THE SYNTHESIS, STRUCTURE, AND CHARACTERIZATION OF ALPHA''-FE16N2 (INVITED)

Controversy about the magnetic properties of α]-Fe16N 2 thin films makes it desirable to examine bulk material. In the original preparation and crystal structure determination of α] in 1951, N-austenite (γ) prepared by nitriding α-iron powder at 700-750°C was quenched to give N-martensite (α') which then gave α+α] on long tempering at 120°C. The final product was a mixture of either α+α] or α+α]+γ with less than 50%α]. Recent repetitions of this preparative method in three different laboratories have given mixtures with varying amounts of α], again never greater than 50%, and with magnetic moments of α] - assessed from measurements made on the mixtures - that show considerable variation. Studies of α' tempering by XRD, and of α] precipitation from supersaturated N ferrite (α) by high-resolution TEM, both show that slow ordering of N atoms to produce α] occurs only after localized regions of α' and α reach the Fe 8N composition by a clustering process. Thus α'-Fe8N can be obtained with a tetragonality c/a equal to that of α], but without the complete N ordering that is characteristic of Fe16N2. This might explain some of the variability in magnetic properties. The inhibiting effect of oxygen and other impurities on the nitriding of iron is emphasized, the existence of the α]-carbonitride Fe16(C,N) 2 is noted, and possible methods for 100% production of bulk α] are suggested.