Development of certain photothermal wave techniques and investigations on the thermal properties of Al20 ASx Te80-x glasses

Abstract

A novel photoacoustic methodology has been developed using rear-surface illumination of samples in the thermally thick regime. The method relies on obtaining the phase data of the normalized signal to measure the thermal diffusivity. This is then used in the signal amplitude to calculate the thermal conductivity. Using the relation between diffusivity, conductivity, and density, the specific heat is calculated. Two important criteria for accurate determination of the thermal conductivity and subsequently the specific heat are:

The normalization procedure should be performed under the same conditions as the experiment on the sample. The frequency range should be chosen such that the sample is thermally thick to the extent that ? << 1.

The instrumentation required for conducting the experiments was developed in the laboratory. The agreement between the measured values of thermal diffusivity, thermal conductivity, and specific heat of the representative samples and the values quoted in literature is encouraging and proves the validity of the methodology developed. The one problem faced in the experiments was that signal-to-noise ratios (SNRs) deteriorated rapidly at higher frequencies, as is expected of thermally thick rear-surface illuminated samples. A photopyroelectric (PPE) instrument was developed to measure the thermal diffusivity (?) of bulk samples. Initially, experiments were conducted to measure ? by determining the slope of the graph of the logarithm of the PPE signal magnitude versus the chopping frequency in the thermally thick regime. However, the most reproducible and accurate measurements were made by finding the frequency at which the sample transitioned from thermally thick to thermally thin as the frequency was decreased. At this frequency, a sharp discontinuity in the slope was observed. This 揷ritical frequency� or 揷rossover frequency� method was used to measure, for the first time, the thermal diffusivities of the Al??As?Te???? system. The extremum at a composition of Al??As??Te?? (?r? = 2.6) in the diffusivity was attributed to the occurrence of the Mechanical Threshold (MT) at that composition. This showed that the MT can be shifted by large values from the Phillips朤horpe value of (r)c = 2.4, due to the ionic effects of aluminum. The extremum at Al??As??Te?? was conjectured to be due to the occurrence of a Chemical Threshold (CT) at that composition. Strong supporting evidence was obtained through molar volume measurements which showed the expected features at the two compositions. The suggestion that the vector-percolation concept might be more sensitively investigated in the liquid state (Tatsumisago et al., 1990) motivated temperature-dependent DSC studies on the Al-As-Te system, as an extension to the room-temperature PPE studies described in Chapter 3. The newly developed Modulated Differential Scanning Calorimeter (MDSC) (TA Instruments Inc., USA) was chosen as the experimental probe to discern the topological thresholds initially revealed from the PPE measurements. The MDSC, however, is a more powerful technique compared with ordinary DSC, 搕hat may well be called the greatest advance in DSC since its inception some 35 years ago� (Wunderlich et al., 1994, pg. 282). The effects of modulation offer better precision in heat capacity-higher by as much as a factor of 10 (Boiler et al., 1994)-and additionally allow for the determination of the 搕rue thermodynamic heat capacity.� Based on the investigations conducted, it was concluded by analysis of the composition dependence that the percolation of rigidity in the Al??As?Te???? system exhibits a stretching over a range of compositions between x = 15 and x = 25 atomic % of arsenic. When these results are juxtaposed with PPE studies of thermal diffusivity and molar volume measurements, it can be concluded that x = 20 corresponds to the median of this transition range. The occurrence of the Chemical Threshold (CT) at x = 30 has also been identified using ?Hnr, which shows distinct maxima at this composition. This chapter not only confirms the results and conclusions of Chapter 3 but also exhibits the power of the MDSC in elucidating aspects of the topological thresholds that cannot ordinarily be resolved using other probes. This chapter describes possibly the first methodology in frequency-domain photopyroelectric techniques where a direct measurement of thermal conductivity is possible. The sample and sensor configuration is unique, and the 2-D mathematical model developed does not rely on the standard 1-D models of Mandelis and Zver (1985) and Chirtoc and Mihailescu (1989), unlike conventional variants of the PPE techniques described in the literature. Also, unlike many conventional PPE experiments, the sensor has been operated explicitly below f? and in the Isothermal-Voltage mode. A study of literature in PPE showed few reports (Antoniow et al., 1997, for example) which explicitly showed measurements made in the region below f?. Most experiments are conducted in the so-called voltage (adiabatic-voltage) and current (adiabatic-current) modes, both of which work at frequencies above f?. In cases where it could be inferred that operating frequencies were below f? (Wang & Mandelis, 1999b, pg. 8368), the effects of the corner frequency were absorbed in an instrumental transfer function normalized by means of frequency scans of samples with well-known responses. The paucity of reports on work in PPE regarding both time constants shows that the study described in this chapter is important in explicitly showing their effects.