CONFERENCE PAPERS

 

A comparative study of hardness of Pt-Cr-V and Pt-Al-V alloys in the as-cast and annealed conditions

 

 

B.O. Odera^{I, II, III}; L.A. Cornish^{II, III, IV}

^IDepartment of Mechanical Engineering, Cape Peninsula University of Technology, Cape Town
^IISchool of Chemical and Metallurgical Engineering, University of the Witwatersrand, Johannesburg
^IIIAfrican Materials Science and Engineering Network (AMSEN), a Carnegie-IAS Network
^IVDST-NRF Centre of Excellence in Strong Materials, hosted by University of the Witwatersrand, Johannesburg

 

 

---

SYNOPSIS

Microhardness values of alloys in the Pt-Cr-V and Pt-Al-V systems were determined using an indenter which incorporated an optical microscope. The values were analysed with respect to the phases identified in each of the alloys. Identification of the phases had been done earlier using scanning electron microscopy with energy dispersive X-ray spectroscopy and X-ray diffraction. The hardness values were determined for samples in the as-cast and annealed conditions and superposed on the solidification projection and isothermal section at 1000°C. The results showed that the hardness depended largely on the phases present in the alloys and generally increased after annealing. However, in a few cases, grain growth and the resulting microstructural coarseness resulted in lower hardness values after annealing. The hardness of the alloys of the Pt-Cr-V and Pt-Al-V systems was also compared with those of Pt-Al-Cr and other Pt-Al based alloys. Comparison was also made with some Pt-Al based alloys of the higher order systems such as Pt-Al-Cr-Ru-V and Pt-Al-Cr-Ru-V-Nb.
In general, higher hardness was exhibited by alloys containing ternary phases.

Keywords: Pt-based ternary alloys, Pr-Cr-V, Pt-Al-V, hardness, microstructure.

---

 

 

Introduction

Platinum-based alloys for high-temperature corrosive environments have been investigated as possible replacements for nickel-based superalloys (NBSAs) in the hotter parts of gas turbines (Wolff and Hill, 2000; Hill et al., 2001b). The NBSAs have excellent mechanical properties due to their γ/γ' structure but are limited by their maximum operating temperatures. Currently, the maximum temperature at which NBSAs operate is approximately 1100°C, which is approximately 85% of their melting temperature (Sims, Stoloff and Hagel, 1987). There is a need to develop new alloys that can operate at higher temperatures. The advantages include increased efficiency, which is related to the combustion temperature and higher temperatures would enable greater thrust, improved fuel efficiency and reduced pollution.

Platinum has the same fcc structure and similar chemistry to nickel, as well as a higher melting point (1769°C for platinum, compared to 1455°C for nickel) and better corrosion resistance than nickel (Süss et al., 2003).

Alloy development work done by other researchers demonstrated that ternary and quaternary Pt-based alloys can exhibit a γ/γ' two-phase structure, similar to nickel-based superalloys (Hill et al., 2001; Cornish Fischer and Völkl, 2003). Although the use of Pt-based alloys as a replacement for Ni-based superalloys is limited due to their higher price and higher density, it is possible that they could be used for critical components, or as corrosion-resistant coatings (Süss et al., 2009).

Some of the quaternary alloys exhibiting the γ/γ' two-phase structure were of the Pt-Al-Cr-Ru system. Alloys of the quinary Pt-Al-Cr-Ru-V system have also been investigated as potential replacement of NBSAs (Odera, 2013; Odera et al., 2015). Vanadium replaced some of the Pt to reduce density and cost, as well as acting as a solid solution strengthener. Pt-Al-V and Pt-Cr-V are two component ternary systems of the quinary and a study of the microstructures of selected alloys of the systems and their ternary phase diagrams formed part of the investigation (Odera et al., 2012b; Odera, 2013; Odera et al., 2014). This paper presents hardness values of selected alloys of the two ternary systems in the as-cast and annealed conditions and compares and discusses these values in relation to the microstructures.

There is an empirical relationship between hardness and strength of a metal. Since hardness testing is easier to perform than tensile strength testing, especially for small samples, it is often used to obtain an indication of the strength of a metal, especially when only small samples are available (Dieter, 1986; Smith, 1990).

 

Experimental procedure

Twenty-four alloy buttons weighing approximately 2 g each were prepared from Pt, Al, Cr of 99.9% purity and V of 99.6% purity. Out of these, 11 samples identified as Alloy 1 to Alloy 11 were of the Pt-Al-V system, while those identified as Alloy 12 to Alloy 24 were of the Pt-Cr-V system, all in the as-cast condition. The annealed samples were identified as 1H to 11H and 12H to 24H respectively. The samples were manufactured by arc-melting under argon atmosphere on a water-cooled copper hearth with a Ti oxygen-getter. Each button was turned over and re-melted three times in an attempt to achieve homogeneity. The samples were then halved and one half of each was metallographically prepared in the as-cast condition, while the other half was annealed for 1500 hours before preparation.

The samples were ground on silicon carbide down to 1200 grit and then diamond polished down to 1 μm. The samples were then etched in a solution of 10 g NaCl in 100 cm3 HCl (32% vol. concentration) (Odera et al, 2012a). Etching was done in a fume cupboard using a DC power supply and a voltage range of 9 V to 12 V gave adequate results. The current density in the electrolyte was approximately 100 A.m-2. The counter-electrode was a stainless steel wire suspended in the electrolyte solution. The indenter used incorporated an optical microscope with a maximum magnification of 1000 times and only an etched surface could be seen clearly in the machine. In the unetched condition, it was difficult to differentiate between the sample and the mounting resin surface. The hardness of the alloys was measured using a Vickers microhardness tester with a load of 300 g. At least five measurements were carried out to obtain an average hardness value.

 

Results

Figure 1a shows an indentation on as-cast Alloy 5, average composition Pt_83.9:Al₆:V10.1 (at.%) and Figure 1b shows an indentation on annealed Alloy 5H of the same average composition. The indentations on as-cast Alloy 24 and annealed Alloy 24H, average composition PW₃₄Cr_57.4 are shown in Figure 2. Figures 3 and 4 show indentations on as-cast Alloys 15 and 18, average compositions Pt_53.0:Cr_22.4:V_24.5and Pt_25.8:Cr_42.9:V_31.3 respectivley. Table I gives hardness values (HVa₃) for as-cast and annealed Pt-Al-V alloys, together with the phases in each of the alloys, while Table II gives the same data for Pt-Cr-V alloys.

 

 

 

 

Discussion

As-castAlloys ofthe Pt-AI-VSystem

Alloys with a Pt content more than 70 at.% had the lowest hardness values (less than 500 HVa₃), while 64% of the alloys had relatively high hardness values of more than 500 HVa₃. The single-phase ternary Alloy 11 had the highest hardness of 886 HV0.3, which would be expected and is likely to be brittle. Generally, the load of 300 g used was too low to cause substantial cracking during indentation on these particular alloys. However, Alloy 11 exhibited cracks before indentation, confirming its brittle nature.

Hill et al. (2001a, 2001b) reported Vickers macrohardness measurements on several Pt-based ternary alloys at room temperature. Eight of these were Pt-Al-Z alloys (where Z was Ni, Ru or Re). Six of the alloys had Vickers hardness less than 500 HV in the as-cast and annealed conditions and the highest value was 530 HV (the load used was not specified). In his investigation of Pt-Al-Cr alloys, Süss (2007) found that 75% of the alloys in the as-cast condition had Vickers hardness over 600 HV₁₀ at room temperature.

Figure 5 shows hardness values of as-cast Pt-Al-V alloys superimposed on the partial solidification projection at the Pt-rich corner derived and reported by Odera (2013). It shows the hardness values increasing away from the Pt-rich corner. The composition of the two ternary phases τ₁and τ₂are approximately V₂₄Pt₅₆Al₂₀ and V₁₈Pt₅₇Al₂s respectively.

 

 

Pt-AI-Valloys annealed at 1000°C for 1500 Hours

The hardness values for annealed alloys of the Pt-Al-V system are given in Table III. Seventy per cent of the annealed alloys had hardness values higher than 500 HV and the rest, which had relatively high Pt contents above 80 at.%, had hardness values below 500 HV_a3. About 45% of the alloys had hardness values over 600 HVa₃. Annealed Alloy 11H cracked and disintegrated while being ground in preparation for polishing and etching, indicating that it was very brittle and it also contained the ternary phase in as the-cast condition. In all of the alloys, there was no cracking or noticeable deformation around the indentations, therefore their toughness could not be quantitatively assessed. However, the absence of notable cracking suggests reasonable toughness (except for Alloy 11 which comprised the ternary phase in the as-cast condition and had cracks before indentation). However, this could also be attributed to the relatively small load used in the microhardness testing. The small load was chosen because of the small size of the samples (approx. 1 g). However, these hardness values for the alloys, whether in the as-cast or annealed condition, are very high compared to that of pure platinum, which is very soft and ductile with a Vickers hardness of 50 HV (the load was not specified) (Murakami, 2008).

These results are comparable to the findings of Süss (2007) in his investigation of Pt-Al-Cr alloys, where just over half of the annealed alloys had hardness values above 600 HVi₀. Only one of the Pt-Al based alloys, Pt6₂:Al₁₉:Ni₁₉, investigated by Hill (2001) had a hardness value of 530 HV (the load used was not specified). The rest had hardnesses below 500 HV.

The phase ~Pt2Al disappeared during annealing of Alloy iH, but the hardness remained more or less the same. In Alloy 2H, two phases, ~Pt2V and ~PtAl, disappeared and a new phase, ~PtV, formed while ~Pt5Al3 was present both in the as-cast and annealed conditions. The hardness of Alloy 2H reduced by approximately 100 HVa₃, suggesting that one or both of the two phases that disappeared contributed to the higher hardness in the as-cast condition.

The hardness of Alloys 3 and 4 remained more or less the same after annealing. In Alloy 3, the two phases ~Pt3V and ~Pt2Al disappeared and two new phases, ~Pt2V and ~PtV, formed while in Alloy 4 there was no change in the phases.

In Alloy 5, a new phase, ~Pt2V, formed during annealing, while the two phases (Pt) and ~Pt3Al were retained. There was a slight increase of hardness from 420±20 HV0.3 to 482±14 HVa₃, suggesting that the new phase was responsible for the increase, as expected. The phases in Alloy 6 remained the same and, as expected, the hardness was also similar after annealing.

The hardness of Alloys 7 and 8 remained the same after annealing if the standard deviations are taken into account. This is not surprising as the phases ~PtV, ~PtV3 and ~Pt5Al3 remained the same before and after annealing.

The hardness of Alloy 9 decreased by approximately i00 HV, while the two phases (Pt) and ~Pt3Al remained the same in the as-cast and annealed conditions. The decrease in hardness may be explained by coarsening of the phases during annealing. Alloy 10 changed from two-phase ~Pt₃Al + ~Pt2Al to single phase ~Pt3Al, while the hardness increased from 462±14 HV_a3 to 535±23 HV_a3. This suggests that ~Pt2Al is a soft phase compared to ~Pt3Al and its disappearance contributed to the increase in hardness.

Alloy 11 consisted of a single ternary phase, τ2 and was the hardest alloy, even in the as-cast condition, with Vickers hardness of 886±23 HV0.3. Further evidence of its hardness and brittleness were the visible cracks around the indentation in the as-cast condition. It disintegrated while being removed from the resin mount and so it was not possible to measure the hardness in the annealed condition.

Figure 6 shows hardness values superimposed on isothermal section of Pt-Al-V at 1000°C, at the Pt-rich corner as derived and reported earlier (Odera et al., 2014). The same trend is seen as with the as-cast alloys, where the hardness values reduced with increasing Pt content.

 

 

As-cast Pt-Cr-Valloys

The alloys having ternary phase dendrites (Alloys 17, 18 and 21) had very high hardness values, as expected. Alloy 18 had the highest hardness value (1109±25 HV₀.₃). The second hardest alloy was Alloy 21, which had a hardness of 1086±28 HVa₃. All the indentations on Alloy 15 were on the eutectic areas and the high hardness value of 936±59 HV0.3 reflects the high hardness of the ternary phase component of the eutectic, rather than the binary intermetallic component. The single-phase Alloys 12 and 13, which had high Pt contents, had the lowest hardness values. Alloy 16, which also had high Pt content, had a similarly low hardness. The rest of the alloys all had relatively high hardness values, except for Alloy 24 which had a low hardness value of 358±19 HV_a3.

The hardness values were comparable to those obtained by Süss (2007) in his investigation of Pt-Al-Cr alloys. Süss (2007) used a load of 10 kg and many of the alloys cracked and deformed around the indentations. However, the highest Vickers hardness values obtained by Süss (2007) were between 800 HV₁₀ and 900 HV₁₀. The hardness of the Pt-Cr-V alloys investigated in this work were also generally much higher than the ternary Pt-Al based alloys investigated by Hill et al. (2001) which had a maximum hardness value of 530 HV (the load used was not specified). These hardness values for the alloys, whether in the as-cast or annealed condition, were much higher than for pure platinum, which is very soft and ductile with a Vickers hardness of approxi mately 50 HV (Murakami ,2008).

Figure 7 shows hardness values superimposed on the solidification projection of the Pt-Cr-V system. It shows that alloys having high Pt content have the lowest hardness and there is an increase in hardness as one moves away from the Pt-rich corner.

 

 

Pt-Cr-Valloys annealed at 1000°C for 1500 hours

There was a general increase in hardness values after annealing the Pt-Cr-V alloys, except for Alloys 15H and 23H, which had lower values and this was probably due to microstructure coarsening. Alloys 17, 18 and 21 (as-cast) had ternary phase dendrites with high hardness values in the as-cast condition, as expected. These three alloys also exhibited extreme brittleness. The annealed Alloys 18H and 21H cracked and disintegrated while being removed from the resin mounting. Alloy 18H cracked, but stayed as one piece and therefore it was possible to undertake microhardness tests. The annealed Alloy 20H also cracked and disintegrated while being removed from the resin mounting. The alloys with high Pt content had the lowest hardness values.

If the three annealed alloys that disintegrated are included among those with hardness values exceeding 600 HV0.3, then approximately 62% of the alloys had Vickers hardnesses higher than 600 HV0.3. This shows that the alloys were generally harder than those investigated by Süss (2007), just over half of which had Vickers hardnesses higher than 600 HV₁₀. Again, if the same three alloys are included among those that had hardness values higher than 750 HV0.3, then slightly more than half the alloys had hardness values more than 750 HV0.3 compared to those investigated by Süss (2007), only three of which (17%) had hardness values higher than 750 HV₁₀. The assumption that the three alloys that disintegrated had hardness values higher than 750 HV0.3 is valid because the hardest annealed alloy whose hardness was measured is Alloy 18H (1086 HV0.3) and an other three alloys had hardness values exceeding 800 HVa₃. These alloys were much harder than the Pt-Al based alloys investigated by Hill (2001), the hardest of which had a Vickers hardness of 530 HV (the load used was not specified).

The hardness of Alloys 12H and 13H remained similar after annealing. Alloy 12H remained single phase ~Pt₃V after annealing, while Alloy 13H changed from a single phase structure of ~Pt3V to a two-phase structure of ~Pt2V and ~Pt3V. This shows that the two phases ~Pt2V and Pt3V have similar hardness. Alloy 16H, which was also single phase ~Pt3V in the as-cast condition, retained the same phase after annealing, with a slight increase in hardness. This is probably due to traces of (Pt) detected by XRD in the as-cast alloy but which disappeared after annealing.

The phases in Alloy 14H were the same before and after annealing and there was not much change in the hardness, as expected. However, there was a substantial decrease in the hardness of Alloy 15H after annealing, although the phases were the same. This decrease is attributed to coarsening.

In Alloy 17H, the phase ~Cr3Pt disappeared and ~Pt2V formed after annealing, while the ternary phase, τ3, was present both in the as-cast and annealed conditions. The increase in hardness from 854±30 HV0.3 to 1085±46 HV0.3 is therefore attributed to ~Pt2V, as expected. Alloy 18, which had the highest hardness value at 1109±25 HV0.3 in the as-cast condition, disintegrated while being removed from the resin mounting after SEM analysis following annealing, therefore its hardness was not measured. However, the phases, τ3, A15(~Cr3Pt) and ~Cr3Pt, remained the same in the as-cast and annealed conditions and the disintegration indicated extreme hardness and brittleness of the alloy.

The phases in Alloy 19H did not change after annealing, but the hardness increased from 796±35 HV0.3 to 923± 57 HV0.3. This was the same with Alloy 20H, where the phases (V,Cr) and ~PtV3 remained the same in the as-cast and annealed conditions. However, Alloy 20H disintegrated while being removed from the resin mounting and it was not possible to measure its hardness after annealing, although the disintegration suggested extreme hardness and brittleness.

Alloy 21 had a hardness value of 1086±28 HV_a3, making it the second hardest alloy in the as-cast condition. It had the same phases ~Cr₃Pt, ~PtV₃, τ₃and (V,Cr) in the as-cast and annealed conditions. It disintegrated while being removed from the resin mounting and as such, its annealed hardness was not measured. The disintegration indicated extreme hardness and brittleness. The phases, (V,Cr) and A15(~Cr₃Pt), in Alloy 22H remained the same in the as-cast and annealed conditions, but the hardness increased from 707±41 HV_a3 to 837±46 HV_a3.

Alloy 23H had the same phases, (V,Cr) and A15(~Cr₃Pt) in the as-cast and annealed conditions, but the hardness decreased from 692±73 HV0.3 to 544±20 HV0.3. This deviated from the trend seen for most of the alloys in this system. Alloy 24H also retained the same phases, ~Cr₃Pt and A15 (~Cr₃Pt), in the as-cast and annealed conditions, but there was a substantial increase in hardness from 358±9 HV0.3 to 838±27 HV0.3. This was because heat treatment produced more of the A15 (~Cr₃Pt), which would be expected to be harder, being an intermetallic phase.

Figure 8 shows hardness values superposed on the isothermal section at 1000°C derived by and reported in Odera (2013). Hardness is lowest at the Pt-rich corner and increases with reduction in Pt content.

 

 

Figure 9 shows a combined plot of hardness values of as-cast and annealed Pt-Al-V alloys, while Figure 10 shows a similar plot for Pt-Cr-V alloys. There is a general increase in hardness values for the Pt-Cr-V alloys after annealing, while there is little change in the values for Pt-Al-V alloys.

 

 

 

 

Comparison with higher order alloys

Table III shows the HV0.3 hardness values of higher order Pt-based alloys (Odera et al., 2015). Pt-Al-V and Pt-Cr-V alloys were generally much harder than the as-cast higher order alloys, because many of the Pt-Al-V and Pt-Cr-V alloys contained hard Pt-V intermetallic phases. The high hardness of some of the Pt-Al-Cr alloys (Süss, 2007) was attributed to ~PtAl₂, ~PtAl and ~Pt₂Al₃.

 

Conclusions

Microhardness values for selected Pt-Al-V and Pt-Cr-V alloys were determined in the as-cast and annealed conditions. Alloys with high Pt content (more than 70 at.%) had the lowest hardness values, while those containing ternary phases had the highest hardness. The single ternary phase Alloy 11 (average composition Pt₅a₉:Al_25.4:V₁₇j) had the highest hardness of 886±23 HVa₃among the Pt-Al-V alloys in the as-cast condition. Alloy 18 (average composition Pt_25.8:Cr_42.9:V_3L3) had the highest hardness value of 1109±25 HVa₃among the Pt-Cr-V alloys in the as-cast condition. These two sets of alloys, especially the Pt-Cr-V alloys, had higher hardness values than the Pt-Al-Cr alloys studied by Süss (2007) and Pt-Al based alloys studied by Hill et al. (2001), as well as the high-order Pt-based alloys studied by Odera et al. (2015).

 

References

Cornish, L.A., Fischer, B. and Völkl, R. 2003. Development of platinum-group- metal superalloys for high-temperature use. MRS Bulletin, vol. 28, no. 9. pp. 632-638.         [ Links ]

Dieter, G.E. 1986. Mechanical Metallurgy. 3rd edn. McGraw-Hill, New York. pp. 329-330.         [ Links ]

Hill, P.J., Cornish, L.A., Ellis, P. and Witcomb, M.J. 2001a. The effects of Ti and Cr additions on the phase equilibria and properties of (Pt)/Pt3Al alloys. Journal of Alloys and Compounds, vol. 322, no. 1. pp.166-175.         [ Links ]

Hill, P.J., Biggs, T., Ellis, P., Hohls, J., Taylor S.S. and Wolff, I.M. 2001b. An assessment of ternary precipitation-strengthened Pt alloys for ultra-high temperature applications. Materials Science and Engeering A, 301, pp. 67-179.         [ Links ]

Murakami, T., Sahara, R., Harako, D., Akiba, M., Narushima, T. and Ouchi, C. 2008. The effect of solute elements on hardness and grain size in platinum based binary alloys. Materials Transactions, vol. 49, no. 3. pp. 538-547.         [ Links ]

Odera, B.O. 2013. Addition of vanadium and niobium to platinum-based alloys. PhD thesis, University of the Witwatersrand, Johannesburg.         [ Links ]

Odera, B.O., Cornish, L.A., Papo, M.J. and Rading, G.O. 2012a. Electrolytic etching of platinum-aluminium based alloys. Platinum Metals Review, vol. 56, no. 4. pp. 257-261.         [ Links ]

Odera, B., Cornish, L.A., Shongwe, M.B., Rading, G. and Papo, M.J. 2012b. As cast and heat-treated alloys of the Pt-Al-V system at the Pt-rich corner. Journal of the Southern African Institute of Mining and Metallurgy, vol. 112, no. 7. pp. 505-516.         [ Links ]

Odera, B., Papo, M.J., Rading, G. and Cornish, L.A. 2014. Experimental solidification projection, liquidus surface projection and isothermal section at 1000° C for the Pt-Cr-V system. Journal of Phase Equilibria and Diffusion, vol. 35, no. 4. pp. 476-489.         [ Links ]

Odera, B., Papo, M.J., Couperthwaite, R., Rading, G., Billing, D. and Cornish, L.A. 2015. High-order additions to platinum-based alloys for high-temperature applications. Journal of the Southern African Institute of Mining and Metallurgy, vol. 115, no. 3. pp. 241-250.         [ Links ]

Sims, C.T., Stoloff, N.S. and Hagel, W.C. 1987. High-temperature Materials/or Aerospace and Industrial Power. Wiley, New York.         [ Links ]

Süss, R., Cornish, L.A., Hill, P.J., Hohls, J. and Compton, D.N. (2003). Properties of a new series of superalloys based on Pt₈₀:Al₁₄:Cr₃:Ru₃. Advanced Materials and Processes for Gas Turbines, pp. 301-307.         [ Links ]

Süss, R. 2007. Investigation of the Pt-Al-Cr system as part of the development of the Pt-Al-Cr-Ru thermodynamic database. PhD thesis, University of the Witwatersrand, Johannesburg.         [ Links ]

Süss, R., Watson, A., Cornish, L.A. and Compton, D.N. 2009. Development of a database for the prediction of phases in Pt-Al-Cr-Ru alloys for high- temperature and corrosive environments: Al-Cr-Ru. Journal of Alloys and Compounds, vol. 476, no. 1. pp. 176-186.         [ Links ] ♦

 

 

This paper was first presented at the AMI Precious Metals 2017 Conference 'The Precious Metals Development Network' 17-20 October 2017, Protea Hotel Ranch Resort, Polokwane, South Africa.