Gyeung-Ho Kim

Stable sealing glass for planar solid oxide fuel cell

Abstract The thermal and chemical stability of glasses in the BaO–Al2O3–La2O3–B2O3–SiO2 system were investigated as well as bonding characteristics and wetting behavior to yttria stabilized zirconia (YSZ) electrolyte, to develop a suitable sealing glass for planar solid oxide fuel cell operating at 800–850 °C. The thermal properties such as glass transition temperature, softening temperature and thermal expansion coefficient were found to depend on the B2O3:SiO2 ratio in glass composition; thus the bonding characteristics of the glass to YSZ were also influenced by this ratio. The glass having a minimum thermal expansion mismatch with YSZ showed an excellent endurance during thermal cycling. No interface reaction was observed for all the glass/YSZ specimens heat-treated at 800–850 °C up to 100 h.

Growth kinetics of three Mo-silicide layers formed by chemical vapor deposition of Si on Mo substrate

Abstract The growth kinetics of three Mo-silicide layers formed by chemical vapor deposition (CVD) of Si on a Mo substrate using SiCl 4 -H 2 gas mixtures were investigated at temperatures between 950 and 1200 °C. Three Mo-silicide layers (Mo 3 Si, Mo 5 Si 3 , and MoSi 2 ) grew simultaneously with a parabolic rate law after an initial nucleation period, indicating the diffusion-controlled growth. The activation energy (130 kJ/mol) for the MoSi 2 layer were in a good agreement with the previous results having low activation energy (130±20∼157 kJ/mol), but its growth rate was higher than the previous results with high activation energy (209∼241±25 kJ/mol). A possible explanation about this difference may be the detrimental effect of impurities such as oxygen on the growth rate of the MoSi 2 layer. The activation energy (350 kJ/mol) for growth of the Mo 5 Si 3 layer was consistent with the prior values (297∼360 kJ/mol) obtained by annealing of the MoSi 2 /Mo diffusion couples, but its growth rate was an order of magnitude lower than the rate measured in the MoSi 2 /Mo diffusion couples. The activation energy (223 kJ/mol) for the growth of the Mo 3 Si layer was similar with the value (199 kJ/mol) obtained from annealed Mo 5 Si 3 /Mo diffusion couple at temperatures between 1250 and 1350 °C. This value was lower than the value (326 kJ/mol) reported at higher temperatures from 1500–1715 °C. This suggests that the rate-limiting step for growth of the Mo 3 Si layer is the grain boundary diffusion-controlled process at low temperatures but volume diffusion-controlled process at high temperatures. The growth rates of the Mo 3 Si layer measured at condition of the simultaneous parabolic growth of three Mo-silicide layers were approximately two orders of magnitude lower than the rates measured in the Mo 5 Si 3 /Mo diffusion couples. The differences in the growth rates of the Mo 5 Si 3 and Mo 3 Si layers depending on the type of diffusion couples were well explained by the multiple layer growth model.

Synthesis and properties of cubic zirconia-alumina composite by mechanical alloying

Abstract Synthesis of 8mol%Y 2 O 3 –ZrO 2 (8YSZ) powder with Al 2 O 3 by mechanical alloying (MA) and the sintering behavior of this powder were investigated. The solubility of Al 2 O 3 in 8YSZ was 1.15 wt.% after MA for 48 h. Large agglomerated particles formed during MA resulted in a slow densification rate of the compact. After eliminating the agglomerates by sedimentation, the relative density of 97% was obtained by sintering the compact at 1400°C for 2 h. Sintered microstructure of mechanically alloyed 3wt.%Al 2 O 3 –8YSZ (3A8YSZ) consisted of 8YSZ grains in the size range of 0.5–1.2 μm and 20–400 nm size Al 2 O 3 particles dispersed uniformly in the 8YSZ. Fracture toughness and hardness of the composite was improved due to intragranular dispersion of Al 2 O 3 particles. Electrical conductivity of the composite was 6.4×10 −5 S cm −1 at 400°C and also exhibited significant increase over monolithic 8YSZ. Improvement in lattice conductivity was unique in this composite and attributed to the role of intragranular Al 2 O 3 particles, which provided strong lattice distortion of 8YSZ matrix.

Study on reaction and diffusion in the Mo–Si system by ZrO2 marker experiments

The dominant diffusing element in the MoSi2, Mo5Si3 and Mo3Si was found to be Si from marker experiments using ZrO2 particles. Based on these marker experiments, the formation mechanism of the Mo5Si3 and Mo3Si layers by reaction and diffusion was also discussed.

Analysis of X-ray diffraction patterns from mechanically alloyed Al-Ti powders

X-ray diffraction (XRD) method is one of the most versatile tools to characterize various forms of materials. Simplicity and wealth of information from the spectrum makes it attractive for the evaluation of mechanical alloyed powders. However, careful interpretation of the solubility of minor phase is necessary due to the effect of particle size on the detection limit in XRD method. In this study, we demonstrate the inaccuracy of solubility from XRD analysis of nanosized particle system using Al-Ti as a model. Using transmission electron microscopy (TEM), it is confirmed that large amount of nanosized Ti in Al matrix is not detected by XRD. The peak disappearance of minor phase can not be used to determine the solubility of mechanically alloyed powders. Lattice parameter change of the major phase should be used to assess the solubility limit of the minor phase in nanosized particle system. In addition, the possible sources of error are addressed when mechanically alloyed powders of Al-Ti system are characterized by the XRD method. Proper XRD analysis methods are suggested to determine the lattice parameter, solubility of minor elements, crystallite size and strain variance in the MA Al-Ti samples. Pure A1 is used as an internal standard to correct instrumental broadening, the Al {111} peak is used to determine lattice parameter of A1, and the lattice parameter of Al is recommended to estimate the solubility of Ti in Al. The calculation of crystallite size and strain variance in the MA powders using Williamson-Hall equation is also discussed in detail.

Simple procedure for phase identification using convergent beam electron diffraction patterns

The use of the primitive cell volume and the zero order Laue zone (ZOLZ) pattern is proposed as a means to identify phases in a complex microstructure. A single convergent beam pattern, containing a higher order Laue zone ring, from a nanosized region is sufficient to calculate the primitive cell volume of the phase, while ZOLZ pattern is used to determine the zone axis of the crystal. A computer program is used to screen out possible phases on the basis of the value of measured cell volume. The indexing of the ZOLZ pattern follows in the program to find the zone axis of the identified phase. Combination of these two methods ensures accuracy and reliability of phase identification from a single CBED pattern. An example of the analysis is given from the rapidly solidified Al‐Al3 Ti system.

Growth Kinetics of MoSi2 Coating Formed by a Pack Siliconizing Process

Growth kinetics of MoSi 2 coating formed by pack siliconizing process of Mo substrate was investigated on the basis of the dependence of its growth rate on Si content in the pack powders. Pack siliconizing was carried out using (1-70) wt % Si-5 wt % NaF-bal SiC packs at 1100°C in a hydrogen atmosphere. The chemical vapor deposition (CVD) of Si on Mo substrate was also performed at 1100°C using SiCl 4 -H 2 gas mixture to determine the maximum growth rate of MoSi 2 coating obtainable in the bulk Si/mo diffusion couple. The growth kinetics of the MoSi 2 coating formed by the pack siliconizing process obeyed a parabolic rate law irrespective of compositions of pack powders. The growth rates of MoSi 2 coating increased with increasing Si content in the pack powders up to 40 wt % Si but attained a constant value in the pack powders with over 40 wt % Si, which was equal to that obtained by the CVD process. The rate-limiting step for growth of MoSi 2 coating formed by the pack siliconizing process was subject to theoretical considerations. Three models such as a gas-phase diffusion controlled model, a solid-state diffusion controlled model, and a dynamic equilibrium model of gas-phase diffusion and solid-state diffusion were evaluated. The theoretically predicted results based on the dynamic equilibrium model and the experimentally obtained results were found to be in good agreement.

Formation of MoSi2–SiC composite coatings by chemical vapor deposition of Si on the surface of Mo2C layer formed by carburizing of Mo substrate

Abstract The formation process of MoSi 2 /β-SiC composite coating by chemical vapor deposition of Si on the Mo 2 C layer formed in advance on a Mo substrate by a carburizing process was investigated using optical microscopy, field-emission scanning electron microscopy cross-sectional transmission electron microscopy and X-ray diffraction. The composite coating was composed of equiaxed MoSi 2 grains with average size of 300 nm and the β-SiC particles with average size of 92 nm, which were mostly located at the grain boundaries of MoSi 2 . The morphology of β-SiC particles was oblate-spheroidal shape and volume percentage was approximately 19.3%. The number of cracks in the composite coating was smaller than that for the monolithic MoSi 2 coating by the reduced mismatch of thermal expansion coefficient with the Mo substrate. The formation mechanism of MoSi 2 /β-SiC composite coating was suggested on the basis of microstructure observation.

Effect of ammonia nitridation on the microstructure of MoSi2 coatings formed by chemical vapor deposition of Si on Mo substrates

Abstract The effect of ammonia nitridation on the microstructure of MoSi 2 coatings formed by chemical vapor deposition (CVD) of Si on Mo substrates at 1100 °C was investigated. The microstructure of monolithic MoSi 2 coating formed without a prior nitridation of Mo exhibited a typical columnar structure perpendicular to the Mo substrate. On the other hand, a two-step deposition process (ammonia nitridation followed by CVD of Si) produced a MoSi 2 /α-Si 3 N 4 composite coating, which consisted of equiaxed MoSi 2 grains with an average size of 0.5 μm, where the α-Si 3 N 4 particles with an average size of 120 nm were mostly located at the grain boundaries of MoSi 2 . The shape of α-Si 3 N 4 particles was oblate-spheroidal type and its volume percentage ranged from 12.9 to 17.7%. The density and average width of cracks in MoSi 2 layer formed by the mismatch of thermal expansion coefficients between the MoSi 2 coating and the Mo substrate was smaller for the two-step deposition process than that for CVD process only. The columnar MoSi 2 coating grew by the diffusion of Si and conversion of Mo 5 Si 3 phase to MoSi 2 phase at the interface of MoSi 2 /Mo 5 Si 3 , while the MoSi 2 /α-Si 3 N 4 composite coating grew by successive displacement reaction of Mo 2 N phase and Si. The growth of MoSi 2 grains was inhibited by the nanosize α-Si 3 N 4 particles.

Multilayer diffusional growth in silicon–molybdenum interactions

Growth kinetics of the Mo–silicide layers formed by chemical vapor deposition of Si on a Mo substrate from the SiCl4–H2 gas mixtures at 1000 °C was investigated using the ‘Wang’ analysis of multilayer diffusional growth. All of the three Mo–silicide phases, tetragonal-MoSi2, Mo5Si3 and Mo3Si in the Mo–Si binary phase diagram were observed by cross-sectional transmission electron microscopy, and obeyed a parabolic rate law indicating diffusion-controlled growth. The intrinsic growth rates of the Mo5Si3 and Mo3Si layers were estimated from their apparent growth rates measured in the Si/Mo diffusion couple. Good agreement was found with the reported values measured from apparent growth rates at the MoSi2/Mo and Mo5Si3/Mo diffusion couples, respectively.