Johnson Matthey Technol. Rev., 2021, 65, (1), 112
Different Deformation Behaviour Between Zirconia and Yttria Particles in Dispersion Strengthened Platinum-20% Rhodium Alloys
Alloy behaviour during hot forging and cold rolling
- Ziyang Wang, Xi Wang, Futao Liu, Faping Hu, Hao Chen, Guobin Wei
- College of Materials Science and Engineering, Chongqing University, No.174 Shazhengjie, Shapingba, Chongqing, 400044, China
- Weiting Liu
- Chongqing Polycomp International Corporation, Jianqiao Industrial Park B, Dadukou District, Chongqing, 400082, China
- Weidong Xie*
- College of Materials Science and Engineering, Chongqing University, No.174 Shazhengjie, Shapingba, Chongqing, 400044, China
Platinum-20% rhodium strengthened by oxides of zirconium and yttrium were prepared by solidification of platinum-rhodium-(zirconium)-yttrium powder which had been internally oxidised. After forging, rolling and annealing, 1 mm plates were obtained. Then the plates were mechanically ground to 50–70 μm from rolling-normal direction, followed by argon ion milling until a hole appeared on the centre of the foil to obtain samples which were characterised by transmission electron microscopy (TEM), combined with thermodynamic analysis. The existence of spherical ZrO2 and Y2O3 particles was verified with platinum and rhodium present as pure metals at the same time. It was found that the deformation behaviour of ZrO2 and Y2O3 particles was quite different during processing, where the former basically maintain their spherical shape and were bonded tightly to matrix, while the latter were compressed along normal direction and form two cracks on both sides of Y2O3 particles along the rolling direction. The differences in hardness and interface bonding properties of these two types of particles are supposed to be the main causes of different deformation behaviour during hot forging and cold rolling.
Platinum and its alloys are very important in industries such as the chemical and glass industries. Some of their main applications are as follows: a functional structural material in the aerospace industry; a catalyst in the nitric acid preparation industry; nozzles for preparing glass fibres; preparing crucibles and utensils that require special properties for use in chemical laboratories. Platinum-rhodium alloys with low rhodium content can also be used as brazing filler metals in metal welding. Although expensive, their high temperature stability and chemical inertia cannot be replaced by other materials (1, 2).
However, the strength and creep resistance of pure platinum would significantly reduce at high temperature because of sharp grain growth. Therefore, it is crucial to find a way to improve the high temperature creep resistance of platinum alloys (3). Many methods of strengthening platinum materials are known. Among these different methods, the best methods are solid solution strengthening and dispersion strengthening. The addition of rhodium has been found to have the best solid solution strengthening effect in high temperature use of platinum. With increasing rhodium content in platinum-rhodium alloys, the normal temperature strength, high temperature endurance strength, creep rupture life and creep activation energy of the alloy increases greatly, so platinum-rhodium alloys are widely used (4, 5). But the high-temperature mechanical properties of platinum-rhodium alloys fail to satisfy the requirement of industrial production with worsening service conditions. Therefore, dispersion-strengthened platinum-rhodium alloys were subsequently developed in Johnson Matthey, UK (6, 7), Engelhard Corporation, USA (8) and Heraeus, Germany (9) from the 1980s to 1990s. They were prepared by adding zirconium, yttrium or scandium oxide particles into the platinum-rhodium alloy under certain process conditions. At present, dispersion-strengthened platinum alloys such as zirconia grain stabilised (ZGS) platinum and oxide dispersion strengthened (ODS) platinum (10, 11), in which Y2O3 and ZrO2 are two commonly used reinforcement particles, are used in industrial applications. According to the experimental results of Hu et al. (12), as the zirconium concentration increases, the platinum-rhodium alloy’s yield stress (0.2% offset) increases significantly, while the elongation decreases slightly. The work hardening rate of particle-reinforced samples increases with the increase of the volume concentration of dispersed particles, which is a typical behaviour of particle dispersion enhancement. In addition, based on a reformulation of the Orowan stress for particle strengthening and by superposing this stress to the matrix stress, a calculated flow stress is in good accord with the experimental value. Though the strength of these particle strengthened platinum alloys can be significantly improved, the further enhancement of the strength is difficult due to the lack of mechanism research and empirical development in the past (12, 13).
Zirconium and yttrium have good plasticity and corrosion resistance. Diffusion strengthening improves the metal strength through adding a second phase or multiple phases and the essence is the interaction of the second phase and the dislocation (13). Nevertheless, the two reinforcement particles, i.e. Y2O3 and ZrO2, are usually thought be to the same as strengthening particles based on Orowan’s equation where the strength increment from the non-deformable dispersed particles is positively correlated with the particle spacing (14, 15). When making samples, we tried to add the yttrium content to 0.1%, and found that the ingot will crack during rolling, so we speculate that different deformation behaviours will have a huge impact on alloy strengthening.
In this study, two kinds of platinum alloys strengthened by Y2O3 and ZrO2, respectively, were prepared and characterised by TEM. A number of differences in deformation behaviour of the two particles during processing were found. These findings are expected to lead to new insights into developing dispersion strengthened high strength platinum alloys.
Nominal composition of the investigated alloys is listed in Table I. The preparation processes have been described in previous research (13) and can be briefly summarised as follows: Pt-20 wt% Rh-0.015 wt% Y without and with 0.1 wt% Zr were smelted in a vacuum induction furnace at 1800–2000°C under argon atmosphere, then cast into a water-cooled copper mould to obtain ingots. The ingots were hot forged to 10 mm plates which were sheared into pieces, followed by mechanical milling to fine powder. The size distribution of the powder was measured using an LS-POP(6) laser particle size analyser (OMEC, China). 97% of the particles were distributed between 5 μm and 60 μm with an average particle size of 25 μm. The powder was sintered and exposed in air to internally oxidise at 1150°C for 4 h. The 30 mm sheets were obtained by hot forging at 1400°C, then rolled to 1 mm sheets at room temperature followed by annealing at 1150°C for 30 min. To prepare the TEM samples, the sheets from rolling-normal direction were mechanically ground to 50–70 μm, followed by argon ion milling at 5 kV until a hole appeared on the centre of the foil. TEM was used to examine the microstructure at 200 kV (JEM-2100 electron microscope (JEOL Ltd, Japan)) and at 300 kV (JEM-3000F field emission electron microscope (JEOL Ltd)). JEM-2100 was used to detect topography of the particles and energy-dispersive X-ray spectroscopy (EDS) was conducted to determine the components; JEM-3000F was used for high-resolution transmission electron microscopy (HR-TEM) observation to determine the structure of the particles. In this paper, Pt-20Rh-0.015Y is referred to as Alloy 1 while Pt-20Rh-0.015Y-0.1Zr is referred to as Alloy 2.
3. Results and Analysis
3.1 Thermomechanical Analysis
During preparation processes, especially at high temperature under oxidising atmosphere, the metals in the alloys may be oxidised and the possible oxides are PtO2, Rh2O3, ZrO2 and Y2O3. Their reaction equations are as follows (Equations (i)–(vi):
According to the Gibbs free energy theorem, a reaction can only occur when the Gibbs free energy is negative and the more negative the Gibbs free energy of a reaction, the more easily the reaction occurs. When the Gibbs free energy is greater than zero, the reaction cannot happen. Figure 1 is the oxidation tendency of platinum, rhodium, zirconium and yttrium, which indicates:
ΔG of all the reactions increases as temperature increases
Within the temperature range we calculated, ΔG of Y2O3·ZrO2, Y2O3 and ZrO2 are much smaller than those of Rh2O3, Rh2O and RhO, which means the formation of Y2O3·ZrO2, Y2O3 and ZrO2 are much easier during processing
ΔG of RhO, Rh2O and Rh2O3 are very close to ΔG = 0 and even greater than 0 at the internal oxidation temperature (1423 K), thus the formation of RhO, Rh2O and Rh2O3 are difficult and can be ignored.
For platinum, when the temperature rises from room temperature, platinum will react with oxygen and form a PtO2 film on the metal surface. The thickness of PtO2 would grow with the rise of temperature. When the temperature reaches about 500°C, this process would stop. If the temperature continues to rise, the PtO2 film will be gradually vaporised. Meanwhile, the higher the temperature, the faster the gasification rate. Samples in this experiment were about 1200°C, the PtO2 film formed by oxidation had been basically vaporised. Therefore, PtO2 is not present in the samples.
Overall, according to the thermomechanical analysis, Y2O3, ZrO2 and Y2O3·ZrO2 are easily formed in the platinum-rhodium-(zirconium)-yttrium alloys. Platinum and rhodium are present as almost pure metals.
3.2 Microstructure and Energy-Dispersive X-ray Spectroscopy Analysis of Platinum-Rhodium-(Zirconium)-Yttrium
Figure 2 is the bright field TEM images of Alloy 1 showing the morphology of particles. It shows that some of the particles still maintained spherical shape and each particle was accompanied by two cracks on two sides along the rolling direction, as shown by particle B. One enlarged image of this kind of particle is shown in Figure 2(b). Some particles had been compressed along the normal direction with two cracks along the rolling direction, such as particle A. A small number of fine particles were severely compressed so that the cracks have closed (particles C). Table II shows the EDS analysis results of the particle in Figure 2(b), which indicates that the particle is mainly composed of yttrium and oxygen atoms. A small amount of iron impurity may have come from the TEM equipment.
Figure 3 is the bright field TEM images of Alloy 2 showing the morphology of the particles. Figure 3(a) shows that the particles (as indicated by red circles) are spherical and uniformly distributed in matrix with a diameter range from 20 nm to 70 nm. Note that the particles show a good bonding with matrix after hot forging and cold rolling during processing. An enlarged image of a particle is shown in Figure 3(b) and its EDS analysis results have been listed in Table III, illustrating that the particle is composed of zirconium and oxygen atoms.
The particles are so small that it is difficult to characterise them using normal selected area diffraction techniques. Thus, HR-TEM was used to investigate the structure of the particles. Binary monolithic ZrO2 is known to exhibit polymorphic transformations between monoclinic (mP12:P 121/c1, ZrO2-b type), tetragonal (tP6:P42/nmc, ZrO2-type) and cubic (cF12:Fm3m, CaF2-type) (16). The monoclinic phase is stable below 1478 K, while the tetragonal phase is stable between 1478 K and 2650 K. The cubic phase is stable from 1796 K to 2993 K, and exhibits some range of homogeneity. The fast Fourier transform (FFT) diagrams of ZrO2 from different zone axes is shown in Figure 4, and when it is compared with the common ZrO2 structure (17, 18), we find that two crystal structures of ZrO2 were monoclinic and tetragonal. The FFT diagrams of Y2O3 from different zone axes is shown in Figure 4. When it is compared with the common Y2O3 structure (19), we find that the structure of Y2O3 is body-centred cubic.
The volume fractions of particles (ZrO2 and Y2O3) are difficult to measure by X-ray diffraction (XRD) due to low content of the particles or by TEM due to the large atomic mass of platinum and rhodium. It is therefore difficult to measure the thickness of the TEM foil. Thus, we approximately calculated the volume fraction of ZrO2 and Y2O3 as 0.42% and 0.0587%, respectively, based on the weight fraction of zirconium and yttrium. Note also that a small amount of zirconium and yttrium may be present as solute atoms in the matrix ascribed to the extremely low solubility of oxygen atoms in the platinum-rhodium alloy. An oxidation rate of 75% was proved to be reasonable, based on the experimental and calculated yield stress of the alloys (20, 21).
3.3 Analysis of Deformation Behaviour Between Zirconia and Yttria Particles
Based on our TEM observations, these two particles have totally different deformation behaviours. Almost all Y2O3 particles have compression deformation along the normal direction with two cracks on the two sides of particles along the rolling direction, while ZrO2 particles basically maintain their spherical shape and are bonded tightly with matrix. The nanoparticle deformation behaviour in particle strengthened metals has been widely researched (22–26). By studying the deformation behaviour of dispersion-strengthened particles in steel, Gove et al. (22, 23) claimed that the formation of voids around the particles and the matrix is due to the fact that the steel matrix cannot flow around the particles while maintaining contact with them. The strength of the inclusion-substrate interface is insufficient to withstand the longitudinal tensile stress caused by the deformation of the surrounding steel, so the interface is separated and voids are generated. As the cavity expands in the rolling direction, the vertical compressive stress is no longer balanced, and the combination of vertical and longitudinal stress causes the steel to partially move into the cavity, creating a tapered cavity. Waudby et al. (24) claimed that the combined force of the stress of the steel matrix flowing with the tangential action of the surface of the particle caused and widened the crack and created a void. Luo et al. (25) proposed that if the resolved normal stress at the interface reaches a critical value, peeling occurs and voids are generated by finite element calculation. Belcheko et al. (26) proposed that the flow of the substrate above and below the undeformed inclusions in the rolling direction results in the formation of conical voids. Zhang et al. (27) argued that although there are subtle differences in the mechanism of void formation proposed by different researchers, it is generally believed that the formation of voids is due to the discontinuity of the interface between the particles and the matrix. Therefore, the interfacial strength between particles and the matrix is the main cause of forming of the adjacent interface voids. It is suggested that the interfacial strength between Y2O3 particles and the platinum matrix is not sufficient to withstand the tensile stresses during processing. A crack will form between the particles and the matrix. As the strain continuously increases, the matrix work hardens, and when the hardness of platinum reaches a critical value, the Y2O3 particles will be deformed. In addition, the agglomeration of the reinforcing phase particles causes an increase in the local volume fraction, which increases the internal stress and causes the destruction of the particles. The coarsening of the particles reduces the stress required for particle damage, and the rate of damage of the particles increases with size. As for ZrO2 particles, no voids are observed because the interfacial strength between ZrO2 particles and the platinum matrix is able to withstand the rolling tensile stresses during processing. ZrO2 is also hard enough to stand the stress form matrix so it will keep its spherical shape.
In platinum-rhodium-(zirconium)-yttrium alloys, yttrium and zirconium are easily oxidised into Y2O3, ZrO2 and Y2O3·ZrO2 and form corresponding oxides while platinum and rhodium are basically present as pure metals. ZrO2 and Y2O3 particles have been observed in platinum-rhodium-zirconium-yttrium and platinum-rhodium-yttrium, respectively, and verified by EDS analysis and HR-TEM observations. The deformation behaviour of these two oxides is quite different during processing, though they have the same deformation history. The ZrO2 particles maintained their spherical shape without any visible deformation, while the Y2O3 particles were compressed along the normal direction with two cracks forming on two sides of the particles. Insufficient hardness of Y2O3 and relatively lower interface strength between Y2O3 particles and matrix were supposed to be responsible for deformation of Y2O3 during processing.
H. Chen, W. Xie, W. Liu, Z. Wang, Z. Yang, H. Tang, Y. Dai, G. Wei and W. Xie, Vacuum, 2019, 160, 445 LINK https://doi.org/10.1016/j.vacuum.2018.12.012
J. V. Pearce, F. Edler, C. J. Elliott, A. Greenen, P. M. Harris, C. G. Izquierdo, Y.-G. Kim, M. J. Martin, I. M. Smith, D. Tucker and R. I. Veltcheva, Metrologia, 2018, 55, (4), 558 LINK https://doi.org/10.1088/1681-7575/aacbf7
B. X. Hu, Y. Ning, L. Chen, Q. Shi and C. Jia, Platinum Metals Rev., 2012, 56, (1), 40 LINK https://www.technology.matthey.com/article/56/1/40-46/
Z. Xie, F. Lu and J. Wang, Precious Met., 1996, 20, 10 LINK http://en.cnki.com.cn/Article_en/CJFDTotal-GJSZ603.001.htm
K. Teichmann, C. H. Liebscher, R. Völkl, S. Vorberg and U. Glatzel, Platinum Metals Rev., 2011, 55, (4), 217 LINK https://www.technology.matthey.com/article/55/4/217-224/
G. L. Selman, J. G. Day and A. A. Bourne, Platinum Metals Rev., 1974, 18, (2), 46 LINK https://www.technology.matthey.com/article/18/2/46-57/
G. L. Selman and A. A. Bourne, Platinum Metals Rev., 1976, 20, (3), 86 LINK https://www.technology.matthey.com/article/20/3/86-90/
Y. Dai, Q. Ma, W. Liu, F. Hu, Y. Zhang and Z. Yang, Mater. Sci. Technol., 2018, 34, (6), 654 LINK https://doi.org/10.1080/02670836.2017.1410355
B. Fischer, D. Freund, A. Behrends, D. Lupton and J. Merker, Precious Met., 1998, 22, 333
J. Stokes, Platinum Metals Rev., 1987, 31, (2) 54 LINK https://www.technology.matthey.com/article/31/2/54-62/
M. S. Rowe and A. E. Heywood, Mater. Design, 1984, 5, (1), 30 LINK https://doi.org/10.1016/0261-3069(84)90173-0
F. Hu, T. Yu, W. Liu, Y. Yang, G. Wei, X. Luo, H. Tang, N. Hansen, X. Huang and W. Xie, Mater. Sci. Eng. A, 2019, 765, 138305 LINK https://doi.org/10.1016/j.msea.2019.138305
G. L Selman and A. S. Darling, Johnson Matthey Plc, ‘Dispersion Strengthening of Platinum Group Metals and Alloys’, US Patent 3709667A; 1973
S. K. Karak, T. Chudoba, Z. Witczak, W. Lojkowski and I. Manna, Mater. Sci. Eng. A, 2011, 528, (25–26), 7475 LINK https://doi.org/10.1016/j.msea.2011.06.039
K. Maruyama, H. Yamasaki and T. Hamada, Mater. Sci. Eng. A, 2009, 510 – 511, 312 LINK https://doi.org/10.1016/j.msea.2008.08.051
N. Sekido, A. Hoshino, M. Fukuzaki, Y. Yamabe-Mitarai and T. Maruko, Mater. Sci. Eng. A, 2011, 528, (29–30), 8451 LINK https://doi.org/10.1016/j.msea.2011.08.011
Yu. E. Gorbunova, V. V. Ilyukhin, V. G. Kuznetsov, A. V. Lavrov and S. A. Linde, Doklady Akademii Nauk SSSR, 1977, 234, (3), 628
G. Katz, J. Am. Ceram. Soc., 1971, 54, (10), 531 LINK https://doi.org/10.1111/j.1151-2916.1971.tb12197.x
F. Frey, H. Boysen and T. Vogt, Acta Cryst., 1990, B46, 724 LINK https://doi.org/10.1107/s0108768190007509
K. Huang, K. Marthinsen, Q. Zhao and R. E. Logé, Prog. Mater. Sci., 2018, 92, 284 LINK https://doi.org/10.1016/j.pmatsci.2017.10.004
H. Yamazaki, Japanese Patent 163069; 2005
K. B. Gove and J. A. Charles, Met. Technol., 1974, 1, (1), 425 LINK https://doi.org/10.1179/030716974803287366
T. J. Baker, K. B. Gave and J. A. Charles, Met. Technol., 1976, 3, (1), 183 LINK https://doi.org/10.1179/030716976803391656
P. E. Waudby, W. J. M. Salter and F. B. Pickering, J. Iron Steel Inst., 1973, 211, (7), 486
C. Luo, Comput. Mater. Sci., 2001, 21, (3), 360 LINK https://doi.org/10.1016/s0927-0256(01)00149-5
C. Luo and U. Ståhlberg, J. Mater. Proc. Technol., 2001, 114, (1), 87 LINK https://doi.org/10.1016/S0924-0136(01)00576-3
Z. Zhang and W. Pantleon, Acta Mater., 2018, 149, 235 LINK https://doi.org/10.1016/j.actamat.2018.02.042
This work was supported by Chongqing Science and Technology Support Project (No. cstc2017zdcy-zdyfX0070, cstc2018jszx-cyzdX0138) and Fundamental Research Funds for the Central Universities (No. 2019CDCGCL316) and National Undergraduates Training Program for Innovation (No. 201810611054).
Ziyang Wang received his Bachelor’s degree in 2020 from Chongqing University, China. He entered Chongqing University, majored in Material Forming and Control Engineering in 2016. His interests include light alloys and composite materials.
Xi Wang received his Bachelor’s degree in 2020 from Chongqing University, China, and he continued to study at Chongqing University for a Master’s degree in the automobile college. Forging and new energy vehicles are his research field.
Futao Liu received his Bachelor’s degree in 2020 from Chongqing University, China. He entered in Chongqing University, majored in Material Forming and Control Engineering in 2016. His dissertation is about the heat treatment of magnesium-lithium alloys.
Faping Hu is a PhD candidate from 2017 in Materials Science and Engineering at Chongqing University. He visited the Technical University of Denmark as a guest PhD for two years. His research is the microstructural characterisation of magnesium alloys during plastic deformation.
Hao Chen is a PhD candidate from 2020 in Materials Science and Engineering at Chongqing University. He studied platinum-rhodium alloys and glass fibre reinforced composites. His current research is on freeze casting.
Guobin Wei received his PhD in Materials Science and Engineering in 2015 from Chongqing University, China. He is vice professor of the School of Materials Science and Engineering. His interests include magnesium-lithium alloys and the simulation of material forming processes.
Weiting Liu is a senior engineer and Vice President of Chongqing International Composite Materials Co Ltd, China, supporting and promoting research and manufacture of platinum-rhodium alloy bushings and glass fibre. His professional affiliation includes Deputy Executive Director of the Functional Materials Association and he obtained a Bachelor’s degree in 1993 from the Department of Mechanical Engineering of Chongqing University, China.
Weidong Xie received his PhD in Materials Science and Engineering in 2008 from Chongqing University, China. He is Vice Dean of the Institute of Scientific Research and Development of Chongqing University and a council member of the Chinese Society for Composite Materials. His interests include light alloys, composite materials, nanomaterials and foundry.