Search

Skutterudites have proven to be a useful thermoelectric system as a result of their enhanced figure of merit (ZT1), cheap material cost, favorable mechanical properties, and good thermal stability. The majority of skutterudite interest in recent years has been focused on binary skutterudites like CoSb3. Binary skutterudites are often double and triple filled, with a range of elements from the lanthanide series, in order to reduce the lattice component of thermal conductivity. Ternary and quaternary skutterudites, such as Co4Ge6Se6 or Ni4Sb8Sn4, provide additional paths to tune the electronic structure. The thermal conductivity can further be improved in these complex skutterudites by the introduction of fillers. The Nd (sub z) Fe (sub x) Co (sub 4-x) Sb (sub 12-y)Ge (sub y) system has been investigated as a p-type thermoelectric material, and is stable up to 600 degrees Centigrade. The influence of Fe and Ge content, along with filler Nd, was investigated on thermoelectric transport properties. In addition to the chemical influence on properties, some processing details of the system will also be addressed.

Skutterudites have proven to be a useful thermoelectric system as a result of their enhanced figure of merit (ZT1), cheap material cost, favorable mechanical properties, and good thermal stability. The majority of skutterudite interest in recent years has been focused on binary skutterudites like CoSb3 or CoAs3. Binary skutterudites are often double and triple filled, with a range of elements from the lanthanide series, in order to reduce the lattice component of thermal conductivity. Ternary and quaternary skutterudites, such as Co4Ge6Se6 or Ni4Sb8Sn4, provide additional paths to tune the electronic structure. The thermal conductivity can further be improved in these complex skutterudites by the introduction of fillers. The Co (sub X) Ni (sub 4-x) Sb (sub 12-y) Sn (sub Y) system has been investigated as both a p- and n-type thermoelectric material, and is stable up to 200 degrees Centigrade. Yb, Ce, and Dy fillers have been introduced into the skutterudite to study the influence of both the type and the quantity of fillers on processing conditions and thermoelectric properties. The system was processed through a multi-step technique that includes solidification, mechano-chemical alloying, and hot pressing which will be discussed along with thermoelectric transport properties.

Thermoelectric materials based on the skutterudite crystal structure have demonstrated enhanced performance (ZT greater than 1), along with good thermal stability and favorable mechanical properties. Binary skutterudites, with single and multiple fillers, have been intensively studied in recent years. Compared to binary skutterudites, the ternary systems have received less attention, e.g. Ni4Sb8Sn4. Ternary skutterudites are isoelectronic variants of binary skutterudites; cation substitutions appear to be isostructural to their binary analogues. In general, ternary skutterudites exhibit lower thermal conductivity. Ternary systems of Ni4Bi8Ge4, Ni4Sb8Ge4, and Ni4Sb8Sn4 were investigated using combined solidification and sintering steps. Skutterudite formation was not achieved in the Ni4Bi8Ge4 and Ni4Sb8Ge4 systems; skutterudite formation occurred in Ni4Sb8Sn4 system. P-type material was achieved by Co substitution for Ni. Thermoelectric properties were measured from 298 K to 673 K for Ni4Sb8Sn4, Ni4 Sb7Sn5 and Co2Ni2Sb7Sn5. N-type Ni4Sb8Sn4 exhibit the highest figure of merit of 0.1 at 523 K.

Skutterudites have proven to be a useful thermoelectric system as a result of their high figure of merit, favorable mechanical properties, and good thermal stability. Binary skutterudites have received the majority of interest in recent years, as a result of successful double and triple filling schemes. Ternary skutterudites, such as Ni4Sb7Sn5, also demonstrate good thermoelectric performance, with high power factor and low thermal conductivity. Ternary skutterudites, as contrasted to binary systems, provide more possibility for tuning electronic structure as substitutions can be studied on three elements. The Co(x)Ni(4-x)Sb(12-y)Sn(y) system has been investigated as both a p- and n-type thermoelectric material, stable up to 200 C. The system is processed through a combination of solidification, mechanical alloying, and hot pressing steps. Rietveld structure refinement has revealed an interesting occupancy of Sn on both the 24g Wyckoff position with Sb as well as the 2a position as a rattler. In addition to thermoelectric properties, detailed processing routes have been investigated on the system.

A set of convenient thermoelectric device solutions have been derived in order to capture a number of factors which are previously only resolved with numerical techniques. The concise conversion efficiency equations derived from governing equations provide intuitive and straight-forward design guidelines. These guidelines allow for better device design without requiring detailed numerical modeling. The analytical modeling accounts for factors such as i) variable temperature boundary conditions, ii) lateral heat transfer, iii) temperature variable material properties, and iv) transient operation. New dimensionless parameters, similar to the figure of merit, are introduced including the device design factor, fin factor, thermal diffusivity factor, and inductance factor. These new device factors allow for the straight-forward description of phenomenon generally only captured with numerical work otherwise. As an example a device design factor of 0.38, which accounts for thermal resistance of the hot and cold shoes, can be used to calculate a conversion efficiency of 2.28 while the ideal conversion efficiency based on figure of merit alone would be 6.15. Likewise an ideal couple with efficiency of 6.15 will be reduced to 5.33 when lateral heat is accounted for with a fin factor of 1.0.

The measurement of Seebeck coefficient and electrical resistivity are critical to the investigation of all thermoelectric systems. Therefore, it stands that the measurement uncertainty must be well understood to report ZT values which are accurate and trustworthy. A detailed uncertainty analysis of the ZEM-3 measurement system has been performed. The uncertainty analysis calculates error in the electrical resistivity measurement as a result of sample geometry tolerance, probe geometry tolerance, statistical error, and multi-meter uncertainty. The uncertainty on Seebeck coefficient includes probe wire correction factors, statistical error, multi-meter uncertainty, and most importantly the cold-finger effect. The cold-finger effect plagues all potentiometric (four-probe) Seebeck measurement systems, as heat parasitically transfers through thermocouple probes. The effect leads to an asymmetric over-estimation of the Seebeck coefficient. A thermal finite element analysis allows for quantification of the phenomenon, and provides an estimate on the uncertainty of the Seebeck coefficient. The thermoelectric power factor has been found to have an uncertainty of +9-14 at high temperature and 9 near room temperature.

In order to provide a complete description of a materials thermoelectric power factor, in addition to the measured nominal value, an uncertainty interval is required. The uncertainty may contain sources of measurement error including systematic bias error and precision error of a statistical nature. The work focuses specifically on the popular ZEM-3 (Ulvac Technologies) measurement system, but the methods apply to any measurement system. The analysis accounts for sources of systematic error including sample preparation tolerance, measurement probe placement, thermocouple cold-finger effect, and measurement parameters; in addition to including uncertainty of a statistical nature. Complete uncertainty analysis of a measurement system allows for more reliable comparison of measurement data between laboratories.

In recent years there has been a renewed interest for high temperature piezoelectrics for both terrestrial and aerospace applications. These applications are limited in part by the operating temperature, which is usually taken as one half of the Curie temperature (Tc), and is 200C for one of the most widely used commercial piezoelectrics, Pb(Zr,Ti)O3 (PZT). In an effort to increase Tc, subsequent research into high temperature Bi(BB)O3 PbTiO3 piezoelectrics led to the discovery of the morphotropic phase boundary (MPB) in the high-Tc BiScO3 PbTiO3 (BS-PT) system with a Tc of 460C and a d33 of 460 pmV. The Tc marks the ferroelectric to paraelectric phase transformation and while, in general, a phase transformation leads to thermal depoling in piezoelectrics with low or moderate Tcs, for high Tc piezoelectrics thermally assisted dipole rotation can lead to randomization of domains at temperatures below Tc. It becomes necessary to determine the depoling temperature (Td) which dictates the actual working temperature range. By doping for Sc and Ti the Td can be shifted while maintaining similar electromechanical properties as a function of temperature. The effect of this B-site doping on depoling temperature has been explored through the characterization of microstructure and weakhigh field measurements.

A variety of piezoelectric applications have been driving the research in development of new high temperature ferroelectrics; ranging from broader markets such as fuel and gas modulation and deep well oil drilling to very specific applications such as thermoacoustic engines and ultrasonic drilling on the surface of Venus. The focus has been mostly on increasing the Curie temperature. However, greater challenges for high temperature ferroelectrics limit the operating temperature to levels much below the Curie temperature. These include enhanced loss tangent and dc conductivity at high fields as well as depoling due to thermally activated domain rotation. The initial work by Eitel et al. [Jpn. J. Appl. Phys., 40 [10, Part 1] 59996002 (2001)] increased interest in investigation of Bismuth containing perovskites in solid solution with lead titanate. Issues that arise vary from solubility limits to increased tetragonality; the former one prohibits processing of morphotropic phase boundary, while the latter one impedes thorough poling of the polycrystalline ceramics. This talk will summarize recent advances in development of high temperature piezoelectrics and provide information about challenges encountered as well as the approaches taken to improve the high temperature behavior of ferroelectrics with a focus on applications that employ the converse piezoelectric effect.

Traditional silicon germanium high temperature thermoelectrics have potential for improvements in figure of merit via nano-structuring with a silicide phase. A second phase of nano-sized silicides can theoretically reduce the lattice component of thermal conductivity without significantly reducing the electrical conductivity. However, experimentally achieving such improvements in line with the theory is complicated by factors such as control of silicide size during sintering, dopant segregation, matrix homogeneity, and sintering kinetics. Samples are prepared using powder metallurgy techniques; including mechanochemical alloying via ball milling and spark plasma sintering for densification. In addition to microstructural development, thermal stability of thermoelectric transport properties are reported, as well as couple and device level characterization.

Two new techniques have been developed to characterize Pb(Mg1/3Nb2/3)O3-PbTiO3 (PMN-PT) based ferroelectric single crystals: (i) electro-thermal imaging, and (ii) single crystal x-ray diffraction in the transmission mode. (i) Electro-thermal imaging is a remote sensing technique that can detect the polarization direction and poling state of a whole crystal slice. This imaging technique utilizes an IR camera to determine the field induced temperature change and does not require any special or destructive sample preparation. In the resulting images it is possible to distinguish regions of 180 deg domains. This powerful technique can be used remotely during poling to determine the poling state of the crystal to avoid over-poling that can result in inferior properties and/or cracking of the crystals. Electro-thermal imaging produced the first direct observations of polarization rotation. Under bipolar field, the domains near the corners were the first to switch direction. As the field increased above the coercive field, domains at the center part of the crystals switched direction. (ii) X-ray diffraction in the transmission mode has long been used in structure determination of organic crystals and proteins; however, it is not used much to characterize inorganic systems. 0.7Pb(Mg1/3Nb2/3)O3-0.3PbTiO3 single crystals were examined by this XRD technique for the first time, and a never-before-seen super-lattice was revealed with a doubling of the unit cell in all three directions, giving a cell volume eight times that of a traditional perovskite unit cell. The significance of the super-lattice peaks increased with poling, indicating a structural contribution to ordering. Lack of such observations by electron diffraction in the transmission electron microscope examinations suggests the presence of a bulk effect.