Thermal Conductivities and Thermal Expansion Coefficients of (Sm0.5Gd0.5)2(Ce1−xZrx)2O7 Ceramics

The (Sm0.5Gd0.5)2(Ce1−xZrx)2O7 oxides were prepared by solid-state reaction, and their phase compositions, microstructures, and thermophysical properties were investigated. Results of x-ray diffraction reveal that pure (Sm0.5Gd0.5)2(Ce1−xZrx)2O7 oxides with fluorite structure are successfully synthesized in the current study. The thermal expansion coefficient decreases with increasing content of ZrO2, which is higher than that of 7 wt.% yttria-stabilized zirconia (YSZ). The substitution of Zr4+ for Ce4+ reduces the thermal conductivity of Sm2Ce2O7 oxide. The thermal conductivity decreases from 1.69 W/m K (x = 0) to 1.22 W/m K (x = 0.3) at 1000 °C. The composition with x = 0.3 exhibits the lowest thermal conductivity at all temperatures, and the thermal conductivity of (Sm0.5Gd0.5)2 (Ce1−xZrx)2O7 ceramics was obviously lower than those of fully dense 7 wt.% YSZ. These results suggested promising potential applications of the (Sm0.5Gd0.5)2 (Ce1−xZrx)2O7 ceramics for high-temperature thermal barrier coatings.


Introduction
Gas turbines operate at high temperatures where phenomena such as oxidation and creep occur readily. Thermal barrier coating (TBC) systems have become common in gas turbines as they lower the temperature of the underlying substrate and provide protection against high-temperature degradation of the substrate materials (Ref 1). TBC systems thus prolong the life of structural parts, as well as increase the gas turbine efficiency by enabling higher combustion temperatures ( . The requirements to improve jet engine efficiencies have proved to be a big driving force to improve the thermal barrier coating technology by novel compositions and improved production techniques for subsequent improvement in microstructure and thus enhance thermo-mechanical performance of TBCs. Yttriastabilized zirconia (YSZ) is the present industrial standard topcoat material in TBC system, which can operate long term at temperatures below 1200°C ( Ref 5,6). However, two flaws, including phase transformation and sintering, have been reported when YSZ is exposed to high temperature (above 1200°C) for long time. Met stable tetragonal phase decomposes into tetragonal and cubic phase above 1200°C. Upon cooling, tetragonal phase transforms to monoclinic phase, causing about 3.5% volume change and resulting in crack formation in the TBCs. In addition, volume fraction of pores decreases due to the significant sintering of YSZ at high temperature, which leads to an increase in the thermal conductivity as well as the in-plane stiffness and thus decreases the strain compliance of TBCs (Ref [7][8][9]. The limited capability of YSZ above 1200°C necessitates its substitution with some novel compositions capable of handling higher temperatures with better stability in advanced gas turbines ( Ref 10,11).
A number of oxide ceramics have been proposed as novel thermal barrier coatings during the last few years. These novel compositions cover especially doped zirconia (

Experimental
In this paper, the polycrystalline samples of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics were synthesized by a solid-state reaction method. Sm 2 O 3 , Gd 2 O 3 , ZrO 2 , and CeO 2 with purity 99.9% were used as starting materials, and all oxides were dried before use. The stoichiometric amounts of precursors were ground in acetone for 6 h to obtain uniformly distributed mixtures. The mixture slurry was dried at 120°C for 2 h for complete removal of acetone. The dried powders were pressed into small columns at 50 MPa followed by cold isostatic pressing with 150 MPa. Finally, the bulks were pressureless sintered at 1600°C for 10 h in air.
The phase structure of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 oxides was identified by x-ray diffraction (XRD, Rigaku D/Max 2500, Japan) with Ni-filtered Cu Ka radiation (0.1542 nm) at the scanning rate of 4°/min. The microstructure of the pellets was observed using field emission scanning electron microscopy (SEM, Model Hitachi S-4800, Japan). The specimens were polished with 1-lm diamond paste, and then thermally etched at 1500°C for 2 h in air for SEM observations. Qualitative xray element analysis of various phases was carried out using SEM equipped with energy dispersive spectroscopy (EDS). For heat-treated samples, the actual densities (q) were measured by using Archimedes method with an immersion medium of deionized water.
The thermal expansion coefficients of these ceramics were determined with a high-temperature dilatometer (Netzsch DIL402C/7, Germany) from ambient temperature to 1000°C at a heating rate of 5°/min in argon atmosphere. The size of sample was 25 mm 9 3 mm 9 4 mm. The specific heat capacities (C P ) were calculated from constituent oxides according to the Neumann-Kopp rule based on the reference specific heat values of Sm 2 O 3 , Gd 2 O 3 , CeO 2 , and ZrO 2 (Ref 26). The thermal diffusivity (k) was measured from 200 to 1000°C using a laser flash apparatus (FlashLineTM3000, USA) in an argon gas atmosphere. The specimen dimension was about 12.7 mm in diameter and about 1.2 mm in thickness. The thermal conductivity was then calculated according to the relation: Because the sintered specimen was not fully dense, the measured thermal conductivity was modified for the actual value k 0 using Eq 2, where / is the fractional porosity and the coefficient 4/3 is used to eliminate the effect of porosity on actual thermal conductivity (Ref 17).

Phase Structure
The XRD patterns of all the products in (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 were recorded in Fig. 1  In the A 2 B 2 O 7 system, the crystal structure is mainly determined by the ionic radius ratio rðA 3þ Þ=rðB 3þ Þ of A and B cations. The stable pyrochlore structure is limited to the range of 1:46 rðA 3þ Þ=rðB 3þ Þ 1:78, and the fluorite oxide will form if the rðA 3þ Þ=rðB 3þ Þ is lower than 1.40 ( Ref 29). For the complex rare-earth cerium oxides, the ionic radius can be estimated from the ionic radius of the component ions and the chemical composition using the following equation.
where 0.5, 1 À x, and x are the composition of each atom. The ionic radii of Sm 3+ and Gd 3+ are 1.079 and 1.053 Å , while the Zr 4+ and Ce 4+ are 0.72 and 0.97 Å , respectively. The values of rðA 3þ av Þ rðB 4þ av Þ for (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 oxides can be calculated by the Eq 3 with above mentioned ionic radiuses, and the values are equal to 1.11, 1.13, 1.156, and 1.189 with x increasing from zero to 0.3. The ionic ratios of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 are smaller than 1.40 due to the co-doping in the Sm and Ce sites, which reveals that the co-doping in Sm 2 Ce 2 O 7 cannot change the crystal structure.

Microstructure
The typical microstructure of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics is shown in Fig. 2. The average grain size of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics is several micrometers. The interfaces between grains are very clean, the gap is very small, and no other inter-phases and unreacted oxides existed in boundaries between grains. The chemical compositions and relative density of the synthesized (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics are listed in Table 1. As can be seen from Table 1 ceramics are very close to their stoichiometries, and each specimen has a high relative density due to the high-temperature sintering.

Thermal Expansion Coefficient
The dilatometric measurement results of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics with calibration are plotted in Fig. 3. The typical linear expansions can be observed for (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics in the temperature range of 20-1000°C. Clearly, there is no phase transformation for (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics in the temperature range. The technical thermal expansion coefficient is defined as where L 0 is the length of the specimen at T 0 (20°C), DL 0 is the change in length at T 0 , and DL k is the corresponding length change at temperature T k . Their calculated technical thermal expansion coefficients against temperature are plotted in Fig. 4, together with the data of YSZ. Fig 4 shows that their thermal expansion coefficients increase smoothly with temperature up to 1000°C, which is attributed to the increasing atomic spacing at elevated temperatures. As also can be seen in Fig. 4, the thermal expansion coefficient decreases gradually with increasing x from 0 to 0.3 where N 0 , A, z, r 0 , e, and n represent Avogadro Õ s number, the Madelung constant, the ionic charge, the inter-ionic distance, the charge of an electron, and Born exponent, respectively. From the lattice parameters listed in Table 1, it can be known that the substitution of Zr 4+ for Ce 4+ results in a slight contraction in the unit cell, indicating a decrease in the average inter-ionic distance, r 0 . This would lead to an increase in the crystal energy of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics, which represents the decrease of thermal expansion coefficient. Alternatively, there are some structure defects, such as substitutional or interstitial cations and other imperfections in the lattice or varying properties in thermal expansion along with different orientations that have significant effect on the overall thermal expansion coefficient of A 2 B 2 O 7 oxides (Ref 31,32). In the fluorite-type A 2 Ce 2 O 7 system, oxygen vacancies are randomly distributed in disordered fluorite structure; therefore, it facilitates formation of the ionic vacancy clustering, which represents to some extent the decrease in thermal expansion. The structure disorder degree of fluorite-type A 2 Ce 2 O 7 is inverse to value of rðA 3þ av Þ rðB 4þ av Þ (Ref 29), and the increasing value of rðA 3þ av Þ rðB 4þ av Þ in (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics represents the increase of thermal expansion coefficient. Hence, the decrease of thermal expansion coefficients with increasing ZrO 2 content in (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics reveals that the influence of crystal energy is greater than oxygen vacancy (Ref 29,33). These thermal expansion coefficients are obviously higher than 7 wt.% yttria-stabilized zirconia [10.7 9 10 À6 /K at 1000°C (Ref 10)] and are good enough to be used as thermal barrier coating materials in the industry condition.

Thermal Conductivity
The specific heat capacities of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics at different temperatures were calculated according to the Neumann-Kopp rule based on the reference specific heat values of Sm 2 O 3 , Gd 2 O 3 , CeO 2 and ZrO 2 . As can be seen in Fig. 5, the specific heat capacity of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics increases with increasing temperature from room temperature to 1000°C, and the calculated specific heat capacity of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics slightly increases from x = 0 to 0.3 at identical temperature levels. Fig 6 shows the composition-dependent thermal diffusivities of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics measured by laser flash method. The values of thermal diffusivity are the arithmetic mean of three measurements. Since the error derived from the mean standard deviation of the three measurements for each specimen is <1.5%, the error bars in Fig. 6 are omitted for all thermal diffusivity date because they are smaller than symbols. As can be seen in Fig. 6, the thermal diffusivity of the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics exhibits inverse temperature dependence, which suggests a dominant phonon conduction behavior in most polycrystalline materials (Ref 32, 33). Furthermore, the thermal diffusivity remarkably decreases with increasing Zr 4+ fraction, and (Sm 0.5 Gd 0.5 ) 2 (Ce 0.7 Zr 0.3 ) 2 O 7 exhibits the lowest thermal diffusivity among all the ceramics investigated.
The thermal conductivities were calculated using Eq 1 and calibrated with Eq 2 to represent fully dense samples, as plotted in Fig. 7. The error bars are omitted because they are smaller where l i ðx; tÞ, l p ðx; tÞ, l v ðx; tÞ; and l gb are the phonon mean free path due to interstitials scattering, point defect scattering, vacancies scattering, and grain boundary scattering, respectively. Since the phonon mean free path is several orders of magnitude smaller than the grain size for the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics, the grain boundary scattering can be omitted (Ref 34 where a 3 is the volume per atom, v the transverse wave speed, x the phonon frequency, c the concentration per atom, J the constant, c the Grü neisen parameter, M and R the average mass and ionic radius of the host atom, respectively, and DM and DR are the differences of masses and ionic radius between the substituted and substituting atoms, respectively. Because the atomic weights of Zr and Ce are 91.22 and 140.1, respectively. The effective ionic radii of Zr 4+ and Ce 4+ are 0.72 and 0.97 Å , and the effective phonon scattering of the substitutional atoms contributes to the lower thermal conductivity. Thus, thermal conductivity of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics decreases with increasing ZrO 2 content at identical temperature levels. The low thermal conductivity indicates that the (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics can be explored as potential candidates for thermal barrier coating applications.

Conclusions
Pure (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics with fluorite structure were successfully prepared by solid-state reaction method and sintered at 1600°C for 10 h using Sm 2 O 3 , Gd 2 O 3 , ZrO 2 and CeO 2 as raw materials. Their microstructure is very dense and there are no other inter-phases or unreacted oxides existing at the boundaries between grains. The thermal conductivity of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics decreases with increasing ZrO 2 content at identical temperature levels, and phonon scattering caused by the substitutional atoms contributes to the low thermal conductivity. The thermal expansion coefficient of (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics decreases with increasing Zr 4+ ions, and these thermal expansion coefficients are obviously higher than 7 wt.% yttriastabilized zirconia. The synthesized (Sm 0.5 Gd 0.5 ) 2 (Ce 1Àx Zr x ) 2 O 7 ceramics can be explored as potential candidates for thermal barrier coatings.