Application of POSS Nanotechnology for Preparation of Efficient Ni Catalysts for Hydrogen Production

POSS (polyhedral oligomeric silsesquioxanes) nanotechnology was applied for preparation of efficient Ni catalysts for hydrogen production through autothermal reforming of methane (ATR of CH4). The novel metal-POSS precursor [Nickel (II) ‒ HeptaisobutylPOSS (C4H9)7Si7O9(OH)O2Ni] of Ni nanoparticles was introduced into Ce0.5Zr0.5O2 support with following calcination and reduction stages of activation. The peculiarity of the genesis of Ni/SiO2/Ce0.5Zr0.5O2 nanomaterials and their characteristics versus deposition mode were studied by X-ray fluorescence spectroscopy, thermal analysis, N2 adsorption, X-ray diffraction, high-resolution transmission electron microscopy and H2 temperature-programmed reduction. The two kinds of supported Ni-containing particles were observed: highly dispersed Ni forms (1‒2 nm) and large Ni-containing particles (up to 50‒100 nm in size). It was demonstrated that the textural, structural, red-ox and, consequently, catalytic properties of ex-Ni-POSS catalysts depend on the deposition mode. The increase of a portion of difficultly reduced Ni2+ species is found upon application of intermediate calcination during Ni-POSS deposition that has detrimental effect on the activity of catalyst in ATR of CH4. The Ni/SiO2/Ce0.5Zr0.5O2 catalyst prepared by one-step Ni-POSS deposition exhibits the highest H2 yield ‒ 80% at T = 800 °C. Article info Received: 14 August 2016 Received in revised form: 2 October 2016 Accepted: 28 November 2016


Introduction
Development of efficient and low cost hydrogen production technologies is an urgent task due to the increased demand of clean energy generation [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. At present more than 50 million tons of hydrogen are produced annually worldwide and much of this hydrogen is used in the chemical and refinery industries [15]. In the future, it is expected that H 2 could be widely used for power generation and in transport by fueling gas turbines, fuel cells and combustion engines [1][2][3][4][5][6][7][8][9][10][11]. The H 2 consumption rate steadily growths in the world and the 30% increase of global H 2 production is predicted in the next 5 years [15]. There are different technologies of hydrogen production using nonrenewable (natural gas reforming, coal gasification) as well as renewable (biomass processes, biological production etc.) resources, nuclear energy (high-temperature water splitting) and electricity (electrolysis of water) [3][4][5][6][7][8][9]. Although the "green" or carbon-neutral path from current fossil-based to future hydrogen economy is preferable and provides sustainable development [1][2][3][4], to date H 2 is usually produced from fossil fuels without CO 2 capture and storage [16,10,13]. Natural gas remains main feedstock for hydrogen production [8,13,14,16,17]. The typical technologies for production of hydrogen from natural gas are steam methane reforming (SMR), partial oxidation (POX) and autothermal reforming (ATR). ATR of CH 4 is actually the combined process of POX and SMR in one reactor. The generated heat from POX is used for an endothermic SMR reaction, thus providing advantages of ATR technology for hydrogen production. Ni-based catalysts have been used and widely studied in methane reforming because of their high catalytic activity and low price [16][17][18][19][20][21][22][23][24]. But under long exposure [44], co-precipitation [45], deposition-precipitation [46], combustion [47,48], sol-gel [49], surfactant assisted and polyol [50] methods. The use of methods which provide the realization of a strong metal-support interaction has positive effect on the long-term stability of catalysts in hydrocarbon reforming processes [51,52]. In particular, it was shown that homogeneous deposition precipitation (HDP) method as compared with pore volume impregnation (PVI) method provides 20 wt.%Ni/MgAl 2 O 4 catalyst with smaller Ni particle size [23]. The average Ni particle size is equal to 4.5 and 8.1 nm for the ex-HDP and ex-PVI catalysts, respectively. In this case, the reaction rate of catalysts in SMR increases linearly with the Ni dispersion, while the content of carbonaceous deposits decreases. Similar conclusion was made at comparative study of Ni catalysts synthesized by the polyol and surfactant-assisted methods [50]. In dry reforming of methane the polyol catalysts displayed the highest activity and selectivity, which can be connected with improved Ni dispersion and presence of stronger metal-support interaction in these catalysts.
As the precursors of the Ni active component, as a rule, inorganic salts of this metal are used. It is established that for higher Ni dispersion using nickel nitrate or nickel acetate as a nickel metal precursor is preferred as against the use of nickel chloride or nickel formate [53]. In order to obtain a high dispersion state for Ni active components, the addition of organic agents such as poly(N-vinyl-2-pyrrolidone) (PVP) [54], citric acid [55], sorbitol or glucose [56] into impregnation solution is made. For example, the application of glucose additive allows decreasing of the average Ni particle size from 7.2 to 3.5 nm in 10 wt.%Ni/Al 2 O 3 samples [56], while the use of poly(N-vinyl-2-pyrrolidone) leads to reduction of the average Ni particle size from 13.0 to 4.3 nm in 5 wt.%Ni/SBA-15 samples [54]. Catalysts prepared with PVP showed better stability in methane reforming. The presence of organic compounds controls the re-distribution of Ni species across the support and their sintering during the activation procedures, thus providing the stabilization of Ni nanoparticles of high dispersion and, consequently, high catalytic properties of materials.
Exploitation of the so-called "POSS nanotechnology" (POSS = polyhedral oligomeric silsesquioxanes) in which metal-containing silsesquioxanes are used as precursors for the preparation of nanomaterials has attracted attention as a tool for synthesis of catalysts with novel properties and functionalities. This method of nanosized catalysts preparation has distinct advantage over traditional methods because the metal atoms remain highly dispersed throughout support [57][58][59]. The study of the Fe-, Cr-, Mg-and Al-containing silicas derived through controlled calcination of appropriate metallsilsesquioxanes showed that nearly homogenous metal dispersion could be achieved [60][61][62]. In the case of Cr-and Fe-containing silicas the small amount of oxide particles was additionally observed.
In this work we have expanded the POSS nanotechnology method for the preparation of Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 catalysts. The peculiarity of the genesis of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 nanomaterials and their performance in the ATR of CH 4 were thoroughly studied. Using X-ray fluorescence spectroscopy, thermal analysis, N 2 adsorption, X-ray diffraction, high-resolution transmission electron microscopy and H 2 temperature-programmed reduction methods the textural, structural and redox properties of the prepared nanomaterials were examined in detail. The results were analyzed in comparison with those for Ni/SiO 2 and Ni/Ce 0.5 Zr 0.5 O 2 catalysts prepared by conventional method.

Support and catalyst preparation
Ce 0.5 Zr 0.5 O 2 support was prepared by polymerizable complex method [44,63]. Aqueous solutions of cerium nitrate Ce(NO) 3 • 6H 2 O and zirconium oxychloride ZrOCl 2 • 8H 2 O taken in the equimolar ratio were added to the solution of citric acid (CA) in ethylene glycol (EG) at 70 °C. Then ethylene diamine (ED) was mixed with this solution at 50 °C, and the mixture was kept at 70 °C until a highly viscous polymeric gel was formed. The molar ratio of the reagents (Ce + Zr):CA:EG:ED was 1:4:11:4. The obtained product was subjected to two-step calcination in air: at 300 °C for 4 h and at 600 °C for 4 h.

Characterization of catalysts
Elemental composition of the catalysts was determined by X-ray fluorescence analysis in a spectrometer ARL PERFORM'X (Thermo Techno Scientific) with a Rh anode of the X-ray tube.
XRD analysis of the catalysts was carried out in an HZG-4C diffractometer (Freiberger Prazisionmechanik) with CoK α radiation The phase composition was determined by X-ray diffraction (XRD) in the 2Theta angle range of 10-80 degrees. The phase identification was carried out using JCPDS database.
The BET specific surface area and porosity texture of support and catalyst samples were determined by nitrogen adsorption measurements at liquid nitrogen temperature with an automatic Micromeritics ASAP 2400 instrument. Before the measurements the samples were degassed at 150 °C for 4 h.
The Brunauer-Emmett-Teller (BET) and Barret-Joyner-Halenda (BJH) methods were applied for calculation of the specific surface area, pore volume and pore size distribution.
The thermal analysis (TA) (thermogravimetric (TG), differential thermogravimetric (DTG) and differential thermal analysis (DTA)) was carried out in a Netzsch STA 449C apparatus. Catalysts were tested over the temperature range from room temperature up to 900 °C at the heating rate of 10 °C/min in air.
The transmission electron microscopy (TEM) images of the catalysts were obtained in a JEM 2010 microscope (JEOL) operated at 200 kV. The structural resolution was 0.14 nm. The samples were deposited on a copper grid supporting a perforated carbon film. Local Energy Dispersive X-ray (EDX) microanalysis was made in an EDAX analyzer equipped with a Si (Li) detector with a resolution of 130 eV.
Hydrogen TPR studies were carried out in a flow quartz reactor with an i.d. of 5 mm. Samples of ca. 50 mg were loaded into the reactor, pre-treated in oxygen flow at 500 °C for 0.5 h and cooled down to room temperature. Then the gas stream was switched to 10%H 2 /Ar. The reactor was heated at a rate of 10 °C/min to 800 °C. Hydrogen consumption was determined quantitatively by a thermal conductivity detector. To minimize local catalysts overheating, the loaded catalyst sample (0.5 g, fraction 0.25-0.50 mm) was diluted by an inert material having higher heat conductivity -1.5 g of β-SiC of the same fraction. Analysis of the initial gas mixture and the reaction products was performed in a QMS 300 mass-spectrometric gas analyzer (Stanford Research Systems) using peak intensity calibrations obtained with model gas mixtures.

Catalytic activity tests
Yields of H 2 and CO were determined as percent of the amounts of products produced by the reaction from maximally possible amounts, using the following equations: where Y H2 is a yield of H 2 , %; V H2out is a molar rate of H 2 at the reactor outlet, mol/min; V CH4in is a molar rate of CH 4 introduced into the reactor, mol/min; V H2Oin is a molar rate of H 2 O fed into the reactor, mol/min; and Y CO = 100%•VCOout/V CH4in , where Y CO is a yield of CO, % V COout is a molar rate of CO at the reactor outlet, mol/min; V CH4in is a molar rate of CH 4 introduced into the reactor, mol/min.
The relative experimental error in the determination of methane conversion and product yields did not exceed 10%. Table 1 shows the sample formula and chemical composition of prepared catalysts. It can be noted that according to actual chemical composition of samples, the Si/Ni atomic ratio in the Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 catalysts differ from Si/Ni atomic ratio in Ni-POSS (namely, 3 vs. 7). Theoretically the Ni content should be lower (ca. 4%) while the Si content should be higher (ca. 13%). The deviation of the obtained values from nominal specified values can be connected with loss of Si-organic fragments during thermal treatment because of its high volatility [64].

Characterization of the catalysts
To study the peculiarity of Ni catalyst genesis, the samples dried at 120 °C were characterized by the thermal analysis. The typical TG, DTG and DTA curves of dried samples are shown in Fig.  1. When the dried Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -1 sample was heated from the room temperature up to 900 °C (Fig. 1, upper left quarter), there were three pronounced exo-effects at 200, 300 and 445 °C on the DTA curve. Meanwhile, it has lost 6.5 wt.% at 150-250 °C, 6.5 wt.% at 250-400 °C and 1.6 wt.% at 400-500 °C. The origin of the weight loss is the decomposition of supported Ni-POSS. The results of TA suggest that it is a multi-stage process, which is accompanied by organic group oxidation with release of carbon oxides and water. The thermal degradation of supported Ni-POSS is completed at 600 °C, and this value of temperature was applied for the calcination of dried Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 samples. It can be also noted that the Ni-POSS thermal degradation behavior in supported Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 and bulk samples is quite similar (Fig.  1, upper and lower left quarters). For both samples three exo-effects accompanied by weight loss are observed. However, for supported Ni-POSS the peak distribution is inverted with a considerable increase of peak 2 in relation to peak 1. This indicates that the thermooxidative degradation of supported Ni-POSS has a more intensive mode in the second stage (ca. 300 °C) while that for bulk Ni-POSS is more pronounced at the first stage (ca. 235 °C).
In addition, the supported Ni-POSS differs by shifting of observed exo-effects in the low-temperature region (200 vs. 235 °C, 300 vs. 360 °C and 445 °C vs. 535 °C). The decrease of temperature of Ni-POSS decomposition in the supported state is probably related to variable valence of cerium (Ce 3+ /Ce 4+ ) in Ce 0.5 Zr 0.5 O 2 support that promotes the reaction of oxidation of organic compounds [65].
In contrast to the degradation of supported and bulk Ni-POSS that goes with the exo-effects, the decomposition of supported Ni nitrate is accompanied by endo-effects (Fig. 1, upper and     The N 2 adsorption-desorption isotherms, pore size distributions and the textural properties of the samples are given in Fig. 2 and Table 2, respectively. It is shown ( Table 2) that introduction of the Ni-POSS into the Ce 0.5 Zr 0.5 O 2 support (S BET = 65 m 2 /g) and further high-temperature calcination of the samples lead to an increase of S BET by ~ 2-2.5 times in the case of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -1 and Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 -2 samples. In the case of three steps deposition of Ni-POSS with intermediate calcination, the S BET of support and catalyst are equal. The average pore diameter decreases during Ni-POSS deposition on the mixed oxide support (D pore = 8.0 nm) and becomes in the range of ~ 3.7-5.5 nm. Impregnation of the Ce 0.5 Zr 0.5 O 2 and SiO 2 (S BET = 325 m 2 /g) supports with the aqueous solutions of Ni precursor and further calcination leads to a decrease of S BET by ~ 20% and ~ 5%, respectively. The average pore diameter practically does not change during Ni nitrate salt deposition on the supports.
The N 2 adsorption-desorption isotherms, pore size distributions and the textural properties of the samples are given in Fig. 2 and Table 2, respectively. It is shown ( Table 2)   The N 2 adsorption on Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 and Ni/Ce 0.5 Zr 0.5 O 2 catalysts shows a type IV isotherm (Fig. 2), H3-type hysteresis loop according to the IUPAC classification and capillary condensation step at relative pressure above 0.4, which are characteristic for mesoporous materials [70]. The Ni/ SiO 2 catalyst is characterized by a type IV isotherm with H1-type hysteresis loop at relative pressure above 0.8, that indicates the existence of mainly textural mesoporosity [71].
The BJH analysis of desorption isotherms revealed that the Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts exhibited narrow pore size distributions with a mean pore size of 3.7-5.5 nm. The Ni/Ce 0.5 Zr 0.5 O 2 and, especially, Ni/SiO 2 samples showed a broader pore size distributions with larger average pore diameters (Fig. 2, Table 2).
The surface area of the Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts which have different deposition strategy decreased from 140 m 2 /g in Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -1 to 65 m 2 /g in Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -3 as a consequence of the increasing of number of deposition-calcination steps. In overall, the value of S BET increases in the order: Ni/Ce 0.5 Zr 0.5 O 2 < Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 < Ni/SiO 2 , that may have effect on the active component dispersion in the support matrix. The higher S BET value of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 samples in comparison with those of Ni/Ce 0.5 Zr 0.5 O 2 is likely connected with the presence of amorphous SiO 2 (~20 wt.%) in the catalyst composition, which is characterized by higher value of S BET . This tendency is typical for Ce 0.5 Zr 0.5 O 2 -SiO 2 composite materials [72].
According to the XRD data (Fig. 3), the ceria-based cubic solid solution (JCPDS 34-0394) and NiO (JCPDS 44-1159) are main phases in the fresh Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 and NiCe 0.5 Zr 0.5 O 2 catalysts after calcination at 600 °C. The characterization by XRD of fresh catalysts prepared using different modes of Ni-POSS deposition does not allow to find differences among them. In the Ni/ SiO 2 catalyst the NiO phase and SiO 2 amorphous phase are observed. The reduction of samples in the hydrogen at 800 °C as a typical stage of catalyst activation for ATR of CH 4 leads to an expected transformation of NiO into metallic Ni phase (Fig. 3).
The average crystallite sizes of active component phases were estimated by the Debye-Scherrer equation and are presented in Table 3. The obtained values of NiO and Ni average particle size lie in the typical range of such values for Ni supported on SiO 2 , CeO 2 -ZrO 2 or SiO 2 -CeO 2 -ZrO 2 [72][73][74]. For the fresh samples, data show that the NiO particles of equal mean size form in all Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts regardless of the deposition mode used in their synthesis. The results also demonstrate that for fresh samples average size of NiO particles increases from 18 to 30 nm in the following order of catalysts: Ni/SiO 2 < Ni/Ce 0.5 Zr 0.5 O 2 < Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 (Table 3). It is noted that there is no straightforward correlation between the value of S BET and the average NiO (or Ni) particle size for the studied samples.
In the catalysts after reduction the average size of Ni particles is equal ~ 25 nm in all studied samples with the exception of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -3 ( Table 3). The average size of Ni-containing particles in reduced Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 samples is comparable with or lower than those obtained for fresh catalysts. So, according to the XRD data, the nanoparticles were stable against sintering during high-temperature activation at 800 °C in the reducing medium. It is known that the strong interaction between metal and support increases the dispersion of Ni-containing species and decreases their particle size, that can be realized in the reduced Ni/SiO 2 / Ce 0.5 Zr 0.5 O 2 -3 catalyst. On the other hand, some agglomeration of Ni-containing particles is observed in the reduced Ni/Ce 0.5 Zr 0.5 O 2 and Ni/SiO 2 catalysts compared to their fresh states. TEM data illustrate that support of the fresh and reduced Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts consists of SiO 2 particles about 500 nm in size and agglomerates of Ce 0.5 Zr 0.5 O 2 crystallites within 5-10 nm in size (Figs. 4a, d). There are intimate contacts between SiO 2 and Ce 0.5 Zr 0.5 O 2 : the CeZeO 2 particles (2-10 nm) were found on the surface of SiO 2 particles (up to 500 nm), while Ce 0.5 Zr 0.5 O 2 crystallites are partly covered by SiO 2 (Fig. 4b). The NiO particles of 1-2 nm in size in the fresh catalysts (Fig. 4c) and Ni particles of 3-4 nm in size in the reduced catalysts (Fig. 4f) are mainly found on SiO 2 . According to the EDX data, highly dispersed Ni species are present on/in the Ce 0.5 Zr 0.5 O 2 region of support (Figs. 4b, e). In addition, the large Ni-containing particles (up to 50-100 nm in size) are also observed.
Further, using TEM for the comparison of nanostructure of Ni/SiO 2 and Ni/Ce 0.5 Zr 0.5 O 2 shows that for both fresh catalysts large NiO particles are observed on the supports: the NiO particle sizes are 10-50 nm for SiO 2 (Figs. 5a, 4b) and 10-100 nm for Ce 0.5 Zr 0.5 O 2 (Fig. 5d). According to the EDX data, in the Ni/Ce 0.5 Zr 0.5 O 2 catalyst the Ni-containing species are also present on Ce 0.5 Zr 0.5 O 2 in the highly dispersed form, or they are included in the Ce 0.5 Zr 0.5 O 2 structure (Fig. 5e). In contrast to Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts, in the case of Ni/ SiO 2 and Ni/Ce 0.5 Zr 0.5 O 2 samples after reduction, formation of the Ni agglomerated particles occurs (Fig. 5c). The particle sizes increase up to 100 nm for Ni/SiO 2 and up to 500 nm for Ni/Ce 0.5 Zr 0.5 O 2 . Nevertheless, EDX data show that dispersed Ni species are retained in the reduced Ni/Ce 0.5 Zr 0.5 O 2 catalyst (Fig. 5f). ii) big Ni-containing particles is polycrystalline solid; iii) wide Ni particle size distribution with a maximum at 20-30 nm takes place.
The H 2 -TPR profiles associated with the studied Ni-based catalysts are given in Fig. 6. In particular, it can be seen that all the Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts present two hydrogen consumption peaks with the temperature maxima T 1 in the range 350-380 °C and T 2 in the range 560-570 °C (Table 4), which can be assigned to reduction of NiO species with weak and intimate metal-support interaction, respectively [75]. While the maximum temperature of peaks is not very different for Ni/ SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts, the peak distribution is changed. As follows data in Table 4, some increase of peak 2 in relation to peak 1 (from 0.6 to 1) is observed in the following order of the samples: Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -1 < Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -2 < Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -3. The increase of peak 2 intensity suggests an increase of the fraction of species with strong metal-support interaction. The metal-support interaction becomes stronger at the application of intermediate calcination during Ni-POSS deposition.
NiO bulk sample is known to exhibit a single reduction peak in H 2 -TPR curve, whose position depends on the experimental conditions, for example, ca. 330 °C [76], 370 °C [77] or 530 °C [75]. In contrast, the reduction of supported NiO led to the appearance of several peaks in the H 2 -TPR profile. Usually this is assigned to different forms of metal-support interaction [75,78,79]. It is suggested [80] that higher NiO reduction temperature can be due to the presence of particles with higher dispersion, interacting stronger with the support, while the lower reduction temperature can be related to larger particles with a lower strength of interaction with the support (temperature of reduction closer to the pure NiO sample). The highly dispersed Ni species (1-2 nm) and large Ni-containing particles (up to 50-100 nm in size) were observed by HR-TEM, which is consistent with two peaks in the H 2 -TPR experiments and indicative of realization of both weak and strong metal interaction with the support.
The profile of H 2 -TPR curve similar to those of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 samples is obtained for the Ni/Ce 0.5 Zr 0.5 O 2 catalyst (Fig. 6). However, in this case the maxima of hydrogen consumption peaks shift to the low-temperature region (Table  4), that can be assigned to the weakening of metal-support interaction and increase of NiO particle size. H 2 -TPR profile associated with the Ni/SiO 2 catalyst shows hydrogen consumption peak with a maximum T 1 at 390 °C and feebly marked shoulder at 475 °C. Usually such peak was assigned to the reduction of "free state" NiO bearing weak interaction with the support [28]. It is noted that value of H 2 /Ni molar ratio is over-stoichiometric (higher than 1) for both Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 and Ni/Ce 0.5 Zr 0.5 O 2 catalysts, which can be associated with the additional reduction of Ce 0.5 Zr 0.5 O 2 solid solution [78,81].
Hence, the reduction of Ni species in the supported nickel catalysts takes place in a broad temperature range from 250 °C to 750 °C and the Ni 2+ reducibility is affected by support composition and catalyst preparation technique. The presence of H 2 consumption at temperatures above 500 °C suggests the existence of Ni species with strong metal-support interaction. As follows from the ratio of area of peak 2 to that of peak 1 (  SiO 2 catalysts was enhanced greatly. The increase in the reducibility and oxygen transfer ability of the support were believed to be the main reasons for the superior performance of Ni/Ce 0.5 Zr 0.5 O 2 -SiO 2 sample as compared to Ni/SiO 2 catalyst in the ATR of CH 4 [84]. In our case the performance of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 -2 and Ni/SiO 2 catalysts in ATR of CH 4 is close to each other. It is expected that the additional differences in catalytic activity will become apparent during prolonged time on stream, resulting from the differences in metalsupport interaction, which will be a topic of further study. Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts were prepared by deposition of novel metal-POSS [Nickel (II) -HeptaisobutylPOSS (C 4 H 9 ) 7 Si 7 O 9 (OH)O 2 Ni] on Ce 0.5 Zr 0.5 O 2 support with following calcination and reduction stages of activation. As a result of a multi-stage process of Ni-POSS decomposition proceeding in a temperature range 200-600 °C the NiO nanoparticles with mean size of 30 nm were stabilized in the support matrix. The nanoparticles were stable against sintering during high-temperature activation at 800 °C in reducing medium. The structural, redox and, especially, textural properties of Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalysts were regulated by deposition mode. Ni/SiO 2 /Ce 0.5 Zr 0.5 O 2 catalyst prepared by one-step Ni-POSS deposition gives the higher H 2 yield in ATR of CH 4 than those prepared by multiple deposition, which may be caused by a decrease of a portion of difficultly reduced Ni 2+ species. The achieved value of H 2 yield 80% at 800 °C confirms that ex-Ni-POSS catalysts have good potential for hydrogen production by the ATR of CH 4 . 262840 (Project DEMCAMER) and has received funding from Russian Academy of Sciences and Federal Agency of Scientific Organizations (project № 0303-2016-0004).