On the Mode of Short-Term Tempering of High-Speed Steel

I. T. Tikhonov

(Presented by the scientific seminar of the departments of physical metallurgy, metal technology, and welding production)

When the temperature of the threefold tempering of hardened high-speed steel is increased, the holding time should be reduced in accordance with the equation

\[ \tau_{\mathrm{e}} = 1.72 \cdot 10^{-18} \cdot e^{\frac{37500}{T}} \ \text{min}, \tag{1} \]

where \(T\) is the tempering temperature in \({}^{\circ}K\) [1]. As a result, the same degree of decomposition of retained austenite will be attained in the hardened steel as after conventional tempering at \(560^{\circ}C\) with holding times of 60 min each. Therefore such short-term temperings were called equivalent to the standard one.

The dependence (1) given above between holding time and tempering temperature was found experimentally for P18 steel hardened from \(1280^{\circ}C\), and for threefold temperings with temperatures in the range from 560 to \(620^{\circ}C\). It was also of practical interest to determine how a change in the hardening temperature would affect this dependence and whether it could be used for high-speed steels of other grades. For this purpose, additional experiments were carried out with specimens of P18 steel hardened from temperatures of 1230 and \(1300^{\circ}C\), and with specimens of P9 steel hardened from \(1230^{\circ}C\). The specimens were tempered three times at \(560^{\circ}C\) with holding times of 60 min each, and at 580, 600, and \(620^{\circ}C\) with holding times of various durations. Tempering was performed in a bath with a liquid aluminum alloy; before immersion in it, the specimens were preheated to \(550^{\circ}C\) in molten lead. The specimens had a cylindrical shape with a diameter of 8 mm and a length of 60 mm. The holding time of equivalent short-term temperings \(\tau_{\mathrm{e}}\) was determined from the magnetic properties of the specimens using a differential-type austenometer according to the previously adopted method [1].

The graph in Fig. 1, whose straight line corresponds to equation (1), shows that, within the accuracy of the experiments, the dependence obtained earlier retains its validity for all selected hardening temperatures from 1230 to \(1300^{\circ}C\). Only the points corresponding to the values of \(\tau_{\mathrm{e}}\) determined after threefold tempering were plotted on the graph. It is significant, however, that the values of \(\tau_{\mathrm{e}}\) obtained after the first and second temperings also give practically the same results. Consequently, when changing the multiplicity of tempering—for example, replacing threefold

tempering double, in equation (1) only the pre-exponential factor will change.

Experiments carried out with R9 steel showed that in this case as well the holding times of threefold equivalent tempers, obtained experimentally and determined from equation (1), have close values. This makes it possible to recommend equation (1) for approximate calculations of the holding times of short-term tempers also for R9 steel.

If it is assumed that there is a direct dependence between the amount of ferromagnetic phase formed and the processes occurring upon heating of hardened high-speed steel during its tempering, then equation (1) can be represented in the form

\[ \tau_{\mathrm{e}} = C \cdot e^{\frac{Q}{RT}} = 1.72 \cdot 10^{-18} \cdot e^{\frac{75000}{RT}} \ \text{min}, \tag{2} \]

where \(R\) is the gas constant, and \(Q\) is the so-called experimental activation energy [2] of the process responsible for the decomposition of retained austenite during tempering of high-speed steel. The obtained value \(Q = 75000\) cal/mole is somewhat greater than the activation energy \(Q = 61000\) cal/mole determined by P. M. Yushkevich [3] for the polymorphic transformation \(\gamma \to \alpha\) in the process of tempering high-speed steel of the R18 type with 1.5% C. In this case the transformation proceeded at holding times whose duration considerably exceeded the duration of the equivalent tempers.

Regarding the possibility of achieving, as a result of short-term tempers, the optimum state and properties of high-speed steel, there are contradictory opinions [4, 5]. The processes occurring during tempering of hardened high-speed steel have not yet been fully studied. However, apparently, it may be considered that the principal task of tempering high-speed steel is the transformation of retained austenite into martensite while attaining high secondary hardness owing to dispersion hardening of the martensite [4]. In this case, deviations from optimum results in connection with replacing ordinary tempering by equivalent short-term tempers of the same multiplicity should evidently be expected only when, with an increase in tempering temperature, the relationship between the rates of the processes causing the decomposition of retained austenite and the dispersion hardening of martensite changes significantly. In other words, the experimental activation energies of these processes will differ substantially. In one of the subsequent works of the author named above

Fig. 1. Dependence of the holding time in threefold tempering of high-speed steel on the value inverse to its absolute temperature: ○ — hardening from 1230°C, ● — hardening from 1300°C.

[6], from the growth rate of the blocks of the \(\alpha\)-phase, determined the activation energy of the process of disordering of the martensite of quenched high-speed steel R18. For temperatures up to \(620^\circ\text{C}\) it amounted to \(60000\ \text{cal/mol}\), and for temperatures above \(650^\circ\text{C}\)—\(80000\ \text{cal/mol}\). The closeness of these values to the activation-energy values cited above makes it possible to suppose that, as a result of equivalent short-term tempers, not only the usual degree of decomposition of retained austenite will be achieved, but also a stage of dispersion hardening of martensite equivalent to standard tempering. This conclusion is confirmed by experimental data, according to which, after short-term tempers, the hardness and red hardness usual for high-speed steel were observed [7, 8]. The cutting properties after short-term tempers, apparently, may also be regarded as being maintained at the usual level [7, 8, 9, 10].

Conclusions

1. The previously proposed equation (1) for determining the holding times of triple equivalent short-term tempers of quenched high-speed steel R18 retains its significance for a broad interval of quenching temperatures from 1230 to \(1300^\circ\text{C}\), and can be recommended for approximate calculations in tempering steel R9.

2. A comparison of the magnitudes of the experimental activation energies of the processes occurring during tempering of quenched high-speed steel shows that, as a result of short-term equivalent tempers, one should expect the attainment of the same properties as after ordinary tempering of the steel at \(560^\circ\).

References

1. I. T. Tikhonov. Izv. TPI, vol. 147, 1965.

2. K. A. Osipov. Certain activated processes in solid metals and alloys. Publishing House of the Academy of Sciences of the USSR, 1962.

3. P. M. Yushkevich. Journal of Technical Physics, vol. 24, no. 4, p. 715, 1954.

4. Yu. A. Geller. Tool Steels. Metallurgizdat, 1961.

5. G. V. Korotushenko. Izv. vuzov SSSR—Ferrous Metallurgy, no. 8, p. 128, 1960.

6. P. M. Yushkevich. Physics of Metals and Metallography, vol. 8, no. 6, p. 896, 1959.

7. K. A. Malinina. Metallography and Heat Treatment of Metals, no. 3, p. 47, 1960.

8. I. T. Tikhonov, N. S. Kilkov, E. P. Filimonova. Izv. vuzov SSSR—Ferrous Metallurgy, no. 10, p. 112, 1964.

9. P. A. Yudkovskii, V. Ya. Bulanov, Yu. N. Zhuravlev, A. P. Shchevel. Machines and Tooling, no. 12, p. 27, 1963.

10. N. S. Kilkov, I. T. Tikhonov, E. P. Filimonova. Izv. TPI, vol. 147, 1965.