J. MATER. CHEM., 1994, 4(8), 1307-1308 Synthesis and Properties of a New p Polymorph of Li,CrO, M. A. K. L. Dissanayake, Susana Garcia-Martin, Regino Saez-Puche,+ H. H. Sumathipala and Anthony R. West Department of Chemistry, University of Aberdeen, Meston Walk, Aberdeen, UK AB9 2UE D-Li,CrO,, isostructural with p-Li3P04, has been synthesized by two routes. Reaction of Li,CO, and Cr203 in a 5 :1 mole ratio in flowing Ar at 800°C yields j3-Li,Cr04 with a small amount of Li,O. The Li2C0, seems to have two functions. It is an essential reactant and acts as a novel oxidant, promoting conversion of Cr"' to Crv. Reaction of Li,CO, and CrO, in a 3 :1 mole ratio in flowing Ar also gives phase-pure P-Li,CrO,. Magnetic measurements show Curie-Weiss behaviour in the range 20-300 K with a magnetic moment of 1.70 pB consistent with Crv in tetrahedral coordination.Conductivity measurements show a modest level of electronic conduction. Stable oxides of CrV are relatively rare. Li,CrO, containing CrV has, however, been prepared by stoichiometric reaction of Li2Cr0, and Li,O.' Subsequently, a single crystal was grown and shown to be isostructural with the higher-temperature y polymorph of Li,P0,.2 Li,PO, is a poly-morphic oxide and the crystal structures of both the high-and low-temperature forms have been determined.,,, The low- temperature p structure has the basic wurtzite structure in which only one set of tetrahedral sites is occupied, but with cation ordering. The high-temperature y structure also has hexagonal close-packed oxygen layers, although the layers are more buckled than in the p structure; in y-Li,PO, the cations are distributed over both sets of available tetrahedral sites.During unsuccessful attempts to synthesize 'Li,CrO,' contain-ing Cr'", we have instead prepared a new p polymorph of Li,CrO, containing CrV. The reaction involved oxidation of Cr"' to CrV in an 0,-free atmosphere, with Li,C03 as the oxidising agent. Subsequently, we have prepared the same p-Li3Cr0, polymorph in a reaction involving reduction of CrV* to CrV. These results are reported here. Experimental P-Li,CrO, was prepared first by reacting powder mixtures of Li2C03 and Cr203 (both analytical grade) in a 5: 1 molar ratio in Au foil boats in flowing Ar.Initial firing was at 500-700°C for a few hours to expel CO,, after which the reaction mixture was reground and heated at 800°C for 12-24 h to give a dark green, somewhat moisture-sensitive product. Subsequently, /?-Li,CrO, was also prepared by reac- tion of a mixture of Li2C03 and CrO, in a 3: 1 molar ratio at 700 'C in Au foil boats in flowing Ar. Exhaust-gas analyses were performed by gas chromatography using a Perkin-Elmer F33 gas chromatograph. Powder X-ray diffraction was carried out using a STOE STAD1,'P diffractometer, with position sensitive detector (PSD) and Cu-Ka, radiation. Silicon was added as an internal standard for accurate d-spacing measurements. Rietveld refinement of the Li3Cr04 structure was not feasible due to the high level of background radiation associated with Cr fluorescence.Pellets for conductivity measurements were cold- pressed at 2 ton cmP2 and sintered at 800 "C for 2-3 h in order to increase their mechanical strength. Electrodes were usually made by coating opposite pellet faces with Engelhard gold paste and gradually heating to 700 "C in air. Conductivity was determined by ac impedance measurements from 0.10 Hz to 13 MHz using Solartron 1250/1286 and HP 4192A impedance analysers in air on a heating cycle. Phase changes t Departamento de Quimica Inorganica, Facultad de Ciencias Quimicas, Universidad Complutense de Madrid, Madrid-28040, Spain. were studied using a Stanton Redcroft DTA 675 instrument, heating rate 5 "C min-'. Magnetic susceptibility measurements in the temperature range 4.2-300 K were made using a DSM-5 pendule magnet- ometer.The set-up was calibrated with Hg[Co(SCY),] and Gd,( S04)3.8H20. Results The synthesis of Li,CrO, from Li2C0, and Cr203 has some unusual features. In flowing Ar, a nearly phase-pure product is obtained provided an excess of Li,CO, (5:1 ratio instead of 3 : 1)is present in the reaction mixture. If a similar reaction is attempted in air or O,, the mixture melts at 700-800°C and no Li,CrO4 forms. If instead a 3: 1 reaction mixture is used, in flowing Ar, then a mixture of Li,CrO, and LiCrO, results. It is clear therefore, that an excess of Li,CO, is required, whose prime function is to act as an oxidant for Cr"' to CrV. Two possible mechanisms for this may be considered.First, Li,O may be the oxidant, giving Li metal which vapourises. We found no direct evidence for this; for instance, on bubbling the exhaust Ar through H,O, no alkalinity developed. Secondly, C02 or carbonate may be the oxidant, giving CO as by-product. Clear evidence for CO formaiion was obtained by gas chromatography of the Ar exhaust gas (a blank experiment, on the synthesis of Li,SiO, from Li2C03 and Si02, in flowing Ar showed no detectable CO). In addition, the excess of Li,O should remain in the reaction mixture and, in fact, a small amount was detected hy XRD. It appears, however, that only C02 generated in situ can act as an oxidant since, on heating 3: 1 mixtures of Li2C0, and Cr203 in flowing CO,, no oxidation to give Li,Cr04 was detected.We conclude that CO, or carbonate rather than Li20 is the principal oxidant, but that the mechanism is not fully understood. Li,Cr04 was also prepared by reaction of Li2C:O3 and CrO, in Ar. In this case, simple reduction of CrV' to CrV occurred during heating; also, the expected 3: 1 ratio of starting materials could be used to obtain a phase-pure product. Since Li3Cr04 can be obtained by two pathways, which involve oxidation and reduction, respectively, of the Cr starting materials, it is concluded that the lattice energy of Li,CrO, must be particularly favourable for stabilising the unusual +V oxidation state of Cr. Of the two methods, that involving reduction of CrO, is preferred since this gives a clean, phase-pure product.Indexed powder X-ray diffraction data for P-Li,C'rO4 are given in Table 1; apart from d-spacing shifts associated with a larger unit cell, the powder data are very similar to those of p-Li3P04, indicating an isostructural relationship. The structure is, therefore, a wurtzite superstructure with the cations ordered over one set of tetrahedral sites within a hexagonal close-packed oxide ion array.5 Table 1 Powder X-rciy diffraction data for P-Li,CrO, [~=5.4305 (0.0009) A, b=6.3136 (0.0008) A, ~=4.9413 (0.0008) A] 28 (obs)/degrees ' ~~ h k 1 int. d (obs)/A d (calc)/A 16.296 100 19.3 5.4348 5.4305 21.555 110 73.9 4.1 192 4.1171 22.828 011 100 3.8923 3.8912 24.325 101 46.3 3.6561 3.6548 28.223 111 35.5 3.1593 3.1630 32.787 120 43.5 2.7292 2.7292 32.948 200 52.8 2.7163 2.7153 35.971 210 7.7 2.4946 2.4944 36.321 002 48.0 2.4714 2.4707 37.624 121 22.7 2.3887 2.3890 37.772 201 43.2 2.3797 2.3797 40.036 102 4.5 2.2502 2.2489 40.484 211 8.8 2.2263 2.2267 42.644 112 11.6 2.1184 2.1185 46.225 130 5.4 1.9623 1.9623 46.664 022 7.7 1.9449 1.9456 46.695 031 7.4 1.9358 1.9362 47.828 221 7.7 1.9002 1.9002 49.740 122 12.8 1.8316 1.8316 49.907 202 7.4 1.8258 1.8274 52.080 212 6.5 1.7546 1.7553 55.181 230 5.7 1.663 1 1.6634 57.785 013 7.1 1.5942 1.5938 58.435 040 13.6 1.5780 1.5784 58.755 320 14.8 1.5702 1.5703 60.139 132 6.3 1.5373 1.5366 60.454 113 4.3 1.5301 1.5293 63.665 023 4.5 1.4604 1.4603 302 1.4602 64.220 141 5.4 1.449 1 1.4490 66.225 123 15.9 1.4100 1.4102 67.827 232 32.4 1.3806 1.3798 68.133 213 32.4 1.3751 1.3745 70.790 042 7.1 1.3299 1.3301 71.050 322 9.1 1.3256 1.3253 71.264 331 6.3 1.3222 1.3223 71.656 241 6.5 1.3159 1.3154 The pentavalent oxidation state of Cr was confirmed by magnetic susceptibility measurements.The magnetic suscepti- bility of Li,CrO, follows a Curie-Weiss behaviour in the temperature range 300-20 K with a magnetic moment of 1.70 pB, Fig. 1. This fully agrees with that expected for the ground term 2E derived from CrV in tetrahedral coordination.The positive value of the Weiss constant, 8 =2.6 K, is indicative of the existence of ferromagnetic interactions in the chromium sublattice, which is confirmed by an increase in the XT value observed below 20 K, Fig. 1 inset. Y "'"E 0 0 0.2h O$ 600 100 200 0-to0f--4001 TIK 0E 0x 0 0 200c no * oo Fig. 1 Temperature dependence of the reciprocal molar susceptibility for Li,CrO,. The inset shows the ~Tvs.T plot. J. MATER. CHEM., 1994, VOL. 4 --3 --5 --7 1-9 Fig. 2 Variation of conductivity with temperature for Li,CrO, The variation of conductivity with temperature, determined by ac impedance measurements is shown in Fig. 2. Modest levels of conductivities, typically 2 x lo-* R-' cm-' at 25 "C rising to 4 x lop5 at 300°C with an activation energy of 0.44 eV were observed.The nature of the impedance -plots, which did not show much evidence of low-frequency electrode polarization effects, indicates the conduction to be primarily electronic. This is consistent with electron hopping associated with the d' electronic state of CrV. In addition, significant numbers of Li' interstitials or vacancies in the a-Li,CrO, structure are not expected and therefore, high Li' ion conduc- tivity in stoichiometric Li,CrO, is not observed. DTA results on /?-Li,CrO, showed an endotherm on heating at 750 "C which was reversible on cooling at 705 "C.This may be associated with the polymorphic phase transition $ y but we could not isolate the high-temperature phase by quenching from e.g.800 "C. The crystal structure report,2 which is for the high-temperature y polymorph, was carried out on a crystal that had been stabilised, by an unknown mechanism, to room temperature. Conclusions A new polymorph of Li,CrO, has been prepared. It appears to be isostructural with P, As and V analogue^,^.^ all of which show a p+y transition with increasing temperature. Li,CrO, may be prepared by two routes, involving either oxidation of Cr"' by Li,C03 or reduction of CrV' in flowing Ar. The structure of Li,CrO,, its magnetic and electrical properties are consistent with the pentavalent oxidation state of Cr. We thank the EC (Human Capital and Mobility Project) for both the grant which supported S.G.M. and for providing financial support (CEC contract C11-CT91-948) to H.H.S. and M.A.K.L.D.; also Dr. I.L. Marr for assistance with the gas chromatography. We are indebted to Dr. Paul Attfield, who suggested that COz or carbonate may be the oxidant for the Cr"' to CrV conversion. References R. Scholder and H. Schwartz, 2.Anorg. Allg. Chem., 1963,1,326. G. Meyer, D. Paus and R. Hoppe, Z. Anorg. Allg. Chem., 1974, 408, 15. J. Zemann, Actu Crystallogr., Sect. B,1960, 13, 863. C. Keffer, A. Mighell, F. Mauer, H. Swanson and S. Block, Inorg. Chem., 1967,6,119. A. R. West, 2.Kristallogr., 1975, 141,422. A. R. West and F. P. Glasser, J. Solid State Chem., 1972,4,20. R. D. Shannon and C. Calvo, J. Solid State Chem.. 1973,6, 538. Paper 4/02012J; Received 5th April, 1994