Jiao Liu *,Chuanyue Zheng,Junrong YueGuangwen Xu3

1 State Key Laboratory of Multi-phase Complex System,Institute of Process Engineering,Chinese Academy of Sciences,Beijing 100190,China

2 China University of Mining and Technology,Xuzhou 221116,China

3 Institute of Industrial Chemistry and Energy Technology,Shenyang University of Chemical Technology,Shenyang 110142,China

Keywords:Methanation Kaolin Catalyst Catalyst support Leaching

ABSTRACT Kaolin as a raw material for mesoporous support was firstly modified by calcination,acid treatment,and then was used to prepare nickel catalysts.The amount of alumina which was activated in kaolin during thermal treatment and then leached out in the acid was different.XRD pattern of the kaolin calcined at 600°C or 900°C exhibited only the diffraction peaks for amorphous silica and quartz while that calcined at 1100°C showed obvious peaks for γ-Al2O3.Therefore,the nickel-based catalysts exhibited different physic-chemical properties.Atmospheric syngas methanation over the catalysts clarified an activity order of CA-1100＞CA-900＞CA-1400＞CA-600＞KA≈0 at temperatures of 350-650°C and a space velocity of 120 L·g-1·h-1.Metallic nickel with small diameter which has medium interaction with the modified kaolin and is well dispersed on the support would have reasonably good activity and carbon-resistance for syngas methanation.

1.Introduction

Nickel-based catalysts have been widely used for syngas methanation,methane reforming or cracking due to its relatively high intrinsic activity and low price.The mesoporous supports with high surface area,such as SiO2[1-4],Al2O3[5-9],TiO2[10,11],ZrO2[12-14]or CeO2[15-17]and other developing materials,such as metal organic frame works(MOF),perovskite,spinel,hydrotalcite,layered double hydroxide(LDH)[18]and Ni or Co phyllosilicate hollow spheres[19]are usually employed.Support affects not only the activity,but also the thermal stability,carbon resistance,strength and the mass/heat transfer efficiency of the catalyst.

Some results suggested that the catalysts with Ni supported on Al2O3presented higher methanation activity than Ni/SiO2catalysts[20,21]and the moderate interaction between nickel and support is necessary to ensure the high activity and stability of the Ni/Al2O3catalyst for syngas methanation at high temperatures[7,8,22].However,γ-Al2O3would transform into γ-AlO(OH)partially which resulted in the increase of average pore diameter,the decrease of surface area and pore volume and then the aggregation of metallic nickel under hydrothermal conditions[23].Silica is a common additive of alumina mainly to stabilize Al2O3against loss of surface area[24,25].The presence of a small amount of silica on the alumina support may also modify the dispersion of the active phases and the final catalyst performance[26,27].

Natural mineral,such as montmorillonite(MMT),volkonskoite,bentonite or diatomite which has abundant slit-like mesopores composed of plate-like clay layers composing Al and Si sublayers is a favorable raw material for the preparation of mesoporous support[28-30].Lu[31]leached the volkonskoite with zirconyl nitrate dihydrate under hydrothermal conditions and found that the zirconia nanoparticles are highly dispersed on the partly damaged clay layers and the catalyst with the higher zirconia content had the larger specific surface area and exhibited better catalytic performance for the CO or CO2methanation.MMT pillared by cetyltrimethylammonium bromide(CTAB)was used as support of nickel based catalysts for naphthalene hydrogenation and the results showed that the organic modification of MMT greatly improved the Ni dispersion,specific surface area,pore volume and then the catalytic activity of the Ni/MMT catalyst[32].Therefore,the layer-damaged clays have a great potential as a cheap and effective support for the nickel catalyst.

Kaolin clay is also a layered silicate mineral,which is composed of an octahedral alumina sublayer joined to a tetrahedral silica sublayer via shared apical oxygen.One application of the kaolin clay is developing zeolite A,Y,ZSM-5[33],MCM-41[34]and SBA-15[35]using two steps(i)transformation of kaolin into highly reactive metakaolin by calcination and(ii)hydrothermal reaction of metakaolin with conventional alkaline zeolite synthesis mixture containing organosilane as a surfactant.Mesoporous silica as adsorbent with high specific surface area could also be synthesized via the successive treatment of natural kaolin by calcination,alkali activation and acid etching[36-38].However,the porous properties and acidities as well as catalytic properties of such material supported catalyst have not been well reported.

Therefore,in the present study natural kaolin was firstly modified by calcination,acid treatment,and then the nickel-based catalyst was prepared by precipitation-deposition method.The supports and catalysts were characterized by N2adsorption-desorption,XRD,XRF,TEM,TPR,NH3-TPD and TPO.The catalytic activity for the CO methanation was also investigated to acknowledge the interactions among the factors of support structure,catalyst physico-chemical property,catalytic activity,carbon-resistance and stability.

2.Experimental

2.1.Catalyst preparation and examination

Natural kaolin from Suzhou,China was used as raw material to prepare the mesoporous supports.Kaolin firstly was calcined at 600,900,1100 and 1400 °C,and denoted as K600,K900,K1100,and K1400,respectively.

Subsequently,the nickel-based catalysts were prepared by precipitation-deposition method.The thermal-treated kaolin was dispersed into a stoichiometric quantity of Ni(NO3)2·6H2O which was dissolved in distilled water to get the acidic solution.After continuously stirring for 6 h at 60°C,1 mol·L-1NaOH as the precipitation agent was added dropwise into the suspension and the pH was adjusted from 0.5 to 9.Then,the formed precipitate was aged at 60°C for 12 h and collected by filtration with thoroughly washing using distilled water.The catalysts were finally obtained by drying at 80°C for 10 h,calcined at 500°C for 4 h and denoted as CA-600,CA-900,CA-1100 and CA-1400,respectively.The catalyst using natural kaolin as support was labeled as KA.For characterization and activity test,all the samples were crushed into 180-230 μm.

The syngas methanation over any of the prepared catalysts was tested at 350-650°C in a quartz fixed bed reactor.Prior to the reaction the catalyst was reduced at 650°C for 4 h in a N2-base gas containing 10 vol%H2.Then,the temperature of the catalyst bed was decreased to the designated reaction-starting temperature.Feeding a gas mixture of H2:CO:N2=3:1:1 in moles in turn started the methanation reaction inside the reactor.The gas product was analyzed on-line using a micro gas chromatograph(Agilent 3000)equipped with TCD.

The CO conversion and selectivity to CH4referred to herein are estimated by

where X is the conversion of CO,S is the selectivity to CH4and f is the volumetric flow rate of the gas species CO or CH4that was determined from the GC-analyzed composition and the total gas flow rate determined by taking the well metered flux of N2as the internal standard.

2.2.Characterization and analysis

The surface area and pore size of the supports and catalysts were estimated from the N2physisorption curve measured at 77 K via Autosorb-1(Quantachrome).For this measurement the catalyst sample was thoroughly degassed at 300°C for 4 h in advance.The crystal structure of the supports and catalysts was analyzed with X-ray powder diffractometry(XRD,X'Pert MPD Pro,Panalytical)at its Cu Kαradiation of λ=0.15408 nm.The transmission electron microscopic measurements(TEM)were carried out in the JEOL electron microscope(JEM-2100)with an accelerating voltage of 200 kV.The composition of catalysts was determined by the X-ray fluorescence method(XRF)using AXIOS(PANalytical B.V.).

The temperature-programmed reduction(TPR),oxidation(TPO),desorption of CO2(CO2-TPD)and hydrogen chemisorption experiment are carried out using an automated chemisorption analyzer(ASAP2920,Micromeritics).During TPR,an Ar-base gas containing 10 vol%H2was used as the reacting agent.A catalyst sample of 50 mg was heated from 50°C to 1000°C at a rate of 10 °C·min-1.For TPO,50 mg of the spent catalyst was first pretreated in a He stream at 200°C for 30 min.After cooling naturally,the oxidation test was performed by heating the sample from 100°C to 650°C at a rate of 10°C·min-1in a N2-base gas flow containing 10 vol%O2and in turn held the sample at 650 °C for 60 min to fully oxidize the carbon species on the catalyst.The emitted CO2in the TPO test was monitored on-line with the MS(TILON).CO2-TPD was performed to identify the basic strength distribution of the catalyst.The pre-reduced sample of 50 mg in a quartz reactor was pretreated with He flow at 200°C(10°C·min-1)for 60 min and then cooled down.CO2was adsorbed at 50 °C for 2 h and weakly adsorbed CO2was removed by purging with He.The sample was heated from 50°C to 800°C at a heating rate of 10°C·min-1in He flow and the desorbed amount of CO2was monitored on-line with MS.Hydrogen chemisorption experiments were conducted to measure the nickel dispersion,nickel surface area,and average nickel diameter of the catalysts.Prior to the chemisorption measurements,50 mg of each catalyst was reduced with a mixed stream of H2and Ar at 650°C for 4 h,and subsequently,it was purged with pure He for 15 min at 300°C.The sample was then cooled to 50°C under a flow of He.The amount of H2uptake was measured by periodically injecting 10 vol%H2/Ar into the reduced catalyst using an on-line sampling valve.Nickel dispersion,nickel surface area,and average nickel diameter were calculated by assuming that one hydrogen atom occupies one surface nickel atom.

3.Results and Discussion

3.1.Characterization

The physisorption results of kaolin calcined at different temperatures are summarized in Table 1.With increasing the calcination temperature to 1100 °C,the BET surface area and pore volume of the catalyst have little difference with the natural kaolin.However,the surface area decreased from about 35 to 8 m2·g-1as the calcination temperature increased to 1400 °C.These results imply that much high calcination temperature(1400°C)results in the collapse of pores and the sintering of the kaolin.

Table 1 Physical properties and chemical composition for kaolin calcined at different temperatures and the corresponding nickel catalysts

Fig.1.XRD patterns of the(a)calcined kaolin(b)calcined and(c)reduced Ni-based catalysts supported on the thermal-treated acid-leached kaolin.

Fig.1(a)shows the XRD patterns of kaolin calcined at different temperatures.The natural kaolin displays characteristic peaks for kaolinite and a little of quartz.Metakaolin,an amorphous solid is obtained by dehydroxylation of kaolin via Eq.(1)at temperatures between 600 and 900°C,so only diffraction peaks for amorphous silica at 2θ=20°-30°and quartz were observed in K600 and K900.Calcination at higher temperature(1100°C)led to the reorganization of the oxide network which was finally composed of some crystallized γ-Al2O3and an amorphous phase rich in silica(Eq.(2)).Further increasing the calcination temperature to 1400°C,the diffraction peaks for γ-Al2O3disappeared and the crystallizations of mullite and cristobalite were enhanced.

The N2physisorption isotherms for the prepared catalysts were all of type IV with a hysteresis loop indicating the mesoporous structure and the main physical properties were also shown in Table 1.It can be seen that leached by acid(aqueous solution of nickel nitrate),the surface area of the catalyst was higher than its corresponding thermal-treated kaolin.It is believed that the amorphous Al2O3inside the calcined kaolin could be etched by H+in accompany with nanopores created in situ.The different calcination temperatures of kaolin result in the varied activation degree of the Al-O network which subsequently leads to the different nanoporosity after the acid leaching.The amount of active Al2O3increased with calcination temperature rising to 900°C,so the surface area of catalyst shows an order of CA-900＞CA-600＞KA.However,when the temperature increased to 1100°C,some crystallized γ-Al2O3was formed and the amount of active Al2O3decreased,so the surface area of the catalyst decreased to 266 m2·g-1.Further raising the calcination temperature,more inactive crystallites appeared and the surface area steadily declined for CA-1400(143 m2·g-1),but this is still much larger than K1400(8 m2·g-1).

Fig.2.(a)H2-TPR and(b)CO2-TPD of the Ni-based catalysts supported on the thermal-treated acid-leached kaolin.

XRD patterns for the catalysts in Fig.1(b)show that the peaks at 37°,43°and 63°ascribed to the diffraction peaks for the(111),(200)and(220)of NiO were observed in all the samples.In KA,the characteristic peaks for kaolinite in natural kaolin disappeared and diffraction peaks for NiO were narrow and sharp due to the layered structure and small surface area of the support.Raising the calcination temperature of kaolin resulted in wider diffraction peaks for NiO in CA-600,CA-900 and CA-1100 due to the well dispersion of NiO on the supports with high surface area as shown in Table 1.However,calcination at 1400°C collapsed the pores of kaolin and the diffraction peaks for NiO in CA-1400 became sharp and narrow again indicating aggregation of the crystallites.

The H2-TPR profiles of all the tested catalysts are presented in Fig.2(a).The nickel oxides deposited on the modified kaolin all exhibited only one reduction band,but the temperature range and width showed great difference.In Fig.2(a),it can be seen that the reduction temperature increased with raising the calcination temperature to 900°C.The reduction temperature of bulk NiO that has no interaction with the support was reported to be at 200-400 °C[39,40],as shown for KA.The peak with Tmaxat 400-700 °C in CA-600,CA-900,CA-1100 and CA-1400 thus can be assigned to the Ni2+of the NiO species interacting with the support.However,the interaction of NiO with the support will decrease its reducibility.The amount of crystallized γ-Al2O3increased as the calcination temperature of kaolin increased from 900 to 1400°C,and then the amorphous Al2O3etched to Al3+decreased.During catalyst preparation,Ni2+was more easily embedded in Al-O network as the Ni2+and Al3+coprecipitated than embedded to crystallized γ-Al2O3,so the interaction between Ni and support decreased as the calcination temperature increased from 900 to 1400°C.

Table 2 Hydrogen chemisorption results for reduced Ni-based catalysts supported on the thermaltreated acid-leached kaolin

After reducing at 650°C for 4 h in 10 vol%H2of N2-base,the five catalysts all showed additional diffraction peaks of metallic Ni at 2θ of 44.5°,51.8° and 76.3° corresponding to its(111),(200)and(220)planes in Fig.1(c),respectively.No diffraction peaks of NiO were detected indicating the catalysts were all reduced completely.However,the narrow and sharp diffraction peaks of Ni for KA and CA-1400 reveal that the Ni-crystallites agglomerated possibly on the small surface area in Table 1.The crystal sizes of metallic Ni calculated using the Scherrer equation for the(111)plane were 26 nm for KA,52 nm for CA-600,26 nm for CA-900,16 nm for CA-1100 and 37 nm for CA-1400,respectively.

CO2-TPD analysis was carried out to characterize the surface basicity of the thermal-treated acid-leached kaolin.Fig.2(b)reveals that in CA-600,CA-900 and CA-1400,two different desorption peaks of CO2could be distinguished at the low-temperature(LT)region(50-250°C)and medium-temperature(MT)region(250-550°C)[15,16].It was noticeable that the intensities of the LT region gradually increased and the peak positions of the MT region progressively shifted to lower temperatures with the increase of the kaolin calcination temperature from 600 to 1100°C.By contrast,the CO2-TPD profile of CA-1400 catalyst showed only a low-temperature peak conductive to the chemisorption on the weak basic sites of modified kaolin[15].

Hydrogen chemisorption measurements were conducted in order to determine the nickel surface area,nickel dispersion,and average nickel diameter in the reduced catalysts and the results are listed in Table 2.The amount of hydrogen uptake on the surface of reduced nickel catalysts increased with increasing the thermal treatment temperature until 1100°C.As a consequence,nickel surface area of reduced catalysts increased.Similarly,average nickel diameter in the reduced catalysts(CA-600,CA-900 and CA-1100)decreased with increasing the calcination temperature of kaolin,whereas average nickel diameter in the reduced CA-1400 catalysts increased.This means that nickel species with weak interaction with support in CA-1400 catalyst was aggregated during the catalyst preparation process,resulting in the formation of large metallic nickel particles with poor dispersion.This was further confirmed by TEM images.

Fig.3.TEM images of the Ni-based catalysts supported on the thermal-treated acid-leached kaolin.

Fig.4.(a)CO conversions and(b)CH4selectivity at different reaction temperatures over the Ni-based catalysts supported on the thermal-treated acid-leached kaolin(SV=120 L·g-1·h-1,P=0.1 MPa,nH2/nCO/nN2=3:1:1).

Fig.3 shows the TEM images of the reduced Ni-based catalysts supported on the thermal-treated acid-leached kaolin.For the catalysts one can see that in KA,the layered structure of kaolin was not fully destroyed so the interaction between NiO and support was weak and after reduction,the dispersion of metallic nickel(black spots)on the support was poor.For the reduced CA-600,CA-900 and CA-1100,the Ni particles were uniformly dispersed on the support and the average size based on 50 Ni particles was found to be about 44 nm,19 nm and 14 nm,respectively.However,due to the small surface area of CA-1400 and the weak interaction between NiO and support,the active sites were assembled on the support.

3.2.Catalytic activity

Fig.4 shows the CO conversion and selectivity to CH4realized by the nickel-based catalysts in a fixed bed reactor operated at different temperatures under atmospheric pressure and an SV of 120 L·g-1·h-1.It can be seen in Fig.4(a)that the CO conversion followed an order of CA-1100＞CA-900＞CA-1400＞CA-600＞KA≈0.For the CA-900,CA-1100 and CA-1400,their CO conversions increased with increasing temperature until 550-600°C and then decreased due to the thermodynamic control.The catalyst CA-1100 had the highest catalytic activity to make its CO conversion over 60% at 550 °C.However,compared with the Ni/Al2O3or the Ni-Mg/SiO2-Al2O3catalyst[41],the CO conversion of the Ni-based catalysts supported on the thermal-treated acidleached kaolin needs to be improved and at 350-550 °C the syngas methanation over such catalysts was subject to kinetic control under high SV.This may be due to the activated silica[42]partially covering the NiO species during calcination.

CO2was found to be the only byproduct gas of syngas methanation.Fig.4(b)demonstrates that the selectivity for these catalysts with CO conversion has little difference and was all about 72%-74%at different reaction temperatures.The CH4selectivity below 100%should be caused by the simultaneous water-gas-shift(CO+H2O→CO2+H2)and Boudouard(2CO→CO2+C)reactions which convert a small part of CO to CO2and deposited carbon,respectively.Even though the thermal-treated acid-leached kaolin fostered high CO2coverage in Fig.2(b),especially for CA-600,CA-900 and CA-1100,their CH4selectivities showed little difference.The different capacities of the catalysts for absorbing CO2on the surface of the catalysts might not promote the methanation activity of CO2,because over these catalysts the reaction is highly selective for the methanation of CO and CO2methanation is essentially inhibited[43]and the mechanisms for CO and CO2methanation were totally different[17,44].

3.3.Spent catalyst characterization

The time-on-stream tests were conducted for those catalysts with better performance in Fig.4(a),which are CA-600,CA-900,CA-1100 and CA-1400.Fig.5 shows the CO conversion varying with time at a temperature of 500 °C but pressures of 0.1 MPa and SVs of 120 L·g-1·h-1.All the catalysts exhibited stable CO conversion in 20 h.

Referring to the previous study,at least three types of deposited carbon would be found on the Ni-based catalysts.The first assigned to the reactive amorphous carbon(Cα)was oxidized below 400°C,the oxidation of carbon whiskers or fibers(Cβ)was at temperature of 400-650°C,and the last contributed to the inactive graphitic carbon(Cγ)only can be oxidized at temperatures higher than 650°C[45].

Fig.5.CO conversions with time-on-series at 500°C of the Ni-based catalysts supported on the thermal-treated acid-leached kaolin(SV=120 L·g-1·h-1,P=0.1 MPa,nH2/nCO/nN2=3:1:1).

Fig.6 shows the TPO curves of all the spent catalysts in Fig.5.It can be seen that the oxidation of the carbon on the catalysts was all at temperatures higher than 400°C.This indicates that both Cβand Cγwere deposited on their surfaces.Meanwhile,the CA-600 had a distinctively large and wide peak of released CO2,especially that at temperatures higher than 650 °C suggesting the largest amount of inactive carbon on CA-600 for its large Ni particles as shown in Fig.3.The amounts of carbon deposited on CA-900 and CA-1400 were only about 1/4 of that on CA-600,whereas there was almost no carbon on the CA-1100 catalyst.The carbon deposition requires some adjacent nickel atoms,and the catalyst with the larger nickel particles is more vulnerable to deposit carbon than that with the smaller particles.It is reported[46]that the inactive carbon Cγis responsible for the deactivation of the methanation catalyst through its encapsulation of the catalyst surface and active sites.Therefore,CA-1100 shows well carbon-resistance stability.

Fig.6.TPO results of spent Ni-based catalysts after atmospheric methanation as shown in Fig.5.

Five nickel-based catalysts supported on thermal-treated acidleached kaolin have been prepared and the effect of calcination temperature of kaolin on the catalyst structure and catalytic performance for syngas methanation was studied.It was found that the calcination temperature strongly affects the interaction between Ni and support,then the active surface area and nickel dispersion,and thus the catalytic performance.Generally,the better dispersion(metallic nickel)the catalyst possesses,the higher CO methanation activity it exhibits.The different calcination temperatures of kaolin result in the varied activation degree of the Al-O network which subsequently leads to the different interactions between nickel and support.The H2-TPR results confirmed the medium interaction between Ni and support in CA-1100.This interaction enabled CA-1100 catalyst to possess well dispersed metallic nickel and small crystallites which was revealed by XRD,TEM and hydrogen chemisorption experiments.The small and uniform Ni metal particles resulted in the high catalytic activity.Additionally,the small Ni metal particles enhanced carbon-resistance.

4.Conclusions

Nickle-based catalyst for syngas methanation was prepared by a precipitation-deposition method using thermal-treated acid-leached kaolin as support.Different calcination temperatures result in varied activation degree of the Al-O network in kaolin which subsequently leads to the different amounts of amorphous alumina leached out in the acidic solution of Ni(NO3)2and then the physico-chemical structure of the catalysts.Atmospheric syngas methanation over the catalysts clarified an activity order of CA-1100＞CA-900＞CA-1400＞CA-600＞KA≈0 at temperatures of 350-650 °C and a space velocity of 120 L·g-1·h-1.The nickel catalyst supported on the kaolin calcined at 1100°C showed the best activity and carbon-resistance stability because the interaction between nickel species and support was medium and metallic Ni with small diameter was well dispersed on the modified mesoporous kaolin.