Journal of Magnetism and Magnetic Materials 310 (2007) e613–e615 Site occupation and local hyperfine fields in s Fe–Cr alloys Jakub Cieslak a , Michael Reissner b,Ã , Walter Steiner b , Stanislaw M. Dubiel a a Faculty of Physics and Applied Computer Science, AGH University of Science and Technology, 30-059 Krako´w, Poland b Institute of Solid State Physics, Vienna University of Technology, Wien A-1040, Austria Available online 13 November 2006 Abstract High field 57 Fe Mo¨ ssbauer and neutron diffraction measurements were performed for a series of s-Fe x Cr 1x samples covering the full concentration range (0.50pxp0.55) of phase formation. Occupation numbers for the individual lattice sites, obtained from the neutron diffraction data, are used to determine the local hyperfine parameters by a new analysis procedure, which simultaneously fits all measured Mo¨ssbauer spectra, thus allowing a drastic reduction of the number of free parameters. r 2006 Elsevier B.V. All rights reserved. PACS: 75.50.Bb; 61.12.Ld; 76.80.+y Keywords: Mo¨ssbauer; s-phase; FeCr; Neutron diffraction; Occupation numbers Whereas for the a-phase of Fe–Cr, strong ferromagnet- ism is found, the s-phase is only weakly magnetic with ordering temperatures below 50 K and magnetic moments below 0.3 m B , both decreasing linearly with increasing Cr-content [1]. The effective moments, however, are much larger, which according to Rhodes and Wohlfarth indicates strongly delocalised moments. With increasing Fe content the itinerant character becomes less pronounced. To gain information about the magnetic ground state we performed neutron diffraction, magnetic, and high field 57 Fe Mo¨ss- bauer investigations on a series of s-Fe x Cr 1x samples, covering the full concentration range in which this phase forms (x ¼ 0.5–0.55). The structure of the s-phase is tetragonal (space group D 4h 14 P4 2 /mnm) with five crystallographically nonequivalent lattice sites (2a, 8i, 8i 0 , 8j, 4f), which differ both in coordination number and local symmetry, thus suggesting that different Fe-moments are formed at each site. Former experiments showed that the distribution of Fe on these sites is nonstatistical [2]. Therefore, the occupation numbers were determined from neutron diffraction mea- surements performed at ILL (D1A) by means of Rietveld analysis (FULLPROF program). In the calculations x was held fixed to the values obtained by microprobe analysis. A preferential occupation was found for Fe on the 2a and the 8i 0 sites, which have the highest local symmetry, whereas Cr occupies preferentially the 4f sites, which have the lowest local symmetry. Changes of the preparation conditions lead only to small alterations of the occupation numbers. Within the same preparation route they increase for Fe linearly with x for sites 4f, 8i and 8j, whereas they remain nearly constant for sites 2a and 8i 0 (Fig. 1). A comparison of the neutron-diffraction data taken at room and He-temperature shows no extra peaks at low temperatures; thus indicating that antiferromagnetic com- ponents in these compounds are highly improbable. To get information on the local hyperfine fields 57 Fe Mo¨ ssbauer spectra were recorded at room temperature and in addition at 4.2 K in fields up to 13.5 T. For the analysis a new scheme was developed, which allows simultaneous fitting of all spectra taken for all samples at different applied fields to reduce the number of free parameters. The procedure takes into account both the five crystallogra- phically different lattice sites and their respective occupa- tion numbers as obtained from the neutron analysis. For each site, it is assumed that a quadrupole splitting (QS) and isomer shift (IS) exist, the former independent of x, because of the similarity of the overall shape of the recorded spectra ARTICLE IN PRESS www.elsevier.com/locate/jmmm 0304-8853/$ - see front matter r 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2006.10.583 Ã Corresponding author. Tel.: +43 1 58801 13772; fax: +43 1 58801 13898. E-mail address: reissner@ifp.tuwien.ac.at (M. Reissner).