lable at ScienceDirect

Renewable Energy 74 (2015) 27e36

Contents lists avai

Renewable Energy

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

From wood wastes to hydrogen e Preparation and catalytic steamreforming of crude bio-ethanol obtained from fir wood

Monica Dan a, 1, Lacrimioara Senila b, 1, Marius Roman b, Maria Mihet a, Mihaela D. Lazar a, *

a National Institute for Research and Development of Isotopic and Molecular Technologies, 67-103 Donath Street, 400293 Cluj-Napoca, Romaniab INCDO-INOE 2000, Research Institute for Analytical Instrumentation, 65 Donath Street, 400293 Cluj-Napoca, Romania

a r t i c l e i n f o

Article history:Received 8 April 2014Accepted 25 July 2014Available online

Keywords:Bio-ethanol from wood wastesSimultaneous saccharification andfermentation (SSF)Crude bio-ethanol steam reformingLa2O3 promoted Al2O3 supported Ni catalystCeO2 promoted Al2O3 supported Ni catalyst

* Corresponding author. Tel.: þ40 264 584037.E-mail address: diana.lazar@itim-cj.ro (M.D. Lazar

1 Authors with equal contribution.

http://dx.doi.org/10.1016/j.renene.2014.07.0500960-1481/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Fir wood wastes were used to produce crude bio-ethanol by two methods: simultaneous saccharificationand fermentation (SSF) and acid hydrolysis followed by the fermentation of the acid hydrolyzate. Themain components of crude bio-ethanol are ethanol and acetic acid. In addition, low concentrations of awide range of alcohols, acids, esters, ethers and aldehydes are also present. Ethanol concentration ishigher in the SSF process than in the acid hydrolysis: 43.69 g/L compared to 37.53 g/L, respectively.Opposite to ethanol concentration, the acetic acid concentration is higher in the acid hydrolysis process:16.36 g/L compared to 10.24 g/L, respectively. The crude bio-ethanol was used to produce hydrogen bycatalytic steam reforming. The tested catalysts were the common Ni/Al2O3 and two rare earth oxidespromoted Ni catalysts: Ni/La2O3eAl2O3 and Ni/CeO2eAl2O3 prepared by successive wet impregnation.The characterization techniques revealed that the addition of rare earth oxides improves the Nidispersion and the reducibility of the promoted catalysts. The best feed rate which assures the optimalratio between conversion and catalyst deactivation is 0.8 mL/min bio-ethanol. The addition of extra oxide(La2O3 and CeO2) to the support improves the ethanol conversion especially at 250 �C, but no significanteffect on the acetic acid conversion was observed. At 250 �C the ethanol conversion is almost 90% for Ni/La2O3eAl2O3 and Ni/CeO2eAl2O3, but the acetic acid conversion is below 30% for all catalysts. At 350 �Cboth ethanol and acetic acid present maximum conversion. At this temperature the best hydrogenproduction is obtained for Ni/La2O3eAl2O3 due to better ethanol conversion and better selectivity forhydrogen formation. At 350 �C the promoted catalysts are stable for 4 h time on stream, different degreesof deactivation being obtained at lower temperatures.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Woody biomass is the most abundant biomass in the world. Inthe actual context of intensive quest for environmental friendlyenergy sources, thewoody biomass and especially the wood wastesresulted from forest activities is of special interest. A superior wayto add value to these wastes is their transformation in the secondgeneration bio-ethanol, as a way for further transformation inhydrogen and/or syngas. In the actual technologies, the hydrogengeneration from lignocelluloses biomass is a two stage process: inthe first step raw biomass is converted to hydrogen substrate (i.e.ethanol) and in the second step this is catalytically transformed intohydrogen [1]. Hydrogen is often called “the fuel of the future”

).

although it is not exactly a fuel but an energy vector. Hydrogen isforeseen by many researchers as a very suitable and environmen-tally friendly energy carrier [2] due to its possibility to be used infuel cells for stationary and mobile power generation with practi-cally no hazardous emissions [3]. The so called “hydrogen econ-omy” seems to be a viable alternative in the future to replace theactual fossil fuel based one [4].

Woody biomass has a complex structure having as main com-ponents cellulose, hemicellulose and lignin. Generally, the processof the bio-ethanol production includes four steps: pretreatment,hydrolysis, fermentation and distillation [5]. Autohydrolysis pre-treatment with compressed hot water is an environmental friendlyprocess in which the selective depolymerization of hemicellulosetakes place [6]. Hemicellulose is easy recovered in the liquid frac-tion as mono- and oligosugars, and cellulose and lignin are sepa-rated in the solid fraction (cellulose is more susceptible tohydrolysis). The remaining lignin fraction has a higher affinity forcellulolytic enzymes components, thus resulting in reduced

M. Dan et al. / Renewable Energy 74 (2015) 27e3628

hydrolysis efficiency. Therefore, the intercalation of an intermedi-ate step of delignification may be an effective method to promoteenzymatic digestibility of wood [7]. The hydrolysis of cellulose isprimarily performed either through a chemical or an enzymaticroute; in this hydrolysis step cellulose is broken into monomers.The acid hydrolysis is much faster than the enzymatic hydrolysis,this representing an economic advantage. Potential acids that canbe used in acid hydrolysis are sulfuric and hydrochloric acids, inconcentrate or dilute form [8]. Previous studies showed that diluteacids significantly improve the cellulose hydrolysis, but requirehigh temperature and neutralization of pH [9]. The resulted sugarsare fermented to ethanol. Enzymatic hydrolysis and fermentationcan be performed together in the so-called simultaneous sacchar-ification and fermentation (SSF) or separately, in two steps: hy-drolysis and fermentation (SHF).

The usual route for ethanol transformation into hydrogen is thesteam reforming (ESR). Being a catalytic process, the importance ofthe catalytic materials is well established. The most used catalystsare noble (Rh, Pt, Pd, Au, Ag) [10,11] and non-noble (Ni, Co, Cu)[12,13] metals supported on various oxides (Al2O3, ZrO2, CeO2).Noble metals have the advantage of high catalytic activity andstability, their main drawback being the high production costs. Nibased catalysts are very active for reformation reactions and oftenused for ESR, their main drawback being the relatively rapiddeactivation due to carbon deposition. A series of approaches havebeen taken to overcome this problem: (i) the addition of smallquantities of noble metal to Ni, the resulting bimetallic catalystspresenting improved stability and selectivity for H2 formation [14];(ii) the modification of support by addition of other oxides (CaO,MgO, CeO2, La2O3, etc) [15e17].

This paper reports for the first time to our best knowledge acomprehensive study of crude bio-ethanol production from firwood wastes and its subsequent transformation in hydrogen byethanol steam reforming. Epron et al. reported the steam reformingof a model raw bio-ethanol mixture obtained by the addition of themost common impurities to a solution of ethanol; the compositionof crude bio-ethanol obtained in our work is very different fromthat reported by the group of Epron [18]. In previous papers, theterm of bio-ethanol usually defines the pure ethylic alcohol ob-tained by crop fermentation followed by distillation. In our paperthe term of crude bio-ethanol defines the liquid part of thefermentation broth, which contains all the fermentation products,not only ethanol.

2. Materials and methods

2.1. Bio-ethanol production

2.1.1. MaterialsThe fir (Albies alba) wood wastes were collected locally and used

as raw material. The dried material was stored in plastic bags atroom temperature. Yeast from Saccharomyces cerevisiae YSC2 andpeptone from animal tissue P5905 were purchased from Sigma-eAldrich (St. Louis, MO, USA). H2SO4 (98%), CH3eCOOH, NaOH,KH2PO4, MgSO4$7H2O, (NH4)2SO4 citric acid (C6H8O7) were

II

Autohydrolysis Fir wood Delignification

Acid hydrolysisin two steps

Fig. 1. Schematic representation of different treatments per

purchased from Merck (Darmstadt, Germany). NaClO2 (80%) waspurchased from Alfa Aesar GmbH & Co (Karlsruhe, Germany). Allchemicals were analytical reagent grade. Ultrapure water was ob-tained from a Milli-Q system (Millipore, Bedford, MA, USA).

The scheme for bio-ethanol production considered in this workis presented in Fig. 1.

2.1.2. Autohydrolysis pretreatmentThe pretreatment of wood was carried out in a steel pressure

Parr reactor with a Parr 4523 temperature controller (Parr In-struments, Illinois, USA). The raw material was loaded into a 1 Lreaction vessel and was supplemented with an appropriate amountof deionized water to obtain a final solid to liquid ratio of 1:7. Themixture was heated at 190 �C and 60 bar pressure for 10 minresidence time. At the end of the reaction, the reactor was cooledand the mixture was separated by filtration into the solid residueand the liquid phase. The pretreatment conditions were selectedbased on preliminary results, according to our previously publishedmethod [19].

2.1.3. Delignification of cellulosic materialsThe solid materials recovered after pretreatment were deligni-

fied using acid-chlorite method, according to Hallac et al. [20]. Theexperiments were carried out in 250 mL conical flasks using so-dium chlorite (0.6 g/g biomass), acetic acid (0.6 mL/g biomass) anddeionized water (84 mL/g biomass). The reactions were carried outat 70 �C for 2 h (repeated for 3 times). The suspension was filtered,washed with water and acetone until neutral pH.

The delignified materials were hydrolyzed with dilute acid fol-lowed by fermentation (route II in Fig. 1), or subjected directly tosimultaneous saccharification and fermentation (SSF) (route I inFig. 1) in order to obtain the crude bioethanol.

2.1.4. Dilute acid hydrolysis followed by fermentation of acidhydrolyzate

Acid hydrolysis experiments of the solid residue recovered af-ter delignification of wood were carried out in two steps: (i) in500 mL conical flasks containing 2% H2SO4 at 100 �C for 60 min,and (ii) in 15% H2SO4 at 100 �C, for 90 min. After the hydrolysisprocess, the liquid fractions were separated from unreacted solidsand then fermented to bioethanol. The amount of acid used forcellulosic fraction hydrolysis was 38 g of H2SO4 to 2 g of dry solid.The fermentation of acid hydrolyzate was performed in 1 L flaskscontaining 700 mL hydrolyzate obtained above, 70 mL nutrientsolution (medium content per liter: 5 g of yeast extract, 20 g ofKH2PO4, 10 g of MgSO4.7H2O, 20 g of (NH4)2SO4 and 1 g ofMgSO4.7H2O), and 70 mL inoculum solution (medium content perliter: 10 g of yeast extract, 20 g of peptone and 50 g of glucose).The initial pH was adjusted to 5.0 by adding NaOH (2 M). All theexperiments were carried out at 30 �C for 72 h. The nutrient so-lution and the inoculum solution were activated and grown at30 �C for 24 h in an incubator. During the fermentation process,samples were collected to determine the content of ethanol andacetic acid.

I Bioethanol in fermentation

broths

SSF process

Fermentation Bioethanol in fermentation

broths

formed for bio-ethanol production from wood wastes.

M. Dan et al. / Renewable Energy 74 (2015) 27e36 29

2.1.5. Simultaneous saccharification and fermentation (SSF)Simultaneous saccharification and fermentation (SSF) processes

were carried out in a 500 mL Erlenmeyer flask equipped with abubble trap to maintain anaerobic conditions and placed in anorbital shaker (250 rpm). The fermentation flask contained 20 mLnutrient solution, 20 mL inoculum solution, mixed with 8% (w/v)solids loadings in citrate buffer (0.05 M) and 0.7 mL/g glucan ofAccellerase 1500 (Genencor, Rochester, NY, USA) at pH 4.8. All SSFexperiments were performed at 45 �C for 96 h.

2.2. Hydrogen production

2.2.1. Catalyst preparationThe Ni/Al2O3 catalyst (denoted as Ni/Al) was prepared by wet

impregnation method. The g-Al2O3 (Merck, Germany) wasimpregnated with the proper amount of aqueous solution obtainedfrom commercially available Ni(NO3)2$6H2O (Alpha Aesar, Ger-many). The impregnated catalyst samples were dried at roomtemperature overnight, calcined in Ar at 650 �C for 4 h and reducedin H2 at 650 �C for another 4 h. The mixed catalysts (Ni/La2O3-eAl2O3 with 6 wt% La2O3 denoted as Ni/LaeAl and Ni/CeO2eAl2O3

with 6 wt% CeO2 denoted as Ni/CeeAl), were prepared in twosuccessive stages. The first stage of the preparation consisted inalumina impregnation with aqueous solution of La(NO3)3$6H2O(Merck, Germany) or Ce(NO3)3$6H2O (Merck, Germany), followedby calcination at 650 �C for 4 h. The second stage consisted infurther impregnation of the freshly modified supports withaqueous solution of Ni(NO3)2$6H2O following the proceduredescribed for Ni/Al. The target concentration of Ni in all catalyst was10 wt%.

2.2.2. Catalysts characterizationN2 adsorptionedesorption isotherms were recorded in order to

determine the total surface area (St), pore volume (Vp) and poreradius (rp) using the Sorptomatic 1990 (Thermo Electron Corpo-ration USA). X-ray power diffraction (XRD) patterns were recordedusing a Bruker D8 advanced diffractometer with CuKa1 radiation,operation voltage 40 kV and current 40 mA. The average Nicrystallite size was assessed using the Scherrer equation. Thereduction profile of the catalyst precursors were investigated bytemperature programmed reduction (TPR) on TPDRO 1100 in-strument from Thermo Scientific, USA. The calcined catalystsamples were placed inside a quartz tube reactor which was thenheated from room temperature to 1100 �C under a gaseousmixture of 10% of H2/Ar with a flow rate of 20 mL/min. The rate ofH2 consumption was measured by the thermal conductivity de-tector (TCD). Temperature programmed oxidation (TPO) was per-formed on the same TPDRO 1100 equipment in order toinvestigate the nature and quantity of carbon deposited on thecatalyst surface during bio-ethanol steam reforming. The catalystsamples were cleaned under He flow (20 mL/min) for 30 min at30 �C, and then heated up to 1100 �C by a temperature rate of10 �C/min, under 20 mL/min flow of 7% O2/He.

2.2.3. Hydrogen production by bio-ethanol catalytic steamreforming

Crude bio-ethanol steam reforming reaction was studied in aMicroactivity Reference Unit (PID Eng & Tech, Spain) in a stainlesssteel reactor (i.d. ¼ 9 mm) under atmospheric pressure. Argon, thecarrier gas, is controlled by a mass flow controller and the liquidreagents are measured and controlled by an LC 6A pump (Shi-madzu, Japan). The liquid mixture passes first through a pre-heaterat 120 �C and then into the reactor. The reaction products collectedat the reactor outlet are a mixture of liquid and gaseous fractions.The liquid fraction is removed through condensation and from time

to time samples are collected and analyzed in the analytical con-ditions presented bellow. The gaseous fraction is dried by passing itthrough silica and then analyzed online using a GC (Sferocarbcolumn 2 m, 80e100 mesh with Ar as a carrier gas at 120 �C)equipped with a TCD.

Prior to experiments, the catalyst samples (1 g) were dilutedwith alumina (1 g) and activated by reduction in H2 for 3 h at350 �C. The catalytic activity experiments were performed between150 and 350 �C, the total flow for Ar was set to 133mL/min, and theliquid feed (crude bio-ethanol) was tested in a range of 0.1e1 mL/min. The ethanol and acetic acid conversions were calculated basedon inlet and outlet concentrations.

XEtOH ¼ CinEtOH � Cout

EtOH

CinEtOH

� 100 (1)

XAcH ¼ CinAcH � Cout

AcH

CinAcH

� 100 (2)

where CinEtOH,C

inAcH and Cout

EtOH, CoutAcH represent the inlet and the outlet

concentrations of ethanol and acetic acid, respectively. Hydrogenproduction is defined as volume of H2 produced in 1 min on 1 g ofcatalyst.

2.3. Analytical methods

Ethanol concentrationwas evaluated using an Agilent 7890A gaschromatograph (Agilent Technologies, Palo Alto, CA, USA) equippedwith a CTC Combi PAL autosampler (CTC Analytics AG, Zwingen,Switzerland) and a flame ionization detector (FID). The methodused for bio-ethanol analysis is full evaporation headspace gaschromatographic method (FEeHSeGC). The advantage of thismethod is that it uses small volumes of samples (10 ml): the liquidsample is introduced into the headspace sample vial and it is thenheated up to a temperature of 105 �C, when a full evaporation canbe achieved within 3 min. The concentration of acetic acid wasdetermined by ion chromatography (IC), using a Metrohm 761Compact IC (Metrohm Ltd., Switzerland) system with suppressedconductivity detection and a Metrosep A Supp 5e100 analyticalcolumn. For the IC system, a mixture of Na2CO3 (0.0027 mol/L) andNaHCO3 (0.001 mol/L) was employed as eluent, with a flow rate of1.2 mL/min.

3. Results and discussion

3.1. Crude bio-ethanol preparation and analysis

The fir wood contains approximately 70% carbohydrates, beingthus a suitable feedstock for bio-ethanol production. The compo-sition of wood samples used in this study was analyzed, and resultsshowed that it is composed of cellulose (46.0%), hemicellulose(24.0%), lignin (28.4%), ash (0.3%) and extractible (1.3%). Bio-ethanolwas prepared from fir wood by two processes using two types ofhydrolysis: (i) acid hydrolysis followed by fermentation of acidhydrolyzate and (ii) simultaneous saccharification and fermenta-tion combined in SSF process. Prior to these processes thedelignification of autohydrolyzed woodwas applied for eliminationof lignin from cellulosic materials. The SSF method applied to firwood wastes is a novelty of this paper and the combination ofautohydrolysis and SSF in order to produce crude bio-ethanol fromfir wood, to our best knowledge, was not reported before.

The material balance for the crude bio-ethanol obtained fromwood wastes using two different hydrolysis procedures is shown inFig. 2.

Water

Hemicellulose – containing liquors

Autohydrolysis

Celulose: 46 g Hemicellulose: 24 g Lignin: 28.3 g Others: 1.7 g

NaClO2/acetic acid

SSF method

Celulose: 36.8 g Hemicellulose: 0.02 g Lignin:1.3 g Others: 0.0 g

Delignification

Celulose: 43.0 g Hemicellulose: 1.25 g Lignin: 28.0 g Others: 0.03 g

Lignin – containing liquors Lignin: 26.7 g

Ethanol: 4.37 g Acetic ac. 1.024g

Waste wood

Acid hydrolysis and fermentation

Ethanol: 3.76 g Acetic ac. 1.63g

Fig. 2. The material balance for ethanol preparation from fir wood.

Table 1The composition of fermentation medium obtained by SSF and acid hydrolysis.Ethanol and acetic acid are the main components of crude bio-ethanol.

Nr. crt. Compound SSF method Acid hydrolysismethod

1 Dimethyl ether 0.008 g/L 0.007 g/L2 Methanol 0.30 g/L 0.28 g/L3 Ethanol 43.69 g/L 37.53 g/L4 n-propanol 0.40 g/L 0.37 g/L5 2-methyl-1-propanol 0.16 g/L 0.15 g/L6 n-butanol 0.16 g/L 0.19 g/L7 3-methyl-1-butanol 0.19 g/L 0.18 g/L8 2-methyl-1-butanol 0.13 g/L 0.10 g/L9 Pentanol 0.19 g/L 0.18 g/L10 Acetic acid 10.24 g/L 16.36 g/L11 2-methyl-1-propanol 0.03 g/L 0.02 g/L12 2,5-dimethylfuran 0.02 g/L 0.02 g/L13 3-methylbutyl acetate 0.01 g/L 0.01 g/L14 cyclohexanol 0.12 g/L 0.12 g/L15 Propanal 15.65 mg/L 12.64 mg/L16 Methyl propanoate 0.74 mg/L 0.52 mg/L17 Ethyl hexanoate 321.56 mg/L 240.82 mg/L18 Ethyl octanoate 1.75 mg/L 1.15 mg/L19 Propanal 15.63 mg/L 13.19 mg/L20 Propanoic acid 625.32 mg/L 423.62 mg/L21 2-methyl-propanal 86.32 mg/L 73.79 mg/L22 Isobutyric acid 10.65 mg/L 8.32 mg/L23 Butanoic acid 4.52 mg/L 2.16 mg/L24 2-methyl-1-butanal 526.30 mg/L 493.90 mg/L25 Pentanal 118.65 mg/L 111.22 mg/L26 Pentanoic acid 4.56 mg/L 2.88 mg/L27 Octanoic acid 0.89 mg/L 0.89 mg/L28 Decanoic acid 3.56 mg/L 2.95 mg/L

M. Dan et al. / Renewable Energy 74 (2015) 27e3630

Autohydrolysis pre-treatment was applied for the removal ofhemicellulose in liquid fraction and recovery of cellulose and ligninin solid fraction. The yield of the solid fraction recovered afterautohydrolysis was 75.0%. The content of cellulose (57.8%) andlignin (37.8%) in the autohydrolyzed wood confirmed that bothcomponents were recovered in solid fraction, only a small amountof hemicellulose (3.3%) being identified in this fraction. In order toreduce the lignin content, delignification with sodium chlorite wasapplied in the present work. The solid yield after delignificationwas43.0%, lignin being almost totally removed.

Acid hydrolysis and SSF method were used for hydrolysis andfermentation of delignified wood to crude bio-ethanol. During SSFmethod the Accelerase 1500 enzymes convert cellulose intomonomeric sugars. These are further fermented to bio-ethanol byyeast cells through glycolytic and fermentation enzymes. It wasreported that the factors that affect the production of bio-ethanolfrom palm oil, in SSF process are: solid loading, pH, yeast concen-tration, the nutrient content, etc. [21]. In order to increase the finalethanol concentration in the SSF process, different solid loadings,temperatures and time of the process were tested. Firstly, the solidloading was 2% (w/v) and the temperature was 35 �C. In this casethe final ethanol concentration was 33 g/L. The maximum ethanolconcentration (43.69 g/L) was obtained for SSF process with 8% (w/v) solid loadings at 45 �C after a period of 94 h. In the acid hydro-lysis method with H2SO4, a high concentration of bio-ethanol(37.53 g/L) was obtained using 2% of acid concentration for60min and 15% of acid hydrolysis for 90min. The temperature in allexperiments was 100 �C. The acid hydrolysis in two stages was usedfor a low degradation of cellulose.

Approximately the same final bio-ethanol concentrations wereobtained in all parallel experiments. In the SSF process, the con-centration of ethanol is slightly increased compared to acid hy-drolysis, while the acetic acid concentration is higher for acidhydrolysis. From these results it can be deduced that ethanol pro-duction is influenced by the hydrolysis method performed onautohydrolyzed wood.

The unpurified fermentation broth obtained from wood wastesby the two described methods is a dark mixture of many differentcompounds containing, besides the liquid fraction, a solid fractioncomposed by nutrients, inoculum solutions and remains of rawmaterial. Some purification is needed before being suitable to beused for steam reforming. For this purpose some methods wereinvestigated and the final purification scheme included: (i) filtra-tion for obtaining a clear solution and (ii) heating the sample fol-lowed by condensation of all liquid products. The clear solutionobtained after filtration still contains dissolved salts and yeastwhich remain as a solid residue after the second step of purifica-tion. The separation of the liquid part recovers all the organiccomponents plus water. The liquid solution obtained after

purification, denominated as crude bio-ethanol, was analyzed ac-cording to the procedure described in the Experimental Section.

Two major compounds were identified in crude bio-ethanol:ethanol and acetic acid (Table 1). In addition, a wide range of al-cohols, acids, esters, ethers and aldehydes were also identified. Thepresence of all these components in crude bio-ethanol wasconfirmed by GC-FID analysis. Their concentration is much lowerthan those corresponding to ethanol and acetic acid and varies withthe hydrolysis method. The acetic acid content is important in allsamples obtained by each preparation method. Its high content isdue to the presence of oxygen during fermentation.

The yeast S. cerevisiae was able to effectively utilize glucose toproduce a significant amount of ethanol, but also a wide spectrumof other biological metabolites. These properties are partially basedon the presence of a highly efficient alcohol dehydrogenase (ADH1)(for oxidizing acetaldehyde to ethanol). The liquid solution pro-duced by acid hydrolysis and by SSF process contain sufficientamount of carbon source, nitrogen, minerals and nutrients to alsoproduce different chemicals. The accumulation of acetyl CoA (as a

0 200 400 600 800 1000

R=0.9997Ni/Ce-Al

Inte

nsity

(a.u

)

Temperature (0C)

R=0.9990Ni/La-Al

Ni/Al R=0.9999

Fig. 3. TPR profiles and their deconvolution for the alumina supported catalysts pre-cursors calcined at 650 �C.

M. Dan et al. / Renewable Energy 74 (2015) 27e36 31

result of limited activity of Krebs cycle enzymes) may explain theaccumulation and release of acetate esters (methyl propanoate,ethyl hexanoate, ethyl octanoate, etc.) during fermentation. Inaddition, the formation of other compounds (higher alcohols: n-propanol, 2-methyl-1-propanol, n-butanol, 3-methyl-1-butanol, 2-methyl-1-butanol, pentanol, 2-methyl-1-propanol, etc.) is due tothe presence of nitrogen sources in fermentation medium. Thisproperty is partially dependent on the fermentation temperature,and is reflected in yeast growth potential [22].

In case of the acid hydrolysis the concentration of acetic acid ishigher (16.36 g/L), representing more than 40% of the ethanolconcentration. The highest ethanol concentration of 43.69 g/L wasobtained in the SSF method, while by acid hydrolysis the concen-tration of ethanol was 37.53 g/L. These values demonstrate a goodyeast performance during the SSF process. The ethanol concen-trations obtained in this work are similar or even higher than thosereported in the literature for a wide range of lignocellulosic feedstocks (Table 2).

The presence of significant concentrations of acetic acid and lowconcentrations of other oxygenated molecules in crude bio-ethanolprepared in this work is not a drawback of the preparation method,as the aim of this study is to produce hydrogen and the reformationof all hydrogen containing molecules is foreseen.

3.2. Catalytic hydrogen production from crude bio-ethanol

The purified crude bio-ethanol was used without any otherchanges for the production of hydrogen by catalytic steamreforming. For this study a common 10 wt% Ni/Al2O3 catalyst ob-tained by impregnation and two promoted alumina supported Nicatalysts were used. Similar catalysts prepared by the samemethods were used by our group for studies on methane steamreforming, being fully characterized by: N2 adsorption-desorption,H2 chemisorption, Transmission Electron Microscopy (TEM), X-rayDiffraction (XRD), Temperature Programmed Reduction (TPR) andInductively Coupled Plasma Mass Spectrometry (ICP-MS) [28,29].In the next section, the main catalysts characteristics are presentedin order to have a general view of the catalytic materials used forbio-ethanol steam reforming.

The addition of La2O3 and CeO2 to the alumina support did nothave a major influence on its porous structure. Total surface areahas very similar values for all catalysts, being 105 m2/g for Ni/Al,103 m2/g for Ni/CeeAl and 100 m2/g for Ni/LaeAl. The pore's meandiameter is 50 Å for all studied catalysts, which places these ma-terials in the mesoporous domain. These textural properties ensuregood access of reagents molecules to the catalyst surface and to thecatalytic active centers.

Temperature programmed reduction (TPR) method was used toevaluate the influence of the additional oxide to the reductionproperties of supported NiO. It is known that the interaction of Ninanoparticles with the support derives from the interaction of NiO

Table 2Ethanol concentration obtained in this work and results reported in related studies.

Raw material Pretreatment condition

Eucalyptus globulus Autohydrolysis at 200 �C and organosolv delignificationEucalyptus globulus Autohydrolysis at 195 �CEucalyptus globulus Autohydrolysis and delignification with 60% ethanolEucalyptus globulus Autohydrolysis at 230 �CEucalyptus grandis 1.2% H2SO4 hydrolysis at 180 �C for 10 minCorn cobs Acid hydrolysis (acid formic, 60 �C, 6 h) and delignificatCorn stover Steam explosion at 200 �C for 4 minFir wood Autohydrolysis at 190 �C and sodium chlorite delignificFir wood Autohydrolysis at 190 �C and sodium chlorite delignific

precursor with the catalysts support. The strength and type of thisinteraction can be assessed from the reduction temperature ofsupported NiO. It is important to notice that for all three catalyststhe TPR profile is a resultant of several reduction peaks, situated atdifferent temperatures (Fig. 3). The overlapping peaks are decon-voluted using Gaussian-type function. Applying the mathematicalfunction to the temperature profile, resulted in the identification ofthe reduction peaks depicted in Fig. 3 and summarized in Table 3.

It can be observed that the addition of the extra oxide to thealumina support leads to a decrease in temperature profile from762 �C in Ni/Al to 749 �C in Ni/CeeAl and 664 �C in Ni/LaeAl. Be-sides, the appearance of two peaks of low intensity situated atlower temperatures, 169 �C for Ni/LaeAl and 275 �C for Ni/CeeAlshould also be mentioned (Fig. 3). These peaks may be attributed tolow amounts of NiO bulk species with very low interactionwith thesupport which generate large Ni particles after reduction of NiO.According to [30] and [31] the peaks situated between 350 and600 �C correspond to NiO species verywell dispersed on the surface

Concentration ofethanol (g/L)

Reference

61.9 [6]51.0 [23]35.0 [24]26.7 [25]28.7 [26]

ion (15% ammonia, 60 �C, 12 h) 29.4 [27]40.6 [27]

ationeSSF 43.69 This workation e acid hydrolysis 37.53 This work

Table 3Deconvolution data of the TPR profiles of alumina supported catalysts.

Catalyst Tmax

(�C)Temperature

Fraction of peak area from total areaa

Ni/Al 762 e 462 �C 600 �C e 673 �C 735 �C 802 �Ce 0.4% 12.1% e 27.3% 45.8% 14.4%

Ni/LaeAl 664 169 �C 368 �C 549 �C 633 �C 669 �C 788 �C e

0.5% 3.2% 7.6% 32.3% 53.7% 2.7% e

Ni/CeeAl 749 275 �C 476 �C 600 �C 739 �C 804 �C3.6% 0.8% 16.6% 70.7% 8.3%

Tmax e the maximum of the temperature profile.a Fraction (%) ¼ peak area � 100/total area.

0.0 0.2 0.4 0.6 0.8 1.0 1.20

60

80

100

Etha

nol c

onve

rsio

n (%

)

Bio-ethanol feed rate (mL/min)

1h3h

Fig. 4. Conversion of ethanol vs. bio-ethanol feed rate; reaction conditions: 1 g cata-lyst, 350 �C, Ar flow 133 mL/min, bio-ethanol prepared by SSF method.

M. Dan et al. / Renewable Energy 74 (2015) 27e3632

of the catalyst and having low to medium interaction with thesupport. The peaks situated between 650 and 800 �C, whichrepresent the main reduction peaks of the studied catalysts, can beattributed to nickel oxide e aluminum oxide nonestoichiometricmixed species. Peakswith values close to the beginning of the rangecan be a nickelealuminum oxide mixture which contains a Ni-richphase with Al3þ ions. In contrast, the peaks that are situated in theupper part of the interval are formed from Al-rich phase with Ni2þ

ions. The interaction of these mixed oxides with the support ismedium to strong. The peaks situated at the highest temperatureresulted from deconvolution are of low intensity (Table 3) and mayvery well correspond to low amounts of NiAl2O4 spinel which hasvery strong interaction with the support [32].

X-ray Diffraction (XRD) was used to evaluate the crystallinity ofthe catalyst components and to estimate the size of Ni nano-particles. The presence of Ni is confirmed in XRD patterns (notshown here, see reference [28]) of all catalysts by its characteristicpeaks situated at 44.5� for Ni (111), 51.8� for Ni (2 0 0) and 76.5� forNi (2 2 0). The peak attributed to NiO and situated at 37.3� and 43.3�

are not present in the Ni/Al diffractograms, suggesting that all NiOwas reduced to metallic state. For Ni/LaeAl the XRD peaks attrib-uted to La-containing phases are absent. This is due to the fact thatlanthanum oxide is very well dispersed on the catalyst surface[28,33]. In Ni/CeeAl diffraction patterns the presence of wellcrystallized CeO2 is confirmed by the peaks situated at: 28.5� forCeO2 (111), 33� for CeO2 (2 0 0), 47.3� for CeO2 (2 2 0) and 56.7� forCeO2 (3 11). The Ni crystallite size was determined from Ni (2 0 0)diffraction peak using Scherrer equation. For Ni/Al the Ni crystallitesize was evaluated at 11.7 nm, for Ni/LaeAl at 8.0 nm and for Ni/CeeAl at 8.8 nm. It may be concluded that the additional oxideimproves the Ni dispersion by lowering the Ni crystallite size.

The conclusion of the characterization section is that the addi-tion of 6wt% of La2O3 or CeO2 to aluminamodifies the interaction ofNi with the catalytic support leading to a better catalyst reducibilityand a better Ni dispersion on the support.

The main reactions which are expected to take place in thecatalytic steam reforming of crude bio-ethanol are:

- Ethanol steam reforming: CH3eCH2eOHþ 3H2O/ 6H2 þ 2CO2(1)

- Acetic acid steam reforming: CH3eCOOHþ 2H2O/ 4H2þ 2CO2(2)

Besides these, a big number of secondary reactions also takeplace involving decomposition, dehydration, dehydrogenation,water-gas shift, reformation of secondary products etc [34].

The reformation of the crude bio-ethanol was performed in astainless steel fixed-bed tubular catalytic reactor (i.d. ¼ 0.9 mm).Two parameters were tested in order to optimize the reactionconditions for maximum hydrogen production: the reaction tem-perature and the feed rate of bio-ethanol.

The study was started with the influence of crude bio-ethanolfeed rate on the conversion. For this, Ni/Al catalyst was tested at350 �C. For this study, the crude bio-ethanol obtained by SSFmethod was used. The feed rate was varied starting from 0.1 to1 mL/min of bio-ethanol, with 133 mL/min Ar flow giving an GHSVof 7780 h�1 for 0.1 mL/min bio-ethanol, 27.050 h�1 for 0.6 mL/minethanol and 42.480 h�1 for 1 mL/min bio-ethanol feeding rate,respectively. Ethanol being the main component of the crude bio-ethanol only its conversion was monitored in this stage of thestudy (Fig. 4). For bio-ethanol feed rate lower than 0.6 mL/min theethanol conversion was 100%. By increasing further the feed flow,the ethanol conversion starts to decrease being approximately 94%for 1 mL/min bio-ethanol flow. This decrease of ethanol conversionmay not be a significant decrease, but more important is the rapiddeactivation of the Ni catalysts at this feeding rate. After only 3 htime on stream the catalyst retains only 90% of its original capacityto convert ethanol.

This response of the whole system at high feed rates can be dueto the “flooding” of the catalyst, in which the catalytic active sitesare not able to convert the whole feedstock, leading to a decrease ofconversion. Considering this influence of the feed rate upon theethanol conversion, the liquid bio-ethanol feed rate, used for allfurther steam reforming experiments was established to 0.8 mL/min.

The energetic efficiency of the hydrogen production processdepends on the reforming temperature; lower temperatures lead tolower energy consumptions. Especially for hydrogen used as energyvector it is important to have production costs at least comparable,if not lower, than the cost of the energy obtained using hydrogenthus produced. The promotion of Ni/Al with rare earth oxidesproved to have a beneficial effect on the efficiency of these catalystsfor ethanol [15,16] and acetic acid [35,36] steam reforming. In thenext section we report the effect of La2O3 and CeO2 addition tocommonNi/Al catalyst used for crude bio-ethanol steam reforming.There are a series of differences between the reformation of ethanolor acetic acid solutions and the reformation of crude bio-ethanolproduced according to the method described in this paper. Thesedifferences are: (i) the simultaneous presence of ethanol and sig-nificant amounts of acetic acid; (ii) the presence of lower amountsof a large number of other organic molecules (Table 1); (iii) thehigher watereorganic molecule ratios than those usually reportedfor reformation of ethanol and acetic acid solutions, due to the fact

Fig. 5. Ethanol conversion (a) and acetic acid conversion (b) versus temperature in crude bio-ethanol steam reforming; 1 g catalyst, 0.8 mL/min bio-ethanol flow; 133 mL/min Arflow, GHSV ¼ 34.770 h�1, bio-ethanol obtained by acid hydrolysis method.

M. Dan et al. / Renewable Energy 74 (2015) 27e36 33

that we used the crude bio-ethanol solution as resulted from thefermentation.

In order to study the influence of temperature on the catalyticsteam reforming of bio-ethanol Ni and promoted Ni catalystshave been tested at temperatures between 150 and 350 �C. Thegaseous products obtained in the steam reforming reaction for allthree catalysts were: H2, CH4, CO and CO2. No significant con-centration of other gaseous product was observed. The formationof very low concentrations of gases that are trapped in silicadrying element cannot be, however, excluded. The collectedliquid phase was also analyzed identifying un-reacted ethanoland acetic acid in more significant amounts, depending on theirconversion at the specific reaction temperature, and also lowconcentrations of acetaldehyde and acetone. The rest of the crudebio-ethanol components were not recovered at the reactor outlet.This can be due to their reformation along with ethanol andacetic acid.

The acetic acid conversion was also quantified besides theethanol conversion, given the important amount of this componentin crude bio-ethanol (Fig. 5).

At 150 �C the conversions of both ethanol and acetic acid arelow. With increasing temperature the conversion increasesapproaching 100% at 350 �C. A beneficial effect of the additional

Ni/Al Ni/La-Al Ni/Ce-Al0

3

6

15

18

H2

prod

uctio

n (m

L/m

in)/g

cat

alys

t

150 °C 250 °C 350 °C

Fig. 6. Hydrogen production by crude bio-ethanol steam reforming; 1 g catalyst,0.8 mL/min bio-ethanol flow, 133 mL/min Ar flow, bio-ethanol obtained by acid hy-drolysis method.

rare earth oxides is observed in ethanol conversion especially atlower temperatures. For Ni/LaeAl important enhancements inethanol conversion compared to Ni/Al is observed at 250 �C andeven at 150 �C. For ceria promoted catalysts a significant increaseof ethanol conversion is observed only at 250 �C. At this tem-perature the ethanol conversion on both promoted catalysts isvery good approaching 90%. The acetic acid conversion instead isvery low at 250 �C, being a major drawback for practical use ofthis reaction temperature in bio-ethanol steam reforming. For theconversion of acetic acid no significant beneficial effects of lan-thana or ceria addition could be observed in our experimentalconditions at any of the investigated temperature. It is importantto notice, however, the lower temperature compared to previousstudies at which acetic acid is fully converted [35,36]. This is dueto high water excess which shifts the reaction equilibrium to-wards the formation of products. The beneficial effect observedfor lanthana and ceria promoted catalysts is primarily due to thesecondary oxide influence on the catalyst surface; these oxidesbehave as a metal dispersant in the Ni-alumina system. Thedirect consequence is the increase of the number of catalyticactive sites which leads to the enhancement of reactants con-version [28,37].

As previously mentioned, the gaseous reaction products detec-ted in significant concentrations are H2, CH4, CO and CO2, with H2 asthe major product. With increasing temperature not only theconversion increases but also the ratio between the productschanges. At 350 �C, at maximum conversion, the outlet gascomposition, without carrier gas, for all the catalysts is given inTable 4.

The addition of the supplementary oxide to the support has animportant effect on the product distribution by: (i) increasing theselectivity for hydrogen production in the detriment of methaneformation, and (ii) favoring the transformation of CO in CO2 by fa-voring the water gaseshift reaction. At lower temperatures, bothcarbon oxides are present in the reaction products mixture for all

Table 4Composition of the gaseous reaction products at 350 �C and maximum ethanolconversion.

Catalysts Concentration (vol.%)

H2 CO CH4 CO2

Ni/Al 55.34 40.11 4.30 0.25Ni/LaeAl 64.49 0 3.21 32.30Ni/CeeAl 63.60 0 3.44 32.96

Ni/AlNi/Ce-AlNi/La-Al

0

10

20

30

40

50

Etha

nol c

onve

rsi o

n(%

)

(a)

Ni/AlNi/Ce-AlNi/La-Al

0

20

40

60

80

100

Etha

nol c

onv e

rsi o

n(%

)

(b)

Fig. 7. The decrease of ethanol conversion due to catalyst deactivation at (a) 150 �C and (b) 350 �C; dark grey e maximum ethanol conversion, light grey e ethanol conversion after4 h time on stream; 0.8 mL/min bio-ethanol flow, 133 mL/min Ar flow, bio-ethanol obtained by acid hydrolysis method.

M. Dan et al. / Renewable Energy 74 (2015) 27e3634

catalysts, but at 350 �C no carbon monoxide was detected for cat-alysts promoted with additional oxide.

Hydrogen production, expressed as volume of H2 produced perminute and per gram of catalyst, is a useful parameter whichcombines the catalyst activity for reactant conversion and theselectivity for H2 formation Fig. 6. It should be emphasized thathydrogen produced and analyzed in this study is obtained byreforming all the components contained by the crude bio-ethanol:ethanol, acetic acid and the other oxygenated organic molecules. At250 �C the contribution of acetic acid to hydrogen production is lowdue to its low conversion on all catalysts. The improved hydrogenproduction obtained at this temperature for ceria and lanthanamodified catalysts is explained by the improved ethanol conversionon these catalysts (Fig. 5a). At 350 �C, hydrogen production isenhanced over all catalysts due to the total conversion of aceticacid. On the promoted catalysts the production of hydrogen is alsoincreased due to: (i) the increased ethanol conversion; (ii) thebetter selectivity for hydrogen formation; and (iii) the enhance-ment of water gaseshift reaction which also produce hydrogen.Among the two oxides, La2O3 seems to give better results for theproduction of hydrogen. It was reported also that the addition ofrare earth oxides to alumina decreases the surface acidity, inhib-iting thus some secondary reactions [18] and favoring the refor-mation reactions.

The catalyst deactivation was expressed as the decrease ofethanol conversion in time. At 350 �C the catalysts are stable, littledeactivation being observed for Ni/Al (conversion decreased from

Fig. 8. TPO profiles of Ni catalysts used in bio-ethanol steam reforming reaction at (a) 150 �Cby acid hydrolysis method.

96% to 93%) after 4 h time on stream. No deactivation appeared forthe promoted catalysts, ethanol conversion being stable at 100%.However, a certain degree of deactivation occurs for all catalyststested in bio-ethanol steam reforming at 150 �C and 250 �C, beingmore accentuated at 150 �C (Fig. 7).

The main cause of catalysts' deactivation during steam reform-ing reactions is the formation and deposition of carbon on thecatalytic active centers, as a result of secondary reactions [34,38].Two methods were used to investigate the carbon deposition inbio-ethanol steam reforming experiments: TPO and XRD. The TPOresults are presented in Fig. 8. The quantity and structure of carbondeposited at 150 �C and 350 �C are very different.

For all three catalysts used at 150 �C the main oxidation peakis situated in TPO spectra around 400 �C, corresponding to car-bon in amorphous state, which usually does not deactivate thenickel catalysts [28]. In order to explain the catalysts deactiva-tion, the XRD measurements were performed on used catalysts(Fig. 9).

As expected, no peak attributed to graphitic carbon, situatedat 26.5�, could be identified. The XRD results revealed instead adecrease in the intensity of the Ni diffraction lines and thepresence of NiO. This indicates another and more probable causeof catalysts deactivation: the oxidation of catalytic active Nispecies to the inactive NiO, which can be related to the combi-nation of high water excess and low temperature used in theseexperiments. For Ni/Al catalyst used in steam reforming experi-ments at 350 �C, TPO measurements revealed two types of

and (b) 350 �C; 0.8 mL/min bio-ethanol flow, 133 mL/min Ar flow; bio-ethanol obtained

Fig. 9. XRD spectra of the spent catalysts in bio-ethanol steam reforming reaction at: (a) 150 �C and (b) 350 �C; 0.8 mL/min bio-ethanol flow, 133 mL/min Ar flow; bio-ethanolobtained by acid hydrolysis method; -- Ni, A- NiO, B-Al2O3 *- CeO2.

M. Dan et al. / Renewable Energy 74 (2015) 27e36 35

deposited carbon: lower quantity of amorphous carbon (oxida-tion temperature 435 �C) and high quantity of crystallized carbonwith maximum oxidation peak situated at 700 �C. For lanthanaand ceria promoted catalysts instead, very low quantity ofdeposited carbon was observed. In XRD spectra the formation ofNiO was not observed (Fig. 9 b). The TPO and XRD results lead tothe conclusion that the deactivation mechanism depends on thereaction temperature. At 150 �C the deactivation is mainly causedby NiO formation and the promotion of Ni catalysts with La2O3and CeO2 has no significant effect on catalysts stability. At 350 �Cthe deactivation of Ni/Al is caused by deposition of crystallizedcarbon on catalytic centers and the promotion with La2O3 andCeO2 drastically decreases the carbon formation and improve thecatalysts stability.

4. Conclusions

Fir wood wastes were used to produce crude bio-ethanol by twomethods: (i) simultaneous saccharification and fermentation (SSF)and (ii) acid hydrolysis followed by the fermentation of acid hy-drolyzate. The main components of the crude bio-ethanol areethanol and acetic acid. In addition, low concentration of a widerange of alcohols, acids, esters, ethers and aldehydes were alsodetermined. The concentration of all components depends on thehydrolysis method. In the SSF process the ethanol concentration ishigher than in the acid hydrolysis: 43.69 g/L compared to 37.53 g/L.Important concentrations of acetic acid were determined in sam-ples obtained by both methods. Opposed to ethanol concentration,the acetic acid concentration is higher in the acid hydrolysis pro-cess: 16.36 g/L compared to 10.24 g/L.

The crude bio-ethanol was used to produce hydrogen by cat-alytic steam reforming. The tested catalysts were the common Ni/Al and two rare earth oxides promoted Ni catalysts: Ni/LaeAl andNi/CeeAl prepared by successive wet impregnation. The charac-terization techniques revealed that the addition of rare earth ox-ides improves the Ni dispersion and the reducibility of thepromoted catalysts. Crude bio-ethanol steam reforming experi-ments were performed using 1 g of catalyst, at atmosphericpressure, in the temperature range of 150e350 �C, with differentbio-ethanol feed rates, and Ar as carrier gas. The best feed ratewhich ensures the optimal ratio between conversion and catalyticdeactivation was established at 0.8 mL/min bio-ethanol. Theaddition of rare earth oxides (La2O3 and CeO2) to the aluminasupport improves the ethanol conversion especially at 250 �C, butno significant effect on acetic acid conversion was observed at anystudied temperature. At 250 �C the ethanol conversion is almost

90% for Ni/LaeAl and Ni/CeeAl, but the acetic acid conversion isbelow 30% for all catalysts. The optimumworking temperature, atwhich both ethanol and acetic acid presents maximum conversionis 350 �C. At this temperature the best hydrogen production isobtained for Ni/LaeAl due to better ethanol conversion and betterselectivity for hydrogen formation. At 350 �C the promoted cata-lysts are stable for 4 h time on stream, while Ni/Al suffers deac-tivation due to graphitic carbon deposition. Higher degrees ofdeactivation are obtained at lower temperatures, mainly caused byNi oxidation to NiO.

Acknowledgments

This work was supported by a grant of the Romanian NationalAgency for Scientific Research, Project Number PN-II-PT-PCCA-2011-3.2-0452:“Hydrogen production from hydroxylic compoundsresulted as biomass processing wastes”.

References

[1] Tanksale A, Beltramini JN, Lu GM. A review of catalytic hydrogen productionprocesses from biomass. Renew Sustain Energy Rev 2010;14:166e82.

[2] Züttel A, Remhof A, Borgschulte A, Friedrichs O. Hydrogen: the future energycarrier. Philos Trans A Math Phys Eng Sci 2010;368:3329e42.

[3] Peter Hoffman. Tomorrow's energy: hydrogen, fuel cells, and the prospects fora cleaner planet. Mass Inst Technol 2012.

[4] Dagdougui H. Models, methods and approaches for the planning and design ofthe future hydrogen supply chain. Int J Hydrog Energy 2012;37:5318e27.

[5] Balat M. Production of bioethanol from lignocellulosic materials via thebiochemical pathway: a review. Energy Convers Manag 2011;52:858e75.

[6] Romaní A, Garrote G, L�opez F, Paraj�o JC. Eucalyptus globulus wood fraction-ation by autohydrolysis and organosolv delignification. Bioresour Technol2011;102:5896e904.

[7] Chung BY, Lee JT, Bai H-W, Kim U-J, Bae H-J, Gon Wi S, et al. Enhancedenzymatic hydrolysis of poplar bark by combined use of gamma ray and diluteacid for bioethanol production. Radiat Phys Chem 2012;81:1003e7.

[8] Hu R, Lin L, Liu T, Liu S. Dilute sulfuric acid hydrolysis of sugar maple woodextract at atmospheric pressure. Bioresour Technol 2010;101:3586e94.

[9] Ruiz HA, Silva DP, Ruzene DS, Lima LF, Vicente AA, Teixeira JA. Bioethanol pro-duction from hydrothermal pretreated wheat straw by a flocculating saccharo-myces cerevisiae strain e effect of process conditions. Fuel 2012;95:528e36.

[10] Bilal M, Jackson SD. Ethanol steam reforming over Rh and Pt catalysts: effectof temperature and catalyst deactivation. Catal Sci Technol 2013;3:754.

[11] Basagiannis AC, Panagiotopoulou P, Verykios XE. Low temperature steamreforming of ethanol over supported noble metal catalysts. Top Catal 2008;51:2e12.

[12] Padilla R, Benito M, Rodríguez L, Serrano A, Mu~noz G, Daza L. Nickel and cobaltas active phase on supported zirconia catalysts for bio-ethanol reforming:influence of the reaction mechanism on catalysts performance. Int J HydrogEnergy 2010;35:8921e8.

[13] Fatsikostas AN, Verykios XE. Reaction network of steam reforming of ethanolover Ni-based catalysts. J Catal 2004;225:439e52.

[14] Dal Santo V, Gallo A, Naldoni A, Guidotti M, Psaro R. Bimetallic heterogeneouscatalysts for hydrogen production. Catal Today 2012;197:190e205.

M. Dan et al. / Renewable Energy 74 (2015) 27e3636

[15] Melchor-Hern�andez C, G�omez-Cort�es A, Díaz G. Hydrogen production bysteam reforming of ethanol over nickel supported on La-modified aluminacatalysts prepared by sol-gel. Fuel 2013;107:828e35.

[16] Sanchez-Sanchez M, Navarro R, Fierro J. Ethanol steam reforming over Ni/MxOyNi/MxOyeAl2O3 (M¼Ce, La, Zr and Mg) catalysts: influence of supporton the hydrogen production. Int J Hydrogen Energy 2007;32:1462e71.

[17] Bussi J, Musso M, Veiga S, Bespalko N, Faccio R, Roger AC. Ethanol steamreforming over NiLaZr and NiCuLaZr mixed metal oxide catalysts. Catal Today2013;213:42e9.

[18] Le Valant A, Can F, Bion N, Duprez D, Epron F. Hydrogen production from rawbioethanol steam reforming: optimization of catalyst composition withimproved stability against various impurities. Int J Hydrog Energy 2010;35:5015e20.

[19] Senila L, Senila M, Costiug S, Miclean M, Cerasel V, Roman C. The autohy-drolysis of albies Alba wood using adaptive neural fuzzy interference systemmathematical modeling. Int J Green Energy 2014;11:611e24.

[20] Hallac BB, Sannigrahi P, Pu Y, Ray M, Murphy RJ, Ragauskas AJ. Biomasscharacterization of Buddleja davidii: a potential feedstock for biofuel pro-duction. J Agric Food Chem 2009;57:1275e81.

[21] Ofori-Boateng C, Lee KT. Same-vessel enzymatic saccharification andfermentation of organosolv/H2O2 pretreated oil palm (Elaeis guineensis Jacq.)fronds for bioethanol production: optimization of process parameters. EnergyConvers Manag 2014;78:421e30.

[22] Jackson RS. Wine science, principles and applications third edition. AcademicPress; 2008. p. 352e417.

[23] Romaní A, Garrote G, Ballesteros I, Ballesteros M. Second generation bio-ethanol from steam exploded Eucalyptus globulus wood. Fuel 2013;111:66e74.

[24] Mu~noz C, Baeza J, Freer J, Mendonça RT. Bioethanol production from tensionand opposite wood of Eucalyptus globulus using organosolv pretreatment andsimultaneous saccharification and fermentation. J Ind Microbiol Biotechnol2011;38:1861e6.

[25] Romaní A, Garrote G, Alonso JL, Paraj�o JC. Bioethanol production from hy-drothermally pretreated Eucalyptus globulus wood. Bioresour Technol2010;101:8706e12.

[26] SilvaNLC, BetancurGJV, VasquezMP,Gomes EDB, PereiraN. Ethanol productionfromresidualwoodchips of cellulose industry: acidpretreatment investigation,

hemicellulosic hydrolysate fermentation, and remaining solid fractionfermentation by SSF process. Appl Biochem Biotechnol 2011;163:928e36.

[27] Zhang J, Chu D, Huang J, Yu Z, Dai G, Bao J. Simultaneous saccharification andethanol fermentation at high corn stover solids loading in a helical stirringbioreactor. Biotechnol Bioeng 2010;105:718e28.

[28] Dan M, Mihet M, Biris AR, Marginean P, Almasan V, Borodi G, et al. Supportednickel catalysts for low temperature methane steam reforming: comparisonbetween metal additives and support modification. React Kinet Mech Catal2011;105:173e93.

[29] Dan M, Lazar MD, Rednic V, Almasan V. Methane steam reforming over Ni/Al2O3 promoted by CeO2 and La2O3. Rev Roum Chim 2011;56(6):643e9.

[30] Zheng W, Zhang J, Ge Q, Xu H, Li W. Effects of CeO2 addition on Ni/Al2O3catalysts for the reaction of ammonia decomposition to hydrogen. Appl CatalB Environ 2008;80:98e105.

[31] Zhao A, Ying W, Zhang H, Ma H, Fang D. La promoted Ni/a-Al2O3 catalysts forsyngas. World Acad Sci Eng Technol 2011;5:2109e13. 2011-11-24.

[32] Zhang J, Xu H, Jin X, Ge Q, Li W. Characterizations and activities of the nano-sized Ni/Al2O3 and Ni/LaeAl2O3 catalysts for NH3 decomposition. Appl Catal AGen 2005;290:87e96.

[33] Yang R, Xing C, Lv C, Shi L, Tsubaki N. Promotional effect of La2O3 and CeO2 onNi/g-Al2O3 catalysts for CO2 reforming of CH4. Appl Catal A Gen 2010;385:92e100.

[34] Mattos LV, Jacobs G, Davis BH, Noronha FB. Production of hydrogen fromethanol: review of reaction mechanism and catalyst deactivation. Chem Rev2012;112:4094e123.

[35] Basagiannis AC, Verykios XE. Catalytic steam reforming of acetic acid forhydrogen production. Int J Hydrog Energy 2007;32:3343e55.

[36] Basagiannis AC, Verykios XE. Reforming reactions of acetic acid on nickelcatalysts over a wide temperature range. Appl Catal A Gen 2006;308:182e93.

[37] Natesakhawat S, Watson R, Wang X, Ozkan U. Deactivation characteristics oflanthanide-promoted solegel Ni/Al2O3 catalysts in propane steam reforming.J Catal 2005;234:496e508.

[38] Dan M, Mihet M, Almasan V, Borodi G, Katona G, Muresan L, et al. ModifiedNi-Cu catalysts for ethanol steam reforming. AIP Conf Proc 2013;1565:208e14.