The development of growing markets for thermoelectric devices strongly depends upon improving the performance of the Peltier effect alloys. Breaking out of the specialty niches requires doubling present figures of merit, Z, of commercial alloys though each incremental gain potentially opens additional specialty niches.

The alloys are polycomponent, heavily doped, N and P type semiconductors. Optimization to the highest Z's requires controlling bulk phase interactions of phase diagram, compositional, crystal growth, and processing variables influencing the imperfection structures impacting on alloy quality. From the first, the system complexity reeds recognition so the performance optimizing variables become clearly identified. This is crucial to commercial production.

The (BiSb)2 (TeSe)3 provide both N and P type alloy model systems useful in understanding 1he general challenge of performance optimization. Illustrations of the imperfection structural chemical limitations on attainable performance by varied technologies are given.

In seeking other superior performance alloys the experimental designs for exploratory research need immediately to address the identification of the dominating imperfection variables in order to recognize quickly the potentials present.