Abstract
Mechanical grinding and alloying techniques are being used in an increasing number of experimental studies, including the mechano-synthesis of alloys [1, 2], studies on chemical kinetics in solid-state reactions [3] and quasi-crystalline systems [4, 5]. Recent studies on several Feand Cu-based oxides have revealed that, in addition to controlling crystallite size, mechanical grinding can induce phase transformations and changes in the oxidation state of the reactants [6–8]. The Mossbauer effect (ME) provides a valuable tool for mechanical grinding research, allowing the observation of local quadrupolar and magnetic hyperfine interactions of the probe atom. These interactions can sense minute changes in the local crystalline structure and magnetic properties of the system. The work reported here was a systematic study, using Mossbauer spectroscopy and X-ray diffraction (XRD), of a 57Fe–Cu–O system with nominal composition (Fe0:005Cu0:995)O subject to controlled mechanical grinding. The main purpose was to follow the evolution of the starting phases, and investigate the effect of mechanical grinding on particle size. Samples were obtained through the nitrate route, starting from powdered CuO (99.999%) and metallic 57Fe enriched to 95% dissolved in dilute HNO3, with nominal composition (Cu0:995Fe0:005)O. The mud obtained was dried at 200 8C for 12 h then thermally treated in air at 1000 8C for 50 h, with intermediate manual grinding in agate mortar. The powder obtained was mechanically ground in a Retsch MM2 vibratory mixer mill for different times, extracting partial amounts in each step. Samples were labelled C0, C1, C2, C11, C15, C21 and C48, where the number indicates the number of hours of mechanical grinding. Characterization by XRD was performed in a Philips PW-1140 diffractometer using Ni-filtered CuKAE radiation. Rietveld profile analysis was used to refine the XRD data using the DBWS-9006PC5 program [9]. ME measurements were taken at room temperature in transmission geometry with a 50 mCi 57Co source in an Rh matrix, with a multiscaler of 512 channels in constant acceleration mode. Samples were placed in circular holders of 1.9 cm (0.75 in.) diameter. The amount of powder calculated to obtain optimal thickness for this composition was 18 mg cmy2. A non-linear least-squares program was used to fit the spectra to Lorentzian line shapes. When site distributions were present, a shapeindependent distribution fitting program [10] was used. Isomer shifts were referred to AE-Fe at 300 K. For all samples, XRD patterns could be indexed assuming a matrix composed only of tenorite (CuO), without any evidence of phase transformations involving the CuO matrix due to mechanical milling (Fig. 1). For sample C48, XRD data displayed broadened diffraction lines from small particles after 48 h mechanical grinding (Fig. 1b). The particle size was estimated from Scherrer’s equation, using a standard sample of known grain size and subtracting instrumental broadening from total broadening. The value of particle diameter, d, obtained in this way for sample C48 was 9.3 nm. After calcinating sample C48 at 800 8C for 2 h, an XRD pattern with sharp lines was recovered, indicating the increase of crystallite size. The iron phases were not detected by XRD due to the low amount of Fe in the system. The Mossbauer spectrum for sample C0 (not milled) is shown in Fig. 2. Two magnetic signals can be observed, whose hyperfine parameters correspond to Fe3 in tetrahedral (t) and octahedral (o) sites of the tetragonal CuFe2O4 spine [11–13]. In accordance with previous works on Cu–Fe–O and Zn–Fe–O systems [14, 15], the spinal was the only iron phase present in sample C0. The absence of other Mossbauer signals allows estimation of an upper
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call