Kinetics of the sigma-to-alpha phase transformation caused by ball milling in near equiatomic Fe-Cr alloys J. Cieślak, 1 B. F. O. Costa, 2 S. M. Dubiel, 1, * and G. Le Caër 3 1 Faculty of Physics and Nuclear Techniques, AGH University of Science and Technology, PL-30-059 Kraków, Poland 2 Department of Physics, University of Coimbra, P-3004-516 Coimbra, Portugal 3 GMCM UMR CNRS 6626, University of Rennes 1, F-35042 Rennes Cedex, France Received 9 February 2006; published 23 May 2006 Kinetics of the sigma-phase decomposition in three series of -FeCr intermetallic compounds caused by ball milling in a vibratory mill was followed by Mössbauer spectroscopy. It can be described by an exponential decrease with milling time. The decomposition leads to the formation of the -phase as well as that of an amorphous phase. The former is expected from the phase diagram, but the latter is not. The amount of the -phase increases with milling time, reaches a maximum characteristic of the milling conditions, and, finally, after a slow decrease, remains constant. The decrease can be explained in terms of the -phase partial amor- phization. The amount of the amorphous phase stays rather constant with a small tendency of increasing for long-term milling times. Its structure seems to evolve with milling time. DOI: 10.1103/PhysRevB.73.184123 PACS numbers: 61.18.Fs, 61.43.Dq, 64.70.Kb, 81.07.Bc I. INTRODUCTION One peculiarity of the Fe-Cr system is the possibility of the formation of a -phase. Its crystallographic structure was definitely identified as a close-packed tetragonal—space group D 4h 14 - P4 2 / mnm—with 30 atoms in the unit cell, 1 which are distributed over five crystallographically nonequivalent lattice sites: I, II, III, IV, and V with occupation numbers 2, 4, 8, 8, and 8, respectively. The sites have rather high coor- dination numbers—12 for sites I and IV, 14 for sites III and V, and 15 for site II—and quite different local symmetries. The phase can be promoted by an isothermal annealing of near equiatomic bulk alloys in the temperature range 870 K– 1100 K Ref. 2and has been both of scientific and technological interest. Its scientific interest follows from properties very different from those of the -phase from which it precipitates, and, in particular, weak magnetism. 3 Its technological interest stems from its high hardness and brittleness which causes deterioration of useful mechanical properties in materials of practical importance. Conse- quently, a major goal is how to avoid, or, at least, to retard its formation. Whatever is the reason for the interest in the -FeCr, the kinetics of its formation and/or decay seems to be of importance both for scientists and engineers. Curiously, in the available literature there is a pretty good deal of papers devoted to the kinetics of the -phase formation both in bulk alloys 4–11 as well as in nanocrystalline ones. 12 However, to our knowledge, there is none devoted to the reversed process, i.e., the kinetics of the decay of the -phase in the Fe-Cr system, despite past studies on the influence of ball milling on the -FeCr behavior. 13–17 The aim of this paper is thus to study the kinetics of the sigma-phase decomposition by ball milling in three series of -FeCr intermetallic compounds, prepared and/or ball-milled in different conditions, using the Mössbauer spectroscopy as the main characterization tool. II. EXPERIMENT Three series S cg1 , S cg2 , and S nc of -FeCr intermetallic compounds were ball-milled in a Fritsch P0 vibratory mill which is a low-energy type of mill. The first two series cor- respond to the milling of a coarse-grained Fe 54 Cr 46 alloy prepared by melting in argon atmosphere appropriate amounts of iron 99.9+% purityand chromium 99.95% purityin an induction furnace. The ingots were homog- enized by remelting them three times. The as-cast alloy had a bcc structure, i.e., was in the -phase. It was transformed into the tetragonal -phase by isothermal annealing at 973 K in vacuum for 100 h. The obtained piece of the -phase was next powdered with a pestle in an agate mortar into particles of 90 m average diameter. A mass of 5 g of such powder was then ball-milled in argon in a Fritsch P0 mill with a hardened steel vial and a hardened steel ball having a diam- eter of 5 cm and a mass of 500 g. The vial’s cover was homemade to permit an easy filling of the vial with argon. The argon was introduced into the vial every 100 min of milling. The mill was working with a vibration amplitude of AM= 2.5 for the first series of samples, and in more energetic conditions with the maximum amplitude, AM= 3, for the second series. The third series consists of nanocrystalline samples of the -Fe 515 Cr 485 alloy and it was ball-milled with the maximum amplitude AM=3. The nanocrystalline -FeCr alloy needed to obtain the starting nanocrystalline -FeCr phase was prepared in a Fritsch P6 planetary mill at a disk rotation of 520 rpm from elemental powders of iron 99.0+ % purity, grain size 60 mand chromium 99.0+% purity, grain size 45 m. The starting mass of each powder was 25 g and the powder-to-ball weight was 1:20. The pow- der was sealed in the vial under argon atmosphere. The total milling time of 16 h was interrupted every 30 min for 10 min. The content of oxygen for this sample was 1.0 at. %. The nanocrystalline bcc alloy was next annealed in vacuum at 973 K for 6 h to transform it into the -phase. For all three series, the transformation of the -into the -phase was verified at room temperature RTby both x-ray diffraction XRDas well as by Mössbauer spectroscopy. The nanocrystalline -FeCr sample was then subjected to the same milling procedure as the previous two samples. A microprobe analysis was used to determine the alloy compositions. The oxygen contents after milling and in the PHYSICAL REVIEW B 73, 184123 2006 1098-0121/2006/7318/1841237©2006 The American Physical Society 184123-1