Thermopower measurements

As noted in Chapter 6 earlier, thermoelectric power is the potential gradient generated between two points or faces of a material when they are subject to a temperature differential. Typically, it is large for semiconductors, while for metals it is small and decreases with decreasing temperature, vanishing near absolute zero. A small decrease of thermopower with temperature would thus likely indicate hopping conduction, whilst a very rapid decrease and very low values (see below) for the thermopower would indicate a quasi-metallic conduction, as for instance observed for highly doped trans-P(Ac). 

To cite a typical thermopower measurement [191], the CP sample is mounted across two single-crystal quartz blocks maintained at two different temperatures, usually a difference of 1°K. A thermocouple is also used to monitor the temperature difference. During the measurement, the temperature difference is achieved slowly (over 10 mins, e.g.). The voltages across the sample and the thermocouple are monitored with a nanovoltmeter. The slope of the thermovoltage of the sample vs. that of the thermocouple yields the sample thermopower, after correction for contact contributions. 

The thermoelectric power behavior of most CPs parallels that of temperature and activation energy relationships. 

Selected thermopower data have been cited in Chapter 6 earlier. 

---

Thermopower measurements due to Kwizera et al (1981) are shown in Fig 6.23. The equation for the thermopower in the metallic state, S=( n2fciT/e)dlnff/d , should not be valid above about 150K it can be seen that d In narrow band in the metallic state, possibly due to Brinkman-Rice enhancement or formation of spin polarons. 

The disappearance of the sharp Verwey transition was discussed by Mott (1979), who suggested that at low temperatures the material is a Wigner glass , the electrons (Fe2 + ions) being frozen into random sites and the whole system stabilized by the fluorine. Discussion of the thermopower measurements show, according to Mott (1979), that a hopping mechanism is operative at low T. Ihle and Lorenz (1985), however, consider that the electrons in the wrong sites move by a small polaron band mechanism. 

As expected for a (DT-TTF)+1/2 cation-radical, thermopower measurements indicate a hole transport for (DT-TTF)2[Au(mnt)2], which behaves like a semiconductor with a slight change in the conductivity regime 220 K. Diffuse X-ray scattering studies show that the donors dimerize along the b stacking direction, two DT-TTF molecules sharing one electron. Since Au(mnt)2 is... 

Wu CC, Mason TO (1981) Thermopower measurement of cation distribution in magnetite. J Am Ceramic Soc 64 520-522... 

Structure calculations give a change-transfer value of 0.8. This value is in agreement with thermopower measurements. Raman spectroscopy studies performed on single crystals also confirm the fractional charge of the TTF units the Vc=c central bond of TTF appears at 1435 cm . From the study performed by Bozio et this frequency... 

Electrical conductivity and thermopower measurements on single crystals of t(NHs) tCk (MGS) and (TTF) NiS C H (TTF = tetrathiafulvalene) illustrate the application of electrical property measurements to the study of intermolecular interactions and electronic structure in planar metal complex systems. MGS is an anisotropic, p-type semiconductor conductivity along the metal-chain direction ohm ... 

The contributions of H. R. Hart, Jr., I. S. Jacobs, J. S. Kasper, G. D. Watkins, and S. H. Wee to the work described herein are gratefully acknowledged. We are indebted, in particular to H. R. Hart, Jr. and W. R. Giard of General Electric Corporate Research and Development for assistance with the conductivity and thermopower measurements. 

Errors in thermopower measurements can arise from several sources. The main source of error in this experimental configuration was an approximately lOZ uncertainty in the temperature difference AT across the samples due to uneven heating of the quartz blocks. For temperatures above 9IK, values of the samples absolute thermopower may be off by as much as 0.25 pV/K,... 

Further information on the transport processes in a-Si H and on the influence of doping can be obtained, e.g., from measurements of the drift mobility (Allan et al., 1977 Moore, 1977), of the photoconductivity (Rehm et al., 1977 Anderson and Spear, 1977), as well as of the magnetic field dependence of the photo- and dark conductivity (Weller et al., 1981). In this chapter, however, we shall confine ourselves mainly to results of conductivity and thermopower measurements. Some results from Hall effect and photoconductivity studies are also discussed. 

Fig. 1. Schematic representation of the temperature dependence of the Fermi energy Ef. Conductivity and thermopower measurements yield apparent Fermi-level positions i RO).

The results of conductivity and thermopower measurements as a function of temperature are shown in Fig. 14 for phosphorus-doped a-Si H films (Beyer et al, 1977a) and in Fig. 15 for boron-doped material (Beyer and Mell, 1977). The plots present log a and S versus reciprocal temperature. We note that the data for phosphorus doping agree qualitatively well with results obtained for a variety of -type dopants, including As, Li, K, and Na obtained by the present authors and by others (Beyer et ai, 1979b,c, 1980 Jan et al, 1979). Our results for boron doping are similar to those of Jan et al (1980) and Tsai eta/. (1977). 

Wu, C. C. Mason, T. O. (1981). Thermopower measurements of cation distribution in magnetite. Journal of the American Ceramic Society, 64, 520-2. 

Pisarkiewicz, T. and Stapinski, T. (1989), Influence of gas atmosphere on thermopower measurements in tin oxide thin-flhns. Thin Solid Films, 174,277-83. 

Tak] X-ray, electrical resistivity and thermopower measurements, dilatometry Electrical resistivity, thermopower, Cr2oFe cNigo-A, at.% 44 < x < 70 from 5 to 60 K... 

In the present section we focus on the various chemical and physical properties of the equiatomic YbTX compounds especially the physical properties, which have been intensively investigated in the last two decades. Besides detailed magnetic susceptibility and electrical resistivity measurements, various other techniques have been used to get deeper insight into the peculiar properties of these intermetallics " Sn and °Yb Mdssbauer spectroscopy, specific-heat data, thermopower measurements, solid-state NMR, photoemission studies, neutron dif action, muon spin relaxation, and a variety of high-pressure experiments. The diverse data are summarized in the following subsections. 

Thermopower measurements indicate a metallic character of these solids [42,43]. Principally, the conductivity of the isolated TT-derivatives is lower (2-7 S cm-i), compared with the TTF-TT compounds (40-150 S cm-i). 

Fig. 2 shows the temperature dependence of the electrical resistivity of (DIET)2(BF4)3 as a typical example. The compound behaves as a metal at room temperature and l low. At lower temperature, a very broad phase transition is indicated in the resistivity data. Fig. 3 summarizes results of the thermopower measurement on this material at different temperatures. The latter data clearly reveal a phase transition at around 50 K. Similar results have been obtained from DIET salts with other anions like HSO4" or N03", as shown in Fig. 4 for the compound (DIET)j (N03)y, the full structure of which could not be solved so far. 

Many theories have been developed (involving solitons, excitons, polarons and bipolarons) [17] to explain the conductivity phenomenon under the assumption that the chains of conductive polymers are being arranged and at least somewhat oriented in fibrils. But now, it must be explained why our dispersed (and later flocculated) polymer showed principally the same transport properties as the fibrillar conductive polymers, as can be concluded from conductivity versus temperature and thermopower measurements. 

The thermopower measurements themselves show that dispersed conductive polymers have essentially the same thermopower behavior as metals it is small (semiconductors show a high S), and decreases practically linearly with decreasing temperature [18]. These properties were known from our previous measurements [19], which showed the linear temperature dependance for a range of 100 K. These results lead us to the assumption that the primary transport mechanism is metallic, but obviously influenced by a barrier mechanism. To overcome the barrier, a temperature-activated transport process had to be active as well. 

Fig. 7.18. (a) Sample holder for thermopower measurements devised by Enderby and Walsh (1966)... 

---