Synthesis and Thermal Stability of Cubic ZnO in the Salt Nanocomposites
P.S. Sokolov, A.N. Baranov, Zh.V. Dobrokhotova, V.L. Solozhenko
SSynthesis and Thermal Stability of Cubic ZnO in the Salt Nanocomposites
P.S. Sokolov,
A.N. Baranov, Zh.V. Dobrokhotova, V.L. Solozhenko
LSPM-CNRS, Université Paris Nord, 93430 Villetaneuse, France Department of Materials Science, Moscow State University, 119991 Moscow, Russia Chemistry Department, Moscow State University, 119991 Moscow, Russia Kurnakov Institute of General and Inorganic Chemistry, RAS,
Abstract
Cubic zinc oxide (rs-ZnO), metastable under normal conditions was synthesized from the wurtzitemodification (w-ZnO) at 7.7 GPa and ~800 K in the form of nanoparticles isolated in the NaClmatrix. The phase transition rs-ZnO → w-ZnO in nanocrystalline zinc oxide under ambientpressure was experimentally studied for the first time by differential scanning calorimetry and high-temperature X-ray diffraction. It was shown that the transition occurs in the 370-430 K temperaturerange and its enthalpy at 400 K is –10.2 ± 0.5 kJ mol -1 . Key words: zinc oxide, high-pressure synthesis, phase transitions *e-mail: [email protected] ntroduction
Hexagonal (P6 mc) zinc oxide with the wurtzite structure, w -ZnO, is a direct-band semiconductor( E g = 3.4 eV) with the highly ionic character of chemical bond [1], whereas cubic (Fm3m) zincoxide with the rock salt structure, rs -ZnO, is an indirect semiconductor with the band gap energy E g ~ 2.7 eV at 11 GPa [2].Under normal conditions, wurtzite ZnO is thermodynamically stable and at pressures of about9 GPa and room temperature transforms into the cubic modification [3]. The reverse transition isobserved only upon the pressure drop down to 2 GPa [4, 5], indicating considerable hysteresisbetween the direct ( w → rs ) and reverse ( rs → w ) phase transitions in ZnO at room temperature. Thehysteresis width decreases with the temperature increase, and above 1200 K the branches of thedirect and reverse transition are brought together under a pressure of about 6 GPa [4, 5], which canbe considered the equilibrium pressure of this phase transition. The cubic phase of zinc oxide isstable only at pressures above 2 GPa and cannot be quenched down to ambient pressure [4, 5].It was experimentally shown [6-8] that for nanocrystalline w -ZnO the direct transition occurs underhigher (> 9 GPa) pressures and the samples with a particle size of ~12 nm treated at pressuredhigher than 15 GPa retain cubic structure after decompression [6, 7]. However, since allexperiments mentioned were carried out in diamond anvil cells, it cannot be excluded that residualpressures are retained in the sample after pressure release.The purpose of the present work is to study the possibility to stabilize cubic zinc oxide synthesizedunder relatively low (~7 GPa) pressures in the sodium chloride matrix. Experimental
Sodium chloride (special-purity grade) and monodispersed nanoparticles of wurtzite zinc oxide withthe average size ~50 nm, synthesized by the thermal decomposition of zinc acetate in diethyleneglycol by the earlier described method [9], were used for the preparation of the starting mixtures. Amixture of w -ZnO and NaCl nanoparticles was ground in an agate mortar and then molded in a steelpress mould under a pressure of 1250 bar.Cubic ZnO was synthesized in a toroid-type high-pressure apparatus [10] at 7.7 GPa in the700-900 K temperature range. The cell was calibrated against pressure using the Bi III-IV phasetransition (7.7 GPa at room temperature), whereas temperature calibration was performed usingPt10%Rh—Pt thermocouple without correction for the pressure effect on the thermocouple emf.old capsules were used to insulate the samples from the high-pressure assembly. The sampleswere gradually compressed at room temperature, and then the temperature was increased up to therequired value. After isothermal heating for 15 min, the samples were quenched by switching offthe power and slowly decompressed down to ambient pressure.X-ray diffraction study of the synthesized samples and precise determination of the latticeparameters were performed on a TEXT 3000 INEL ( λ = 1.54056 Å) diffractometer. A sample ofLaB ( a = 4.15695 Å) was used as standard for detector adjustment. Lattice parameters weredetermined by the full-profile analysis with allowance for corrections to X-ray absorption and theshift of the sample surface from the diffraction plane. The size of ZnO crystallites was determinedby the Debye—Scherrer equation taking into accounts the instrumental function.The quenched samples were studied in situ at temperatures up to 1000 K in vacuo by X-raydiffraction on a Rigaku D/MAX 2500 diffractometer (Cu K α radiation) in the 30-65º (2 Θ ) rangeusing an HT-1500 high-temperature attachment. The 2 Θ scan rate was 5 K min -1 at a linear heatingrate of 2 K min -1 , which corresponded to the change in the sample temperature by 15 K duringcollection of each diffraction pattern.The differential scanning calorimetry (DSC) study of the quenched samples in the 300-900 Ktemperature range was carried out using a Netzsch DSC 204 F1 calorimeter at continuous heatingwith rates of 2, 5 and 10 K min -1 in high-purity (> 99.998%) argon. Aluminum ampules were usedas sample holders. The calorimeter was calibrated by the phase transitions of the standardsubstances (Hg, KNO , In, Sn, Bi, and CsCl with purity not lower than 99.99%). The experimentaldata were processed using the NETZSCH Proteus Analysis program package. Results and discussion
At 7.7 GPa and temperatures below 1000 K when pristine w -ZnO nanopowder was used as astarting material, the quenching resulted in the formation of a mixture of rs -ZnO and w -ZnO invarious ratios rather than single-phase cubic zinc oxide. Therefore, all further experiments werecarried out only with w -ZnO in the NaCl matrix.Cubic ZnO was synthesized at 700—900 K which allowed us to ensure the completeness of the w -ZnO → rs -ZnO phase transition at 7.7 GPa, and to avoid an increase in the ZnO nanoparticlesizes at higher temperatures.ccording to the X-ray diffraction data of the quenched samples, the complete stabilization of thecubic structure of zinc oxide nanoparticles in the sodium chloride matrix is observed only startingfrom the ZnO/NaCl ratio 1 : 3 (see Table 1). The diffraction pattern of the rs -ZnO/NaClnanocomposite (1 : 3) with the maximum content of zinc oxide (25 wt. %) is presented in Fig. 1. Allreflections in the diffraction patterns are assigned either to rs -ZnO (JCPDS No. 77-0191; Fm3m, a = 4.28 Å) or to NaCl (JCPDS No. 05-0628; Fm3m, a = 5.642 Å); no reflections of other phases,including w -ZnO, are observed. The average sizes of coherent scattering areas of rs -ZnO, estimatedfrom the reflections broadenings in the diffraction patterns by the Debye-Scherrer equation, arealmost the same for all salt nanocomposites listed in Table 1, being ~50 nm. This indicates that thesizes of the ZnO nanoparticles do not change during the phase transition at high pressures andtemperatures.When w -ZnO/NaCl initial mixture contains more than 33 wt. % zinc oxide, the quenched samplesrepresent rs-ZnO and w-ZnO mixtures, i.e. only partial stabilization of the cubic structure of zincoxide in the salt nanocomposites is observed after quenching. If w -ZnO micropowders (99.99%,Aldrich, 325 mesh) with an average particle size of ~44 µm are used as the starting material, thecubic ZnO phase can not be quenched down irrespective to the w -ZnO/NaCl ratios in the startingmixtures. This indicates that zinc oxide is characterized by some critical particle size, above whichno stabilization of rs -ZnO can be achieved in the salt nanocomposites under normal conditions. Theresults obtained allow the conclusion that the salt matrix plays the key role in the stabilization of the rs -ZnO nanoparticles synthesized at high pressures and temperatures after their quenching.In a special series of experiments we have shown that the dissolution of the salt matrix of the rs -ZnO/NaCl nanocomposites in distilled water initiates the reverse phase transition rs -ZnO → w -ZnO, and after the complete removal of NaCl the sample represents w -ZnO with traceamounts of rs -ZnO.At room temperature the rs -ZnO nanoparticles in the sodium chloride matrix exhibit no tendency tothe transition to w -ZnO for at least 5-6 months. For a longer storage period (1 year), the partialreverse transition rs -ZnO → w -ZnO is observed in the samples, and finally, after two years, thenanoparticles of cubic zinc oxide are completely transformed into wurtzite zinc oxide. Thus, it canbe asserted that at room temperature the salt matrix kinetically hinders the reverse phase transitionof zinc oxide from the metastable cubic modification to the thermodynamically stable wurtzite one.The thermal stability of the as-synthesized rs -ZnO/NaCl nanocomposites of different composition(see Table 1) at ambient pressure was studied in situ by X-ray diffraction and differential scanningalorimetry. The diffraction patterns of the rs -ZnO/NaCl nanocomposite 1 : 3 collected in situ arepresented in Fig. 2. It can be seen in the left part of Fig. 2 that the ( ) reflection of w -ZnO appearsat 365 K and its intensity increases during further temperature increase. The right part of Fig. 2shows the corresponding decrease in the intensity of the ( ) reflection of rs -ZnO, whichdisappears completely at 425 K. Thus, two phases of zinc oxide in the NaCl matrix coexist in the365–425 K temperature range.Fig. 3 shows the temperature dependence of the integral intensity of ( ) reflection of rs -ZnOnormalized to the corresponding 300 K value. The curve has the classical S-like shape, i.e. , at lowtemperature the transition rate is low, then it increases sharply with temperature, passes through amaximum at ~384 K, and further decreases down to zero at temperatures above 430 K.Upon further heating of the nanocomposite in the temperature range from 430 to 970 K, theintensities of w -ZnO reflection increase noticeably, most likely, due to an increase in degree ofcrystallinity of the phase. However, according to the DSC data, this process is not accompanied byany thermal effect.The results of the thermoanalytical study of the 1 : 3 and 1 : 5 salt nanocomposites are presented inFig. 4 and Table 1. At temperatures above 380—430 K, the DSC curves contain the pronouncedexothermic effect, which can unambiguously ascribed to the irreversible structural transition ofcubic zinc oxide into the wurtzite phase, which is in good agreement with the results of high-temperature in situ X-ray diffraction study. As can be seen from the Table 1, for all rs -ZnO/NaClnanocomposites the increase in the heating rate is accompanied by an increase in the peaktemperature.Along with the main exothermic effect, the thermoanalytical curve of the nanocomposite with themaximum content of rs -ZnO (1 : 3) contains the weak exothermic effect at 350-370 K(see Fig. 4, b ), which can be attributed to the phase transition of submicronic aggregates of the rs -ZnO nanoparticles. It can be assumed that the rs -ZnO → w -ZnO transition in such aggregatesoccurs at lower temperatures than in the case of rs-ZnO nanoparticles isolated in the salt matrix. Fora more "dilute" composition 1 : 4 , this effect is less pronounced and is completely absent for thecomposition with rs -ZnO/NaCl ratio 1 : 5 (see Fig. 4, a). No other processes accompanied by theheat release or absorption were observed up to 870 K. Thus, based on the data of DSC and high-temperature X-ray diffraction, we can conclude that the structural transition Fm3m → P63mcobserved for zinc oxide in the sodium chloride matrix at ambient pressure is the monotropic first-order phase transition [11] that occurs without formation of any intermediate phase.or the 1 : 3 and 1 : 4 rs -ZnO/NaCl nanocompositions , the thermal effect of the phase transition ata given heating rate was determined as the sum of two observed exothermic effects taking place at350-370 K and 380—430 K, respectively. The thermal effect values obtained at different heatingrates were averaged for each composition (see Table 1). The enthalpy of the phase transition at400 K was determined by averaging over all experimental values of thermal effects and is equal to-10.2±0.5 kJ mol -1 .The enthalpy of the reverse rs -ZnO → w -ZnO phase transition determined by us differs by an orderof magnitude from the enthalpy values of the direct w -ZnO → rs -ZnO phase transition estimatedfrom the data on ZnO solubility in borate melts ( ∆ tr H º
297 K = 24.5±3.6 kJ mol) [12, 13] andtemperature dependences of the emf of the electrochemical cells with ZnO-based electrodes( ∆ tr H º = 17.4±1.3 kJ mol -1 ) [14]; as well as from the ∆ tr H º
297 K = 3.3 kJ mol -1 value [3],calculated by the Clapeyron-Clausius equation from the experimental data on high-pressuressynthesis of rs -ZnO. Unlike the direct calorimetric determination of the enthalpy of the rs -ZnO → w -ZnO phase transition performed in the present work, the enthalpy values reportedearlier [3, 12-14] were obtained from indirect thermodynamic data and, hence, should be consideredas estimates. Conclusions
Thus, in the present work we have shown for the first time that the cubic phase of zinc oxide can besynthesized in the form of salt nanocomposites, which are kinetically stable under ambient pressureup to temperatures of about 370 K. This provides new possibilities for studying the physical andchemical properties of cubic ZnO. According to the DSC data, at ambient pressure the rs -ZnO → w -ZnO phase transition in the salt nanocomposites occurs in the 370-430 K range, andits enthalpy at 400 K is –10.2±0.5 kJ mol -1 . Acknowledgements
We thank O.O. Kurakevych for help in high-pressure experiments and V.A. Mukhanov for valuablecomments. P.S. Sokolov is grateful to the French Government for financial support (Bourse deco-tutelle no. 1572-2007).This work was financially supported by the Russian Foundation for Basic Research (Project No. 09-03-90442-Ukr_f_a). eferences
1. C. Klingshirn,
Phys. Status Solidi B , 2007, . 3027.2. J.A. Sans, A. Segura, F.J. Manjon, B. Mari, A. Munoz,
Microelectron. J. , 2005, , 928.3. C.H. Bates, W.B. White, R. Roy, Science, , 993.4. K. Kusaba, K. Syono, T. Kikegawa,
Proc. Jpn Acad. Ser. B , 1999, , 1.5. F. Decremps, J. Zhang, R.C. Liebermann, Europhys. Lett. , 2000, , 268.6. J.Z. Jiang, J.S. Olsen, L. Gerward, D. Frost, D. Rubie, J. Peyronneau , Europhys. Lett. , 2000, , 48.7. F. Decremps, J. Pellicer-Porres, F. Datchi, J.P. Itie, A. Polian, F. Baudelet, J.Z. Jiang, Appl. Phys. Lett., , 4820.8. E. Grzanka, S. Gierlotka, S. Stelmakh, B. Palosz, T. Strachowski, A. Swiderska-Srode,G. Kalisz, W. Lojkowski, F. Porsch, Z. Kristallogr. Suppl., , 337.9. E.W. Seeling, A. Yamilov, H. Cao, R.P.H . Chang,
Mater. Chem. Phys. , 2003, , 257.10. L.G. Khvostantsev, V.N. Slesarev, V.V. Brazhkin, High Press. Res. , 2004, , 371.11. J.B. Clarke, J.W. Hastie, L.H. . Kihlborg, R. Metselaar, M.M. Thackeray, Pure Appl. Chem. ,1994, , 577.12. A. Navrotsky, A. Muan, J. Inorg. Nucl. Chem., , 35.13. P.K. Davies, A. Navrotsky, J. Solid State Chem., , 264.14. J.S. Kachhawaha, V.B. Tare, Solid State Ionics , 1981, , 575. able 1. Results of the thermoanalytical study of the rs -ZnO/NaCl nanocompositesT max (K) at v T c
ZnO/NaCl a Q (kJ mole -1 ) b a Weight ratio b Q is the thermal effect c T max is the peak temperature of the main thermal effect at different heating rates, v T / K min -1 . ig. 1. Diffraction pattern of the rs -ZnO/NaCl nanocomposite (1 : 3) synthesized at 7.7 GPa and 800 K.
30 35 40 45 50 55 60 65 Θ (deg.) ZnO (Fm3m)NaCl ig. 2.
Regions of the diffraction patterns of the rs -ZnO/NaCl nanocomposite (1 : 3) taken in situ inthe course of linear heating with a rate of 2 K min –1 . Note: The reflections ( ) of rs -ZnOand ( ) of w -ZnO were chosen as the most intense non-overlapping reflections. ZnO (P6 mc) ZnO (Fm3m) Θ (deg.)
305 K320 K335 K350 K365 K380 K395 K410 K425 K440 K ig. 3.
Temperature dependence of the integral intensity of ( ) reflection of rs- ZnO normalizedto the corresponding 300 K value.
300 320 340 360 380 400 420 440 4600.00.20.40.60.81.0
Temperature (K) ∆ A T / ∆ A K ig. 4 . DSC curves of the rs -ZnO/NaCl nanocomposites of the 1 : 5 ( a ) and 1 : 3 ( b ) compositionrecorded at the heating rate 2 ( ), 5 ( ), and 10 ( ) K min –1 ( Q is the specific heat flux).
300 320 340 360 380 400 420 440 460-0.35-0.30-0.25-0.20-0.15-0.10-0.050.000.050.100.15 (b)Exo
Temperature (K)
Q/mW mg -1
300 320 340 360 380 400 420 440 460-0.35-0.30-0.25-0.20-0.15-0.10-0.050.000.050.10 Temperature (K) Q/mW mg -1-1
300 320 340 360 380 400 420 440 460-0.35-0.30-0.25-0.20-0.15-0.10-0.050.000.050.10 Temperature (K) Q/mW mg -1-1 Exo (a)
300 320 340 360 380 400 420 440 460-0.35-0.30-0.25-0.20-0.15-0.10-0.050.000.050.100.15 (b)Exo
Temperature (K)
Q/mW mg -1
300 320 340 360 380 400 420 440 460-0.35-0.30-0.25-0.20-0.15-0.10-0.050.000.050.10 Temperature (K) Q/mW mg -1-1