Catalysis Communications 58 (2015) 127131

Contents lists available at ScienceDirect

Catalysis Communications

j ourna l homepage: www.e lsev ie r .com/ locate /catcomShort CommunicationSolgel synthesis and characterization of silica supported nickel ferritecatalysts for dry reforming of methaneRafik Benrabaa a,b, Axel Lfberg c,, Jess Guerrero Caballero c, Elisabeth Bordes-Richard c, Annick Rubbens c,Rose-Nolle Vannier c, Hamza Boukhlouf a,d, Akila Barama a

a Laboratoire de Matriaux Catalytiques et Catalyse en Chimie Organique, Facult de Chimie, USTHB, BP32, El-Alia, Bab Ezzouar, 16111 Alger, Algrieb Universit 20 Aot-Skikda, Facult de Technologie, Dpartement de Ptrochimie &Gnie des Procds, BP 26, Route El-Hadaiek, 21000 Skikda, Algriec Unit de Catalyse et de Chimie du Solide, UMR CNRS 8181, Universit Lille 1, Sciences et Technologies, Bt. C3 Cit Scientifique, 59655 Villeneuve d'Ascq, Franced Sonatrach, Institut Algrien du Ptrole, Les Platanes Filfila, 21101 Skikda, Algrie Corresponding author. Tel.: +33 3 20 43 45 27; fax: +E-mail address: Axel.Lofberg@univ-lille1.fr (A. Lfberg

http://dx.doi.org/10.1016/j.catcom.2014.09.0191566-7367/ 2014 Elsevier B.V. All rights reserved.a b s t r a c ta r t i c l e i n f oArticle history:Received 18 April 2014Received in revised form 2 September 2014Accepted 11 September 2014Available online 19 September 2014

Keywords:NiFe2O4Solgel methodIsopropanol decompositionDry reforming of methaneSilica-supportedNiFe2O4 spinelwas prepared by solgelmethod using tetramethyl orthosilicate as a precursor ofsilica. B.E.T., XRD, MEBEDS, TEM, XPS and Raman scattering techniques were used for its characterization. Thereducibility by hydrogenwas investigated by TPR andHT-XRD. These properties are compared to those of unsup-ported NiFe2O4. Both acidic and redox sites were found by studying the decomposition of isopropanol. First ex-periments in the dry reforming of methane by CO2 showed that owing to more acidic properties supportingNiFe2O4 on silica provides a more active and selective catalyst that seems less prone to coking.

2014 Elsevier B.V. All rights reserved.1. Introduction

Though steam reforming of methane to syngas is industrially prac-ticed over theworld, dry reforming ofmethane (DRM) attractsmuch at-tention from both environmental and industrial concerns. Nickel is themain active component of most catalysts, but upon sintering particlesbecome large and easily deactivated by coking. A means to reduce thesize of Ni particles is to control the reduction of a nickel-containing ox-idic matrix [1]. Spinels, perovskites, pyrochlores are typical structuralfamilies examined for this purpose [26]. Recently we studied severalmethods of syntheses of NiFe2O4, some of them allowing the formationof nanoparticles [79]. The acid-base properties of NiFe2O4 determinedby decomposition of isopropanol were examined [1012]. Propene wasformed on acidic sites while basic and redox sites led to acetone forma-tion [13,14].

Nanosized particles of ferrites are gaining considerable interest fortheir properties, beginning by magnetism [1519]. However, it is diffi-cult to produce uniform and reproducible nanoparticles using tradi-tional methods. In addition, nanoparticles have a strong tendency toagglomerate. A way to solve this problem is to disperse nanoparticlesin an oxidic matrix by using solgel technique. The latter allows lowsynthesis temperature, homogenous dispersion and a good control of33 3 20 43 65 61.).stoichiometry. Tetramethyl orthosilicate (TMOS) as a precursor of silicais an excellent host for supporting different kinds of guest nanoparticles[20]. TMOS facilitates the hydrolysis and condensation in the presenceof dilute HCl agent, giving homogeneous monolithic and transparentgels [21]. In addition, the porous nature of silicaminimizes the phenom-enon of aggregation of hosted nanoparticles. We have prepared solgelderived nickel ferrite/SiO2 to examine the effect of dispersion on the cat-alytic properties inDRM reaction. The textural and structural properties,the acid-base character and the catalytic properties are compared tothose of unsupported ferrite particles.

2. Experimental

2.1. Catalysts preparation

Silica supported NiFe2O4 was synthesized by solgel method,using TMOS as the precursor of silica. Ni(NO3)26H2O (2.9 g) andFe(NO3)39H2O (8.1 g) were dissolved in 40mL ethyl alcohol; this solu-tionwas added to 60mLof TMOS inwater (Ni:Fe:SiO2= 1:2:9). TenmLof HCl (0.1M)was added up to pH ~ 2. Thewet sols were allowed to gelat room temperature for three weeks. The 10-NiFe2O4/90-SiO2 gel wasdried at 200 C and calcined at 750 C for 2 h (NF/SiO2). For compara-tive purposes, unsupported NiFe2O4 samples were prepared by hydro-thermal (NFHT) and by solgel (NFSG) syntheses, using the sameprotocols as previously reported [79].

http://crossmark.crossref.org/dialog/?doi=10.1016/j.catcom.2014.09.019&domain=pdfhttp://dx.doi.org/10.1016/j.catcom.2014.09.019mailto:Axel.Lofberg@univ-lille1.frUnlabelled imagehttp://dx.doi.org/10.1016/j.catcom.2014.09.019Unlabelled imagehttp://www.sciencedirect.com/science/journal/15667367www.elsevier.com/locate/catcom

Table 1Textural data of NF/SiO2, NFHT and NFSG.

Catalyst SSA (m2/g) Ap (m2/g)a Vp (cm3/g)b rp ()c

NF/SiO2 152 108 0.049 10NFHT 184 6.4 0.001 26NFSG 52 77

a Micropore area.b Micropore volume.c Average pore radius.

128 R. Benrabaa et al. / Catalysis Communications 58 (2015) 1271312.2. Characterization

The specific surface area and porosity were determined by nitrogenadsorption at 196 C using BET and BJH methods. SEMEDS, TEM,XPS, XRD and Raman scattering techniques were used for textural andstructural characterization of catalysts. The reducibility of catalystswas examined by TPR-H2 and by H2-HTXRD. Details of the operatingconditions may be found in [79].

2.3. Catalytic experiments

The catalytic conversion of isopropanol was studied between200 and 400 C in the absence of oxygen, at atmospheric pressure(ca. 0.05 g of catalyst) [7]. The DRM reaction was investigated using100 mg of catalyst mixed with SiC loaded in a conventional fixed bedflow reactor. Samples were pre-reduced under 5% H2/He at 400 C.Effluents were analyzed by mass spectrometry. The diluted reactants(CH4:CO2:He:Ar = 20:10:10:50) were flowed at 90 mL/min rate andGHSV 54 Lh1gcat1 (contact time 0.6 s) at 800 C.

3. Results and discussion

3.1. Textural properties

The specific surface area (SBET) of NF/SiO2 after calcination at 750 Cand other porosity data are gathered in Table 1. They are comparedto unsupported NF samples prepared by hydrothermal synthesis(NFHT) which delivered directly the spinel [7,8] and by solgel meth-od after calcination at 400 C (NFSG) [9]. As expected, the value is larg-er for NFHT, and SBET decreases along the NFHT N NF/SiO2 N NFSGseries. The value of porous volume calculated by BJH method is in har-mony with SEM picture (Fig. 1-a) which shows aggregates of particlesof 0.2 to 5 m, some of themhaving a crystalline aspect. In contrast, par-ticles of NFHT and NFSG are nanosized (Fig. 1-b, c). Thus the porosityexhibited by NF/SiO2 seems to be mostly due to silica.

3.2. Structural properties

The diffractograms of NF/SiO2, NFHT and NFSG are presented inFig. 2. Though the pattern of NF/SiO2 is blurred, all lines could beassigned to NiFe2O4 (PDF 00-054-0964); silica which is amorphouswas recognized by a hump at ca. 2= 22.5 [22]. These figures suggestFig. 1. SEM picture of NF/SiO2 (a); TEM pthat particles of NiFe2O4 have nucleated in the silica matrix. No line cor-responding to nickel or iron oxides, quartz, cristobalite or mixed com-pounds like SiFe2O4 could be observed.

The Raman spectra of NF/SiO2 calcined at 750 C and of NF samples(Fig. 3) show the bands located at 335, 459, 488, 570, 589, 664 and701 cm1 characteristic of NiFe2O4. The strong band at 701 cm1,assigned to the symmetrical stretching motion of [FeO4] tetrahedralunit, is thefingerprinting of NiFe2O4 inverse spinel [23,24]. These resultsare in good agreement with XRD analysis.

3.3. Surface properties

The surface composition was determined by XPS (Table 2). Thepresence of Ni2+, Fe3+ and Si4+ was detected by Ni 2p3/2 (854.9855.8 eV), Fe 2p3/2 (710.1710.7 eV) and Si 2p3/2 (103,7103,9 eV)photopeaks, respectively [25,26]. Surface atomic ratios were calculated.In NF/SiO2, Fe/Ni is smaller than 2, the expected value for NiFe2O4, ascompared to bulk NF samples (Table 2). About a third of Fe3+ lacksand conversely two thirds of surface specieswould be nickel (II) species.Si/Fe is higher but Si/Ni is lower than the respective theoretical values.As a conclusion, Ni2+ species (may be as nanosized particles of NiO un-detectable byXRD) andNiFe2O4 particles aremostly embedded, and notproperly supported on silica.

3.4. Reducibility in hydrogen

The reducibility of NF/SiO2 examined by H2-TPR is clearly differentfrom that of NFHT and NFSG catalysts (Fig. 4). The amount of con-sumed H2 is 1617 mol/g for both NFHT and NFSG whereas it issmaller for NF/SiO2 (2.6 mol/g). The TPR profiles also are different.Well-defined steps of reduction are observed during the reduction ofbulk NiFe2O4 (NFHT and NFSG), reactions being assigned to them inFig. 4. For NF/SiO2, these peaks are replaced by a broad one located inthe 300800 C range which is attributed to the reduction of Ni2+andFe3+ species and/or of NiFe2O4 nanoparticles dispersed inside silicamatrix. Though XPS indicated that Ni2+ species are more numerouson surface (Fe/Ni = 0.6) than for NFSG and NFHT (Fe/Ni = 2.3 and1.8, respectively) it is difficult to put in evidence their own reduction.

In situ H2-HTXRD of NF/SiO2 was performed up to 800 C. The tem-peratures of reduction of Ni2+ and of the formation of NiFe alloyare not that different from those of unsupported nickel ferrite [8,9].The diffractograms (Fig. 5) show that silica supported NiFe2O4 is stableup to 400 C. At 450 C, the line at 2 ~ 44 ofmetallic Ni species starts togrow, as well as the formation of NiFe, identified by the peaks at2 ~ 43 and ~50. Metallic nickel disappears at ca. 600 C.

3.5. Catalytic decomposition of isopropanol

Table 3 shows the conversion of isopropanol, the distributionof products and the rates of formation of propylene and acetone ofNF/SiO2 and NFHT. Propylene and acetone are the main products.Isopropanol decomposition begins at 200 C, its conversion beinghigher for NF/SiO2 than for NFHT. The rate of formation of propyleneictures of NFHT (b) and NFSG (c).

image of Fig.1

10 20 30 40 50 60 70 80

0

2000

4000

6000

8000In

tens

ity (a

.u.)

S: NiFe2O4

NF/SiO2

2 ()

S

S

S

S

S

S

S

S

NF-HT

NF-SG

Fig. 2. XRD patterns of silica-supported and unsupported NiFe2O4.

Table 2XPS data of NF/SiO2, NFHT and NFSG.

Catalyst Binding energy (eV) Atomic ratio (XPS)

Fe Ni Si Fe/Nia Si/Feb Si/Nic

NF/SiO2 711.2 854.9 103.9 0.6 6 3.8NFHT 710.7 854.9 1.8 NFSG 711.0 854.8 2.3

a Theoretical ratio Fe/Ni = 2.b Theoretical ratio Si/Fe = 4.5.c Theoretical ratio Si/Ni = 9.

129R. Benrabaa et al. / Catalysis Communications 58 (2015) 127131rProp on NF/SiO2 is greater than for NFHT and the formation of ace-tone is very minor. The ratio rAcet/rProp (or SAcet/SProp) in NFHT variesfrom 1.2 to 1.7 depending on temperature. Consequently the surfaceof NF/SiO2 is mostly acidic in the 200400 C rangewhereas both acidicand redox/basic sites are present in NFHT. The acidity of oxides canbe compared using the optical basicity , which allows comparison ofthe acidity of cations by taking into account their coordination and sym-metry in the oxide [2729]. The optical basicity of silica ( = 0.35)is smaller than those of direct spinel NiAl2O4 ( = 0.636) and ofinverse spinel NF ( = 0.771). For cations, it increases along the Si4+

(tet = 0.35) b bFe3+ (tet = 0.659) b Fe3+ (oct = 0.756) b Ni2+

(oct = 0.953) series (tet and oct for tetrahedral and octahedral coordi-nation, respectively). The experimentally stronger acidic character ofNF/SiO2 compared to that of NF can be attributed to the large amountof surface Si4+ as observed by XPS analysis (Table 2). Compared tothese Brnsted SiOH species (at low temperature), the sites presenton NiFe2O4 that could be pictured as Fe3+OH and Ni2+OH are lessacidic or even basic. Fe3+ (Fe2+ Fe) and Ni2+ (Ni) species areobviously potential redox sites.3.6. Catalytic properties in the dry reforming of methane

The DRM reaction was investigated using unsupported NFSGand NFHT and NF/SiO2 catalysts (Table 4). For the sake of comparison[8,9], and because in all cases the activity decreases with time, the300 400 500 600 700 800 900 1000-1000

0

1000

2000

3000

Ram

an In

ten

sity

(a.

u.)

S

S SS

S: NiFe2O4

SS

S

NF/SiO2

NF-HT

NF-SG

Wavenumber (cm-1)

Fig. 3. Raman spectra of silica-supported and unsupported NiFe2O4.values of CH4 and CO2 conversions and of H2/CO and H2/CH4 (molar ra-tios) are given for the same time on stream (one hour at 800 C).

The conversions of unsupported and unreduced catalysts are small,but larger for NFSG than for NFHT. They decrease with time, probablydue to sintering as the instantaneous carbon mass balance reaches100%, although some coking cannot be excluded. The temperature ofprereduction by H2 (400 C) was chosen after H2-TPR experimentswith the purpose of avoiding the reduction of iron, which is thoughtto contribute to the reverse water gas shift (RWGS) reaction [8,9].Except for NFHT, the prereduction modifies the catalytic behavior.Conversions of methane and CO2 obtained with NF/SiO2 catalystare modest (XCH4 = 10 and XCO2 = 33%) when compared to those ofNFSG but it should be remembered that the supported catalyst con-tains only 10% of active NF phase.

Again, activities decrease with time. In the case of NF/SiO2 catalyst,the instantaneous carbonmass balance during reaction reaches approx.95%whereas it reaches only 75% in the case of NFSG. This indicates thatthe NF catalyst is less prone to coking when supported than withoutsilica. Further investigations are nevertheless necessary to better under-stand the deactivation process, and in particular the respective contri-butions of coking and sintering to this process.

The selectivity, represented by H2/CO and H2/CH4 ratios, is muchlower than the stoichiometric values when NF is not prereduced. ForNF/SiO2 and NFSG, the prereduction considerably improves the selec-tivity with the best results obtained for the supported catalyst (Table 4).

The differences in the catalytic reactivity depend mostly on theamount, size and dispersion of the metallic particles created duringthe prereduction step and/or the reaction itself. When the catalyst isnot prereduced, the reaction at 800 C provokes probably the reductionof both Ni2+ and Fe3+ to Fe and NiFe metallic species. Iron catalyzesRWGS, which contributes to low selectivity. The presence of iron doescontribute to stabilize the catalysts against coking but it considerably af-fects selectivity through RWGS reaction.200 400 600 800 1000

0,00

-0,02

-0,04

-0,06

-0,08

-0,10

-0,12

TCD

sig

nal (

a.u.

)

Fe3+

Fe

FeO Fe

Fe3O4 FeO

408

Fe3+ Fe3O4Ni2+ Ni

266

324

345

700

537

NF-SG

NF-HT

Temperature (C)

NF/SiO2

555

Fig. 4. H2-TPR profiles of silica-supported and unsupported NiFe2O4.

image of Fig.2image of Fig.3image of Fig.4

30 40 50 60 70 80

0

400

800

1200

1600

2000

2400

2800

3200

Inte

nsity

(a.u

.)

2 ()

Nix: substrate (Pt)x

Ni-Fe

100200300350

800

700

400

450

600650

550

500

750

Ni-Fe T (C)

Fig. 5. HT-XRD under 3%H2/N2 of NF/SiO2.

Table 4Catalytic properties in DRM reaction of NFHT, NFSG and NF/SiO2 at 800 C, 1 h of reac-tion time.

Catalysts XCH4(mol%)

XCO2(mol%)

H2/CO H2/CH4 C balance(%, 3%)

NFSG 9 36.5 0.3 1.0 100NFHT 4.2 22 0.2 0.8 100NF/SiO2 red a 10 33 0.7 1.7 95NFHT red a 5.5 26 0.2 0.8 100NFSG red a 66 93 1.1 1.3 75

a "red" for prereduced.

130 R. Benrabaa et al. / Catalysis Communications 58 (2015) 127131NFSG behavior changes dramatically before and after reduction.This prereduction may contribute to segregate Ni and Fe species. Theformer may be reduced to Ni0 whereas the latter may be present bothin reduced and partially oxidized states. Indeed, thermodynamic calcu-lations show that CO2 and H2O can contribute to oxidize Fe0 to Fe2+

and/or Fe3+. The reduction of Ni(II) can explain the increase in conver-sion of CH4 and CO2 from 9% to 66% and from 36.5% to 93%, respectively(Table 4). The prereduction has also a great influence on selectivity, H2/CO sharply increasing from 0.3 to 1.1. This can be explained by CO con-sumption via Boudouard reaction (2CO C+ CO2), and/or by the pro-duction of H2 viamethane cracking (CH4 C+2H2)which contributesto coking as the low carbon balance suggests. Nevertheless, part of theproduced H2 is still oxidized to H2O by RWGS as indicated by the ratherlow H2/CH4 ratio (1.3 instead of the optimal ratio of 2).

Thus in the case of NFSG, the prereduction of Ni species consider-ably increases the activity but the selectivity remains poor (H2/CO N 1and H2/CH4 b b 2) due to the combined RWGS and coking reactions.

In the case of NF/SiO2, H2-TPR and XRD do not give indications ofany reduction of Fe3+ at 400 C as it was for NFHT, for which theprereduction does not affect the activity nor the selectivity. The higheractivity of NF/SiO2 as compared to NFHT, especially when consideringthe lower NiFe2O4 loading in the supported catalyst, could be explainedby a better dispersion of particles.

The higher selectivity to hydrogen of NF/SiO2 is themost interestingfeature of this catalyst. The surface acidity of silica could contribute tostabilize more selective species. It could also contri\bute to the dryreforming reaction itself, e.g. by decreasing the relative contribution ofRWGS and by improving the methane activation with respect to thatof CO2.Table 3Isopropanol decomposition on NFHT and NF/SiO2 catalysts.

Catalysts T (C) XISOa (%) SPropb (%) SAcetb

NF/SiO2 200 28 80 20NF-HT 200 20 46 54NF/SiO2 250 59 99 1NF-HT 250 47 46 54NF/SiO2 300 86 99 1NF-HT 300 86 41 59NF/SiO2 350 94 99 1NF-HT 350 100 39 61

a Isopropanol conversion.b Selectivity to propene or to acetone.c Rate of formation of propylene or acetone.d Ratio of rates of formation (ratio of selectivity in brackets).4. Conclusion

For the first time, nickel ferrite catalyst was coprecipitated whileTMOS was hydrolyzed to silica to deliver NF/SiO2 catalyst. Structuralanalyses show that nanoparticles of NiFe2O4 were indeed formed butthat they are mostly embedded in SiO2 matrix instead of being sup-ported on it. At first sight, the comparison of its reducibility with thatof bulkNiFe2O4 does not seem in favor of NF/SiO2 but the higher amountof Ni2+ species on the near surface as observed by XPS is an advantagein the DRM reaction. As expected, NF/SiO2 is more acidic than unsup-ported catalysts. Isopropanol dehydration to propylene predominateson NF/SiO2 while the formation of acetone and propylene is favoredon bulk NiFe2O4. In the DRM reaction, NFSG exhibits the highest cata-lytic activity after prereduction but, considering the lower amount of ac-tive phase in NF/SiO2, supporting nickel ferrite is promising as it maycontribute to a better dispersion of NF species. Moreover, supportingsuch active species brings a remarkable improvement in terms of selec-tivity to hydrogen as compared to bulk NF. This improvement could beascribed to the surface acidity that would limit the relative contributionof RWGS reaction.

Acknowledgments

Dr. Christine Lancelot is gratefully acknowledged for the discussionsand TEM experiments.

References

[1] N. Sahli, C. Petit, A.C. Roger, A. Kiennemann, S. Libs, M.M. Bettahar, Catal. Today 113(2006) 187193.

[2] J. Guo, H. Lou, H. Zhao, D. Chai, X. Zheng, ACA 273 (2004) 7582.[3] V.R. Choudhary, K.C. Mondal, A.S. Mamman, U.A. Joshi, Catal. Lett. 100 (2005)

271276.[4] H. Provendier, C. Petit, C. Estournes, S. Libs, A. Kiennemann, ACA 180 (1999)

163173.[5] G.S. Gallego, C. Batiot-Dupeyrat, J. Barrault, E. Florez, F. Mondragon, ACA 334 (2008)

251258.[6] S. Gaur, D. Pakhare, H.Y. Wu, D.J. Haynes, J.J. Spivey, Energy Fuels 26 (2012)

19891998.(%) rPropc

(molh1 g1)rAcetc

(molh1 g1)rAcet/rPropd

(SAc/SProp)

4000 983 0.25 (0.25)1465 1987 1.3 (1.2)7616 59 0.008 (0.01)3556 4169 1.2 (1.2)5748 30 0.005 (0.01)2564 4352 1.7 (1.4)3554 36 0.01 (0.01) (1.5)

http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0005http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0005http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0010http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0015http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0015http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0020http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0020http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0025http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0025http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0030http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0030image of Fig.5

131R. Benrabaa et al. / Catalysis Communications 58 (2015) 127131[7] R. Benrabaa, H. Boukhlouf, S. Barama, E. Bordes-Richard, R.-N. Vannier, A. Barama,Catal. Lett. 48 (2012) 4249.

[8] R. Benrabaa, H. Boukhlouf, A. Lfberg, A. Rubbens, R.-N. Vannier, E. Bordes-Richard,A. Barama, J. Nat. Gas Chem. 21 (2012) 595604.

[9] R. Benrabaa, A. Lfberg, E. Bordes-Richard, A. Rubbens, R.-N. Vannier, A. Barama,Catal. Today 203 (2013) 188195.

[10] A. Auroux, A. Gervasini, J. Phys. Chem. 94 (1990) 63716379.[11] M.A. Aramendia, V. Borau, C. Jimenez, J.M. Marinas, A. Porras, F.G. Urbano, J. Catal.

161 (1996) 829838.[12] S.A. Bocanegra, A.D. Ballarini, O.A. Scelza, S.R. De Miguel, Mater. Chem. Phys. 111

(2008) 534541.[13] C. Ancion, G. Poncelet, Appl. Catal. A Gen. 108 (1994) 3140.[14] J.C. Luy, J.M. Parera, Appl. Catal. A Gen. 13 (1984) 3948.[15] B. Baruwati, K. Reddy, S. Manorama, R. Singh, O. Parkash, Appl. Phys. Lett. 85 (2004)

28332835.[16] M.B. Miroslaw, H. Krzysztof, J. Eur. Ceram. Soc. 27 (2007) 723727.[17] J. Hu, I.M.C. Lo, G. Chen, Sep. Purif. Technol. 56 (2007) 249256.[18] H. Xianghui, C. Zhenhua, Mater. Res. Bull. 40 (2005) 105113.[19] Kim Yeong Il, Kim Don, Sub Lee Choong, Physica B 337 (2003) 4251, http://dx.doi.org/10.1016/S0921-4526(03)00322-3.

[20] C.J. Brinker, G.W. Scherer, J. Non-Cryst. Solids 70 (1985) 301322.[21] C.J. Brinker, G.W. Scherer, E.P. Roth, J. Non-Cryst. Solids 72 (1985) 345368.[22] L. Khanna, N.K. Verma, Mater. Sci. Eng. B 178 (2013) 12301239.[23] J.A.C. De Paiva, M.P.F. Graca, J. Monteiro, M.A. Macedo, M.A. Valente, J. Alloys Compd.

485 (2009) 637641.[24] Y. Shi, J. Ding, Z.X. Shen, W.X. Sun, L. Wang, Solid State Commun. 115 (2000)

237241.[25] http://www.lasurface.com/accueil/index.php.[26] V.K. Mittal, S. Bera, R. Nithya, M.P. Srinivasan, S. Velmurugan, S.V. Narasimhan, J.

Nucl. Mater. 335 (2004) 302310.[27] P. Moriceau, A. Lebouteiller, E. Bordes, P. Courtine, Phys. Chem. Chem. Phys. 1 (1999)

57355744.[28] E. Bordes-Richard, Top. Catal. 50 (2008) 8289.[29] E. Bordes-Richard, P. Courtine, Optical basicity: a scale of acidity/basicity of solids

and its application to oxidation catalysis, in: J.L.J. Fierro (Ed.), Metal Oxides: Chem-istry and Applications, Marcel Dekker, 2005.

http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0035http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0035http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0130http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0130http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0040http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0040http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0045http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0050http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0050http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0055http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0055http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0060http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0065http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0135http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0135http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0075http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0080http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0085http://dx.doi.org/10.1016/S0921-4526(03)00322-3http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0145http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0150http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0100http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0155http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0155http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0110http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0110http://www.lasurface.com/accueil/index.phphttp://refhub.elsevier.com/S1566-7367(14)00377-X/rf0115http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0115http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0120http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0120http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0125http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0165http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0165http://refhub.elsevier.com/S1566-7367(14)00377-X/rf0165

Solgel synthesis and characterization of silica supported nickel ferrite catalysts for dry reforming of methane1. Introduction2. Experimental2.1. Catalysts preparation2.2. Characterization2.3. Catalytic experiments

3. Results and discussion3.1. Textural properties3.2. Structural properties3.3. Surface properties3.4. Reducibility in hydrogen3.5. Catalytic decomposition of isopropanol3.6. Catalytic properties in the dry reforming of methane

4. ConclusionAcknowledgmentsReferences