Experimental Techniques

In the stress-rupture test, a load is applied to a sample of the material and the time to rupture of the sample is determined. The values of nominal stress are plotted against the time to rupture on a double-logarithmic scale to obtain the stress-rupture curve; this normally approximates to a straight line.

As equipment for stress-rupture testing at extremely high temperatures was not commercially available, a special facility for testing high-melting metals was developed and built where:

Using this test facility, comparative investigations were conducted on various platinum, rhodium and iridium materials. The samples of platinum materials for the tests were in the form of strips, 0.5 × 3.0 × 120 mm, supplied from normal production of sheet from Heraeus. For tests on rhodium and iridium the materials were electron-beam melted, then rolled and cut into wire samples of size 1.0 × 1.0 × 120 mm. Comprehensive metallographic and scanning electron microscope (SEM) investigations of the microstructure, the surface structure and the fracture faces were carried out on the samples after testing, to obtain a better understanding of their behaviour during the test. The results of these investigations are summarised here with some examples.

High Temperature Properties of Platinum and Platinum Alloys

Stress-rupture curves at 1700°C for platinum alloys containing 10 to 30 per cent of rhodium and iridium and at 1600°C for pure platinum are shown in Figure 1 (20). The stress-rupture curve of pure platinum was included for comparison, because at 1700°C platinum can neither be tested in the stress-rupture apparatus nor used as a structural material. The individual values obtained showed an excellent degree of reproducibility and lie almost ideally on straight lines in the stress-rupture diagrams. By interpolation from the stress-rupture curve, it is possible to define the stress-rupture strength for a particular time, for example, for 10 hours, to facilitate comparison of different materials.

In Figure 1, it can be seen that solid-solution strengthening of platinum leads to an enormous increase in the stress-rupture strength. As would be expected, the effect increases with increasing amounts of the alloying addition. The alloys investigated have a higher stress-rupture strength at 1700°C than pure platinum at 1600°C.

However, alloys with alloying additions of more than about 30 per cent are not generally used as structural materials because of problems that arise in their processing. Values for the stress-rupture strengths of alloys with 20 per cent rhodium and 20 per cent iridium do not differ very greatly, although the platinum-rhodium (Pt-Rh) alloy has a slight strength advantage. Increasing the alloying addition by a further 10 per cent leads to only a moderate increase in stress-rupture strength for the alloy platinum 30 per cent rhodium (Pt-30%Rh), whereas the platinum-30 per cent iridium (Pt-30%Ir) alloy shows a substantial strength increase, see, respectively, Figures 2(a) and 2(b). At a test temperature of 1700°C, for example, the increase in strength for Pt-20%Rh on addition of a 10 per cent increase in rhodium is only ∼ 0.3 N mm ² compared with ∼ 2.5 N mm^-2 for the Pt-Ir alloys, that is, the strength of these alloys is doubled. Similar relationships have been found in studies of high temperature elastic properties of these alloys (21).

In practice alloys with more than 20 per cent rhodium are rarely used because the increase in strength does not compensate for increased problems in fabrication. Care must also be taken in using Pt-Ir alloys with high iridium contents because they can become embrittled in service at intermediate temperatures due to the precipitation of primary iridium in the platinum matrix.

Figure 3 shows the 10-hour stress-rupture strengths as a function of temperature for pure platinum and the alloys over the temperature range 1400 to 1700°C. At all temperatures the Pt-30%Ir alloy has the highest strength. The stress-rupture strength of pure platinum even at 1400°C is lower than that of any of the alloys investigated at 1700°C. For all the alloys the relative decrease in stress-rupture strength resulting from a temperature increase from 1400 to 1700°C is very similar. However, the strength of pure platinum, although low at all temperatures, decreases more slowly as the temperature is increased.

Using alloys at high temperatures under conditions of mechanical loading leads to a change in their dimensions by creep, that is by diffusion-controlled deformation processes that are time and rate dependent. Figure 4 shows creep curves of the platinum alloys investigated. The stress was selected for each sample so that a stress-rupture life of approximately 10 hours was achieved. The change in the overall length of the samples was measured as a function of the testing time. The elongation values therefore relate to the nominal gauge length of 100 mm. However, because samples were direct resistance heated, only the central part of each gauge length reaches the test temperature and it is cooler near the grips. The true elongation in the central portion of the sample is therefore considerably greater than the values shown. In the secondary, steady-state range of creep, all alloys show similar creep rates in the range 2 to 5 μm s^-1.

The Pt-Ir alloys give higher values for rupture elongation than the Pt-Rh alloys. The Pt-30%Ir alloy, which has the highest stress-rupture strength, also shows the greatest elongation to fracture. It also demonstrates excellent ductility at very high temperatures, despite the very high level of solid solution strengthening and the tendency of this alloy to embrittlement at lower temperatures.

The high fracture elongation of the Pt-30%Ir samples in the stress-rupture tests was accompanied by a very considerable degree of necking. After short test times slip bands could be seen on the surface of the material in the zone close to the fracture, some of which can be seen in Figure 5(a).

At longer test times, intergranular cracks, creep pores and superficial structures resulting from the selective evaporation of iridium were observed, see Figure 5(b). A certain amount of loss of iridium by evaporation in the form of volatile oxides had no measurable influence on the high temperature mechanical properties at the test durations studied.

High Temperature Properties of Dispersion Hardened Platinum Materials

In a number of applications solid solution strengthening is insufficient, for example, when maximum mechanical stability is required at very high temperatures. The use of solid solution additions may also be limited for other reasons, such as, they have insufficient oxidation resistance or there is embrittlement of the alloy after exposure. In these cases the dispersion hardening of platinum materials can be used to its full advantage.

One example of this is the melting and processing equipment used for the manufacture of quality high grade optical glass, when it is not possible to use the Pt-Rh and Pt-Ir alloys discussed above. If trace quantities of the alloying elements dissolved in the glass melts it would be severely detrimental to the optical properties of the glass, causing discoloration of the glass and a reduction in the optical transmission. Less noble alloying elements cannot be considered for the solid solution strengthening of platinum in these applications because they would be detrimental to the corrosion resistance of platinum in the molten glass and the corrosion products would lead to even greater contamination.

Platinum used for melting equipment for optical glass must therefore be strengthened by other means. A strengthening effect can be achieved by the introduction of finely dispersed stable particles into the alloy matrix; this is known for other materials, such as the alloys of nickel, copper and aluminium. In order to ensure that the dispersion hardening is effective up to the highest possible temperatures, the dispersed phase must have a very high thermodynamic stability, a higher melting temperature than the matrix and must be insoluble in the matrix even at the highest temperatures. For this reason oxides are preferable to carbides and silicides. Furthermore, oxygen is harmless to platinum, in contrast to both carbon and silicon.

Oxides of zirconium and yttrium have proved particularly effective for the oxide dispersion hardening of platinum. Provided that the oxide particles are very small (< 1 μm) and the inter-particle spacing is also small (< 10 μm), the dispersion of particles will hinder the movement of dislocations in the matrix. During cold-working of the semi-products, high dislocation densities are created. These lead to a significant hardening of the material. The presence of particles at grain boundaries also hinders the movement of the grain boundaries at high temperatures and thus restricts grain coarsening. In this way, it is possible to achieve a stable finegrained microstructure in the platinum which is maintained during long-term service at high temperatures.

The use of dispersion hardening is not limited only to pure platinum. It is also possible to increase the strength of solid-solution strengthened platinum alloys further, especially at very high temperatures. The development of dispersion hardened platinum and platinum alloys has permitted a substantial increase in the range of possible applications of platinum materials at high temperatures.

During the present work, platinum materials which had been dispersion hardened with zirconium-yttrium oxide particles were investigated (22). These materials are designated DPH here. High temperature properties were determined on dispersion hardened platinum (Pt DPH) and on the dispersion hardened alloys Pt-10%Rh DPH and Pt-5%Au DPH. The materials Pt DPH and Pt-10% Rh DPH are already widely used in the glass industry and many other fields where long term reliability is essential. They can be readily processed, and their good forming properties and excellent weld-ability, in particular, are important features in the manufacture of equipment and components.

The micrograph in Figure 6(a) shows finely distributed particles both in the matrix and at the grain boundaries of Pt-10%Rh DPH. The lower-magnification micrograph in Figure 6(b) shows elongated grains resulting from cold rolling of the sheet. After the stress-rupture test the microstructure has recrystallised, see Figure 6(c). However, despite the high test temperature of 1600°C it has remained very fine grained. The severe deformation of the material, which largely occurs at high temperatures by grain boundary sliding, has resulted in numerous creep pores at the grain boundaries. This microstructure is much more favourable than that of the conventional alloy Pt-10%Rh in which similar conditions lead to a coarse microstructure with fewer, much larger pores at the grain boundaries. The dispersion hardened material has necked down substantially during the stress-rupture test, see Figure 6(d), which shows its good ductility. The severe deformation in the necked zone has led to a marked elongation of the grains.

It is possible to control the stress-rupture strength of the dispersion hardened platinum materials by adjusting the content of the dispersed phase and modifying the production process. As would be expected, maximum strength is associated with a reduced level of ductility.

However, our experience with customers has shown that for most applications material with maximum ductility is preferred to material with the highest level of strength. The stress-rupture curves in Figure 7 and the following diagrams refer to DPH materials which are optimised with regard to their ductility.

It is particularly interesting that all DPH materials have significantly greater stress-rupture strengths than the three conventional alloys. At the longer test times Pt DPH has a higher stress-rupture strength than the Pt-10%Rh alloy. It is significant that the stress-rupture curves of the dispersion hardened materials show a more gradual decrease in strength with increasing rupture time than those of the conventional materials. This means that the dispersion hardened materials show an even greater strength advantage when they are used for components with long service lives. The relatively short test times, which are dictated by the available testing capacity, are completely adequate to compare different materials. However, the design of components for use, for example, in the glass industry requires values determined over longer test times and such tests are currently in progress.

The main application for the Pt-5%Au DPH alloy is in the production of crucibles, dishes and other items for the analytical laboratory, where its stable fine-grained structure is of particular advantage along with its well known anti-wetting characteristics. This material has not yet achieved significant use in structural components for glass processing. However, as can be seen in Figure 7, the alloy also has interesting high temperature mechanical properties and its strength exceeds that of conventional Pt-10%Rh.

In Figure 8 is shown the 10-hour stress-rupture strength as a function of temperature for the dispersion hardened materials Pt DPH and Pt-10%Rh DPH in comparison to the conventional materials. The advantage of the dispersion hardened materials, especially at the highest temperatures, is quite clear. The temperature limit for the use of platinum materials, for example in the glass industry, is currently ∼ 1600°C. However, Figure 8 shows that Pt-10%Rh DPH has a higher level of strength than the conventional alloy even at considerably higher temperatures.

Examples of creep curves for Pt-10%Rh DPH, compared to the conventional alloy, are shown in Figure 9 for a test temperature of 1600°C and two test stresses. The higher strength of the dispersion hardened material produces a much longer time to rupture at the same stress and a much lower creep rate in the linear range of secondary creep. For example, the creep rate of Pt-10%Rh at 1600°C and a test stress of 7.0 N mm^-2 is reduced by a factor of ten (from 12.9 μm s^-1 to 1.3 μm s^-1) by dispersion hardening.

Structural components which are manufactured from the dispersion hardened material therefore show greater stiffness and resistance to deformation, so it is therefore possible to increase their service life.

In addition to the strength increase resulting from dispersion hardening it is also important to consider the ductility of the materials. It can be seen in Figure 9, for example, that the fracture elongation of Pt-10%Rh has been little reduced by being dispersion hardened (from 16.3 to 14.4% at 1600°C and 7.0 N mm^-2). This means that it is not necessary to sacrifice ductility to obtain the strengthening effect of dispersion hardening.

High Temperature Properties of Rhodium and Iridium

In Figure 10 are shown stress-rupture curves for pure rhodium (1600 to 1800°C) and pure iridium (1800 to 2300°C) (23). The measurements on iridium were carried out under a protective atmosphere of argon to prevent the evaporation of volatile oxides which would lead to a reduction in cross-section. For comparison, the stress-rupture curve of Pt-10%Rh at 1600°C is included. It can be clearly seen that rhodium and, in particular, iridium still have substantial stress-rupture strengths which can be made use of at the very high temperatures where platinum materials cannot be considered. The stress-rupture curves for rhodium at temperatures up to 1800°C and for iridium up to 2200°C lie considerably higher than the stress-rupture curve for Pt-10%Rh at 1600°C which is generally considered to be the maximum temperature for its application.

The photographs in Figure 11 show the surface of an iridium sample after various test times during stress-rupture testing at 2000°C. The grain boundaries are clearly visible and they become broader with increasing test time. Despite the high test temperature the material shows no significant grain coarsening. Shortly before rupture the sample shows first indications of necking and the cracking initiates at a broadened grain boundary.

Further Investigations

Important uses of platinum materials are in complex components which may contain numerous welds. The welds tend to be areas of weakness in the structure, particularly when dispersion hardened materials are used. In many applications, for example, in glass making and processing, components are exposed to significant corroding effects. The properties of platinum materials in the welded state and after exposure to corrosive glass melts are currently being investigated and will be the subject of a subsequent report.