摘要:
Paper Temperature-resolved, in-situ powder X-ray diffraction of silver iodide under microwave irradiation{ G. R. Robb,* A. Harrison and A. G. Whittaker Chemistry Department, University of Edinburgh, Kings Buildings, West Mains Road Edinburgh, UK EH9 3JJ. E-mail: grr@ed.ac.uk Received 25th July 2002, Accepted 27th August 2002 Published on the Web 9th September 2002 Some unexpected properties of the ionic conductor, silver iodide, when heated by 2.45 GHz microwave radiation are investigated using temperature-resolved, in-situ powder X-ray diffraction. The lowering of the phase transition temperature can be explained by multi-phonon interactions with the microwaves and low-lying transverse optic modes. 1. Introduction In solid state chemistry, diffraction is arguably the most incisive technique for detailed structural analysis.This is particularly true for microwave chemistry as it involves volumetric heating (as opposed to heating from the surface inwards) and diffraction measures bulk properties, rather than surface properties (as opposed to most forms of spectroscopy for solids). Microwave chemistry, however, presents particular challenges to the experimental chemist. In order for a microwave vessel to heat a sample effectively, the walls must be smooth, continuous and conducting, i.e. metal. Holes are permissible, but these must have a maximum dimension of less than a quarter of the radiation wavelength (l ~ 122 mm at standard frequency of 2.45 GHz).1 These complications have meant that many diffraction studies have applied ex-situ analysis techniques and used these to make informed guesses as to what occurs during microwave heating.2–4 However, no matter how quickly a measurement is taken after heating has stopped, inevitably some cooling occurs.Furthermore, any microwave-specific field effects disappear virtually instantaneously (allowing for dielectric relaxation). Information gathered in this way is at best a distortion of the true nature of the system. It is far better to perform the measurements during heating in order to get a clear picture of what occurs. The use of neutrons as the diffracted radiation5 is one solution as they can penetrate many metals, but this is an expensive option. Here we present a more practical solution that is compatible with a commercially available, laboratory X-ray diffractometer.It is known that microwave irradiation can affect ionic diffusion properties6–8 and here we investigate such phenomena for the archetypal fast ion conductor, silver iodide. 2. In-situ diffraction Silver iodide has a significant ionic conductivity at room temperature. The thermodynamically stable phase at 300 K is the wurtzite-type, b-AgI, although the metastable c-AgI is found as the product of cubic stacking faults in the main b-phase.9 At elevated temperatures, the structure undergoes a phase transition (Tc ~ 420 K) to the bcc structure a-AgI phase with a statistical distribution of silver ions across multiple sites. The structures are shown in Fig. 1. The prototype microwave applicator was designed for in-situ powder X-ray diffraction, Fig.2. A thin (10 mm) aluminium foil window fulfils the requirements of being a conducting surface, yet allows the passage of up to 70% of the X-ray flux, permitting measurements to take place. Microwaves are transmitted down a coaxial cable and re-emitted within the applicator. The sample holder is a microwave-transparent ceramic. Temperature measurement is by a fluoroptic probe (Nortech Fibronic). The applicator internal diameter is 20 mm: well below the cut-off diameter for 2.45 GHz microwaves, which would normally result in significant power attenuation. However, the { This paper was originally presented as a poster at the Faraday Discussion 122 meeting. It is prepared here as a regular article, expanding on that presented previously. Fig. 1 Structures of silver iodide (Ag ~ grey, I ~ magenta).(a) Superionic, a-AgI: silver sites are 1/6 occupied. (b) Wurtzite-type, b-AgI. (c) Metastable, zinc-blend type, c-AgI. Click images or (a)–(c) to access 3D representations. DOI: 10.1039/b207258k PhysChemComm, 2002, 5(19), 135–137 135 This journal is # The Royal Society of Chemistry 2002introduction of a conducting antenna makes the entire cavity act as a coaxial waveguide with good transmission efficiency. A reasonably steady temperature rise of about one Kelvin per minute was achieved. In order to obtain good time and temperature resolution, a relatively narrow angle range (41u v 2h v 48u) was studied, the Bragg peaks in this region having been unambiguously identified. There are three intense diffraction peaks in the investigated region of the diffraction pattern.Below Tc the b-AgI phase has reflections at 42.5u (103) and 46.2u (112) in the range studied. Above Tc there is only one peak in this region, belonging to the a-AgI phase, at 43.5u (211). Diffraction data were recorded between 300 K and 450 K with a step size of approximately 10 K. Heating was performed, first with microwaves (ASTeX 2.45 GHz 0–1 kW variable power source) and then repeated with convection heating (Steinel 300–1000 K variable temperature heat gun) for comparison. The resulting series of diffraction patterns are plotted as contour maps, Fig. 3. The phase transition temperatures, Tc, were calculated to be 380 ¡ 10 K for microwave heating and 412 ¡ 2 K for convection heating. The phase change under microwave irradiation is therefore depressed by more than 30 K.3. Discussion Thought must be given to whether the temperature being reported is representative of the sample as a whole. This is a difficult question, particularly as regarding a sample with high surface area, such as a powder. Studies have shown that microwave field intensity is concentrated at surfaces, defects, and other micro-structural features.10,11 It is therefore difficult to predict details of the field pattern within the sample. Moreover, the sample sits near the end of the applicator, next to the metal end-choke, where—by definition—the tangential component of the electric field, Etan ~ 0. Field intensity will therefore drop sharply as it approaches the end of the applicator. However, it is thought that this will not be a significant effect across the width of the sample (small relative to the radiation wavelength). Thus, the cause of the change in transition temperature must be a real physical phenomenon.In order to interpret the result it is necessary to consider exactly how microwave heating works. Its nature is very different to conventional modes of heating. Normally heat is provided through convection, conduction or radiative processes. The energy is imparted as a broad spectrum, chiefly in the infrared region, and a band of vibrational modes of a solid are excited in amounts specified by the normal distribution. With microwave radiation energy is supplied with a specific frequency. The frequency does not correspond to a particular resonant phonon mode of a solid, these having typical frequencies in the terrahertz region, but interacts via a multiphonon process8 to activate a small number of phonon modes, e.g.Vfield ~ vphonon1 2 vphonon2 Energy is redistributed through all phonons via anharmonicity Fig. 2 (a) Applicator attached to diffractometer, ready for use. (b) Schematic of the applicator. Fig. 3 (a)Diffraction peak evolutionwithmicrowave heating, Tc~380¡10K. (b)Diffraction peak evolutionwith convection heating, Tc~412¡2K. 136 PhysChemComm, 2002, 5(19), 135–137in the crystal and phonon–phonon coupling. The rate of redistribution depends on the phonon density of states and the degree of coupling between phonons. The phonon density of states has been calculated by Bu� hrer and Bru� esch for the wurtzite-type, b-phase of silver iodide, and has been shown to be in good agreement with experimental results.12 Their calculated phonon dispersion curves show that the phonon spectrum is dominated by strong anharmonic effects,12 and that there are several low-lying transverse optic (TO) modes which behave as largely decoupled from the remainder of the phonon spectrum.Itthese modes that give the major contribution to the thermal motion because they have a high density of states and are easily thermally activated due to the low energy of around 500 MHz. In terms of microwave heating, this low-lying band of closely related TO modes should allow facile multi-phonon interactions with the microwave field. Furthermore, owing to the large separation between TO and higher frequency bands it is a more difficult process to redistribute this energy quickly.This may result in a non-classical distribution of energies within the phonon spectrum. Another important point is that the silver ion performs most of the movement in these low-energy modes. Though lighter, the silver atom is smaller and it is this movement which leads to the formation of Frenkel defects in the structure and allows ionic conduction of the silver ions to take place.13 The nature of the b–a phase change in silver iodide has been studied in some depth.9,12–16 Bu� hrer and Bru� esch12 postulate that the phase change is driven by entropy (an order–disorder transition). However, there is an energy cost for the creation of Frenkel defects, effectively limiting the increase in entropy.Transformation into the a-AgI phase results in many more vacant sites and a statistical distribution of silver ions across them. However, Yoshiasa and co-workers propose that the aforementioned low-lying TO mode is responsible not only for large Ag ion movements, but also for small I ion displacements which lead directly into the phase transformation (displacive transition).14 As was previously stated, the silver iodide system is highly anharmonic. Anharmonicities in atom vibrations increase as the temperature approaches the phase transition. The low energy TO modes are primarily bending modes, with apical Ag–I bond angles changing significantly. This movement is accompanied by small changes in the apical Ag–I bond length, see Fig. 4.At temperatures significantly lower than Tc, the apical Ag–I bond is observed to be longer than the ideal structure would suggest, owing to a large covalent component in the bonding. This bond length becomes equal to the basal bond lengths on heating and as the phase transition is approached the apical Ag–I distance becomes shorter whilst the three basal Ag–I distances become slightly larger. This structure change can be explained by anharmonicity in the atomic thermal vibrations, especially for the iodine atom, as demonstrated by Yoshiasa.14 As the anharmonic effect increases, it is observed that the magnitudes of iodine displacement are greater in the three antibonding directions. This makes the structure closer to the bcc arrangement of a-AgI. Hence, the increase in anharmonic thermal motion is consistent with a further transformation to the superionic a-phase.The hypothesis is that microwave energy interacts directly with low-lying TO modes via a multi-phonon process. Redistribution of this energy is slow with respect to the rate at which energy goes into the system from the microwave field, and a non-classical distribution of internal energies is the result. The phase transition occurs when enough energy is in the relevant mode to promote the structural change (be that through the creation of sufficient interstitial Ag ions, or through displacement of I ions). However, the average internal energy in the system is lower than at the equivalent point during conventional heating. Hence, the temperature measured is lower.Further work is necessary in order to test this theory. Individual atomic displacement parameters (ADPs) would be particularly useful. This level of accuracy, however, is beyond the scope of this technique. 4. Conclusion Temperature resolved, in-situ powder X-ray diffraction has been used to study the b to a phase change in the ionic conductor silver iodide under the influence of microwave radiation. Multi-phonon interactions with the microwave field have been used to provide a consistent theory to explain why the phase change occurs at a much lower temperature (~380 K) when heated with microwaves rather than heated conventionally (~412 K). Acknowledgements The authors would like to thank the skills of the mechanical workshop of the school of chemistry at the University of Edinburgh. References 1 R.Meredith, Engineers’ Handbook of Industrial Microwave Heating, The Institution of Electrical Engineers, London, 1998. 2 I. N. Lin, W. C. Lee, K. S. Liu, H. F. Cheng and M. W. Wu, J. Eur. Ceram. Soc., 2001, 10–11, 2085. 3 G. Wang, A. G. Whittaker, A. Harrison and L. Song, Mater. Res. Bull., 1998, 33(11), 1571. 4 S. S. Park, K. S. Jung, B. W. Kim, S. E. Lee and H. C. Park, Glass Technol., 2002, 43(2), 70. 5 A. Harrison, R. Ibberson, G. R. Robb, A. G. Whittaker, C.Wilson and D. Youngson, Faraday Discuss., 2002, 122, in press. 6 K. I. Rybakov, V. E. Semenov, S. A. Freeman, J. H. Booske and R. F. Cooper, Phys. Rev. B: Condens. Matter Mater. Phys., 1997, 55(6), 3559. 7 S. A. Freeman, J. H. Booske and R. F. Cooper, J. Appl. Phys., 1998, 83(11), 5761. 8 Y. V. Bykov, K. I. Rybakov and V. E. Semenov, J. Phys. D: Appl. Phys., 2001, 34, R55. 9 J. E. Maskasky, Phys. Rev. B: Condens. Matter Mater. Phys., 1991, 43(7), 5769. 10 B. S. Meng, B. D. B. Klein, J. H. Booske and R. F. Cooper, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 53(19), 12777. 11 A. Birnboim, J. P. Calame and Y. Carmel, J. Appl. Phys., 1999, 85(1), 478. 12 W. Bu� hrer, R. M. Nicklow and P. Bru� esch, Phys. Rev. B: Condens. Matter Mater. Phys., 1978, 17(8), 3362. 13 R. J. Cava and E. A. Rietman, Phys. Rev. B: Condens. Matter Mater. Phys., 1984, 30(12), 6896. 14 A. Yoshiasa, K. Koto, F. Kanamaru, S. Emura and H. Horiuchi, Acta Crystallogr., Sect. B: Struct. Sci., 1987, 43, 434. 15 C. A. Rains and J. R. Ray, Phys. Rev. B: Condens. Matter Mater. Phys., 1991, 44(17), 9228. 16 J. L. Tallon, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 38(13), 9069. Fig. 4 Ag4I tetrahedron form wurtzite-type, b-AgI. PhysChemComm, 2002, 5(19), 135&ndash
ISSN:1460-2733
DOI:10.1039/b207258k
出版商:RSC
年代:2002
数据来源: RSC