Thermal Metrology for Waste Heat Systems: Thermoelectrics to Phase Change Materials

Collier S Miers, Purdue University

Abstract

Accurate measurements are essential to further the advancement of a technical field. This sentiment is particularly true for the thermal sciences, where due to the indirect nature of temperature, the most critical physical quantity that we discuss is often grossly mismeasured. This leads to substantial spread in reported experimental data, which impedes progress. Here I present two systems for which a gap in current measurement systems or experimental practices exists. First, I detail the design and fabrication of a high-temperature Z-Meter which simultaneously measures pertinent thermoelectric properties. Then I present work done to improve the design of phase change material passive thermal management solutions and a new measurement platform allowing the direct comparison of performance across a wide range of thermal management designs. In the first part of my dissertation, I design a Z-Meter system to evaluate the performance of thermoelectric materials under large temperature differences. Few systems exist for simultaneous measurement of thermal, electrical, and thermoelectric properties at high-temperatures (> 400℃), which is crucial for high-temperature waste heat recovery applications. The Z-Meter simultaneously measures the three thermoelectric properties from which the figure of merit, ZT, is calculated. Traditionally, this technique has been employed utilizing a very small temperature difference across the sample (∆T = 1-2℃) and test temperatures between 20-400℃; however, thermoelectric materials are typically subjected to large temperature gradients for favorable energy conversion. Additionally, high operating temperatures (> 400℃) provide access to higher grade waste heat than is available at low temperatures. It is critical that materials be characterized under conditions close to the targeted operating conditions, because the relevant properties possess strong temperature dependencies. In this work, I focus on extending the measurement capabilities for a Z-Meter to higher temperatures (∼ 1000℃). In order to accomplish this, I develop detailed 2D axisymmetric multi-physics finite element analysis (FEA) simulations of the system components and measurement process to extract simulation values at the probe locations in the system. This simulated “experimental” data is passed as the input for the uncertainty quantification (UQ) model to statistically compare design options, evaluate the impact of component modifications on the overall system performance, and to provide the quantitative data required to deselect options from the design space and determine the final Z-Meter design which minimizes the measurement error for ZT. The UQ model for the meter bars includes effects from radiative losses, contact resistances, thermoelectric effects, and interaction of electrical measurement signals in the bar. The detailed UQ model combined with the fully-coupled multi-physics simulations results provide a powerful platform to investigate the sensitivity of ZT to specific design variables (e.g., component positioning, material selection, or sensor type) which is invaluable for determining the correct system design. The system, as designed, is capable of sample hot side temperatures of 1000℃ with temperature gradients on the order of 500℃. The elevated temperature capability is necessary to fill a gap in characterization equipment for high-temperature thermoelectric applications. In order to maximize the accuracy and repeatability of high-temperature measurements, I instrumented the meter bars with fine gauge typeS thermocouples. The measurement system is capable of operating in high-vacuum to suppress convection losses or it can be backfilled with an inert gas permitting atmospheric control for testing under specific environments.

Degree

Ph.D.

Advisors

Marconnet, Purdue University.

Subject Area

Design|High Temperature Physics|Materials science|Packaging|Physics|Thermodynamics

COinS