J O U R N A L O F C H E M I S T R Y Materials Self propagating high-temperature synthesis of chromium substituted magnesium zinc ferrites Mg0.5Zn0.5Fe2-xCrxO4 (0x1.5) Maxim V. Kuznetsov,a Quentin A. Pankhursta and Ivan P. Parkin*b aDepartment of Physics and Astronomy, University College London, Gower Street, London, UK WC1E 6BT bDepartment of Chemistry, University College London, 20 Gordon Street, London, UK WC1H 0AJ.E-mail: i.p.parkin@ucl.ac.uk Received 29th June 1998, Accepted 4th September 1998 Magnesium ferrite MgFe2O4 and chromium substituted magnesium–zinc ferrite Mg0.5Zn0.5Fe2-xCrxO4 (0x1.5) have been made in air by self-propagating high temperature synthesis (SHS), a combustion process involving the reaction of magnesium, zinc, iron(III ) and chromium(III ) oxides with iron or chromium metal powders and sodium perchlorate.This produced an orange–yellow propagation wave of velocity 2–3 mm s-1. Two series of SHS samples were produced: series 1, SHS followed by annealing at 1400 °C for 2 h and series 2, SHS in a magnetic field of 1.1 T followed by annealing at 1400 °C for 2 h. X-Ray data showed that in all cases nearly phase pure cubic spinel ferrites were produced.Changes in the cubic lattice parameter were seen as a function of Zn and Cr content. Room temperature and 80 K Mo� ssbauer data showed a significant change in sublattice occupancy with Cr content. Magnetic hysteresis data for series 1 and 2 showed that the coercive force of doped samples is higher than pure compositions whilst magnetisation is lower.It was also shown that the use of a magnetic field during SHS can influence the microstructure and magnetic properties of the final material. ferrites of formula MgFe2-xCrxO4 are valuable for the long Introduction wave part of the high-frequency range (10–14 cm) as they As part of an on-going research programme aimed at exploring have very low coeYcient losses.5 Pure and Cr-substituted the potentials and problems associated with innovative routes Mg-ferrites also demonstrate catalytic activity.6 to ceramic materials, we are currently interested in synthesising In this paper we present the first SHS preparation of materials by self-propagating high temperature synthesis chromium substituted Mg and Mg–Zn ferrites and investigate (SHS).1 This is a non-conventional method for materials the eVects of the presence of an applied magnetic field during synthesis which makes use of an exothermic chemical reaction.the combustion synthesis. SHS processes rely on a balance between the heat generated and dissipated in a chemical reaction. It involves mixing diVerent reactant powders, such as oxides, metals and non- Experimental metals, which on initiation produce an exothermic chemical reaction that is self-sustaining due to a positive energetic All reagents were obtained from Aldrich Chemical Company balance.2 The reaction produces rapid heating (up to 3000 °C (UK) and used as supplied.Manipulations, weighings and in 1–2 s) and cooling and is propagated by a combustion wave grindings were performed under a nitrogen atmosphere in a (also known as thermal flash, synthesis wave or solid flame) SaVron Scientific glove box.SHS reactions were carried out that moves out from the source of initiation. SHS processes in air on pre-ground powders on a ceramic tile using sodium are fast, do not require external energy such as from a furnace perchlorate as an oxygen source. Initial reaction compositions and can be used to produce complex oxides, intermetallics and are given in Table 1.The reactions were initiated by a heated composite materials. SHS reactions produce materials with filament (ca. 800 °C). Sintering was carried out in a Carbolite unusual often porous microstructure. rapid heating furnace with heating and cooling rates of Single phase magnesium ferrite MgFe2O4 with the spinel 20 °Cmin-1.Samples were ground after the SHS reaction and structure cannot be made from stoichiometric combination of also after sintering, and all measurements were recorded on MgO and Fe2O3. The spinel structure can be made by using powder samples. For the applied field SHS reactions a permaa superstoichiometric amount of MgO (1.092 equivalents) to nent magnet Halbach cylinder purchased from Magnetic help combat oxygen loss from the ferrite.3 Unsubstituted Solutions Ltd was used.This 20 mm bore cylinder, comprising magnesium ferrites have low magnetic penetration and rela- eight NdFeB magnets, provided a field of 1.1 T transverse to tively small specific resistance. They have only limited indus- the cylinder axis. For the applied field SHS reactions a ceramic trial use.In contrast doped magnesium ferrite is known tile, containing the green mixture, was placed inside a quartz commercially as ‘Ferramic-A’ and has comparably good mag- tube which was in turn inserted into the Halbach cylinder netic properties up to 100 MHz (Bs=1400 G; r=4.5 g cm-3).4 prior to initiation of the combustion process. Much better properties are obtained for the series of solid X-Ray diVraction was performed in reflection mode on a solutions of Mg–Zn ferrite.The ferrite Mg0.5Zn0.5Fe2O4 is Philips X-Pert using unfiltered Cu-Ka radiation (l1=1.5405 A° , known commercially as Ferroxcube-2 and has a saturation l2=1.5443 A° ). Vibrating sample magnetometry was carried magnetic induction Bs=3500 G and a coercivity Hc= out on a Aerosonic 3001 magnetometer at room temperature 600 G Oe-1 (r=4.7 g cm-3). These ferrites are used widely, in applied fields up to 7.5 kOe. 57Fe Mo� ssbauer spectra were for example in mobile telephones. Chromium substituted recorded with a Wissel MR-260 constant acceleration specmagnesium and magnesium–zinc ferrites are suitable not only trometer with a triangular drive waveform.Spectra were folded for fundamental studies of structural and magnetic properties to remove baseline curvature and were calibrated relative to a-iron at room temperature. FTIR spectra were obtained as but also for industrial application. Chromium substituted iron J. Mater. Chem., 1998, 8, 2701–2706 2701Table 1 Molar ratio of components used in self-propagating high-temperature synthesis (SHS) of Mg0.5Zn0.5Fe2-xCrxO4, and nominal composition of the end product, using a notation where parentheses denote tetrahedral sites and square brackets denote octahedral sites x Fe2O3 Fe Cr2O3 Nominal composition 0 0.50 1.00 0 (Zn0.5Fe0.5)[Mg0.5Fe1.5]O4 0.3 0.35 1.00 0.15 (Zn0.5Fe0.5)[Mg0.5Fe1.2Cr0.3]O4 0.6 0.20 1.00 0.30 (Zn0.5Fe0.5)[Mg0.5Fe0.9Cr0.6]O4 0.9 0.05 1.00 0.45 (Zn0.5Fe0.5)[Mg0.5Fe0.6Cr0.9]O4 1.2 0.40 1.00 (Cr) 0.10 (Zn0.5Fe0.5)[Mg0.5Fe0.3Cr1.2]O4 1.5 0.25 1.00 (Cr) 0.25 (Zn0.5Fe0.5)[Mg0.5Cr1.5]O4 KBr pellets on a Nicolet 205. SEM/EDAX measurements were Two diVerent series of samples were prepared: series 1, the zero field SHS product subsequently sintered at 1400 °C for performed using a Hitachi S570. 2 h, and series 2, SHS product of an applied field synthesis of 1.1 T followed by sintering at 1400 °C for 2 h.Preparation of Mg0.5Zn0.5Fe2-xCrxO4 (0x1.5) The same reaction scale and procedure was used for all Characterisation reactions illustrated here for Mg0.5Zn0.5Fe1.4Cr0.6O4. Relative molar ratios of the reactants are given in Table 1. X-Ray powder diVraction showed that nearly single phase cubic spinel structures were produced for all the sintered Magnesium oxide (0.200 g, 5.0 mmol), iron oxide (0.638 g, 4 mmol), chromium oxide (0.916 g, 6 mmol), zinc oxide products.Representative diVractograms are shown in Fig. 1, and the deduced lattice parameters are given in the Table 2. A (0.204 g, 2.5 mmol), iron metal (1.116 g, 20 mmol) and sodium perchlorate (0.458 g, 3.75 mmol) were ground together in a pestle and mortar. The resultant solid was placed on a ceramic tile (ca. 1 cm by 7 cm strip) in air and a reaction initiated by means of a heated filament at one end. This produced an orange–yellow propagation wave of velocity ca. 2–3 mm s-1. The resultant black powder was washed with distilled water (2×1l ) filtered through a Buchner funnel and air dried. The powder was reground and sred at 1400 °C for 2 h.Yields in all reactions were essentially quantitative. The resultant powder was analysed by X-ray powder diVraction, Mo�ssbauer spectroscopy, vibrating sample magnetometry, FTIR and SEM/EDAX. Results Sample preparation SHS reactions were performed using various starting mixtures of Fe2O3, Fe, MgO, ZnO, Cr2O3 and NaClO4. The molar ratio of each reagent was chosen to conform with the desired stoichiometry in the product, Table 1.At high degrees of chromium substitution (Mg0.5Zn0.5Fe2-xCrxO4; x=1.2 and 1.5) Cr metal powder was used in the SHS process instead of Fe metal powder. The overall reaction is driven by the Fig. 1 Representative X-ray powder diVraction patterns obtained from exothermic oxidation of Fe or Cr metal by oxygen which was the sintered products of the SHS reactions of: (1) MgO, Fe2O3, Fe evolved by the decomposition of sodium perchlorate at 600 °C.7 and NaClO4 in the ratio 1.0050.5051.0050.375 to produce MgFe2O4; In the case of Mg0.5Zn0.5Fe1.7Cr0.3O4 the reaction scheme was: (2) MgO, ZnO, Fe2O3, Fe and NaClO4 in the ratio 0.5050.5050.5051.0050.375 to produce Mg0.5Zn0.5Fe2O4, and 0.5 MgO+0.5 ZnO+0.35 Fe2O3+0.15 Cr2O3+Fe (3) MgO, ZnO, Fe2O3, Fe, Cr2 O3 and NaClO4 in the ratio 0.5050.5050.5051.0050.4550.375 to produce Mg0.5Zn0.5Fe1.1Cr0.9O4.+0.375 NaClO4�Mg0.5Zn0.5Fe1.7Cr0.3O4+0.375 NaCl Also shown for comparison is a reference stick pattern for MgFe2O4. It should be noted that all reactions were carried out in air using solid oxidisers. Sodium perchlorate acts as the oxidising Table 2 Cubic lattice parameter a for two series of MgFe2O4 and agent.On decomposition it co-produces sodium chloride which Mg0.5Zn0.5Fe2-xCrxO4 samples produced by self-propagating highis easily removed from the product by washing with water. temperature synthesis (SHS): series 1, zero field SHS followed by The SHS preparation of pure and chromium substituted annealing at 1400 °C for 2 h; series 2, SHS conducted in a magnetic field of 1.1 T followed by annealing at 1400 °C for 2 h magnesium and magnesium–zinc ferrites is much quicker than standard conventional methods.8,9 Using SHS instead of a aa/A° furnace enables the first step in ferrite production to be reduced Mg0.5Zn0.5Fe2-xCrxO4 to ca. 30 s. In standard ceramic technology this stage requires x Series 1 Series 2 some hours heating at 1200–1350 °C.The SHS reaction however does not completely alleviate the need to sinter the 0 8.416 8.415 0.3 8.408 8.399 material. After the initial SHS reaction the product contained 0.6 8.398 8.379 about 90% of the required ferrite mixed in with partially 0.9 8.375 8.369 combined starting materials and sodium chloride. Subsequent 1.2 8.373 8.367 washing with water followed by sintering of the powder at 1.5 8.362 8.334 1400 °C for 2 h produces virtually single phase ferrites. The MgFe2O4 8.381 8.377 sintering process is relatively quick as the SHS material has aError limit: ±0.004 A° .good contact between reacting grains. 2702 J. Mater. Chem., 1998, 8, 2701–2706the spectra at 620 cm-1 with increasing chromium substitution.The absorption bands are all characteristic of metal oxygen stretches. 57Fe Mo� ssbauer absorption spectra were recorded for series 1 and 2 samples of MgFe2O4 at room temperature (see Fig. 2), and of Mg0.5Zn0.5Fe2-xCrxO4 at room temperature and at 80 K (see Fig. 3). The spectra were least squares fitted using Lorentzian lineshapes and a first order perturbation approach to the combination of electric quadrupole and magnetic interactions. For the MgFe2O4 samples two subcomponent spectra, attributable to Fe atoms in the octahedral and tetrahedral sites, were fitted.For the Mg0.5Zn0.5Fe2-xCrxO4 spectra broad lines were observed, indicating the occurrence of distributions in the environments of the 57Fe nuclei, with corresponding distributions in the Mo�ssbauer hyperfine parameter.Such distributions are to be expected in the chromium substituted materials where the Cr atoms disrupt the long range magnetic interactions between the Fe atoms. Consequently these spectra were analysed using a 200 box histogram model for the distribution of hyperfine fields experienced by the 57Fe nuclei. For simplicity a single isomer shift parameter was used for all the histogram subspectra, and the quadrupole splitting was set to zero.The linewidths of the histogram subspectra were fixed at an appropriate experimental minimum, namely Fig. 2 Room temperature Mo�ssbauer spectra for samples of MgFe2O4 0.28 mm s-1. Smoothing and endpoint constraints were prepared by SHS in zero field and in a field of 1.1 T, after sintering for 2 h at 1400 °C.Solid lines represent a least squares fit with two applied to the allowed histogram solutions. Further details of subcomponents corresponding to Fe atoms in octahedral and the histogram fitting program are available elsewhere.13 The tetrahedral sites in the spinel structure. fitted spectra are shown as solid lines in Fig. 2 and 3 and selected parameters obtained from these fits are given in general decrease in a parameter with increasing chromium Table 3 and 4.content was noted. This mirrors the same results for conven- Hysteresis loops were recorded on all the samples in fields tionally prepared materials.10,11 This variation in unit cell may up to 7.5 kOe at room temperature. Representative data are be attributed to the smaller ionic radii of six-coordinate Cr3+ shown in Fig. 4. The maximum magnetisation smax, remanent ions compared to those of six-coordinate high spin Fe3+ magnetisation sr and coercive force Hc are listed in Table 5. ions.12 The chromium preferentially replaces iron from octa- For all samples prepared in magnetic field (series 2) the hedral sites because of favourable crystal field eVects (Cr3+ magnetic parameters are greater than those prepared in the 6/5 Do, Fe3+ 0 Do).absence of a magnetic field. A general decrease in smax and sr For series 1 materials the lattice parameters were consistently and an increase in Hc was observed with chromium larger than for series 2, albeit at the limits of experimental substitution. resolution. This eVect is minor, and might be explained by diVerences in the degree of Cr substitution achieved in the two Discussion series.However, from our observations both series of products appear to be almost single phase, with no signs of unreacted Mo�ssbauer measurements chromium oxide. We surmise therefore that this observation may be an eVect of the applied field during synthesis. It is well known that magnesium ferrite MgFe2O4 has an inverse spinel structure with Mg occupying octahedral sites EDAX measurements for Mg0.5Zn0.5Fe2-xCrxO4 showed the presence of magnesium, iron, chromium and zinc only and Fe equally occupying octahedral and tetrahedral sites: Fe[MgFe]O4.In contrast zinc ferrite is a normal spinel with (oxygen was below the machine cut oV ). The elemental ratios were within experimental error (1–2%) identical for the same Zn occupying the tetrahedral positions and Fe occupying the octahedral sites: Zn[Fe2]O4.Mg–Zn ferrite is a solid solution sample across many surface spots indicating that a homogeneous powder was formed. The atom percentages mirrored between these, with Zn2+ ions occupying tetrahedral sites and Mg2+ ions on octahedral sites.Chromium substitution into very closely the expected values, for example the sample of formula Mg0.5Zn0.5Fe0.5Cr1.5O4 gave an average elemental this species occurs preferentially on the octahedral sites. From the current set of Mo�ssbauer spectra these site ratio of Mg0.5Zn0.5Fe0.6Cr1.6. The FTIR spectra of MgFe2O4 and Mg0.5Zn0.5Fe2-xCrxO4 occupancies may be tested for the pure Mg ferrites, as shown in Fig. 2. As can be seen from the data, reasonable quality fits mirrored previous literature measurements, showing three broad overlapping bands: a shoulder at 685 cm-1 and two have been obtained for both the zero field and applied field SHS samples assuming an equal spectral area for the two bands at 520 and 410 cm-1. An additional band grows into Table 3 Room temperature Mo�ssbauer parameters for sintered MgFe2O4 ferrites prepared v performed in zero field and in an applied field of 1.1 T (series 1 and 2): isomer shift d (±0.01 mm s-1); quadrupole shift 2e (±0.01 mm s-1); linewidths C and DC (±0.01 mm s-1) and hyperfine field Bhf (±0.1 T).Two equal area subspectra were fitted, corresponding to Fe atoms at the octahedral and tetrahedral sites.Preferential line broadening due to distributions in local environments was modelled by ascribing linewidths of C+DC, C and C-DC to the outer, middle and inner pairs of lines in the sextets Octahedral sextet Tetrahedral sextet d 2e C DC Bhf d 2e C DC Bhf Series 1 0.43 0.06 0.51 0.16 48.1 0.16 -0.08 0.49 0.14 47.6 Series 2 0.45 0.04 0.54 0.15 48.3 0.17 -0.08 0.49 0.11 47.9 J.Mater. Chem., 1998, 8, 2701–2706 2703Fig. 3 Mo�ssbauer spectra measured at room temperature and at 80 K for samples of Mg0.5Zn0.5Fe2-xCrxO4 (x=0, 0.3 and 0.6) prepared by SHS in zero field and in an applied field of 1.1 T, after sintering for 2 h at 1400 °C. Solid lines represent least squares fits to the data: for the magnetically split spectra a probability distribution of hyperfine fields was used, otherwise a combination of one or two quadrupole split doublets was used.subcomponent sextets. This is consistent with the equal occu- field SHS sample are larger than in the zero field SHS counterpart. pancy of octahedral and tetrahedral sites by the Fe atoms. The parameters obtained from the fits, given in Table 3, are Mo�ssbauer spectra of some selected Mg0.5Zn0.5Fe2-xCrxO4 samples recorded at room temperature and 80 K are shown in also consistent with this assignment.Comparison of the Mo�ssbauer parameters for the two samples shows only minor Fig. 3.We note in passing that further Mo�ssbauer experiments, such as at liquid helium temperatures in the presence of large diVerences, although it may be significant that the hyperfine fields at both the octahedral and tetrahedral sites in the applied external magnetic fields, could be used to garner information Table 4 Mo�ssbauer parameters at room temperature and 80 K for selected examples of sintered Mg0.5Zn0.5Fe2-xCrxO4 ferrites prepared via SHS reactions in zero field and in an applied field of 1.1 T (series 1 and 2): isomer shift d (±0.01 mm s-1) and mean hyperfine field Bhf (±0.5 T).The spectra were modelled with a 200 box histogram probability distribution of subcomponent sextets, each of which had the same isomer shift d, a quadruple shift of zero, and a hyperfine field covering the range from zero to 65.0 T. In some cases the spectra comprised one or two paramagnetic doublets, in which case the parameters given are the isomer shift d (±0.01 mm s-1) and quadrupole splitting D (±0.01 mm s-1) Room temperature spectra Spectra at 80 K Series 1 Series 2 Series 1 Series 2 x d D Bhf d D Bhf d D Bhf d D Bhf 0.0 0.32 — 31.7 0.33 — 31.4 0.43 — 49.9 0.42 — 48.0 0.3 0.33 — 19.7 0.35 — 27.4 0.42 — 46.1 0.42 — 47.2 0.6 0.31 0.90 — 0.38 — 36.5 0.35 0.42 — 1.5 0.33 0.49 — 0.44 0.52 — 2704 J.Mater. Chem., 1998, 8, 2701–2706Magnetic measurements As is evident from the data in Table 5, the introduction of Cr3+ ions into magnesium zinc ferrite strongly aVects the room temperature magnetic parameters of the system. The decrease in both the maximum magnetisation smax and the remanent magnetisation sr mirrors the fall in Mo�ssbauer hyperfine fields, and may be ascribed to the increasing fluctuations in the Fe moments as the increasing Cr concentration disrupts the interatomic exchange interactions. Systematic diVerences in the magnetisation parameters are seen for the series 1 and 2 samples, with the applied field samples having consistently higher smax and sr parameters.This is most probably indicative of a degree of magnetic texturing, generated during the applied field SHS processing, remaining in the sintered powders.This is an interesting result and indicative of possible microstructural eVects which may be related to the Fig. 4 Representative room temperature hysteresis loops of samples lattice parameter changes seen earlier in the X-ray diVraction of Mg0.5Zn0.5Fe2-xCrxO4 prepared by SHS in zero field (ZF) and in experiments.In contrast the coercive force Hc data for both an applied field (AF) of 1.1 T, after sintering at 1400 °C for 2 h. series 1 and series 2 samples are similar. Table 5 Bulk magnetic properties of sintered MgFe2O4 and EVect of magnetic field on the synthesis Mg0.5Zn0.5Fe2-xCrxO4 prepared by SHS performed in zero field and in an applied field of 1.1 T (series 1 and 2): maximum magnetisation The only diVerence between the samples prepared in series 1 smax (±0.1 emu g-1), remanent magnetisation sr (±0.1 emu g-1) and and 2 was the application of an external magnetic field during coercive force Hc (±0.5 Oe).Measurements were made at room the SHS step. Despite this, the two series of samples show temperature in applied fields up to 7.5 kOe some small diVerences in X-ray and Mo�ssbauer parameters, Series 1 Series 2 and major diVerences in magnetic parameters.The changes in Mg0.5Zn0.5Fe2-xCrxO4 unit cell dimension may indicate diVerent levels of crossx ss sr Hc ss sr Hc substitution or defects in the two series. These eVects may have arisen as a consequence of a slightly faster propagation 0 59.8 1.42 9.4 82.6 1.80 9.3 rate, coupled with an increased reaction temperature, in the 0.3 46.8 1.34 9.9 59.1 1.37 9.8 0.6 12.2 0.31 12.5 24.5 1.10 11.8 applied field SHS reactions: visual inspection of the reactions 0.9 2.45 0.19 16.7 10.9 0.56 16.2 indicated that the synthesis wave moved faster and glowed 1.2 0.99 0.14 39.5 2.78 0.26 21.7 with a more yellow coloration in the applied field series.In 1.5 0.61 0.10 82.0 1.17 0.24 77.4 the applied field some pre-organisation of the precursor pow- MgFe2O4 34.6 1.97 19.8 39.6 2.42 18.3 ders was observed prior to initiation of the SHS reaction, most likely due to the alignment of the iron and iron oxide components along the magnetic flux lines.It is reasonable to on the cation distributions in the Cr substituted materials.14 suppose that this pre-organisation leads to better surface However, in the present work we choose to limit our consider- contacts between the reacting powders, hence giving rise to ation to the general features of the Mo�ssbauer spectra of these the hotter and faster reaction conditions.samples, and to a comparison of the data obtained from the series 1 and 2 samples. On inspection of Fig. 3 it is apparent that in the case of Conclusions Mg0.5Zn0.5Fe2O4 there is little diVerence between the series 1 Self propagating high temperature synthesis allows the rapid and 2 samples either at room temperature or at 80 K, while formation of near single phase chromium substituted mag- there are discernible diVerences between the two series for nesium and magnesium zinc ferrite.Despite the short synthesis Mg0.5Zn0.5Fe1.7Cr0.3O4 at room temperature, but less so at time solid solutions of the ferrites were formed. The ferrites 80 K. The fitted parameters in Table 4 bear this out, with the had magnetic orders of merit equal to those prepared by main diVerence being a higher mean hyperfine field in the conventional synthesis. Mo�ssbauer spectroscopy showed that applied field SHS x=0.3 sample at room temperature than in for all samples the chromium metal was substituted on the its zero field counterpart.We tentatively ascribe this to a octahedral sites within the spinel. Use of a magnetic field possibly less disruptive distribution of substituted Cr atoms in during SHS synthesis increases the rate of reaction propa- the series 2 sample, with a commensurate increase in the gation. The materials prepared by SHS in a magnetic field samples Curie temperature. However, such a result needs showed increased saturation magnetisation compared to those further investigation, beyond the scope of the present study, prepared in the absence of a field.It is likely that these before this can be known for certain, especially since a slightly enhanced magnetic eVects are a consequence of chanvel of Cr doping in the two samples might give rise product microstructure rather than in spinel site occupancy.to a similar eVect. Representative data for the higher Cr dopings of x=0.6 and 1.5 are also shown in Fig. 3 for series 1 samples. These References spectra are paramagnetic doublets at room temperature, indicating that in these samples the Cr substitution is suYciently 1 I.P. Parkin, Chem. Soc. Rev., 1996, 199; I. P. Parkin, G. E. Elwin, A. V. Komarov, Q. T. Bui, Q. A. Pankhurst, L. Fernandez high to disrupt the Fe ordering enough that the Curie tempera- Barquin and Y. G. Morozov, J. Mater. Chem., 1998, 8, 573. tures of the samples fall below room temperature. Indeed, for 2 A. G. Merzhanov, Proc. Technol., 1996, 56, 222. the x=1.5 sample the Curie temperature falls below 80 K, so 3 A. E. Padalino, J. Am. Ceram. Soc., 1960, 43, 183; that even at that temperature a doublet is measured. For the Y. D. Tretyakov and N. I. Oleinikov, Inorg. Mater., 1965, 1, 254. x=0.6 sample at 80 K a broad hyperfine split spectrum is 4 Ferrites, ed. L. A. Rabkin, S. A. Soskin and B. S. Epshtein, observed, with parameters that are in line with those seen for Energy, Leningrad, 1968, p. 384. 5 H. Kojima, in Ferromagnetic Materials: A Handbook of the the lower Cr content samples. J. Mater. Chem., 1998, 8, 2701–2706 2705Properties of Magnetically Ordered Substances, ed. 11 F. C. Romeijn, Philips Res. Rep., 1953, 8, 304. 12 R. Shanon, Acta Crystallogr., Sect. A, 1976, 32, 751. E. P. Wohlfarth, North-Holland, Amsterdam, 1982, vol. 3. 6 V. S. Darshane, S. S. Lokegaonkar and S. G. Oak, J. Phys. IV Fr., 13 Q. A. Pankhurst, S. Suharan andM. F. Thomas, J. Phys. Condens. Matter, 1992, 4, 3551. 1997, 7, C1. 7 M. V. Kuznetsov, Y. G. Morozov, M. D. Nersesyan and 14 P. M. A. de Bakker, E. De Grave, D. GryVroy, R. E. Vandenberghe and P. Moens, Mater. Sci. Forum, 1991, T. I. Ignateva, Inorg. Mater., 1995, 31, 1125. 8 L. C. F. Blackman, Trans. Faraday Soc., 1959, 55, 391. 79–82, 777. 9 A. H. Morrish and P. E. Clark, Phys. Rev. B, 1975, 11, 278. 10 E. W. Gorter, Philips Res. Rep., 1954, 9, 295. Paper 8/04942D 2706 J. Mater. Chem., 1998, 8, 2701–2706