Nucleation of the isothermal martensitic transformation in iron-nickel-manganese alloys

Abstract

By means of quantitative metallography and electrical resistance measurements, the incubation period (time to form a detectable amount of martensite) and the initial nucleation rate have been determined as a function of subzero reaction temperature and austenitic grain size in three iron-nickel manganese alloys containing 23-25 percent nickel, 2-3 percent manganese and 0.02-0.04 percent carbon. The isothermal martensitic transformation in these alloys has been found to occur over a wide temperature range of at least -1960 to -200 C, and the martensitic structure over the entire range is body centered cubic. The incubation period and nucleation rate as a function of reaction temperature at various austenitic grain sizes show typical C-curve kinetics. The true initial nucleation rate (not the apparent initial nucleation rate during the incubation period) is independent of grain size, and so the effect of grain size on the incubation period arises merely because of the size of the first-formed plates. Accordingly, it may be concluded that grain boundaries do not predominate as preferred nucleation sites for the martensitic transformation in the alloys studied. Calculations of the relevant thermodynamic functions indicate that 1 atomic percent manganese is thermodynamically equivalent to 1.6 atomic percent nickel in the driving force for the martensitic transformation. However, alloys of equivalent thermodynamic driving force exhibit striking differences in kinetic behavior due to slight differences in their manganese and carbon contents, suggesting that these elements reduce the potency of the embryos. The observed activation energies of the isothermal martensitic transformation are in very good agreement with the calculated activation energies based on a nucleation model. Moreover, in these alloys, the potency of the active embryos remains almost constant during the course of the isothermal transformation, notwithstanding the progressive generation of new embryos due to autocatalytic factors. The best fit between theory and experiment is obtained on the assumption that the number of pre-existing embryos in the parent austenite is 10⁷ per cm³ . The probability of detecting such embryos by transmission electron microscopy is less than 1 in 10⁵. The increase in the rate of isothermal martensitic transformation as a function of time has been shown to be due to the formation of new plates by auto catalysis, while the subsequent retardation is attributable to the partitioning of the austenite by the martensitic plates. The model used provides quantitative agreement with the course of the isothermal transformation up to 12 percent martensite.