Park, G.W., Choi, J., Kang, J.W., Kim, S.M., Lee, W., Woo, S.I., Chang, H.N.,2011. Steam reforming of volatile fatty acids (VFAs) over supported Pt/Al2O3

catalysts. Int. J. Hydrogen Energy 36, 7505–7515.Marquevich, M., Czernik, S., Chornet, E., Montane, D., 1999. Hydrogen from biomass:

steam reforming of model compounds of fast-pyrolysis oil. Energy Fuels 13,1160–1166.

Medrano, J.A., Oliva, M., Ruiz, J., Garcıa, L., Arauzo, J., 2009. Catalytic steamreforming of model compounds of biomass pyrolysis liquids in fluidized bedreactor with modified Ni/Al catalysts. J. Anal. Appl. Pyrolysis 85, 214–225.

Medrano, J.A., Oliva, M., Ruiz, J., Garcıa, L., Arauzo, J., 2011. Hydrogen from aqueousfraction of biomass pyrolysis liquids by catalytic steam reforming in fluidizedbed. Energy 36, 2215–2224.

Mohanty, P., Pant, K.K., Parikh, J., Sharma, D.K., 2011a. Liquid fuel production fromsyngas using bifunctional CuO–CoO–Cr2O3 catalyst mixed with MFI zeolite. FuelProc. Tech. 92, 600–608.

Mohanty, P., Pant, K.K., Naik, S.N., Das, L.M., Vasudevan, P., 2011b. Fuel productionfrom biomass: Indian perspective for pyrolysis oil. J. Sci. Ind. Res. 70 (08), 668–674.

Patel, M., Pant, K.K., Mohanty, P., 2012. Renewable hydrogen generation by steamreforming of acetic acid over Cu–Zn–Ni supported calcium aluminate catalysts.In: Dalai, A.K. (Eds.), Nanocatalysis for fuels and Chemicals. ACS SymposiumSeries, Washington, DC 20036, 1092, pp. 111–137, doi:http://dx.doi.org/10.1021/bk-2012-1092.ch009.

Pefia, M.A., Gomez, J.P., Fierro, J.L.G., 1996. New catalytic routes for syngas andhydrogen production. Appl. Catal. A: General 144, 7–57.

Segner, J., Campbell, C.T., Doyen, G., Ertl, G., 1984. Catalytic oxidation of CO on Pt(111): the influence of surface defects and composition on the reactiondynamics. Surf. Sci. 138, 505–523.

Slinn, M., Kendall, K., Mallon, C., Andrews, J., 2008. Steam reforming of biodieselby-product to make renewable hydrogen. Bioresour. Technol. 99, 5851–5858.

Takanabe, K., Aika, K., Seshan, K., Lefferts, L., 2004. Sustainable hydrogen frombiooil-steam reforming of acetic acid as a model oxygenate. J. Catal. 227, 101–108.

Takanabe, K., Aika, K., Seshan, K., Lefferts, L., 2006a. Catalyst deactivation duringsteam reforming of acetic acid over Pt/ZrO2. Chem. Eng. J. 120, 133–137.

Takanabe, K., Aika, K., Seshan, K., Lefferts, L., 2006b. Steam reforming of acetic acidas a biomass derived oxygenate: bifunctional pathway for hydrogen formationover Pt/ZrO2 catalysts. J. Catal. 243, 169–263.

Tien-Thao, N., Zahedi-Niaki, M.H., Alamdari, H., Kaliaguine, S., 2007. Effect of alkaliadditives over nanocrystalline Co–Cu based perovskites as catalysts for higher-alcohol synthesis. J. Catal. 245, 348–357.

Vagia, E.C., Lemonidou, A.A., 2007. Thermodynamic analysis of hydrogen productionby steam reforming of bio-oil components. Int. J. Hydrogen Energy 32, 212–223.

Vagia, E.C., Lemonidou, A.A., 2008. Thermodynamic analysis of hydrogen productionvia autothermal steam reforming of selected components of aqueous bio-oilfraction. Int. J. Hydrogen Energy 33, 2489–2500.

Xun, H., Gongxuan, L., 2007. Investigation of steam reforming of acetic acid tohydrogen over Ni–Co metal catalyst. J. Mol. Catal. A 261, 43–48.

Xun, H., Gongxuan, L., 2010. Comparative study of alumina-supported transitionmetal catalysts for hydrogen generation by steam reforming of acetic acid. Appl.Catal. B: Environ. 99, 289–297.

Zhang, J.G., Wang, H., Dalai, A.K., 2008. Effects of metal content on activity andstability of Ni–Co bimetallic catalysts for CO2 reforming of CH4. Appl. Catal. A:General 339, 121–129.