ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose

ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2 Available online at www.sciencedirect.com ScienceDi...

NAN Sizes 0 Downloads 2 Views

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose Robert Ryczkowski a, Marcin Je˛drzejczyk a, Beata Michalkiewicz b,  ski d, Agnieszka M. Ruppert a, Grzegorz Słowik c, Witold Kwapin Jacek Grams a,* _ Institute of General and Ecological Chemistry, Faculty of Chemistry, Lodz University of Technology, Zeromskiego 116, 90-924, Lodz, Poland b Institute of Inorganic Technology and Environment Engineering, West Pomeranian University of Technology, ul. Pułaskiego 10, 70-322, Szczecin, Poland c University of Maria Curie-Sklodowska, Faculty of Chemistry, Department of Chemical Technology, pl. M. CurieSklodowskiej 3, 20-031, Lublin, Poland d Carbolea Research Group, Department of Chemical and Environmental Science, University of Limerick, Limerick, Ireland a

article info

abstract

Article history:

The goal of this work is to determine an impact of the modification method of Ni/ZrO2

Received 11 July 2018

catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of

Received in revised form

cellulose to hydrogen-rich gas. MexO-ZrO2 supports (where Me ¼ Ca, Mg, Na or K) were

2 October 2018

prepared by impregnation, precipitation and sol-gel methods. The obtained results reveal

Accepted 7 October 2018

that an introduction of dopants to the zirconia support considerably enhances the H2 yield

Available online xxx

in comparison to unmodified catalyst. An increase in the hydrogen formation is accompanied by a rise in the total volume of the produced gases. It is demonstrated that the

Keywords:

highest amount of hydrogen is formed in the presence of the catalysts containing CaO-ZrO2

Nickel catalyst

support followed by Na doped materials. This phenomenon can be attributed to more

Alkali metal

efficient incorporation of Ca2þ and Naþ cations in the zirconia lattice making it more stable

Alkaline earth metal

in the reaction conditions. Moreover, it is observed that an activity order of the investigated

Hydrogen

catalysts is consistent with the changes in the basic character of their surface.

Biomass conversion

© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Cellulose

Introduction Dependence on fossil fuels and growing consumption of energy are one of the current challenges that need to be resolved. Various methods of the replacement of non-renewable energy sources have been investigated during the recent years. One of

them is high temperature conversion of lignocellulosic biomass [1,2]. The thermo-chemical treatment of cellulose (main component of lignocellulosic feedstock) can lead to the formation of useful products, such as hydrogen, synthesis gas, bio-oil, activated carbon, etc. [3]. Although, this process is very complex and consists of a large number of chemical reactions

* Corresponding author. E-mail address: [email protected] (J. Grams). https://doi.org/10.1016/j.ijhydene.2018.10.059 0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

2

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

(i.e. cracking, reforming, water-gas shift, dehydration, partial oxidation, decarboxylation, decarbonylation or oligomerization, among others) [4]. Therefore, it is extremely important to improve the yield and selectivity of valuable products in order to increase the industrial competitiveness of thermal treatment of biomass and obtain substances which can be directly used as fuels or chemical reagents for various industrial processes. That can be achieved by an application of supported metal catalysts [5]. However, the efficiency of the catalysts may be reduced due to coking or poisoning taking place during high temperature decomposition of lignocellulosic feedstock [6e8]. In this work we focused on the thermo-chemical conversion of cellulose (as a model molecule) into hydrogen rich gas. One of the most popular catalysts used in this process are nickel based materials [9]. The main advantages of Ni catalysts are relatively high activity and considerably low price in comparison to the systems containing noble metals. Although, nickel catalysts can suffer due to formation of carbon deposit or sintering related to high temperature treatment. One of the solutions of this problem is the use of support that increases the thermal resistance and slows down the deactivation of the catalyst. The promising candidate allowing for fulfillment of the mentioned requirements is zirconium oxide. Previous works [10] confirm its positive impact on the increase in the hydrogen production during thermal treatment of cellulose. On the other hand literature data shows that an addition of various dopants to ZrO2 can further increase the activity and thermal resistance of the catalyst [11]. The positive effect of alkali and alkaline metals addition was confirmed in steam reforming [12], combustion [13], dry reforming of CH4 [14], and methanation of CO2 [15]. The dopants promote the reaction between CO2 and carbon accumulating on the catalyst surface which allow for the removal of carbon deposit decreasing reaction efficiency [16]. Wang et al. [17] applied mesoporous Ni-CaO-ZrO2 catalyst to dry reforming of methane. The formation of CaCO3 due to the adsorption of CO2 on the basic CaO sites was observed during the performed reaction which limited carbon deposit formation and make the nickel phase accessible for reaction intermediates. The similar results were obtained by Lan-jie et al. who investigated the Ni-CaO-ZrO2 system in trireforming of methane [18]. In this case the results of the performed experiments exhibited that an application of high pH during the synthesis of the catalyst facilitated the formation of the tetragonal ZrO2 phase with small particle size and enhanced NiO dispersion. Moreover, it was demonstrated that high calcination temperature (700  C) and pH value in the range of 10e12 was responsible for the increase in strength of the interaction between nickel species and zirconia. The prepared catalyst allows for conversion of methane exceeding 70% during the mentioned process. On the other hand Chen et al. [19] investigated the influence of the preparation conditions on the performance of Ni/ ZrO2 doped by CaO in CO2 reforming of CH4 and suggested that an additional aging of the samples in hot water (reflux digestion) after co-precipitation step had direct impact on the texture and activity of the studied catalysts. Moreover, it was demonstrated that such kind of a treatment allowed to obtain smaller crystallite size and tetragonal ZrO2. Owing to the

application of aging process it was possible to create a new type of the interface between support and metal phase. In this case Ni species were partially decorated with ZrO2 which had positive impact on the stabilization of nickel in high temperature treatment. Secondly, the interaction between CaO and ZrO2 led to the formation of a specific ZrO2 surface layer with oxygen vacancies, which was responsible for the CO2 activation and in turn more efficient removal of the carbon species. The same phenomenon was observed by Lertwittayanon et al. [20] for Ni/CaO-ZrO2/Al2O3 catalyst used in methane steam reforming. The literature data shows that not only calcium but also the effect of the addition of other alkali or alkaline earth metals have an impact on the catalytic performance of the materials containing nickel. The studies performed by Liang et al. [21] demonstrated that an addition of sodium or potassium in form of nitrates via sol-gel method had beneficial effect on catalytic performance of 70%CeO2-30%ZrO2 used in the soot oxidation process. The highest activity was observed for the material containing 15% of dopant. Its presence resulted in the formation of a larger number of oxygen vacancies on the catalyst surface which in turn increased the yield of CO2 produced during the oxidation reaction. Barroso-Quiroga et al. [22] investigated nickel catalyst supported on ceramic oxides in dry reforming of methane. The authors of the mentioned work showed that introduction of 0,5% of potassium to the 10%Ni/CeO2 enhanced the stability of the catalyst over reaction conditions in comparison to unmodified sample. Similar results were presented by Nagaraja et al. [23]. They reported that an addition of 0.5% of K to 8%Ni/ MgO-ZrO2 catalyst improved its activity in dry reforming of methane. However, it was noted that only small concentration of potassium was beneficial, for samples exceeding 0.5% K, (concentration up to 1.9% K was studied) effect was diminished. In contrast the studies performed by Xu et al. [24] showed that the introduction of dopants to the catalyst structure might not be always advantageous. The results of the investigations of an impact of MgO addition on the activity of Ni/TiO2 catalyst in CH4/CO2 reforming demonstrated that magnesium oxide formed solid solution with NiO making the crystallites of an active phase larger and less active as compared to the catalyst without dopant. Although, it is worth noting that an addition of MgO reduced coking rate of the catalyst in the studied reaction. The positive influence of the addition of the compounds contained alkali or alkaline earth were also observed in biomass conversion processes. Stonor et al. [25] demonstrated that the presence of group I or group II hydroxides increased the efficiency of Ni catalyst in the hydrogen production with the simultaneous limitation of the formation of gaseous side products. NaOH and Na2CO3 were also used for desilication of ZSM-5 which resulted in the formation of hierarchical catalysts allowing for the increase in the content of aromatics in the liquid phase and decrease in the amount of formed coke during fast pyrolysis of cellulose [26]. The preliminary results of our investigations confirmed the beneficial impact of the addition of selected alkali and alkaline earth metals on the activity of Ni/ZrO2 catalyst in the thermal conversion of cellulose into hydrogen rich gas. However, in that case the dopants were introduced onto the surface of the

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

support only by impregnation method. The obtained results and literature data encouraged us to investigate more deeply the effect of the modification of Ni/ZrO2 catalyst by alkali/ alkaline earth metals on its activity in thermo-chemical conversion of cellulose. Therefore, in this work we focused on the studies of an impact of the kind of dopant introduction method on the properties of the synthesized catalyst and composition of gaseous products formed in thermo-chemical conversion of cellulose (in particular on the enhancement in the yield of the produced hydrogen).

3

Synthesis of Ni catalysts The supported 20 wt% Ni catalysts (about 5 g in each case) were prepared by the impregnation method which was described in Ref. [27]. The synthesized materials were calcined before activity test under the same conditions as supports. It is believed that NiO phase undergo reduction to the metallic nickel during the reaction due to reducing conditions existing in the reactor [28]. In our case this phenomenon was confirmed by TPR and XRD measurements performed for the samples subjected to reaction conditions. The synthesized catalysts were denoted as follows (Table 1):

Experimental Characterization methods Synthesis of supports Zirconium oxide was precipitated by NaOH. Then the mixture was heated to 104  C and stirred for 24 h. The precipitate was filtered and washed with aqueous solution of NH4NO3 (Chempur, pure (min. 99%)), and then with distilled water until neutral pH. Subsequently, it was dried in air at 110  C overnight and calcined in air flow at 700  C for 3 h (with the temperature ramp e 1  C/min up to 400  C and then 10  C/min up to 700  C). In the further steps of the synthesis of supports following modifying agents were used: Ca(NO3)2$4H2O (Chempur p.a.), Mg(NO3)2$6H2O (Chempur p.a.), Na2CO3 (Chempur p.a.) and of K2CO3 (Chempur p.a.). Their amount was calculated based on metal oxide contribution in the total weight of the support. MexO-ZrO2 (where M ¼ Ca, Mg, Na, K, abbreviated as PREC) was prepared by the same method as pure ZrO2 with the exception that the calculated amount of alkali/alkaline earth precursor was mixed with 25.78 g ZrOCl2$8H2O dissolved in water and later added dropwise to 70 ml of 5M NaOH. MexO-ZrO2 (where M ¼ Ca, Mg, Na, K, abbreviated as IMP) was synthesized by dissolving the calculated amount of the precursor in the small amount of water and adding it to the beaker containing a portion of earlier prepared ZrO2. The mixture was stirred for several minutes and aged for 24 h at room temperature. Then water was evaporated and support was calcined in air flow at 700  C for 3 h (with the temperature ramp e 1  C/min up to 400  C and then 10  C/min up to 700  C). MexO-ZrO2 (where M ¼ Ca, Mg, Na, K, abbreviated as SOL) was formed by sol-gel method. First, 20 ml of Zr(OC3H7)4 (70 wt % in 1-propanol) was mixed with 26.6 ml of n-propanol in the beaker that was placed on a magnetic stirrer (rate of stirring 400 rpm). Then the calculated amount of the precursor was dissolved in 3 ml amount of distilled water and slowly added dropwise for 5 min to the first solution to induce hydrolysis and create xerogel. The formation of gel prevented the stirring to be possible. Then the gel was aged for 1 h. Subsequently, beaker was placed in heating mantle and the gel was heated to temperature above 75  C to allow n-propanol to boil and evaporate from gel pores. The next step consisted of gel drying at 120  C for 12 h and finally calcination in air flow at 700  C for 3 h (with the temperature ramp e 1  C/min up to 400  C and then 10  C/min up to 700  C). Two MexO-ZrO2 supports with different alkali/alkaline metal oxides loading (5 wt% and 1 wt%) were prepared by the methods described above.

The specific surface area and porosity of the catalysts were determined with surface area and porosity analyzer Micrometrics ASAP 2020. Prior to the analysis, the samples were degassed under vacuum at 200  C for 3 h and then the low temperature N2 adsorption-desorption measurements were carried out. The BET specific surface area was calculated from the N2 adsorption isotherm. Powder X-ray diffractograms (XRD) were collected using a PANalytical X'Pert Pro MPD diffractometer. The X-ray source was a copper long fine focus X-ray diffraction tube operating at 40 kV and 30 mA. Data were collected in the 5e90 2Q range with 0.0167 step. Crystalline phases were identified by references to ICDD PDF-2 (ver. 2004) database. All calculations were performed with X'Pert HighScorePlus computer program. The calculation of NiO crystallite size was based on the Scherrer equation. Temperature-programmed reduction (TPR) was performed on AMI1 system from Altamira Instruments equipped with a thermal conductivity detector and used for examining the reducibility of the catalysts. In the experiments, mixture of 5 vol% H2 and 95 vol% Ar was applied with the flow rate of 30 ml/min and linear temperature ramp of 10  C/min from 40  C to 700  C. X-ray photoelectron spectroscopy (XPS) surface analysis of the investigated catalysts was performed on Kratos AXIS 165 spectrometer using Al mono Ka X-ray. The samples were fixed onto the sample holder by double-sided adhesive tape. The energy shift due to electrostatic charging was subtracted using the carbon C 1s band at 284.6 eV. Infrared (IR) spectra of CO2 adsorption on the surface of the studied catalysts were collected using Nicollets 6700 spectrometer equipped with a diffuse reflectance environmental chamber. Prior to CO2 adsorption the calcined catalysts were heated at 500  C under argon flow for 1 h to remove impurities from the surface. In the first case an adsorption of carbon dioxide was carried out at 100  C for 30 min and then the samples were cooled down to room temperature under CO2 flow and the IR spectra were recorded at room temperature. In the second case the samples were initially heated from ambient temperature to temperature at which catalytic reaction was carried out under CO2 flow and next IR measurements were performed. Scanning electron microscope (SEM) UHR FE-SEM Hitachi SU8020 and attached energy-dispersive X-ray spectroscopy (EDS) system were used for the investigation of the morphology

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

4

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Table 1 e Abbreviations of the catalysts used in the text. Catalyst

Preparation method

Abbreviation

20%Ni/1%CaO-ZrO2 20%Ni/1%CaO-ZrO2 20%Ni/1%CaO-ZrO2 20%Ni/1%MgO-ZrO2 20%Ni/1%MgO-ZrO2 20%Ni/1%MgO-ZrO2 20%Ni/1%Na2O-ZrO2 20%Ni/1%Na2O-ZrO2 20%Ni/1%Na2O-ZrO2 20%Ni/1%K2O-ZrO2 20%Ni/1%K2O-ZrO2 20%Ni/1%K2O-ZrO2 20%Ni/ZrO2

precipitation impregnation sol-gel precipitation impregnation sol-gel precipitation impregnation sol-gel precipitation impregnation sol-gel impregnation (nickel) precipitation (zirconia)

Ni/1CaZr-PREC Ni/1CaZr-IMP Ni/1CaZr-SOL Ni/1MgZr-PREC Ni/1MgZr-IMP Ni/1MgZr-SOL Ni/1NaZr-PREC Ni/1NaZr-IMP Ni/1NaZr-SOL Ni/1KZr-PREC Ni/1KZr-IMP Ni/1KZr-SOL Ni/Zr

and composition of the catalysts’ surface. The measurements were taken at the acceleration voltage of 5.0 kV. Several samples were analyzed with the Quanta 3D FEG FEI microscope with EDAX EDS system, equipped with Si(Li) detector. Acceleration voltage in the range of 5.0e20 kV was used. Thermogravimetric analysis (TGA-DTA-MS) has been performed using derivatograph SETSYS 16/18, Setaram and mass spectrometer ThermoStar, Balzers. The TGA-DTA and MS spectra were recorded in the air flow (40 cm3/min) for the range of temperature 20e900  C with heating rate of 10  C/ min. The mass sample was varied between 5 and 20 mg and the samples were weighted in the corundum crucible. This studies were used to determine the content of carbon deposit.

Catalytic activity measurements The activity of the investigated Ni/MexO-ZrO2 catalysts was tested in stirred batch reactor (with a volume of 250 ml) close to the atmospheric pressure at 700  C for 4 h with the heating rate 15  C/min. The minimal flow of Ar was used in order to direct the formed gases from the reactor to the gas chromatograph and maintain the total gas flow at 15 ml/min. The reaction temperature was chosen based on preliminary measurements [29]. The conversion of the model biomass sample - a-cellulose (SigmaeAldrich, pure) was conducted in the presence of nickel catalysts. In each case 5 g of a-cellulose and 0.2 g of the catalyst were used. An analysis of the reaction products exhibited the formation of gaseous products, liquid fraction and carbonaceous residue. An amount of permanent gases such as hydrogen, methane, carbon oxide and carbon dioxide was determined using gas chromatograph (GCHF 18.3, Chromosorb 102 column) equipped with a thermal conductivity detector (TCD). The gaseous products were collected every 30 min, the measurements were repeated three times.

Results and discussion Physicochemical properties of Ni/MexO-ZrO2 catalysts prepared by different methods The measurements of physicochemical properties of the studied catalysts allowed to find that the choice of the

preparation method has direct impact on the surface area of the analyzed samples. Generally, the precipitation method enabled to obtain relatively high surface area about 100e150 m2/g while in the case of sol-gel synthesis it was much lower, usually about 10 m2/g. Surface areas of the investigated catalysts are summarized in Table 2. A comparison of the results obtained for the catalysts containing supports prepared by impregnation method showed that samples doped with calcium and magnesium had very low surface area (only a few m2/g). In contrast, the materials doped with sodium and potassium possessed noticeably higher areas e 128 m2/g and 148 m2/g, respectively. According to the literature the latter metals are usually used as precipitating agents and high surface area can be attributed to their positive influence on developing the 3D structure of the zirconia in high pH [30]. A clear relation between pore volume, pore radius and kind of the support preparation method was also observed. The precipitation and impregnation methods allowed for formation of higher surface area and smaller pores in comparison to sol-gel which led to creation of lower surface area and arising of noticeably larger pores (Table 2). In the next step of the studies XRD experiments were carried out (Fig. 1). The diffraction lines at 2Q values of 37.2 , 43.2 , 62.8 , 75.3 and 79.4 were ascribed to the presence of nickel oxide phase [31]. On the other hand, in the case of reference Ni/ZrO2 catalyst the reflexes at 30.4 , 35.3 and 50.4 confirmed the presence of t-ZrO2 phase. Those signals were slightly shifted in comparison to literature (30.2 , 35.2 , 50.3 [32]). However, this resulted from the presence of some amount of sodium into zirconia lattice which was introduced during catalyst preparation (by the method using NaOH as precipitating agent). The similar effect was observed for modified catalysts (2Q values in the range of 30.3 e30.8 , 35.4 e35.5 and 50.3 e50.9 , respectively) which can be attributed to partial incorporation of dopants in the structure of zirconia and substitution of Zr4þ cations in the solid solution [33]. The estimation of NiO crystallites size performed using Scherrer equation (Table 2) revealed that the dimension of nickel oxide species present on the surface of the investigated samples ranged between 19 nm and 36 nm. However, this

Table 2 e Selected physicochemical properties of Ni/ZrO2 and Ni/MeOx-ZrO2 catalysts. Catalyst

Ni/1CaZr-PREC Ni/1CaZr-IMP Ni/1CaZr-SOL Ni/1MgZr-PREC Ni/1MgZr-IMP Ni/1MgZr-SOL Ni/1NaZr-PREC Ni/1NaZr-IMP Ni/1NaZr-SOL Ni/1KZr-PREC Ni/1KZr-IMP Ni/1KZr-SOL

BET surface area [m2/g]

NiO crystalline size [nm]

Pore volume (cm3/g)

Pore radius (nm)

138 2 11 91 7 6 129 128 8 146 148 9

23 36 19 25 31 28 27 21 31 23 23 27

0.21 0.01 0.07 e e 0.03 0.28 0.19 0.05 e e 0.04

2.4 4.1 7.4 e e 21.6 3.7 2.7 10.2 e e 7.5

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

5

Fig. 1 e XRD patterns of reference and Ca, Mg, Na and K modified catalysts.

does not exclude the existence of a certain number of smaller crystallites that may be located inside the pores of the catalysts [28]. TPR profiles presented in Fig. 2 exhibit severe differences in the reduction behavior of NiO supported on parent ZrO2 and mixed MexO-ZrO2 prepared by impregnation, precipitation and sol-gel methods. The reduction of nickel oxide begins from the temperature of about 300  C. In the case of Ni/ZrO2 sample only one reduction peak with the maximum of hydrogen consumption at 550  C is observed. It can be ascribed to the reduction of NiO particles weakly interacting with the support [34]. The reduction behavior of modified with alkali and alkaline earth metals samples is different. Broadening of the reduction range of nickel oxide species and the presence up to three maxima of hydrogen consumption during the TPR measurements are observed in this case. A shift of the maxima of hydrogen consumption rate towards lower temperature observed for the samples prepared by the precipitation and impregnation methods indicates that an incorporation of metal oxide in inorganic precursor lattice increases reducibility of an active phase. On the other hand, in the case of the catalysts synthesized by sol-gel method, where organic zirconium source was used, the intense signals with maximum shifted towards higher temperature range were noted. However, the samples modified with Ca and Mg exhibited lower reduction temperature of NiO in comparison to Na and K doped catalysts. The surface composition of the studied catalysts was also investigated by X-ray photoelectron spectroscopy (Table 3, Fig. SI4 and Fig. SI5). The XPS spectrum collected from the surface of the unmodified sample revealed the presence of Ni 2p3/2 signals at 854.0 eV and 856.0 eV confirming the existence of Ni in the þ2 oxidation state in the form of not only nickel oxide but also nickel hydroxide [35]. An analysis of the spectra obtained for Ca and Na modified materials exhibited a slight shift of the maximum of Ni 2p3/2 signals which ranged from 853.8 to 854.3 eV and from 855.9 to 856.3 eV most likely due to differences in the oxygen content in the vicinity of nickel atoms.

The values of Zr 3d5/2 and 3d3/2 binding energies of 182.3 eV and 184.7 eV (with chemical shift of 2.4 eV) noted for Ni/Zr catalyst were characteristic for the zirconium oxide phase [36]. A comparison of the spectra collected for doped samples demonstrated that they remained unchanged besides those which were obtained for the materials modified by calcium and prepared by impregnation or sol-gel methods (Zr 3d5/2 and 3d3/2 binding energies - 181.7 eV and 184.1 eV, respectively) and catalyst modified by sodium, synthesized by sol-gel method (Zr 3d5/2 and 3d3/2 binding energies - 181.4 eV and 183.8 eV, respectively). It may be connected with the formation of zirconium suboxides and deficiency of oxygen atoms in the lattice which in turn leads to the formation of oxygen vacancies as reported previously [37]. The binding energies of calcium observed for the catalysts containing this metal suggested that in the case of the material prepared by precipitation method (Ni/1CaZr PREC) CaCO3 is present on the surface (Ca 2p3/2e347.1 eV), while for the samples prepared by impregnation and sol-gel methods Ca could exist rather in the form of calcium oxide (Ca 2p1/ 2e346.6 eV and 346.8 eV). However, the presence of CaZrO3 mixed phase (Ca 2p1/2 binding energy of 350.3 eV) could not be also excluded [38]. The existence of calcium carbonate can be also confirmed by the signals observed in the carbon region. However, the intensity of C 1s2 peak with binding energy between 288 and 289 eV is rather weak, what indicates that only small part of alkali ions is involved in the formation of surface carbonates. An analysis of the shift of the signal originating from sodium indicated that in the case of the catalysts prepared by impregnation and precipitation methods (Ni/1NaZr IMP and Ni/1NaZr PREC) Na was mainly present in the form of sodium oxide (Na 1s binding energy e 1072.2 eV) [39]. Although, for the sample synthesized by precipitation a weak signal ascribed to sodium carbonate or sodium bicarbonate was noted. In the case of the catalyst prepared by sol-gel method (Ni/1NaZr SOL) Na existed rather as Na2CO3 or NaHCO3 (Na 1s binding energy e 1071.3 eV). However, according to Chen et al. [40], a shift of Na 1s binding energy towards lower values might be also

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

6

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Fig. 2 e TPR profiles of unmodified and Ca, Mg, Na and K modified catalysts.

Table 3 e The results of XPS measurements of Ni/CaO-ZrO2 and Ni/Na2O-ZrO2 catalysts prepared by different methods. Catalyst

Ni/Zr Ni/1CaZr PREC Ni/1CaZr IMP Ni/1CaZr SOL Ni/1NaZr PREC Ni/1NaZr IMP Ni/1NaZr SOL

Position (eV) Ni 2p3/2 1

Ni 2p3/2 2

Zr 3d5/2

Zr 3d3/2

Ca 2p3/2

Ca 2p1/2

854.0 853.9 854.0 854.1 854.2 854.3 853.8

856.0 856.0 856.1 856.1 856.3 856.3 855.9

182.3 182.2 181.7 181.7 182.2 182.4 181.4

184.7 184.6 184.1 184.1 184.6 184.7 183.8

e 347.1 346.6 346.8 e e e

e 350.6 350.1 350.3 e e e

attributed to the presence of Na2ZrO3 phase. The formation of metal carbonates on the surface of the catalysts could not be only depended on the kind of the preparation method but also might result from the conversion of metal oxides with CO2 when the materials were exposed to air, both during the synthesis or storage. IR spectroscopy was applied for the evaluation of the surface character of investigated catalysts and their ability to carbon dioxide adsorption which was used as a probe mole€ nsted or cule. It is known that CO2 can be adsorbed on the Bro Lewis basic centers located on the surface of the catalysts with simultaneous formation of bicarbonate species (HCO 3 ) or ), respectively [41]. surface carbonates (CO2 3 The IR spectra collected for Ni/Zr and samples modified with various metals (prepared by sol-gel method) are showed

Na 1s 1

Na 1s 2

e e e

e e e

1071.1 e 1071.3

1072.2 1072.2 e

C 1s 1

C 1s 2

e 284.6 284.6 284.6 284.6 284.6 284.6

e 288.2 288.4 288.4 288.7 288.3 288.7

in Fig. 3. An analysis of the results of CO2 adsorption on the surface of unmodified nickel catalyst at room temperature exhibited the presence of intensive band at 1670 cm1 with a shoulder at 1570 cm1 ascribed to the presence of bicarbonate 2HCO 3 and bidentate carbonate b-CO3 , while the band at 1 1360 cm was attributed to the existence of monodentate carbonate m-CO2 3 group. Moreover, a low intensity band at 1215 cm1 attributed to bicarbonate HCO 3 species was also observed [41,42]. An introduction of alkali and alkaline earth metals into the structure of Ni/Zr catalyst resulted in the change in the intensity and location of the bands on IR spectra. Furthermore, the mentioned spectra were more complex than those obtained for unmodified material. In the case of Ni/1KZr SOL and Ni/1NaZr SOL samples the signals with the maxima at

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Fig. 3 e CO2 adsorption on the surface of the catalysts modified by different metals measured (A) at room temperature and (B) after thermal treatment.

7

1215 cm1 completely disappeared. The band at 1370 cm1 not only decreased its intensity with a rise in the temperature of CO2 adsorption but also shifted its maximum to lower wavenumber - 1320 cm1. The latter band was assigned to bidentate carbonate b-CO23 and indicated a transformation of monodentate carbonate species recorded at 25  C to b-CO23 [44,45]. The similar phenomenon was also observed for the modified catalysts (Fig. 3). However, it was noted that in the case of CO2 adsorption on the surface of Ni/1KZr SOL and Ni/ 1NaZr SOL at elevated temperature IR spectra revealed higher intensity of the bands at 1680 cm1 and below 1300 cm1 € nsted basic centers), while which were attributed to HCO 3 (Bro the results obtained for Ni/1CaZr SOL and Ni/1MgZr SOL exhibited the existence of stronger signals in the range begroups tween 1600 cm1 and 1300 cm1 coming from CO2 3 (Lewis basic centers). A comparison of the efficiency of CO2 adsorption (at both room and elevated temperatures) on the surface of Ni/1CaZr SOL catalysts prepared by different methods demonstrated that in spite of a similar shape of IR spectra the bands ascribed 2 to the presence of HCO 3 and CO3 groups possessed various intensity which increased in the following order: Ni/1CaZr PREC < Ni/1CaZr IMP < Ni/1CaZr SOL (Fig. 4). SEMeEDS measurements were performed in order to determine morphology of the modified catalysts. In the case of samples prepared by both precipitation and sol-gel methods the obtained results exhibited the presence of dopants in the form of larger clusters, while the material synthesized by impregnation method possessed more uniform distribution of alkali or alkaline earth metal species. The exemplary SEM images obtained for samples modified by calcium oxide are presented in Fig. 5. The comparison of the results of surface composition of the investigated materials confirmed that generally calcium and sodium were incorporated more efficiently in the structure of the catalysts than magnesium and potassium, respectively (Table 4).

Influence of the surface properties of the studied catalysts on their catalytic activity 1670 cm1, 1590 cm1, 1510 cm1, 1380 cm1 (coming from 222respectively) were noted HCO 3 , b-CO3 , m-CO3 , m-CO3 [41e43]. Moreover, the bands (ascribed to b-CO23) at 1280 cm1 (Ni/1KZr SOL) and at 1300 cm1 (Ni/1NaZr SOL) were visible. The mentioned shift in the wavelength between these two signals is probably connected with differences in the electronegativity of potassium and sodium. On the other hand, the spectra of Ni/1CaZr SOL and Ni/1MgZr SOL exhibited the presence of bands at about 1625 cm1, 1425 cm1, 1225 cm1 1 1 (attributed to HCO (ascribed to b-CO23 ), 1540 cm , 1325 cm 3 ) 1 2and 1390 cm (originated from m-CO3 ). A decrease in the efficiency of CO2 adsorption after thermal treatment (conducted at the temperature of catalytic reaction) in comparison to the experiment performed at room temperature indicated reduction of the amount of basic centers on the catalyst surface at higher temperature [41,42]. In the case of Ni/Zr sample IR spectra demonstrated a drop in the intensity of bands at 1675, 1550 and 1370 cm1 whereas band at

Thermo-catalytic conversion of cellulose was conducted in inert gas atmosphere with the use of series of 20%Ni/MexOZrO2 (Me ¼ Ca, Mg, Na, K) catalysts with different alkali/ alkaline metal oxides loading (5 wt% and 1 wt%) and unmodified 20%Ni/ZrO2 reference material. However, the samples containing higher amount of dopants did not lead to higher yields of the tested reaction. Due to that in the following part of this work we focused only on the presentation of the results obtained for the Ni/ZrO2 systems modified by 1% of dopants (data achieved for Ni/5%MexO-ZrO2 are included in Supporting Information e Tables SI1-SI3 and Figs. SI1-SI3). Gaseous mixture obtained during thermal treatment of cellulose mainly consisted of hydrogen, carbon oxide, carbon dioxide and methane. Their yield was summarized in Table 5. The reaction performed without the catalyst allowed only for production of 6.1 mmol H2/g cellulose. An addition of unmodified 20%Ni/ZrO2 sample increased the hydrogen yield to

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

8

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Fig. 4 e CO2 adsorption on the surface of the catalysts doped with CaO via different methods measured (A) at room temperature and (B) after thermal treatment.

the value of 10.3 mmol H2/g cellulose. However, the formation of the highest amount of H2 was observed in the presence of the catalyst modified by alkali/alkaline earth metals. The comparison of the amount of hydrogen produced with the use of the catalysts doped with various alkali or alkaline earth metals demonstrated that an addition of calcium and sodium increased the H2 yield to greater extent than addition of magnesium or potassium. It was observed that the highest amount of hydrogen was formed in the presence of Ni/CaOZrO2 sample prepared by sol-gel method (17.8 mmol H2/g cellulose) followed by Ni/Na2O-ZrO2 catalyst synthesized in the same manner (16.9 mmol H2/g cellulose). This phenomenon can be attributed to more efficient incorporation of Ca2þ and Naþ cations in the zirconia lattice due to the size of their ionic radius (0.099 nm and 0.095 nm, respectively). The radius size of these ions fits better the parameters needed to obtain more stable tetragonal lattice of ZrO2. On the other hand, Mg2þ (0.065 nm) and Kþ (0.133 nm) ions have ionic radius which do not quite much the parameters of the mentioned structure. It results in generation of the strains deforming optimal tetragonal structure of the support and shifting positions of ions in

the lattice [19]. A substitution of Zr4þ inside ZrO2 lattice with Ca2þ cations contributes to the transformation of unstable monoclinic phase of zirconium oxide into stable tetragonal phase of zirconia through the formation of solid solution. That in turn makes the support more resistant to sintering occurring in high temperature of the cellulose conversion process. An introduction of a Ca2þ ions into ZrO2 lattice produces also new vacancies in the O2 sublattice. The substitution of Zr4þ with lower charge cation results in the formation of oxygen vacancies in the structure of zirconia with neutral charge. Those vacancies enhance the oxygen mobility and oxygen ionic conductivity. It results in the promotion of in situ removal of carbon deposit formed on the catalyst surface during thermochemical treatment of cellulose [46]. It is worth noting that hydrogen yield obtained with the use of the catalysts doped by calcium was higher than that observed in the case of the process carried out with Ni/ZrO2 modified by CeO2 [27]. An analysis of the results obtained for the catalysts prepared by different methods revealed that hydrogen yield depends not only on the kind of the introduced dopant but also on the way of the incorporation of alkali/alkaline earth metals into zirconia structure. The highest amount of hydrogen was obtained in the presence of catalysts containing the support synthesized by sol-gel method followed by the materials prepared by impregnation and precipitation methods (Fig. 6). This effect can be associated with the fact that the catalyst prepared by sol-gel method possessed considerably higher pore radius than other systems. It allows for more efficient penetration of its structure by the heavier intermediates of cellulose decomposition process. Due to that cellulose can be decomposed more efficiently to smaller molecules. Moreover, a comparison of the stability of the catalysts doped by Ca revealed that the material prepared by sol-gel method is the most stable and possesses Ni species the most resistant to sintering (19 nm and 18 nm before and after high temperature reduction, respectively). In the case of the catalysts containing supports prepared by other methods about 20e25% difference between the size of nickel species before and after treatment at high temperature (700  C) was noted. The results of IR studies performed for the samples modified by calcium demonstrated that an activity order of the investigated catalysts was also consistent with basic character of their surface. This suggests that the most active catalyst possesses the highest number of basic sites, what is connected with the ability of adsorption of the highest amount of CO2. The adsorbed carbon dioxide may shift the equilibrium of the reactions occurring during high temperature treatment of cellulose which facilitates the formation of hydrogen. On the other hand adsorbed CO2 can be decomposed and the released oxygen takes part in the removal of carbon deposited on the catalyst surface during the studied process. A comparison of the amount of other gaseous products formed in high temperature treatment of cellulose showed that differences in the yield of CO, CO2 and CH4 are not as significant as in the case of H2. Generally, an increase in the hydrogen formation was accompanied by a slight rise in the total volume of the produced gases. However, the comparison of the values of H2/CO and H2/CH4 ratios demonstrated that hydrogen content in the gaseous mixture was the highest for the catalysts modified with the use of calcium and sodium in

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

9

Fig. 5 e SEM images of unmodified and Ca doped catalysts prepared by different methods.

Table 4 e Surface composition of modified 20%Ni/ZrO2 catalyst determined via SEM-EDS measurements. Preparation method

Content of dopant [wt %]

Precipitation Impregnation Sol-gel

Ca

Mg

Na

K

1.53 0.95 0.50

0.79 0.60 0.36

0.59 1.50 0.31

0.01 0.62 0.51

Table 5 e Production of H2, CO, CH4 and CO2 in thermochemical conversion of cellulose conducted at 700  C for 4 h. Sample

without catalyst Ni/Zr Ni/1%CaZr PREC Ni/1%CaZr IMP Ni/1%CaZr SOL Ni/1%MgZr PREC Ni/1%MgZr IMP Ni/1%MgZr SOL Ni/1%NaZr PREC Ni/1%NaZr IMP Ni/1%NaZr SOL Ni/1%KZr PREC Ni/1%KZr IMP Ni/1%KZr SOL

Yield [mmol/g cellulose] H2

CO

CH4

CO2

6.1 10.3 13.1 16.9 17.8 8.6 13.6 15.0 12.5 13.8 16.9 12.0 12.8 14.7

7.1 7.7 6.4 9.1 7.4 7.3 7.7 7.5 6.4 5.6 7.6 6.6 6.2 6.9

1.6 2.5 1.7 2.4 1.9 2.0 2.5 1.7 1.3 1.3 2.3 0.8 1.7 1.7

12.3 11.0 10.3 13.3 13.4 11.8 11.7 12.3 10.0 10.6 18.7 8.3 10.7 11.9

contrast to materials doped by magnesium and potassium, respectively (Table SI4). It is worth noting that mechanism of thermo-chemical conversion of cellulose can be very complex. First of all the thermal decomposition of the feedstock and subsequent cracking, dehydratation, decarbonylation and decarboxylation of the reaction intermediates take place. Moreover, undesirable reactions may occur, such as oligomerization and formation of carbon deposit. All of them have a direct impact on the yield of permanent gases production [47]. Role of nickel as a catalyst for the cleavage of CeO and CeC bonds was previously reported in Ref. [27]. This results in the facilitation of hydrogenation reaction leading to more efficient hydrogen formation. The presence of residual water in the reaction mixture can promote water-gas shift (WGS) reaction to some extent which leads to the production of small amount of CH4 among others. The studies described by M. Yung et al. [48] and our preliminary investigations showed that the activity of catalysts containing nickel does not really depend on the initial oxidation state of metal. It is connected with the possibility of the reduction of NiO to Ni by H2 and/or CO generated in situ during high temperature treatment of lignocellulosic feedstock. As mentioned earlier, Ni0 species can strongly facilitate hydrogenation of primary products of pyrolysis. However, the catalyst containing nickel in the reduced state is subjected to continuous deactivation by carbon deposition. The presence of small amount of NiO (which left after initial reduction process or was formed during the interaction between Ni0 and reaction intermediates (e.g. oxygenates or carbon dioxide))

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

10

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

Fig. 6 e Influence of preparation method on the efficiency of modified catalysts.

may increase the amount of Lewis acid sites on the catalyst surface or limit the rate of coke formation by the release of oxygen which in turn can participate in carbon gasification process. ZrO2 is known for the ability to enhance the stability of the active phase of the catalyst due to interaction with Ni species and their partial decoration [10]. Zirconia can be also responsible for the generation of some amount of Lewis basic sites on the surface of the catalyst. However, as it was demonstrated earlier an introduction of nickel on the surface of ZrO2 modified by CaO led to almost 60% increase in hydrogen production in comparison to Ni/ZrO2 catalyst. It should be noted that an addition of alkali or alkaline earth metal oxides can be responsible for providing extra oxygen facilitating in situ regeneration of the active phase and limiting carbon deposition (according to the results of thermogravimetric analysis the amount of coke for the most active Ni/ CaO-ZrO2 catalyst was about 20% lower than that observed for unmodified Ni/ZrO2 sample). It was connected with the presence of mobile oxygen atoms generated by the formation of oxygen vacancies in the lattice of the support. The enhanced resistance to coking resulted in the increase in the efficiency of permanent gases formation [16,49]. An interpretation of the obtained results exhibited also that morphology and pore size of the synthesized catalysts are very important factors. The most active samples, prepared by sol-gel method, had larger pore radius (in the range of 7.4 nme21.6 nm) than the catalysts supported on the materials synthesized by impregnation or precipitation (pore radius between 2.4 nm and 4.1 nm). In addition to the possibility of more efficient penetration of the pore structure by the intermediates of cellulose decomposition, the use of the support containing larger pores hinders their blockage by carbon deposit and increases accessibility of the nickel species present on the studied catalyst.

Conclusions The investigations of the product distribution of thermochemical conversion of cellulose revealed that an

introduction of alkali and alkaline earth metals to the zirconia support considerably enhanced the H2 yield in comparison to unmodified Ni/ZrO2 sample. The obtained results demonstrated that the most important factors impacting an activity of the catalysts were ionic radius of dopant and type of method of its introduction leading to obtain optimal distribution of alkali/alkaline earth metals on the support surface and desired porous structure of the catalyst. The highest amount of hydrogen was formed in the presence of the catalyst containing CaO-ZrO2 support prepared by sol-gel method.

Acknowledgement The authors would like to thank Dr Waldemar Maniukiewicz for the XRD measurements and Dr Karolina Chałupka for TGDTA-MS analysis.

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijhydene.2018.10.059.

references

[1] Chen D, He L. Towards an efficient hydrogen production from biomass: a review of processes and materials. ChemCatChem 2011;3:490e511. https://doi.org/10.1002/ cctc.201000345. ~ a MA, Fierro JLG. Hydrogen production [2] Navarro RM, Pen reactions from carbon feedstocks: fossil fuels and biomass. Chem Rev 2007;107:3952e91. https://doi.org/10.1021/ cr0501994. [3] Demirbas‚ A. Mechanisms of liquefaction and pyrolysis reactions of biomass. Energy Convers Manag 2000;41:633e46. https://doi.org/10.1016/S0196-8904(99)00130-2. [4] Collard FX, Blin J. A review on pyrolysis of biomass constituents: mechanisms and composition of the products obtained from the conversion of cellulose, hemicelluloses

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

and lignin. Renew Sustain Energy Rev 2014;38:594e608. https://doi.org/10.1016/j.rser.2014.06.013. Bulushev DA, Ross JRH. Catalysis for conversion of biomass to fuels via pyrolysis and gasification: a review. Catal Today 2011;171:1e13. https://doi.org/10.1016/j.cattod.2011.02.005.  ralski J, Grams J, Paryjczak T, Rzez nicka I. Investigation of the Go coke deposit on Ni-Al2O3 and Co-Al2O3 catalysts. Carbon 2002;40:2025e8. https://doi.org/10.1016/S0008-6223(02)00145-8. Ruppert AM, Grams J, Matras-Michalska J, Chełmicka M, Przybysz P. ToF-SIMS study of the surface of catalysts used in biomass valorization. Surf Interface Anal 2014;46:726e30. https://doi.org/10.1002/sia.5370.  ralski J, Szczepaniak B. Time-of-flight secondary Grams J, Go ion-mass spectrometry as a new technique for the investigations of the deactivation process of hydrodechlorination catalysts. Russ J Phys Chem A 2007;81:1515e20. https://doi.org/10.1134/S0036024407090336. Lu Y, Li S, Guo L, Zhang X. Hydrogen production by biomass gasification in supercritical water over Ni/gAl2O3 and Ni/ CeO2-gAl2O3 catalysts. Int J Hydrogen Energy 2010;35:7161e8. https://doi.org/10.1016/j.ijhydene.2009.12.047.  ski W. Ruppert AM, Niewiadomski M, Grams J, Kwapin Optimization of Ni/ZrO2 catalytic performance in thermochemical cellulose conversion for enhanced hydrogen production. Appl Catal B Environ 2014;145:85e90. https://doi.org/10.1016/j.apcatb.2013.01.010. Takeguchi T, Furukawa S-N, Inoue M, Eguchi K. Autothermal reforming of methane over Ni catalysts supported over CaOeCeO2eZrO2 solid solution. Appl Catal A Gen 2003;240:223e33. https://doi.org/10.1016/S0926-860X(02) 00449-0. Yang RX, Chuang KH, Wey MY. Hydrogen production through methanol steam reforming: effect of synthesis parameters on Ni-Cu/CaO-SiO2 catalysts activity. Int J Hydrogen Energy 2014;39:19494e501. https://doi.org/10.1016/ j.ijhydene.2014.09.140. Vamvuka D, Tsamourgeli V, Galetakis M. Study on catalytic combustion of biomass mixtures with poor coals. Combust Sci Technol 2014;186:68e82. https://doi.org/10.1080/ 00102202.2013.846331. Sengupta S, Deo G. Modifying alumina with CaO or MgO in supported Ni and Ni-Co catalysts and its effect on dry reforming of CH4. J CO2 Util 2015;10:67e77. https://doi.org/ 10.1016/j.jcou.2015.04.003. Zheng Q, Farrauto R, Chau Nguyen A. Adsorption and methanation of flue gas CO2 with dual functional catalytic materials: a parametric study. Ind Eng Chem Res 2016;55:6768e76. https://doi.org/10.1021/acs.iecr.6b01275. Nichele V, Signoretto M, Pinna F, Menegazzo F, Rossetti I, Cruciani G, et al. Ni/ZrO2 catalysts in ethanol steam reforming: inhibition of coke formation by CaO doping. Appl Catal B Environ 2014;150e151:12e20. https://doi.org/10.1016/ j.apcatb.2013.11.037. Wang C, Sun N, Zhao N, Wei W, Sun Y, Sun C, et al. Coking and deactivation of a mesoporous Ni-CaO-ZrO2 catalyst in dry reforming of methane: a study under different feeding compositions. Fuel 2015;143:527e35. https://doi.org/10.1016/ j.fuel.2014.11.097. Si L, Wang C, Sun N, Wen X, Zhao N, Xiao F, et al. Influence of preparation conditions on the performance of Ni-CaO-ZrO2 catalysts in the tri-reforming of methane. J Fuel Chem Technol 2012;40:210e5. https://doi.org/10.1016/S18725813(12)60011-5. Chen QJ, Zhang J, Jin QW, Pan BR, Kong WB, Zhao TJ, et al. Effect of reflux digestion treatment on the catalytic performance of Ni-CaO-ZrO2 nanocomposite catalysts for CO2 reforming of CH4. Catal Today 2013;215:251e9. https:// doi.org/10.1016/j.cattod.2013.06.011.

11

[20] Lertwittayanon K, Atong D, Aungkavattana P, Wasanapiarnpong T, Wada S, Sricharoenchaikul V. Effect of CaO-ZrO2 addition to Ni supported on g-Al2O3 by sequential impregnation in steam methane reforming. Int J Hydrogen Energy 2010;35:12277e85. https://doi.org/10.1016/ j.ijhydene.2010.08.098. [21] Liang H, Wu S, Hong Y, Li S, Chen Y, Yu X, et al. Influence of alkali metals with different ionic radius doping into Ce0.7Zr0.3O2 on the active oxygen. Catal Lett 2014;144:685e90. https://doi.org/10.1007/s10562-014-1195-7. [22] Barroso-Quiroga MM, Castro-Luna AE. Catalytic activity and effect of modifiers on Ni-based catalysts for the dry reforming of methane. Int J Hydrogen Energy 2010;35:6052e6. https://doi.org/10.1016/j.ijhydene.2009.12.073. [23] Nagaraja BM, Bulushev DA, Beloshapkin S, Ross JRH. The effect of potassium on the activity and stability of Ni-MgOZrO2 catalysts for the dry reforming of methane to give synthesis gas. Catal Today 2011;178:132e6. https://doi.org/ 10.1016/j.cattod.2011.08.040. [24] Xu Z, Li Y, Zhang J, Chang L, Zhou R, Duan Z. Bound-state Ni species - a superior form in Ni-based catalyst for CH4/CO2 reforming. Appl Catal A Gen 2001;210:45e53. https://doi.org/ 10.1016/S0926-860X(00)00798-5. [25] Stonor M, Ferguson T, Chen J, Park AA. Biomass conversion to H2 with substantially suppressed CO2 formation in the presence of Group I & Group II hydroxides and a Ni/ZrO2 catalyst. Energy Environ Sci 2015;5:1702e6. https://doi.org/ 10.1039/c4ee04103h. [26] Qiao K, Shi X, Zhou F, Chen H, Fu J, Ma H, et al. Catalytic fast pyrolysis of cellulose in a microreactor system using hierarchical ZSM-5 zeolites treated with various alkalis. Appl Catal Gen 2017;547:274e82. https://doi.org/10.1016/ j.apcata.2017.07.034. [27] Grams J, Niewiadomski M, Ryczkowski R, Ruppert AM,  ski W. Activity and characterization of Ni catalyst Kwapin supported on CeO2-ZrO2 for thermo-chemical conversion of cellulose. Int J Hydrogen Energy 2016;41:8679e87. https:// doi.org/10.1016/j.ijhydene.2015.11.140. [28] Grams J, Potrzebowska N, Goscianska J, Michalkiewicz B, Ruppert AM. Mesoporous silicas as supports for Ni catalyst used in cellulose conversion to hydrogen rich gas. Int J Hydrogen Energy 2016;41:8656e67. https://doi.org/10.1016/ j.ijhydene.2015.12.146. [29] Matras J, Niewiadomski M, Ruppert A, Grams J. Activity of Ni catalysts for hydrogen production via biomass pyrolysis. Kinet Catal 2012;53:565e9. https://doi.org/10.1134/ S0023158412050096. [30] Chuah GK, Jaenicke S. The preparation of high surface area zirconia d influence of precipitating agent and digestion. Appl Catal A Gen 1997;163:261e73. https://doi.org/10.1016/ S0926-860X(97)00103-8. [31] Lu H, Yang X, Gao G, Wang K, Shi Q, Wang J, et al. Mesoporous zirconia-modified clays supported nickel catalysts for CO and CO2 methanation. Int J Hydrogen Energy 2014;39:18894e907. https://doi.org/10.1016/ j.ijhydene.2014.09.076. [32] Ali TT, Narasimharao K, Ahmed NS, Basahel S, Al-Thabaiti S, Mokhtar M. Nanosized iron and nickel oxide zirconia supported catalysts for benzylation of benzene: role of metal oxide support interaction. Appl Catal A Gen 2014;486:19e31. https://doi.org/10.1016/j.apcata.2014.08.012. [33] Takano H, Shinomiya H, Izumiya K, Kumagai N, Habazaki H, Hashimoto K. CO2 methanation of Ni catalysts supported on tetragonal ZrO2 doped with Ca2þ and Ni2þ ions. Int J Hydrogen Energy 2015;40:8347e55. https://doi.org/10.1016/ j.ijhydene.2015.04.128. [34] Maia TA, Assaf EM. Catalytic features of Ni supported on CeO2-ZrO2 solid solution in the steam reforming of glycerol

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059

12

[35]

[36]

[37]

[38]

[39]

[40]

[41]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 2

for syngas production. RSC Adv 2014;4:31142e54. https:// doi.org/10.1039/c4ra02886d. Grosvenor AP, Biesinger MC, Smart RSC, McIntyre NS. New interpretations of XPS spectra of nickel metal and oxides. Surf Sci 2006;600:1771e9. https://doi.org/10.1016/ j.susc.2006.01.041. Liu J, Liao M, Imura M, Tanaka A, Iwai H, Koide Y. Low onresistance diamond field effect transistor with high-k ZrO2 as dielectric. Sci Rep 2014;4:2e6. https://doi.org/10.1038/ srep06395. Sinhamahapatra A, Jeon JP, Kang J, Han B, Yu JS. Oxygendeficient zirconia (ZrO2-x): a new material for solar light absorption. Sci Rep 2016;6:1e8. https://doi.org/10.1038/ srep27218. Teterin YA, Vukchevich L, Ivanov KE, Utkin IO, Teterin AY, Maslakov KI, et al. XPS studies of ceramics with pyrochlor structure for radiactive waste disposal. In: Proc 5th Int Yugosl Nucl Soc Conf (YUNSC-2004) Belgrade, Serbia Montenegro, Sept 27-30, 2004, vol. 2; 2005. p. 253e8. Wu¨rz R, Rusu M, Schedel-Niedrig T, Lux-Steiner MC, € vecker M, et al. In situ X-ray photoelectron Bluhm H, Ha spectroscopy study of the oxidation of CuGaSe2. Surf Sci 2005;580:80e94. https://doi.org/10.1016/j.susc.2005.01.054. Chen X, Nouri A, Li Y, Lin J, Hodgson PD, Wen C. Effect of surface roughness of Ti, Zr, and TiZr on apatite precipitation from simulated body fluid. Biotechnol Bioeng 2008;101:378e87. https://doi.org/10.1002/bit.21900. Pokrovski K, Jung KT, Bell AT. Investigation of CO and CO2 adsorption on tetragonal and monoclinic zirconia. Langmuir 2001;17:4297e303. https://doi.org/10.1021/la001723z.

[42] Hertl W. Surface chemistry of zirconia polymorphs. Langmuir 1989:96e100. https://doi.org/10.1021/la00085a018. [43] Di Cosimo JI, Dıez VK, Xu M, Iglesia E, Apesteguia CR. Structure and surface and catalytic properties of Mg-Al basic oxides. J Catal 1998;178:499e510. https://doi.org/10.1006/ jcat.1998.2161. [44] Bachiller-Baeza B, Rodriguez-Ramos I, Guerrero-Ruiz A. Interaction of carbon dioxide with the surface of zirconia. Langmuir 1998:3556e64. https://doi.org/10.1021/la970856q. [45] Solymosi F, Knozinger H. Infrared spectroscopic study of the adsorption and reactions of CO2 on K-modified Rh/SiO2. J Catal 1990;122:166e77. https://doi.org/10.1016/0021-9517(90) 90268-O. [46] Sun N, Wen X, Wang F, Peng W, Zhao N, Xiao F, et al. Catalytic performance and characterization of Ni-CaO-ZrO2 catalysts for dry reforming of methane. Appl Surf Sci 2011;257:9169e76. https://doi.org/10.1016/ j.apsusc.2011.05.127. [47] Wang K, Kim KH, Brown RC. Catalytic pyrolysis of individual components of lignocellulosic biomass. Green Chem 2014;16:727e35. https://doi.org/10.1039/C3GC41288A. [48] Yung MM, Starace AK, Mukarakate C, Crow AM, Leshnov MA, Magrini KA. Biomass catalytic pyrolysis on Ni/ZSM-5 : effects of nickel pretreatment and loading. Energy Fuels 2016;30:5259e68. https://doi.org/10.1021/ acs.energyfuels.6b00239. [49] Bellido JDA, De Souza JE, M'Peko JC, Assaf EM. Effect of adding CaO to ZrO2 support on nickel catalyst activity in dry reforming of methane. Appl Catal A Gen 2009;358:215e23. https://doi.org/10.1016/j.apcata.2009.02.014.

Please cite this article in press as: Ryczkowski R, et al., Impact of the modification method of Ni/ZrO2 catalyst by alkali and alkaline earth metals on its activity in thermo-chemical conversion of cellulose, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.10.059