Solar thermoelectric power conversion : materials characterization to device demonstration

• Main Author: Kraemer, Daniel, Ph. D. Massachusetts Institute of Technology
• Other Authors: Gang Chen.
• Format: Others
• Language: English
• Published: Massachusetts Institute of Technology 2016
• Subjects: Mechanical Engineering.
• Online Access: http://hdl.handle.net/1721.1/103490

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016. === Cataloged from PDF version of thesis. === Includes bibliographical references (pages 268-289). === Meeting the ever growing global energy demand with mostly fossil fuel based energy technologies is not sustainable, pollutes the environment and is the main cause of climate change threatening our planet as we know it. Solar energy technologies are a promising, sustainable and clean alternative due to the vast abundance of sunlight. Thus far, photovoltaic solar cells and concentrated solar power are considered to be the most promising approaches. Solar cells directly convert sunlight into electricity by photon induced electron-hole pair generation. Concentrated solar power captures the sunlight in form of heat which is then converted to electricity by means of a traditional mechanical power block. In this thesis, we explore solar thermoelectric generators (STEGs) as an alternative way to convert sunlight to electricity. Similar to concentrated solar power STEGs capture the sunlight in form of heat. However, the captured heat is directly converted to electricity by means of a thermoelectric generator. This solid-state direct heat-to-electricity conversion significantly simplifies the system, reduces cost and maintenance and enables transient operation and system scalability without affecting the performance. Therefore, STEGs have the potential to be deployed as small scale solar power converters in remote areas and on rooftops and as large scale concentrated solar power plants. While the concept of solar thermoelectric power conversion has been proposed over a century ago, most successful experimental efforts reported in, the literature have been limited to below 1 % for STEGs without optical concentration and to approximately 3 - 5 % with optical concentration. Theoretical STEG performances as modeled and discussed in this thesis predict significantly higher efficiencies. A detailed STEG model is introduced to theoretically investigate various parasitic losses and how to minimize their effect to obtain highest and most realistic performance predictions. Additionally, a methodology to optimize a photovoltaic-thermoelectric hybrid system based on spectral splitting is introduced. The optimization and performance prediction of a STEG is only accurate if the relevant material properties are known with high accuracy. However, typical spectroscopy techniques to determine the optical properties, namely the solar absorptance and infrared emittance, of a solar absorber have shortcomings which can lead to significant errors. Similarly, typical commercial equipment to measure the properties of thermoelectric materials including the Seebeck coefficient, the electrical resistivity and the thermal conductivity are prone to large errors. Therefore, we introduce in this thesis novel experimental techniques to measure all relevant properties with improved accuracies in particular the techniques to measure the total hemispherical emittance of a surface and a material's thermal conductivity. A record-low total hemispherical emittance of 0.13 at 500 °C is demonstrated for an Yttria-stabilized-Zirconia-based cermet solar absorber with solar absorptance of 0.91 and thermal stability up to 600 °C. Furthermore, a method was developed to directly measure the efficiency of a thermoelectric leg. Using this method a record-high thermoelectric efficiency of 8.5 % is demonstrated at a relatively small temperature difference of 225 °C for a novel MgAgSb-based compound with hot-pressed silver contact pads. By increasing the temperature difference to a material's compatible 275 °C a thermoelectric efficiency of 10 % is achievable which, thus far, has only been achieve at almost twice the temperature difference. The third main contribution of this thesis is the experimental demonstration of solar thermoelectric power conversion. A record-high STEG efficiency of 4.6 % is demonstrated at AM1.5G (1 kW/m 2) conditions which is 7 times higher than previously reported best values. The performance improvement is achieved by using a STEG with nano-structured bulk thermoelectric materials, a spectrally-selective solar absorber and taking advantage of large thermal concentrations under a vacuum. Despite the vacuum environment and the use of a low-temperature spectrally-selective solar absorber the optimal hot-junction operating temperature is limited to approximately 200 °C due to increasing thermal radiation heat loss. In order to substantially increase the operating temperature difference and STEG efficiency, larger incident solar power densities are required. Furthermore, the STEG requires segmented thermoelectric legs and a high-temperature stable solar absorber. The optimized STEGs are fabricated and tested at moderate and high optical solar concentration. Efficiencies of close to 8 % at 38 suns and close to 10 % at 211 suns, measured based on the solar flux at the absorber, are demonstrated for a STEG with a spectrally-selective solar absorber. The maximum demonstrated solar-to-electricity CSTEG efficiency is 7.5 %. Furthermore, the performance of a STEG at moderate optical concentration with a high-temperature stable black paint solar absorber and a directionally-selective solar receiver cavity is demonstrated to be comparable to a STEG with a spectrally-selective surface at similar insolation. === by Daniel Kraemer. === Ph. D.