Practical electrical contact resistance measurement method for bulk thermoelectric devices
- PDF / 659,778 Bytes
- 7 Pages / 612 x 792 pts (letter) Page_size
- 101 Downloads / 240 Views
Practical electrical contact resistance measurement method for bulk thermoelectric devices Rahul P. Gupta, Robin McCarty, Jim Bierschenk, and Jeff Sharp Marlow Industries Inc., a subsidiary of II-VI Inc., Dallas TX 75238 ABSTRACT As thermoelectric (TE) element length decreases, the impact of contact resistance on TE device performance grows more significant. In fact, for a TE device containing 100-µm tall Bi2Te3TE elements, the figure of merit ratio (ZTDevice/ZTMaterial) drops from 0.9 to 0.5 as the contact resistivity increases from 5 x 10-07 to 5 x 10-06 Ω-cm2. To understand the effects of contact resistance on bulk TE device performance, a reliable experimental measurement method is needed. There are many popular methods to extract contact resistance such as Transmission Line Measurements (TLM) and Kelvin Cross Bridge Resistor method (KCBR), but they are only wellsuited for measuring metal contacts on thin films and do not necessarily translate to measuring contact resistance on bulk thermoelectric materials. The authors present a new measurement technique that precisely measures contact resistance (on the order of 5 x 10-07 Ω-cm2) on bulk thermoelectric materials by processing stacks of bulk, metal-coated TE wafers using TE industry standard processes. One advantage of this technique is that it exploits realistic TE device manufacturing techniques and results in an almost device-like structure, therefore representing a realistic value for electrical contact resistance in a bulk TE device. Contact resistance measurements for metal contacts to n- and p-type Bi2Te3 alloys are presented and an estimate of the accuracy of the measurements is discussed. INTRODUCTION Cooling power density quantifies the capacity of a TE device to pump heat. High watt density devices can dramatically change the historical standards of cost per watt of heat pumped. Maximum cooling power density per unit area for a single element (Qmax) is derived by differentiating total heat flux (Q) for Peltier cooling [1] and is given by Equation 1, (1) where L is the device leg length, ρcont is the contact resistance , ρ is the bulk resistivity, k is the thermal conductivity, Tc is the cold side temperature, and α is the Seebeck coefficient. From Equation 1, cooling power density is inversely proportional to the leg length, L of the device; therefore cooling power density can be increased by reducing the leg length. Bulk material is currently the predominant commercial option to obtain the required performance because of the high cost of thin film material processing and limitations on film thickness. Micro TE coolers, bulk TE coolers with very small element lengths are becoming the norm in the telecommunications industry where the TE cooler temperature stabilizes the laser diodes that send the optical signals down the fiber optic cables [2]. The impact of contact resistance on micro TE cooler performance is quite significant for devices with element length of 200 um or less.
One of the requirements to quantify the effects of contact resistance on
Data Loading...
