Jan W. Vandersande

High temperature electrical conductivity measurements of natural diamond and diamond films

Abstract The electrical conductivity of diamond films and diamond-like carbon films were measured from room temperature to 1200°C. These results are the first known measurements for diamond films in this temperature range. The measurements were compared with the conductivity of two natural type IIa diamonds measured with the same apparatus. The activation energy for the two natural type IIa diamonds was found to be 1.4 eV, in agreement with earlier measurements on type I diamonds. The room temperature conductivities of the two diamond films were found to be higher than that of natural IIa diamond, but were found to approach or even just fall below that of natural IIa diamond at elevated temperatures. This difference, and the difference in slope of the conductivities of the various samples, are indicative of the difference in type and number of impurity and structural defects accounting for conductivity in the different samples. These results offer promise that diamond films could be fabricated which have an electrical resistivity exceeding that of natural IIa diamond.

Effect of high temperature annealing on the thermoelectric properties of GaP doped SiGe

Silicon-germanium alloys doped with GaP are used for thermoelectric energy conversion in the temperature range 300-1000 C. The conversion efficiency depends on Z = S-squared/rho lambda, a materials parameter (the figure of merit), where S is the Seebeck coefficient, rho is the electrical resistivity and lambda is the thermal conductivity. The annealing of several samples in the temperature range of 1100-1300 C resulted in the power factor P (= S-squared/rho) increasing with increased annealing temperature. This increase in P was due to a decrease in rho which was not completely offset by a drop in S-squared suggesting that other changes besides that in the carrier concentration took place. SEM and EDX analysis of the samples indicated the formation of a Ga-P-Ge rich phase as a result of the annealing. It is speculated that this phase is associated with the improved properties. Several reasons which could account for the improvement in the power factor of annealed GaP doped SiGe are given.

Accurate determination of specific heat at high temperatures using the flash diffusivity method

The flash diffusivity method can be extended, very simply, to measuring simultaneously thermal diffusivity and specific heat and thus obtaining the thermal conductivity directly. This was accomplished by determining the amount of heat absorbed by a sample with a well-known specific heat and then using this to determine the specific heat of any other sample. The key to using this technique was to have identically reproducible surfaces on the standard and the unknowns. This was achieved earlier by sputtering the surfaces of the samples with a thin layer of graphite. However, the accuracy in determining the specific heat was within ±10% and there was considerable scatter in the data. Several improvements in the technique have been made which have improved the accuracy to ±3% and increased the precision. The most important of these changes has been the introduction of a method enabling the samples to be placed in exactly the same position in front of the light source. Also, the control of the thickness and the application of the graphite coating have turned out to be very important. A comparison of specific heats obtained with this improved technique and with results obtained using other techniques has been made for two materials.

Determination of the thermal diffusivity and specific heat using an exponential heat pulse, including heat-loss effects

The one-dimensional heat diffusion equation has been solved analytically for the case of a heat pulse of the form F(t) = exp(−t/τ)/τ applied to the front face of a homogeneous body including the effects of heat loss from the front and back faces. Approximate expressions are presented which yield a simple, accurate technique for the determination of the thermal diffusivity and specific heat, suitable to a wide range of heat-pulse time constant and heat-loss parameters, without recourse to graphical techniques or requiring further computer analysis. A procedure is described for the determination of an effective time constant to allow application of the present results to the case of a nonexponential heat pulse. Experimental results supporting the theoretical analysis are presented for five samples of silicon germanium alloys of various thicknesses, determined using a xenon flash tube heat-pulse exhibiting an exponential dependence. Proper consideration of the experimental heat pulse shape is shown to lead to reliable corrections to the apparent thermal diffusivity, even for relatively long heat-pulse times.

Using high-temperature electrical resistivity measurements to determine the quality of diamond films

Abstract The electrical resistivity of undoped diamond films has been measured between room temperature and 1200 °C. The films were grown by either microwave plasma CVD or combustion flame at various different companies. It was found that the room temperature resistivities were all around 1015–1016 Ω cm, which has been shown to be the apparatus-limited value (higher resistivities cannot be measured). Hence these resistivity measurements cannot indicate which of the films, which all have very similar Raman spectra, are of the best quality. Also, the sample treatment (such as as-fabricated, heat treated, cleaned) will affect the room temperature electrical resistivity because of different surface conditions. On the other hand, high-temperature measurements up to 1200 °C clearly do show differences for samples that have the same electrical resistivity at room temperature. The high-temperature resistivities varied from about one order of magnitude lower than that for natural type IIa diamond to about two orders of magnitude greater over the whole temperature range, with activation energies between 1.5 and 1.6 eV. These high-temperature measurements are thus very helpful in determining the quality of undoped diamond films.

Thermoelectric Devices and Diamond Films for Temperature Control of High Density Electronic Circuits

The increased speeds of integrated circuits is accompanied by increased power levels and the need to package the IC chips very close together. Combined, these spell very high power densities and severe thermal problems at the package level. Conventional packaging materials have difficulty dealing with these thermal management problems. However, it is possible to combine both active and passive cooling by using thin film bismuth‐telluride thermoelectric coolers (microcoolers) and diamond substrates for the temperature control of these high density electronic circuits. The highest power components would be mounted directly onto thin film thermoelectric elements, which would maintain the temperature of these components from a few degrees to tens of degrees below that of the diamond substrate. This allows these components to operate within their required temperature range, effectively manage temperature spikes and junction temperatures, and increase clockspeed. To optimize the design of the thermoelectric coo...

Thermal conductivity of La3−xTe4☆

Thermal conductivities of a series of compositions of LaTey 1.33 < y < 1.6 were measured by a flash-diffusivity method. The results have been correlated with composition and with microstructure. They have also been compared with results previously reported on the sulfide system and with telluride conductivities reported by earlier workers. Careful electrical and morphological sample characterization is crucial for meaningful interpretation of the thermal conductivity results.

Thermoelectric properties of hot‐pressed fine particulate powder SiGe alloys

Several research groups have tried to reduce the thermal conductivity of thermoelectric materials in order to improve the thermoelectric’s figure‐of‐merit and conversion efficiency (Pisharody and Garvey 1978). Some of these efforts have successfully reduced thermal conductivity, but also have decreased the electrical properties of the thermoelectric. Hence there has been not net gain in figure‐of‐merit. During the past, year of novel material fabrication technique has been applied to the production of silicon germanium thermoelectric decreased the electrical properties of the thermoelectric. Hence there has been material. Ultra‐fine particulates (50 A to 100 A) have been hot pressed into boron doped, p‐type, 80/20 silicon germanium. The initial results have been promising. When compared to standard silicon germanium, a reduction in thermal conductivities of up to 40% and an increase in figure‐of‐merit of 10% to 15% has been achieved.

Improved n‐type SiGe/GaP thermoelectric materials

Experimental and theoretical work has been conducted at the Jet Propulsion Laboratory in order to improve the thermoelectric properties of n‐type SiGe materials. Particular emphasis has been placed upon the understanding of the differences in dopant solid solubilities when multidoping with both Ga and P instead of P alone as in standard SiGe alloys. A set of various experimental techniques for obtaining heavily‐doped hot‐pressed samples coupled with thermodynamic theoretical considerations was used to relate microstructure and composition to electrical and thermal transport properties. A straightforward application of this work resulted in substantial improvements of the thermoelectric figure of merit on several SiGe/GaP, with average Z values close to 1×10−3 K−1 over the 600–1000°C temperature range. New experiments show that values 10% higher are even possible as the thermal conductivity can be decreased while conserving a high power factor.

Thermal Conductivity Reductions in Sige via Addition of Nanophase Particles

Transport models have predicted that the thermal conductivity of SiGe alloys could be appreciably reduced by incorporating a discrete 40A particles with the SiGe grains. Such a thermal conductivity reduction would lead to substantial improvements in the figure-of-merit of thermoelectric materials. This paper reports on recent results on adding 40A particles to SiGe via a spark erosion process. Thermal conductivity reductions consistent with the transport models have been achieved, however, the improvement in figure-of-merit has not been as large as predicted.