Anion vacancy aggregates

Dynamic ETEM studies of the nucleation and growth of CS planes in reducing environments provide direct experimental proof of the motion of CS planes. They are consistent with the model of anion vacancies aggregating into discs... 

We thus ask What causes CS planes to nucleate (i.e. what are the reasons for anion vacancy aggregation and collapse in an oxide catalyst) and grow. We examine the response of defects in oxidizing atmospheres and, in particular, the role of anion vacancy concentrations in catalytic oxides. The EM results have led to novel concepts in oxidation catalysis (Gai 1981, 1992-1993, Gai et al 1982). 

Anion vacancy aggregate M "+ (m = 0,1) Two (or more) adjacent oxygen vacancies... 

Anion vacancy aggregates (Sect. 2.3.9). Under some conditions, aggregates of O vacancies can form in the bulk of MgO. The smallest aggregate is the M center, consisting of two adjacent O vacancies. 

Particle irradiation effects in halides and especially in alkali halides have been intensively studied. One reason is that salt mines can be used to store radioactive waste. Alkali halides in thermal equilibrium are Schottky-type disordered materials. Defects in NaCl which form under electron bombardment at low temperature are neutral anion vacancies (Vx) and a corresponding number of anion interstitials (Xf). Even at liquid nitrogen temperature, these primary radiation defects are still somewhat mobile. Thus, they can either recombine (Xf+Vx = Xx) or form clusters. First, clusters will form according to /i-Xf = X j. Also, Xf and Xf j may be trapped at impurities. Later, vacancies will cluster as well. If X is trapped by a vacancy pair [VA Vx] (which is, in other words, an empty site of a lattice molecule, i.e., the smallest possible pore ) we have the smallest possible halogen molecule bubble . Further clustering of these defects may lead to dislocation loops. In contrast, aggregates of only anion vacancies are equivalent to small metal colloid particles. 

It was Ziman [77] who has noted that there is little hope, at least at present, to develop an experimental technique permitting the direct measurement of these correlation functions. The only exception are the joint densities x / (r> ) information about which could be learned from the diffraction structural factors of inhomogeneous systems. On the other hand, optical spectroscopy allows estimation of concentrations of such aggregate defects in alkali halide crystals as Fn (n = 1,2,3,4) centres, i.e., n nearest anion vacancies trapped n electrons [80]. That is, we can find x mK m = 1 to 4, but at small r only. Along with the difficulties known in interpretating structure factors of binary equilibrium systems (gases or liquids), obvious specific complications arise for a system of recombining particles in condensed media which, in its turn, are characterized by their own structure factors. 

The temporal evolution of spatial correlations of both similar and dissimilar particles for d = 1 is shown in Fig. 6.15 (a) and (b) for both the symmetric, Da = Dft, and asymmetric, Da = 0 cases. What is striking, first of all, is rapid growth of the non-Poisson density fluctuations of similar particles e.g., for Dt/r = 104 the probability density to find a pair of close (r ro) A (or B) particles, XA(ro,t), by a factor of 7 exceeds that for a random distribution. This property could be used as a good aggregation criterion in the study of reactions between actual defects in solids, e.g., in ionic crystals, where concentrations of monomer, dimer and tetramer F centres (1 to 3 electrons trapped by anion vacancies which are 1 to 3nn, respectively) could be easily measured by means of the optical absorption [22], Namely in this manner non-Poissonian clustering of F centres was observed in KC1 crystals X-irradiated for a very long time at 4 K [23],... 

Shear Plane-Point Defect Equilibria.—The question of the existence of point defects in compounds where extended defects are known to occur has been controversial. Indeed, it has occasionally been claimed that point defects cannot form in such phases and that they will always be eliminated with the formation of extended structures. We reject these latter arguments as thermodynamically unsound. From a thermodynamic standpoint, the formation of extended defects can be viewed as a special mode of point defect aggregation as such, shear planes will be in equilibrium with point defects, with the position of the equilibrium depending on both temperature and the extent of the deviation from stoicheiometry. Thus, if we assume, as is suggested by our calculations, that anion vacancies are the predominant point defects in reduced rutile (a further point of controversy as mentioned above) then there will exist an equilibrium of the type... 

Hematite forms by a combination of aggregation-dehydration-rearrangement process for which the presence of water appears essential. Structural details about this process at 92 °C were obtained from EXAFS (Combes et al. 1989 1990) face-sharing between Fe octahedra developed before XRD showed any evidence for hematite. It is followed by internal redistribution of vacancies in the anion framework and by further dehydration. The dehydration process involves removal of a proton from an OH group and this in turn leads to elimination of a water molecule and formation of an 0X0 linkage. The local charge inbalance caused by proton loss is compensated for by migration and redistribution of Fe " within the cation sublattice. 

---