Zincblende phase

STRUCTURE. CdS Can exist in three different crystal structures hexagonal (wurtzite), cubic (zincblende)— both tetrahedrally coordinated and cubic (rock-salt), which is sixfold coordinated. Except in a few cases, the rocksalt modification of CdS has been observed only at very high pressures CD films of this phase have never been reported. The other two phases have been reported to occur in CD films under various conditions. The wurtzite phase is thermodynamically slightly more stable, and invariably forms if the zincblende phase is heated above 300-400°C. The low-temperature CD method therefore can allow the formation of the zincblende phase, and this phase is commonly obtained in CD CdS films. Very often, a mixture of wurtzite and zincblende phases has been reported in the literature. There are many variables that affect the crystal structure, including the nature of the complex, the substrate, and sometimes even stirring. 

In the original study by Padam and Gupta [112], the deposition solution contained triethanolamine/ammonia-complexed CdAci and the Te source was TeOi with hydrazine hydrate as a reducing agent. The nature of the TeOi solution was not clear, since TeOi is only slightly soluble in water it may have been dissolved in a hydroxide solution, in which it is much more soluble. The deposition was carried out at 90°C. Electron diffraction of the as-deposited films showed both zincblende and wurtzite phases of CdTe (the zincblende phase is the more stable and commonly encountered one). 

No XRD pattern of the films was found, but the powder precipitated in the solution was zincblende phase with very broad peaks (>3 mn coherence length). In contrast to other measurements, where the Zn Se ratio was greater than 1, the... 

The cubic zincblende phase of GaN (lattice parameter of about 4.5 A) is a metastable one and observed only for heteroepitaxial layers on highly mismatched cubic substrates (OOl)-oriented, for example GaAs [11], Si [12], MgO [13] and p-SiC [14],... 

In the case of highly mismatched substrates, such as hexagonal sapphire, cubic GaAs(lll) and Si(lll), there is usually a certain amount of zincblende phase of the nitrides [12,13], separated by crystallographic defects from the wurtzite phase. 

Again, many calculations were undertaken to understand the electronic and structural properties of GaN, both in the wurtzite and the cubic zincblende phase. The results concerning the mechanical properties are summarised in TABLE 6. 

Complete conversion of the ZnO columns to ZnS columns has mostiy been carried out in sulphur vapour. X-ray diffraction shows that the newly formed ZnS is in the zincblende phase, which is the stable phase at room temperature. From a detailed comparison of micrographs it was conclnded that the ZnS column volume is measurably larger than that of the ZnO columns. This change in size was attributed to a different size and geometry of the ZnO and ZnS unit cells (Dloczik et al, 2003). 

The best fit to experimental XRD pattern was obtained in the case of two phase B3 and R4 mixture in the CdS nanopowder (Fig. 1). The main sharp peak at 27° which is characteristic for the experimental pattern appears in the simulation if the ratio of 65 at. % of wurtzite phase and 35 at. % of zincblende phase is fulfilled. From the one hand, such ratio of CdS phases could be hardly understood, but from the other hand, the fit for the disordered structure demonstrates the better result (Fig. 1). These circumstances tend to accept the disordered model. 

In addition to hexagonal (wurtzite) and cubic (zincblende) phase. 

Buckley used the same technique to deposit films for photovoltaic cells [113], only with CdCli as Cd source and apparently a lower Cd concentration. Electron diffraction of these films showed a predominantly zincblende structure with some wurtzite phase. The films were p-type with a resistivity of 20 5 fl-cm. These values, and the subsequent photovoltaic cells, apparently refer to as-deposited films no reference to annealing was made in this stndy. 

TABLE 1 Lattice parameters of III-N compounds (hexagonal-wurtzite and cubic-zincblende structures). For GaN bulk crystals, the errors indicate variations between various samples, as the measurement accuracy was of about 5 parts per million. For cubic AIN and InN, the given lattice parameters are estimated from bond-lengths of the wurtzite phase. For all epitaxial layers, the given values are relaxed lattice parameters calculated from the measured ones using EQN (1),... 

EPR measurements were first performed on wurtzite GaN in 1993 by Carlos and co-workers [2-4] and on cubic GaN by Fanciulli and co-workers at about the same time [5], The primary resonance in the wurtzite films is slightly anisotropic (gy = 1.9510 and gi = 1.9483) with a width 0.5 mT at 4.2 K and generally acknowledged to be due to a band of delocalised effective mass (EM) donor electrons. The average g value is consistent with the expectations of a 5-band k.p analysis and is also similar to that obtained by Fanciulli [5] for a much broader line (—10 mT) in their conduction electron spin resonance experiments on zincblende films. With this exception all of the work discussed in this Datareview is on the wurtzite phase. 

FIGURE 3 HREM image taken along the [1120] zone axis of GaN grown (by MBE) on c-plane sapphire. The zincblende (zb) GaN phase is found near the interface within a wurtzite GaN matrix. (From [7].)... 

Minor (<5%) quantities of cubic phase (a0 = 6.24 A, zincblende sphalerite structure, space group F 43m) may sometimes be present. Further annealing of the sample will cause transformation of all the material into the hexagonal phase. 

---