1 / 49

Solid State Energy Conversion

Solid State Energy Conversion. Cronin B. Vining ZT Services, Inc. Auburn, Alabama USA Phone: (1) (334) 887-2404 FAX: (1) (334) 887-2604 vining@zts.com http://www.zts.com. Auburn University October 9, 1998. Dilbert 10-10-1993. Outline. Introduction

meryl
Télécharger la présentation

Solid State Energy Conversion

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Solid State Energy Conversion Cronin B. Vining ZT Services, Inc. Auburn, Alabama USA Phone: (1) (334) 887-2404 FAX: (1) (334) 887-2604 vining@zts.com http://www.zts.com Auburn University October 9, 1998

  2. Dilbert 10-10-1993

  3. Outline • Introduction • Alternative energy conversion technologies • Modern ideas for new thermoelectric materials • Summary

  4. Show Cassini

  5. Show Unicouple

  6. Igloo-Brand Cooler

  7. Production Cost Reduction • Production costs have decreased steadily • Significant consumer markets have opened • Picnic Baskets use >500,000 modules/year (Igloo, Coleman, etc...) • Reliability is very high • Efficiency remains near 1960 levels After R.J. Buist, 1993

  8. Physical Origin of ZT • Any efficiency calculation will involve the same factors shown here • S, r and lalways occur together in the end • ZT (not just Z) is the preferred quantity • this is the only unitless combination of S, r and l. • ZT also occurs in thermodynamics

  9. 60% Carnot 50% 40% Efficiency 30% 20% Today 10% 0% 0 2 4 6 8 10 12 14 ZT • For Power Generation • For a single stage cooler:

  10. Applications Drive R&D • In the ‘50-60’s semiconductors were new, hopes were high • From the mid-1960s until the mid-1990s, most US R&D has been directed to support Space Nuclear Power • Systems & hardware oriented • Materials work focused on ‘modest’ improvements in SiGe • Evolutionary, not revolutionary • Space Power R&D has essentially ended • New ideas being supported mostly by new sponsors • Emerging R&D hopes to significantly improve efficiencies to approach mechanical engines

  11. Trends in the US & the Future

  12. Experimental ZT Results Essentially unchanged since the beginning of the SPACE AGE. Why?

  13. Outline • Introduction • Alternative energy conversion technologies • Thermionic • Thermophotovoltaic • AMTEC - Alkali Metal Thermal to Electric Conversion • Thermoelectric • Modern ideas for new thermoelectric materials • Summary

  14. Thermionic Converter • Plates emit electrons at different rates • Work function values of several eV • Te ~ 1500-2000 K • Tc ~ 1000 K • Recently Mahan pointed out refrigeration is possible with sufficiently small work functions • Borealis is pursuing • Or, replace vacuum with a solid to reduce the work function • receiving real attention from Angrist, 1982

  15. Outline • Introduction • Alternative energy conversion technologies • Thermionic • Thermophotovoltaic • AMTEC - Alkali Metal Thermal to Electric Conversion • Thermoelectric • Modern ideas for new thermoelectric materials • Summary

  16. Thermophotovoltaics • A solar cell is a heat engine • The heat source is 1 AU away and at 6000 K • Thermophotovoltaics use • A lower temperature heat source, 1000-1400 K • Insulation/reflectors to return unused radiation back to the source • Narrow spectrum emitters • Lower band-gap semiconductors From Angrist, 1982

  17. Outline • Introduction • Alternative energy conversion technologies • Thermionic • Thermophotovoltaic • AMTEC - Alkali Metal Thermal to Electric Conversion • Thermoelectric • Modern ideas for new thermoelectric materials • Summary

  18. Thermodynamic Cycle for AMTEC Liquid-Fed AMTEC a-d: liquid vaporizes, isothermal d-e: gas cools e-f: gas condenses, isothermal f-g: liquid pressurized g-a: liquid heated • Two converters in one: • 1) Condensable gas cycle turns heat into mechanical (=Pressure ) • 2) BASE turns Pressure Drop • (= concentration difference) into electrical Schematic of a liquid-fed AMTEC cell

  19. AMTEC P-V and T-S Plots • a-d: liquid vaporizes, isothermal • a-b: de-pressurization of liquid • b-c: vaporization - saturatied liquid • c-d: isothermal expansion • d-e: gas cools • e-f: gas condenses, isothermal • f-g: liquid pressurized • Na is so incompresible it doesn’t matter • Pump efficiency is not critical • g-a: liquid heated • Liquid-Fed AMTEC • - illustration, small T • Cycle: a-b-c-d-e-f-g-a • Heat in: g-a-b-c-d

  20. Outline • Introduction • Alternative energy conversion technologies • Thermionic • Thermophotovoltaic • AMTEC - Alkali Metal Thermal to Electric Conversion • Thermoelectric • Modern ideas for new thermoelectric materials • Summary

  21. Outline • Introduction • Alternative energy conversion technologies • Modern ideas for new thermoelectric materials • Summary

  22. TE Basics Thermoelectricity : • Coupled electrical and thermal transport • Heat Flow resulting from an Electrical Current • Electrical Voltages from Thermal Gradients, or • Important applications: • Peltier Cooling, Power Generation, Thermometry • Existing technology is reliable, but inefficient • Efficiency is related to a single material property: ZT • Best known materials have at best ZT~1 • Efficiency limited to about 1/6 of Carnot • If ZT~1 is the best possible: Why? • If ZT>>1 is possible: How to achieve it?

  23. BACKGROUND • Thermoelectricity dates to 1822 (or before) and matured to present status in the 1950’s-1960’s, along with other semiconductor technologies • The coefficient of performance (e) for a thermoelectric power generator or cooler depends on the active thermoelectric material through the Figure of Merit: where • Neither equilibrium thermodynamics nor non-equilibrium thermodynamics place any upper bound on ZT.

  24. Definitions The Definitions of Transport Coefficients of Interest in Thermoelectricity: Thermoelectric Property Definition Under Condition Type Electrical Conductivity Direct Thermal Conductivity Direct Seebeck Coefficient Cross Peltier Coefficient Cross Generalized Ohm’s Law, Good Under All Conditions: Symmetrical Version, Currents are created by Forces

  25. Single Band Model

  26. Today: Doping and Alloying are the Major Effects

  27. More Detailed Model • Today, increasingly detailed and quantitative models are possible • Any transport or thermodynamic property of interest can be calculated self-consistently • Quantitative agreement with known materials is excellent

  28. How To Increase ZT • All approaches fall into one of three categories: • 1. Decrease the lattice thermal conductivity • Focus on phonons • Larger unit cell and higher mass to decrease sound velocity • Increase disorder to decrease phonon mean free path • 2. Increase the carrier mobility • New, covalently bonded materials • Heterostructures to physically separate carriers from scattering centers • 3. Increase the thermopower • Larger effective mass materials • Barriers to inhibit transport of low energy carriers • Novel band structures and/or scattering mechanisms

  29. ZT=0.1 ZT=3 Mobility Edge Mobility Edge h h f I0=s I1=sS I2=l + sS2T Schematic ZT Calculation Energy Energy

  30. There is no theoretical basis for a limit near ZT~1 • Discovery of a theoretical limit could have broad implications • Estimates indicate thermoelectric performance can triple, or more • Even incremental progress is significant • A systematic approach to advanced thermoelectric materials is very promising • Utilize recent advances in experimental and theoretical methods • Explore new materials discovered since the 1960’s The challenge is to accurately evaluate many possibilities

  31. Outline • Thermoelectric Technology Today • Thermoelectric Fundamentals and Physical Phenomena • Thermoelectric Materials of the Future • Summary

  32. Where do we start to look? • Today’s materials are based on: Bi2Te3, PbTe, SiGe • might also include BiSb, TAGS and FeSi2 • These will not be replaced in the near future • Mature device technologies available • Current markets are too small to develop new technologies quickly • By establishing a deeper understanding of today’s materials we lay the foundation for new materials • Use well understood materials to test novel ideas

  33. Conventional Semiconductors • Are there semiconductors which “work” according to conventional rules, but have more favorable parameters? • Large meff, & m • Small lph (approach the minimum possible) • Eg > 4kT • Binary Compounds • Most (but not all) binary compounds have already been studied • Novel binary compounds studied in recent years: • B4C, La3-xS4, La3-xTe4 • Ru2Si3, Ir3Si5, IrSi3, Ru2Ge3, Re3Ge7, Mo13Ge23, Cr11Ge19, CoGe2 • RuSb2, IrSb2, IrSb3, and CoSb3

  34. Conventional Semiconductors • Slack has surveyed all the binary compounds! • in CRC Handbook on Thermoelectricity • Emphasis on small electronegativity difference for high mobility values • 28 candidate binary compounds tabulated! • Particularly promising: IrSb3, Re6Te15, and Mo6Te8

  35. Skutterudites - IrSb3 • Large family of compounds like MX3 • M=Co, Rh, Ir; X=P, As, Sb, many other, more complex substitutions also possible • High p-type mobility values reported • m~1000-1200 cm2/V-s for IrSb3, m>8000 cm2/V-s for RhSb3 • Low thermal conductivity values • as low as 0.008 W/cm-K for selected alloys

  36. IrSb3: Premature Breakthrough NASA Tech Briefs, December 1994, p. 54 • Reports of ZT~1.5-2 for IrSb3 have been retracted • Annealing causes formation of a ‘skin’ of IrSb2 • High ZT measurements now considered unreliable • interesting physics, but much more work is required

  37. Skutterudite Crystal Structure

  38. Groups Working on Skutterudites • Slack, Nolas, Rensealear Polytechnical Institute • Experimental • Caillat, Fleurial, Borshchevsky at JPL • Everything! • Cook, Canfield at Ames Laboratory • Experimental • Morelli at GM • Tritt, Gillespie, Ehrlich at Naval Research Laboratory • Experimental • Reineke, Naval Research Laboratory • Band Structure, Theory • Matsubara et al, Yamaguchi University (now moved). • Experimental

  39. Heterostructures • Apply modern fabrication techniques to thermoelectric materials • allows materials and properties not previously possible • extensively applied to control electronic properties • extension to thermal and thermoelectric properties is only starting

  40. Heterostructures • Hicks and Dresselhaus: Quantum wells • ZT increases with decreasing size of quantum well • Factor of 14 increase in ZT predicted for Bi2Te3! • Another factor of 2 increase predicted for 1D quantum wires • Theoretical work also at Naval Research Lab, Oak Ridge • Harman at MIT Lincoln Labs is pursuing this type of approach by Molecular Beam Epitaxy • - Some early experimental results consistent with theory have been reported • Note criticisms by Sofo, Mahan and Lyon (Oak Ridge) and by Whitlow • Other effects could also enhance ZT • Mobility enhancement due to physical separation between carriers and ionized impurities • Phonon scattering and/or Bragg reflection at heterostructure boundaries

  41. Energy E>m, S>0 E<m, S<0 x Heterostructures • Moizhes and Nemchinsky: Barriers enhance the Seebeck • Carriers below the chemical potential degrade the Seebeck • Energy barriers allow “good carriers” to pass, inhibit bad carriers

  42. Ternary and More Complex Compounds • Vast number of ternary compounds known • Thousands studied for superconductivity, but TE data is rare • Copper Oxides - evaluated by Mason • only low ZT expected, due to poor mobilities • Mn4Al3Si5 - studied by Marchuk et al • Such anomalous RH and S results are always worth careful study • HfNiSn - studied by Dashevsky et al • 67% metal and still a promising semiconductor! • This approach is being pursued at Mich. St. by Kanatzidis

  43. Unconventional Semiconductors • Not all semiconductors work the same way • In hopping conductors, carriers interact so strongly with “phonons” that the lattice distorts around the carrier • In some materials, charge carriers interact with each other so strongly that electrons cannot be considered as “independent” • Conventional selection criteria may fail for such materials Pursue the anomalies

  44. Strong Carrier-Lattice Interaction • n-type FeSi2 is a hopping conductor • b for n-type FeSi2 is about 50 times smaller than b for SiGe • but ZTmax~0.4 for FeSi2, less than 3 times small than for SiGe • low cost and “anomalous” behavior are good reasons for further studies • BxC has ZT~0.4-0.5 • too small mobility (~ 1 cm2/V-s), too high carrier concentration (~1021 cm-3) • Very high melting point and composed of very light elements • All conventional rules suggest this material has no promise • Still, it is within 2-3 of the very best

  45. Strong Carrier-Carrier Interaction • U3Pt3Bi4 - suggested by Slack • many isostructural compounds, such as Ce3Pt3Bi4 • so-called “heavy fermion semiconductor” • carriers behave as if they have large effective mass • Should have high Seebeck values • Related materials suggested by Louie and Radebaugh • (Ce1-xLax)Ni2, (Ce1-xLax)In3, CePd3, and CeInCu2 • This General Approach is being pursued at: • Cornell (DiSalvo), Ames Lab (Cook, Vining)

  46. Organic Conductors • Many organic polymers with high electrical conductivity are now known • Doped polyacetylene can have electrical conductivity comparable to good metals • At low doping levels, high Seebeck values (>1000 mV/K) have been observed • Sometimes, electrical mobility values can be quite good • Give the low cost and the great ability to modify organic materials, some closer attention seems justified

  47. Outline • Thermoelectric Technology Today • Thermoelectric Fundamentals and Physical Phenomena • Thermoelectric Materials of the Future • Summary

  48. Status of New Materials Research • Large ZT values have not yet been confirmed • There is no easy path to large ZT • But there are many plausible approaches that have yet to be tried • Persistent efforts are bound to yield exciting results The challenge is not the generation of plausible ideas, but the rapid and accurate evaluation of those ideas

  49. The Goal is not New • The Westinghouse Thermoelectric Generator Program goal for efficiency was “only 35%” because • “Frankly, I wish the goal to be one that we can attain.” • From C. Zener, 1959

More Related