High Thermal Stability of La2O3- and CeO2-Stabilized Tetragonal

Feb 15, 2016 - School of Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, Indiana 47907, United States. ∥ Energy Sys...
0 downloads 3 Views 6MB Size
Article pubs.acs.org/IC

High Thermal Stability of La2O3- and CeO2‑Stabilized Tetragonal ZrO2 Shichao Wang,† Hong Xie,† Yuyuan Lin,† Kenneth R. Poeppelmeier,*,† Tao Li,‡ Randall E. Winans,‡ Yanran Cui,§ Fabio H. Ribeiro,§ Christian P. Canlas,∥ Jeffrey W. Elam,∥ Hongbo Zhang,⊥ and Christopher L. Marshall⊥ †

Center for Catalysis and Surface Science, Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States ‡ X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States § School of Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, Indiana 47907, United States ∥ Energy Systems Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States ⊥ Chemical Sciences and Engineering Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States ABSTRACT: Catalyst support materials of tetragonal ZrO2, stabilized by either La2O3 (La2O3−ZrO2) or CeO2 (CeO2− ZrO2), were synthesized under hydrothermal conditions at 200 °C with NH4OH or tetramethylammonium hydroxide as the mineralizer. From in situ synchrotron powder X-ray diffraction and small-angle X-ray scattering measurements, the calcined La2O3− ZrO2 and CeO2−ZrO2 supports were nonporous nanocrystallites that exhibited rectangular shapes with a thermal stability of up to 1000 °C in air. These supports had an average size of ∼10 nm and a surface area of 59−97 m2/g. The catalysts Pt/La2O3−ZrO2 and Pt/CeO2−ZrO2 were prepared by using atomic layer deposition with varying Pt loadings from 6.3 to 12.4 wt %. Monodispersed Pt nanoparticles of ∼3 nm were obtained for these catalysts. The incorporation of La2O3 and CeO2 into the t-ZrO2 structure did not affect the nature of the active sites for the Pt/ZrO2 catalysts for the water−gas shift reaction.



ture.20,21 Numerous technical routes have been developed for the synthesis of homogeneous, stable t-ZrO2 nanocrystals with a large surface area, including sol−gel,22,23 hydrothermal,13,24,25 coprecipitation,26−28 and thermal decomposition29 methods. The usual preparative procedures are based on precipitation with hydroxide from a solution containing inorganic zirconium salts (e.g., zirconyl chloride or zirconium nitrate) followed by high-temperature treatment. Among the wet chemical preparation methods, the hydrothermal route has been recognized as a time and energy saver with crystallization kinetics that are faster than those of classic sol−gel or coprecipitation methods. The main advantages of hydrothermal synthesis are related to the homogeneous nucleation processes and nanocrystal powders with well-controlled sizes and shapes. In an earlier study, we reported the synthesis of Na-stabilized t-ZrO2 supports and its application for the water−gas shift (WGS) reaction.30 Sodium was found to play a crucial role in the formation of the stable t-ZrO2 upon incorporation into the internal crystal structure of ZrO2 during the hydrothermal synthesis. Moreover, Na promoted the WGS reaction. However, the Na-stabilized t-ZrO2 was phase-stable only up to 600 °C in air,30 which limits its application for high-

INTRODUCTION Zirconia (ZrO2), because of its high mechanical and thermal stability as well as unique redox and acid−base properties, has received much research attention as both a catalyst and a catalyst support material1 in numerous reactions such as alcohol dehydration,2,3 CO4−6 and olefin7 hydrogenation, olefin isomerization,8,9 methanol synthesis,6 steam re-forming,10 and olefin epoxidation.11 At atmospheric pressure, pure zirconia exhibits three polymorphs with monoclinic (P21/c space group, 20−1170 °C), tetragonal (P42/nmc space group, 1170−2370 °C), and cubic (2370−2680 °C, Fm3-m space group) symmetries.12 The crystal phase of ZrO2 (monoclinic and tetragonal) significantly influences its catalytic activity and selectivity,6,8,13−16 and it has been found that the tetragonal phase (t-ZrO2) is the most efficient phase for some reactions.17 The sensitivity of the catalytic behavior to the crystal phase for ZrO2 renders it necessary to synthesize t-ZrO2 with a high phase purity. Compared to the monoclinic phase (m-ZrO2), tZrO2 is thermodynamically metastable at room temperatures and tends to transform to m-ZrO2 upon thermal treatment or contact with water vapor,18,19 which complicates the synthesis of pure t-ZrO2 in the form of nanosized, large-surface area powders. The formation of metastable t-ZrO2 nanocrystals at low temperature has been correlated with a critical size and the existence of amorphous ZrO2 and surface hydration struc© XXXX American Chemical Society

Received: December 3, 2015

A

DOI: 10.1021/acs.inorgchem.5b02810 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry temperature reactions.31−34 For example, t-ZrO2 needs to maintain its tetragonal phase stability at temperatures as high as 1000 °C to be used as a support for commercial three-way catalysts in a closely coupled converter. Until now, developing t-ZrO2 materials with high thermal stability has presented a challenge when aiming for a low-cost, uncomplicated synthesis route. Solid solution formation,35 generally achieved by introducing larger isovalent cations (e.g., Ce4+)31 or by doping aliovalent cations (e.g., La3+, Y3+, Sc3+, Pr3+, Mg2+, and Ca2+)12,36−46 into the ZrO2 lattice, has been established as an effective way to enhance the thermal and kinetic stability of ZrO2 into tetragonal phases at room temperature. These cations are believed to substitute for Zr4+ in the cation framework, thus creating oxygen vacancies for charge compensation.47 Here, we report the preparation of La2O3- and CeO2stabilized tetragonal ZrO2 with high thermal stability via hydrothermal synthesis and subsequent calcination. An inexpensive and nontoxic inorganic material, ZrOCl2·8H2O, was used as the Zr precursor. The La2O3- and CeO2-stabilized tetragonal ZrO2 thus prepared were used to prepare supported Pt catalysts. The water−gas shift48 reaction was used as a model reaction to investigate the effect of La2O3 and CeO2 on the Pt/ t-ZrO2 catalysts for WGS reaction.



continuous increase in the viscosity of the mixture. After being vigorously stirred for 30 min, the suspension was transferred to a 125 mL Teflon-lined autoclave and heated at 200 °C for 20 h. The autoclave was subsequently allowed to cool to room temperature. The autoclave reaction mixture was then separated by centrifugation at 6500 rpm, washed exhaustively using DI water until the mixture was free of chloride ions, and air-dried at 70 °C for 12 h. Light yellow powders were then collected and calcined in air at 700 °C for 3 h to form the final 20% CeO2−ZrO2 products. Synthesis at lower temperatures (150 and 180 °C) and calcination at low temperatures (600 and 650 °C) did not change the particle size or morphology. Syntheses of 5, 10, 30, and 40% La2O3−ZrO2 were conducted using a similar method by changing the amount of Ce precursor used. Pt/La2O3−ZrO2 and Pt/CeO2−ZrO2 Preparation. Alternating exposures to MeCpPtMe3 and O2 for Pt ALD49,50 were executed to deposit Pt on the 5, 10, and 20% La2O3−ZrO2 and 10, 20, and 30% CeO2−ZrO2 supports. This process used a stainless steel, viscous flow reactor with an ultrahigh purity (UHP) N2 carrier gas at a pressure of 1 Torr and a mass flow rate of 200 sccm.51 Approximately 1.0 g of each metal oxide support was placed into the ALD flow tube and allowed to equilibrate for 20 min at the ALD temperature, after which an ozone treatment of 10 min was performed to clean the support surface of any organic residues. Next, the Pt ALD was performed at a deposition temperature of 300 °C using 300 s exposures to MeCpPtMe3, 200 s exposures to O2, and 50 s N2 purge times between each precursor exposure. A flow of 30 sccm UHP N2 was passed through the 50 °C stainless steel MeCpPtMe3 bubbler during the Pt precursor exposures to transport the MeCpPtMe3 vapor to the metal oxide supports. To prevent condensation of the MeCpPtMe3 vapor, the inlet lines were heated to 150 °C. The resulting catalysts from the ALD procedure are summarized in Table 3. Pt loadings were expressed in Pt weight percent. Powder X-ray Diffraction (XRD). The XRD patterns of the samples were recorded at room temperature on a Rigaku UltimaIV powder X-ray diffractometer (Cu Kα radiation, Ni filter, 40 kV, 20 mA, 2θ = 20−70°, 0.1° step size, and 2 s count time). High-Resolution Transmission Electron Microscopy (HRTEM). The HRTEM images were obtained using a JEOL JEM2100F transmission electron microscope operated at 200 keV. The TEM samples were prepared by dispersing nanoparticles in ethanol to form a suspension. The suspension was sonicated for ∼30 min and was deposited onto a 300 mesh Cu grid with lacey carbon coated. Elemental Analysis. The semiquantitative energy dispersive X-ray spectroscopy (EDS) analyses of Pt/La2O3−ZrO2 and Pt/CeO2−ZrO2 were obtained using a Hitachi S-3400 scanning electron microscope equipped with a Princeton Gamma Tech (PGT) energy dispersive Xray analyzer. Data were acquired with an accelerating voltage of 25 kV and an accumulation time of 300 s. Nitrogen Sorption. Surface area measurements were determined via N2 physisorption using a Micromeritics ASAP 2020 system. Each sample was degassed by being heated at 300 °C in vacuo for 2 h prior to surface area measurement. Specific surface areas were calculated using BET (Brunauer−Emmett−Teller) theory in a range of relative pressures (P/P0) of 0.05−0.3. In Situ Temperature-Dependent Synchrotron Powder X-ray Diffraction. For the in situ heating experiment, fine powders of Nastabilized t-ZrO2,30 10% La2O3−ZrO2, or 20% CeO2−ZrO2 samples were pressed into thin films and placed onto a TS 1500 heating stage (Linkam Scientific Instruments). In situ synchrotron XRD (λ = 0.1127 Å) patterns were recorded from room temperature to 1000 °C at a rate of 20 °C/min using the 11-ID-C beamline with an X-ray energy of 110 keV at the Advanced Photon Source. Ex Situ and In Situ Synchrotron Small-Angle X-ray Scattering (SAXS). Small-angle X-ray scattering measurements were conducted at the 12-ID-B beamline at the Advanced Photon Source. Data were acquired in transmission mode using X-rays at 14 keV (λ = 0.8857 Å) and a Pilatus 2M detector (Dectris Ltd.) with typical exposure times in the range of 0.1−1.0 s. The sample−detector distance was 1.9 m. Ex situ samples were measured in 1.5 mm quartz capillaries. For in situ experiments, samples were pressed into wafers and held on a TS 1500

EXPERIMENTAL SECTION

Synthesis of La2O3- and CeO2-Stabilized t-ZrO2. Materials included ZrOCl2·8H2O (99.5%, Sigma-Aldrich), NH4OH (28−30% NH3, Sigma-Aldrich), absolute ethanol (Macron Chemicals), La(NO3)3·6H2O (99.999%, Sigma-Aldrich), Ce(NO3)3·6H2O (99.99%, Sigma-Aldrich), tetramethylammonium hydroxide (TMAOH, 97%, Sigma-Aldrich), and trimethyl (methylcyclopentadienyl) platinum (MeCpPtMe3, 98%, Sigma-Aldrich). All chemicals were used as received. Synthesis of La2O3−ZrO2. To mitigate the impact of Na ions on hydrothermal synthesis, NH4OH and TMAOH were used as the mineralizers to synthesize the La2O3−ZrO2 and CeO2−ZrO2 nanoparticles by a hydrothermal method, respectively. La2O3−ZrO2 (10 mol %) was synthesized via a hydrothermal treatment in the presence of ethanol. Typically, 4.5 mmol of ZrOCl2· 8H2O and 1 mmol of La(NO3)3·6H2O were dissolved in 10 mL of deionized (DI) H2O before being mixed together to yield a transparent clear solution. Then 30 mL of absolute ethanol was added to the solution. The reaction mixture was ultrasonicated for 1 h. Next, the pH was adjusted to ∼10 by adding 20 mL of NH4OH, which led to a continuous increase in the viscosity of the mixture. After being vigorously stirred for 30 min, the suspension was transferred to a 125 mL Teflon-lined autoclave and heated at 200 °C for 20 h. The autoclave was subsequently allowed to cool to room temperature. The autoclave reaction mixture was then separated by centrifugation at 6500 rpm and washed exhaustively using DI water until the mixture free of chloride ions. The chlorine-free mixture was subsequently dried in air at 70 °C for 12 h. White powders were then collected and calcined in air at 900 °C for 3 h to form the final 10% La2O3−ZrO2 support. Synthesis at lower temperatures (150 and 180 °C) and calcination at low temperatures (800 and 850 °C) did not change the particle size or morphology. Syntheses of 5, 20, and 30% La2O3−ZrO2 were conducted using a similar method by changing the amount of La precursor used. Synthesis of CeO2−ZrO2. CeO2−ZrO2 (20 mol %) was synthesized via a hydrothermal treatment in the presence of ethanol. Typically, 4 mmol of ZrOCl2·8H2O and 1 mmol of Ce(NO3)3·6H2O were dissolved in 10 mL of DI H2O before being mixed together to yield a transparent clear solution. Then 40 mL of absolute ethanol was added to the solution, before the reaction mixture was ultrasonicated for 1 h. Then the pH was adjusted to ∼10 by dropwise addition of 2 g of TMAOH (dissolved in 12 mL of DI H2O), which led to a B

DOI: 10.1021/acs.inorgchem.5b02810 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. Powder XRD pattern of (a) La2O3−ZrO2 and (b) CeO2−ZrO2 samples.

Figure 2. SAXS data for La2O3−ZrO2 and CeO2−ZrO2 showing (a and c) raw data and (b and d) the resulting particle size (volume) distributions.



Linkam stage for measurement in air from room temperature to 1000 °C at a rate of 20 °C/min. The data were fitted using Irena software.52 Water−Gas Shift Reaction Tests. The WGS kinetics were measured in an automated setup with four independent parallel tubular plug flow reactors as described elsewhere.53 For each run, ∼300 mg of catalyst was loaded into the reactor. The catalysts were pretreated by being reduced in a 25% H2/Ar mixture at 300 °C for 2 h. The reduction flow rate was 50 sccm. The catalysts were stabilized under our standard WGS conditions (7% CO, 22% H2O, 8.5% CO2, 37% H2, and balance Ar) with a flow rate of 75.4 sccm at 300 °C for 20 h until a stable conversion was observed. The temperature was lowered after stabilization to reach a conversion of Zr4+), a slight decrease in the lattice parameter of CeO2−ZrO2 C

DOI: 10.1021/acs.inorgchem.5b02810 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry from 68.70 Å3 to 69.77 Å3 to 71.03 Å3 also confirms the ionic substitution. Thus, 10, 20, and 30% CeO2−ZrO2 were selected as the supports for the subseqeunt ALD catalyst preparation and testing. SAXS measurements were performed at room temperature to obtain the particle size distributions of La2O3−ZrO2 and CeO2−ZrO2 after the calcination. The raw SAXS data of La2O3−ZrO2 and CeO2−ZrO2 (Figure 2a,c) was used to calculate the particle size distributions with a log-normal model (Figure 2b,d). The particle size results with very good standard deviations are listed in Table 1. There is good agreement between the SAXS and TEM results.

fringes in the HRTEM images (Figure 3b,d) confirm the high crystallinity of these two samples. According to the P42/nmc space group symmetry, the aforementioned solid solution nanoparticles are terminated with the (100) and (111) lattice planes, in agreement with the theoretical calculations.55 In situ temperature-dependent synchrotron powder XRD and SAXS were performed to further investigate the thermal stability and the particle size change of La2O3−ZrO2 and CeO2−ZrO2 solid solutions. Panels a and b of Figure 4 show the temperature dependence of the synchrotron XRD patterns of 10% La2O3−ZrO2 and 20% CeO2−ZrO2 supports during heating and cooling treatment in air, respectively. A temperature ramp rate of 20 °C/h and a 2θ range of 1−8° were used. The synchrotron XRD patterns at room temperature confirm the purity of the tetragonal phase. During the process of temperature ramping from room temperature to 1000 °C, a slight shift in the peak positions to the small angles was observed, which indicates that the cell volume of the aforementioned supports increased continuously as a function of temperature. The gradual increase in the intensities of diffraction peaks with increasing temperature was due to an increase in crystallite size. No phase transition can be observed during both heating and cooling processes. Clearly, La2O3- or CeO2-stabilized tetragonal ZrO2 supports had thermal stability higher than that of Na-stabilized t-ZrO2 supports, which transform to monoclinic phase at higher temperatures as shown in Figure 4c. Figure 4d shows the particle sizes of temperature dependence for 10% La2O3−ZrO2, 20% CeO2−ZrO2, and Nastabilized t-ZrO2 supports on heating and cooling based on synchrotron SAXS data. In situ SAXS measurements showed that the particle size of 10% La2O3−ZrO2 and 20% CeO2− ZrO2 supports increased marginally after the heating and cooling cycle. However, the particle size of Na-stabilized t-ZrO2 increased to ∼45 nm after the heating and cooling cycle. N2 adsorption−desorption was conducted for the calcined La2O3−ZrO2 and CeO2−ZrO2 samples. All these samples showed similar N2 adsorption−desorption isotherms (Figure 5a,b). These isotherms correspond to a type IV isotherm, according to IUPAC, with a H3 hysteresis loop observed in solids exhibiting platelike particles. The BET surface areas for these samples are summarized in Table 2. The BET surface areas of La2O3−ZrO2 and CeO2−ZrO2 samples are between 50 and 100 m2/g. These values are comparable to that of Nastabilized t-ZrO2 (55 m2/g).30 Characterization of Pt-Loaded La 2O3−ZrO 2 and CeO2−ZrO2 Catalysts. One ALD Pt cycle was performed on the 5, 10, and 20% La2O3−ZrO2 and 10, 20, and 30% CeO2−ZrO2 samples. The XRD patterns of Pt/La2O3−ZrO2 and Pt/CeO2−ZrO2 are shown in Figure 6. No changes in the XRD patterns for the Pt/La2O3−ZrO2 and Pt/CeO2−ZrO2 samples were observed compared with those of the samples before Pt ALD, suggesting that the Pt was mainly deposited on the sample surfaces without altering the structure of the solid solutions. Furthermore, no Pt peaks were detected by XRD for these samples, suggesting that the ALD Pt nanoparticles were