Experimental Method

The purities of the ruthenium powder and the nickel from which the alloys were prepared were 99.95 and 99.99 per cent, respectively; iodised titanium was also used. In order to avoid sputtering and losses during alloying, the ruthenium powder was sintered in vacuum at 1200°C and then melted in an arc furnace. Titanium-nickelruthenium alloys containing 0.5, 1, 2 and 5 atomic per cent ruthenium along the TiNi-TiRu section were prepared from melts of their component parts in an arc furnace under an argon gas atmosphere. In order to ensure a complete melt, the ingots were remelted four times. The weight losses which occurred on melting amounted to only 0.05 weight per cent, so the final compositions were considered to be identical to the initial ones. Investigations were carried out on both as-cast and annealed specimens.

Measurements of microhardness were carried out by a standard method at a load of 50 g. Other determinations of the temperature dependence of the electrical resistance and thermal expansivity were performed using custom designed equipment (6). A modified differential thermal analysis method was used to record the solid state transformations in the alloys, from liquid nitrogen temperatures up to room temperature.

Experimental Results and Discussion

The constitution of the TiNi-TiRu section of the titanium-nickel-ruthenium ternary system is important for the present work. It is necessary to know the state of the parent phase which then transforms upon cooling. During our investigations we found that the titanium-nickel equiatomic alloy formed a pseudo-binary continuous solid solution with the titanium-ruthenium equiatomic alloy, TiNi-TiRu.

Data showing the dependence of the electrical resistance upon temperature for alloys which contain 0.5, 1, 2 and 5 atomic per cent ruthenium, and measurements of thermal expansion with temperature for alloys containing 0.5, 1 and 2 atomic per cent ruthenium between room temperature and liquid nitrogen temperatures, are shown in Figures 1 and Figures 2, and also in the Table.

Transformations in the TiNi-TiRu system occur in two stages, in a similar way to the transformations in titanium-nickel based alloys containing additives of iridium, rhodium and iron (1, 3, 4, 7). This can be clearly seen from the shapes of the electrical resistance curves for alloys containing 1 and 2 atomic per cent ruthenium. The sequence B2 → R → B19’ is the same as for the alloys mentioned above, see Figures 1(b) and Figures 1(c)). The stages are associated with rhombohedral (R) and monoclinic (B19’) lattice distortions. T_R is the temperature at which the parent B2-phase starts to transform into the intermediate R-phase and at which the electrical resistance sharply increases. At M_s the martensitic transformation from the R-phase starts, causing a decrease in the resistance. M_f is the temperature at which the martensitic transformation is complete. On heating the alloys from liquid nitrogen temperatures the reverse transitions take place. A_s and A_f are the temperatures at the start and finish of the B19’ → R transformation, respectively, for alloys containing 0.5 and 1, and 2 atomic per cent ruthenium.

One of the characteristics of the electrical resistance/temperature curves is that they reveal a pronounced hysteresis for R ↔ B19’ transformation, while the R-phase transformation, R → B2, does not. Thus the cooling and heating curves for alloys containing 1 and 2 atomic per cent ruthenium almost coincide at temperatures above A_f, see Figures 1(b) and Figures 1(c).

The B2 → R transformation causes a decrease in the volume of the sample, that is, the alloy is observed to shrink, while the R → B19’ transformation results in an increase in the volume of the sample, that is, thermal expansion occurs, see Figures 2(b) and Figures 2(c).

The observed variations in electrical resistance and thermal expansion with temperature show that the R → B19’ transformation appears to represent a first order transition. This correlates well with the DTA data, which demonstrate the corresponding thermal effects on heating and cooling (the thermal curves in Figure 3 show only the effects caused by heating, because the data on cooling obtained in our experiments reveal the same transition temperature regardless of composition).

The B2 → R transformation is a reversible process, according to the electrical resistance data shown in Figure 1, and almost reversible, as shown by the thermal expansion data in Figure 2, and therefore can be considered as a transformation close to a second order transition. However the DTA data shown in Figure 3 illustrate that the B2 → R transformation is accompanied by thermal effects, as is the R → B19’ transformation.

It is worth noting that the intermediate phase formed during the B2 → R transformation in titanium-nickel based alloys with added iron (as well as in titanium-palladium based alloys with added nickel) is divided into ranges of incommensurate and commensurate phases with different crystal structures (10,11). The first range (incommensurate phase) corresponds to the second order transition and the next range (commensurate phase) corresponds to the first order transition. This could explain the thermal effect which was observed in the B2 → R stage for the TiNi-TiRu alloys.

The solid state transformation in titanium-nickel-ruthenium alloy containing 5 atomic per cent ruthenium, see
Figure 1 (d), starts at -20°C. The R-phase exists over a wide temperature range, down to liquid nitrogen temperatures.

The martensitic transformation scheme, Figure 4, shows that the B2-phase is approximately stabilised at room temperature in alloy containing 2 atomic per cent ruthenium. The substitution of ruthenium for nickel results in a lowering of all critical transformation temperatures, but the decrease in the T_R temperature occurs less acutely than the decrease in the temperatures which indicate the transformation range of the phase having the B19’ crystal structure.

The microhardness of the titanium-nickel-ruthenium alloys varies with composition and is at a minimum for an alloy containing about 2 atomic per cent ruthenium, and at a maximum for an alloy containing about 1 atomic per cent ruthenium, see Figure 5. A comparison of these findings with those obtained from the electrical resistance and thermal expansion analyses indicates that the minimum is in accordance with the absence of a martensitic transformation in the alloys, and that the maximum is associated with accumulation of the R-phase.

Comparison of the results of the present investigation with data from the literature on the influence of platinum metals on the martensitic transformation in titanium-nickel based alloys enables the following observations to be made:

Conclusions

The substitution of ruthenium for nickel in titanium-nickel-ruthenium ternary alloys along the TiNi-TiRu section, as examined by physical and chemical analyses, results in the stabilisation of the high temperature B2-type phase at room temperature for alloys containing at least 2 atomic per cent ruthenium. Alloys containing 0.5 to 2 atomic per cent ruthenium undergo a two stage transformation.

In continuation of our work with the platinum group metals, we intend to study phase equilibria in the titanium-nickel-ruthenium system both experimentally and by calculation, and following that calculate the effect of ruthenium on the martensitic transformation in equiatomic titanium-nickel compound. Data obtained from the titanium-nickel-ruthenium alloys will make we expect that it will sharply decrease the it possible to forecast the effect of osmium on transformation temperatures in a way similar to martensitic transformations in TiNi-based alloys; that of ruthenium.