Summary

As described in the previous discussion, despite the advances in quasicrystals for thermoelectrics, there is still a long road ahead if quasicrystals are to be viable for thermoelectrics. It is encouraging that theoretical predictions by Macia indicate that high values of ZT may be possible in this class of materials (Macia, 2000). Cyrot-Lackmann is also investigating quasicrystals as possible thermoelectric materials and has a patent on quasicrystals for this application (Cyrot-Lackmann, 1999b). A systematic approach in relation to doping, composition, processing, and other factors along with subsequent measurement of the transport properties will be necessary. Certainly this data coupled with ideas on how to further enhance the thermopower could greatly advance our knowledge of these materials. Quasicrystals closely match the concept that a good thermoelectric should behave as a glasslike material in relation to phonons and a metal in relation to electronic transport. AlPdMn quasicrystals have thermal conductivity that closely resembles that of an amorphous solid. The “tunability” in the electrical conductivity and thermopower allows for many compositional, impurity, damage, and additional element studies to be performed in an effort to better understand the transport in this system and optimize the figure of merit for potential thermoelectric application. Obviously, more work is required to fully address the feasibility of these materials for thermoelectric applications. However, along the way much information related to the interplay of many of the parameters to the electrical and thermal transport in these systems will be gained. Thus, a more fundamental understanding of the electrical and thermal transport mechanisms related to the quasicrystalline materials may become evident. It is the strong belief of one of us (TMT) that a new higher performance thermoelectric material will be found and it will truly change the world around us. Where will it be? Will it be in a quasicrystal? The current data tends to indicate that it probably will not be in these materials, since the hurdle of enhancing the thermopower by a factor of 4 seems too great; yet theoretically high values (ZT>1) have been predicted to be possible. This is especially difficult without the ability to systematically dope and “tune a bandgap” as in the semiconductor thermoelements, which were described by Ioffe nearly 50 years ago. Quasicrystals are truly a fascinating class of materials, and whether or not subsequent research efforts and time determine their feasibility (or even impracticality) for thermoelectrics, they will still retain their most unusual properties, about which we have much to learn. Most likely, there are a few surprises left in these materials (or similar classes of intermetallic materials), and probably even more applications of quasicrystals will become evident over the next few years.