Material and manufacturing cost considerations for thermoelectrics

https://doi.org/10.1016/j.rser.2013.12.030Get rights and content

Abstract

Interest in thermoelectrics for waste-heat recovery and localized cooling has flourished in recent years, but questions about cost and scalability remain unanswered. This work investigates the fabrication costs and coupled thermal and electrical transport factors that govern device efficiency and commercial feasibility of the most promising thermoelectric materials. For 30 bulk and thin film thermoelectric materials, we quantify the tradeoff between efficiency and cost considering electrical and thermal transport at the system level, raw material prices, system component costs, and estimated manufacturing costs. This work neglects the cost of heat, as appropriate for most waste-heat recovery applications, and applies a power generation cost metric in $/W and a cooling operating cost metric in $/kWh. The results indicate material costs are too high for typical thermoelectric power generation applications at mean temperatures below 135 °C. Above 275 °C, many bulk thermoelectric materials can achieve costs below $1/W. The major barrier to economical thermoelectric power generation at these higher temperatures results from system costs for heat exchangers and ceramic plates. For cooling applications, we find that several thermoelectric materials can be cost competitive and commercially promising.

Introduction

Thermoelectric devices are used in power generation and cooling applications to either convert heat into electricity or to pump heat. To date, thermoelectric technology has been constrained to applications that include vehicle waste-heat recovery prototypes, space vehicle power sources, seat coolers, solid-state refrigerators, and temperature control in laboratory equipment. Thermoelectric conversion has received renewed interest due to the development of better-performing materials and their potential to improve the efficiency of combustion systems through waste-heat recovery [1], [2], [3]. Improvements in material performance are ongoing [4], [5], [6] since conversion efficiencies of typical thermoelectric materials remain below 10% [3], [7].

Thermoelectric devices are attractive because they have no moving parts and can be silent, reliable, and versatile. Since multiple thermoelectric n- and p-type couples can be connected in series, a thermoelectric device can be smaller than a computer chip or larger than a solar panel. Nevertheless, considerable technical challenges remain. Existing thermoelectric converters have lower conversion efficiencies than alternatives such as organic Rankine cycles. While some applications are well established, in particular power sources for space vehicles, a variety of terrestrial applications – such as high temperature waste-heat recovery – are yielding new concerns about reliability and durability. Challenges related to sublimation, oxidation, thermal and electrical interface degradation, and thermal expansion mismatch remain critical for applications of thermoelectric devices [8], [9], [10], [11], [12], [13], [14], [15].

Thermoelectric technologies face several additional commercialization challenges. The cost of many thermoelectric materials may be prohibitively high, largely due to the use of tellurium and germanium in the most common contemporary thermoelectric applications [16]. Safety concerns preclude toxic materials such as lead [17]. The weight and specific power of thermoelectric devices are particularly important for mobile applications such as vehicle waste-heat recovery [18], yet few strategies for weight reduction have been proposed. While government funding for thermoelectric technology research and development has expanded significantly in the past decade, the relative lack of private sector familiarity with the technology made early stage financing for companies comparatively slow to follow [19]. There has been a rapid development of materials promising higher efficiencies, in many cases through the use of nanostructuring and novel crystal structures [1], [4], [6]. Recent prototypes demonstrated the feasibility of thermoelectric devices for widespread, terrestrial, scalable applications [20], [21], [22], [23]. Multiple start-up companies have received recognition and funding while also forming partnerships with established academic and industrial research teams [15], [18], [24], [25], [26]. The continued development and deployment of scalable thermoelectric devices depend on the device cost and energy conversion performance [1], [27].

There is an urgent need for a comprehensive assessment of the commercial feasibility of the most promising thermoelectric materials while also considering relevant manufacturing and system costs. Cost-performance analyses of this type have been conducted for other energy technologies such as batteries and photovoltaics by considering material extraction costs [28], [29]. For thermoelectric devices, there have been multiple evaluations of the coupled thermal and electrical module performance without considering cost [30], [31]. The importance of cost was considered by normalizing the material figure of merit by the raw material cost, and the costs for specialized applications such as a marine waste incinerator have also been determined [17], [27], [32], [33]. A method for determining cost using the thermal energy cost and the module construction cost for specified material thicknesses has been presented [32]. A recent investigation of the cost-efficiency tradeoffs of thermoelectric generators first determined the geometry that optimizes the thermal and electrical performance of the device and then calculated the material cost associated with the optimized dimensions [27]. The work highlighted the importance of optimizing modules in conjunction with the heat source and sinks in the overall system. An important extension to the optimization approach is made by evaluating the coupled thermal, electrical, and economic factors simultaneously [34]. This method captures the ability of a moderately efficient device with low capital cost to perform better in the market than a highly efficient device with higher capital cost. While this prior work made progress on coupling efficiency and cost, the actual values and thus large impacts of manufacturing costs and the costs of key system components have not been considered. This is particularly important for the newer bulk and thin film materials for which manufacturing costs can be high, and new system designs might be required to optimize heat exchange.

The present work provides a detailed investigation of the materials, manufacturing, and system costs for 30 leading thermoelectric materials [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61] considering the conversion efficiency. We combine these using cost metrics that account for the coupled cost and system design implications and trade-offs [34]. These metrics enable an assessment of which materials are most promising for heat recovery and cooling and incorporate an optimization of material thickness and fill factor. The present work sets targets for thermoelectric cooling and power generation. The cost of power generation, in $/W, is reported by combining material properties, device physics, and raw material costs, all of which are reported for seven material classes presented in Table 1 and in Appendix F. The analysis is performed for five operating temperatures shown in Table 2 to reflect the myriad of potential thermoelectric applications. The difference in manufacturing costs for bulk and thin film materials influences the decision on whether to use novel materials. A comparison to existing, competitive technologies highlights the cost of current state-of-the-art thermoelectric technologies and device performance.

Section snippets

Methods

This work determines the appropriate device geometry based on cost and performance. The fill factor, F, and the leg length, L, of a thermoelectric device are the dominant design parameters. The fill factor is the ratio of the area covered by the active thermoelectric material to the plate area, A, as illustrated in Fig. 1. While it goes against traditional thermoelectric module architecture, it is possible to envision devices with F>1 (see Section 3.1). The inverse of the fill factor is known

Results and discussion

The results specifically address the cost of candidate materials, the cost to process those materials into thermoelectric legs for a module, and the cost of the heat exchangers making up the full thermoelectric device. This enables realistic assessments of current and future materials' potential for use in power generation and cooling applications by linking performance with estimated material, manufacturing, and system costs.

Conclusion

This work applies a new thermoelectrics cost analysis [34] which incorporates material properties, device physics, material costs, manufacturing costs, and system costs. The analysis and resulting cost values provide a tool for thermoelectric device research and development. Optimization of thermoelectric leg length and fill factor minimizes the ratio of cost to performance, as opposed to optimization of efficiency and power output alone. This work makes progress on evaluating the status of

Acknowledgments

We gratefully acknowledge funding support through the NSF/DOE Partnership on Thermoelectric Devices for Vehicle Applications (Grant No. 1048796). S. LeBlanc would like to acknowledge the Sandia National Laboratories Fellowship and the Stanford DARE fellowship. S. K. Yee would like to acknowledge support from the John and Fannie Hertz Foundation and the Big George Fellowship. We would also like to thank Profs. Arun Majumdar, Gang Chen, and G. Jeffrey Snyder for their insight and discussion of

References (87)

  • TM. Tritt

    Thermoelectric phenomena, materials, and applications

    Annu Rev Mater Res

    (2011)
  • GS Nolas et al.

    Recent developments in buk thermoelectric materials

    MRS Bull

    (2006)
  • GJ Snyder et al.

    Complex thermoelectric materials

    Nat Mater

    (2008)
  • JR Sootsman et al.

    New and old concepts in thermoelectric materials

    Angew Chem Int Ed Engl

    (2009)
  • A. Shakouri

    Recent developments in semiconductor thermoelectric physics and materials

    Annu Rev Mater Res

    (2011)
  • Salvador JR, Cho JY, Ye Z, Moczygemba JE, Thompson AJ, Sharp JW, et al. Thermal to electrical energy conversion of...
  • Li F, Huang X, Jiang W, Chen L. Microstructure and contact resistivity of (Bi, Sb)2Te3/Sb interface. AIP Conf Proc...
  • Battiston S, Boldrini S, Fiameni S, Famengo A, Fabrizio M, Barison S, Multilayered thin films for oxidation protection...
  • J Leszczynski et al.

    Studies on thermal decomposition and oxidation of CoSb3

    J Therm Anal Calorim

    (2011)
  • Bennet GL, Skrabek EA. Power performance of U.S. space radioisotope thermoelectric generators. In: 15th International...
  • Goodson KE, Asheghi M. Novel nanostructured interface solutions for automotive thermoelectric modules application. 3rd...
  • KE. Goodson

    Automotive thermoelectric modules with scalable thermo- and electro-mechanical interfaces. 2011 Thermoelectrics applications workshop

    (2011)
  • Y Gao et al.

    Nanostructured interfaces for thermoelectrics

    J Electron Mater

    (2010)
  • A. Miner

    The industrialization of thermoelectric power generation technology. 3rd Thermoelectrics applications workshop

    (2012)
  • GG Yadav et al.

    Nanostructure-based thermoelectric conversion: an insight into the feasibility and sustainability for large-scale deployment

    Nanoscale

    (2011)
  • J. Fairbanks

    Vehicular Thermoelectrics: a new green technology. Thermoelectrics Applications Workshop

    (2011)
  • Straub P. Clarement creek ventures. (personal...
  • DT. Crane

    Thermoelectric generator performance for passenger vehicles. 3rd Thermoelectrics applications workshop

    (2012)
  • L. Aixala

    RENOTER project. 3rd Thermoelectrics applications workshop

    (2012)
  • DT Crane et al.

    High heat flux thermoelectric module using standard bulk material. 3rd Thermoelectrics applications workshop

    (2012)
  • G. Meisner

    Skutterudite thermoelectric generator for automotive waste heat recovery. 3rd Thermoelectrics applications workshop

    (2012)
  • Y Gao et al.

    Impact of nanotube density and alignment on the elastic modulus near the top and base surfaces of aligned multi-walled carbon nanotube films

    Carbon

    (2012)
  • K Yazawa et al.

    Cost-efficiency trade-off and the design of thermoelectric power generators

    Environ Sci Technol

    (2011)
  • C Wadia et al.

    Materials availability expands the opportunity for large-scale photovoltaics deployment

    Environ Sci Technol

    (2009)
  • SW. Angrist

    Direct energy conversion

    (1965)
  • DM. Rowe

    Thermoelectrics handbook

    (2006)
  • DM Rowe et al.

    Design theory of thermoelectric modules for electrical power generation.

    IEE Proc – Sci Meas Technol

    (1996)
  • NR Kristiansen et al.

    Waste heat recovery from a marine waste incinerator using a thermoelectric generator

    J Electron Mater

    (2012)
  • S.K. Yee et al.

    $ per W metrics for thermoelectric power generation: beyond ZT

    Energy Environ Sci

    (2013)
  • Parker WJ, Jenkins RJ. Thermal conductivity measurements on bismuth telluride in the presence of a 2 MEV Electron Beam....
  • ND Lowhorn et al.

    Development of a Seebeck coefficient Standard Reference Material™

    J Mater Res

    (2011)
  • W Xie et al.

    High thermoelectric performance BiSbTe alloy with unique low-dimensional structure

    J Appl Phys

    (2009)
  • J Zhou et al.

    Thermoelectric properties of individual electrodeposited bismuth telluride nanowires

    Appl Phys Lett

    (2005)
  • Cited by (401)

    View all citing articles on Scopus
    1

    Indicates joint first authorship.

    View full text