Decomposition peritectic

Well-known examples of peritectic behavior occur in the binary Na/K system (from decomposition of Na2K) and the Na2S04/H20 system (from decomposition of the decahydrate). 

Preparation of a peritectic compound requires a solid state (solid-solid) or solid-liquid reaetion. A solid state reaction requires transport of matter (diffusion) in the direetion of a chemical potential gradient (i.e., a chemical composition gradient). Thus solid state reactions are diffusion controlled, with diffusion fastest across grain boundaries. Within a crystal, diffusion is enhanced by defects. Clearly the preparation of a peritectic alloy should be carried out at the highest possible temperature just below the peritectic decomposition temperature. The speciality of powder metallurgy exploits the optimum conditions for solid state reactions. 

Initial crystal growth experiments using the more common low temperature fluxes such as B203, KF and PbO were unsuccessful. Subsequent DTA measurements suggested that YBC decomposes peritectically at about 1020°C. To overcome this difficulty, we searched for a liquidus field for crystal growth below the decomposition temperature. We concentrated our efforts on the pseudoternary YBC-BaCu02-CuO system. 

The vertical section of the Nd-Fe-B ternary system which passes through the Fe comer and the phase Nd2Fe14B is shown in fig. 4a. Schneider et al. emphasize that this is not a pseudo-binary section. The dominant feature of this vertical section is the peritectic reaction L + Fe —> at 1180° C. It also follows from the results shown in fig. 4a that cooling of a liquid whose composition corresponds to leads to the formation of primary crystallized Fe. The concentration limit beyond which no primary Fe crystals are formed is at 77 at.% Fe. This is very close to the overal composition of commercial magnets, as will be discussed in more detail in section 3.2. Schneider et al. note that the vertical section of fig. 4a represents the stable situation which applies only to melts that were kept near the liquidus temperatures for a sufficiently long time. For superheated alloys the vertical section is quite different and corresponds to a metastable situation (fig. 4b). A comparison of the two vertical sections reveals that the liquidus temperatures, and the temperature at which the univariant reaction L - + tj begins, are unaltered, but that the temperature at which the 4> phase forms is lower in fig. 4b than in fig. 4a. Furthermore, one notices a new phase in fig. 4b (x) which is formed peritectically at 1130 °C. The latter temperature is below the temperature of the stable reaction L + Fe - 4> (1180 °C). Schneider et al. note that the primary crystallization of is suppressed in the metastable sequence (fig. 4b), in favour of Fe. In the microstructure one now observes that primary Fe is surrounded by a shell of Fe + which is the decomposition product of x- The x phase was identified by Grieb et al. (1987), as a compound of the 2 17 structure type. 

LBO also undergoes a peritectic decomposition (834 °C) and must be grown by a flux method. The original phase diagram for the system Li2 0-B203 indicated that LBO decomposed into Li2B407 and Li2B80i3 below 595 °C, but this was subsequently disproved. A stable melt can be... 

The liquidus curve and the invariant temperatures are known to good accuracy. This result will be an aid for understanding the thermodynamic behaviour of the ternary and quaternary phases. It has been possible to calculate the temperature of the metastable melting point of CoSbj (1283K instead of 1210K for the peritectic decomposition of the 1 2 phase). One can observ e that these temperatures are not so far and it could explain the problems occuring in the fabrication of the materials as well as the lack of 1 3 compounds for iron and nickel systems. 

Transforms into metallic yS-Sn slightly below RT Peritectic decomposition temperature at p = 35 bar... 

---