«Metallurgy, Thermal Stability, and Failure Mode of the Commercial Bi-Te-based Thermoelectric Modules Nancy Yang and Alfredo Morales Prepared by ...»
Printed February 2009
Metallurgy, Thermal Stability, and
Failure Mode of the Commercial
Bi-Te-based Thermoelectric Modules
Nancy Yang and Alfredo Morales
Sandia National Laboratories
Albuquerque, New Mexico 87185 and Livermore, California 94550
Sandia is a multiprogram laboratory operated by Sandia Corporation,
a Lockheed Martin Company, for the United States Department of Energy’s
National Nuclear Security Administration under Contract DE-AC04-94AL85000.
Approved for public release; further dissemination unlimited.
Issued by Sandia National Laboratories, operated for the United States Department of Energy by Sandia Corporation.
NOTICE: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government, nor any agency thereof, nor any of their employees, nor any of their contractors, subcontractors, or their employees, make any warranty, express or implied, or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represent that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government, any agency thereof, or any of their contractors or subcontractors. The views and opinions expressed herein do not necessarily state or reflect those of the United States Government, any agency thereof, or any of their contractors.
Printed in the United States of America. This report has been reproduced directly from the best available copy.
Available to DOE and DOE contractors from U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831 Telephone: (865) 576-8401 Facsimile: (865) 576-5728 E-Mail: firstname.lastname@example.org Online ordering: http://www.osti.gov/bridge Available to the public from U.S. Department of Commerce National Technical Information Service 5285 Port Royal Rd.
Springfield, VA 22161 Telephone: (800) 553-6847 Facsimile: (703) 605-6900 E-Mail: email@example.com Online order: http://www.ntis.gov/help/ordermethods.asp?loc=7-4-0#online SAND2009-0758 Unlimited Release Printed February 2009 Metallurgy, Thermal Stability, and Failure Mode of Commercial Bi-Te-based Thermoelectric Modules Nancy Yang and Alfredo Morales
Abstract Bi–Te–based thermoelectric (TE) alloys are excellent candidates for power generation modules.
We are interested in reliable TE modules for long-term use at or below 200°C. It is known that the metallurgical characteristics of TE materials and of interconnect components affect the performance of TE modules. Thus, we have conducted an extensive scientific investigation of several commercial TE modules to determine whether they meet our technical requirements. Our main focus is on the metallurgy and thermal stability of (Bi,Sb)2(Te,Se)3 TE compounds and of other materials used in TE modules in the temperature range between 25°C and 200°C.
Our study confirms the material suite used in the construction of TE modules. The module consists of three major components: AlN cover plates; electrical interconnects; and the TE legs, P-doped (Bi8Sb32)(Te60) and N-doped (Bi37Sb3)(Te56Se4). The interconnect assembly contains Sn (Sb~1wt%) solder, sandwiched between Cu conductor with Ni diffusion barriers on the outside.
Potential failure modes of the TE modules in this temperature range were discovered and analyzed. The results show that the metallurgical characteristics of the alloys used in the P and N legs are stable up to 200°C. However, whole TE modules are thermally unstable at temperatures above 160°C, lower than the nominal melting point of the solder suggested by the manufacture.
Two failure modes were observed when they were heated above 160°C: solder melting and flowing out of the interconnect assembly; and solder reacting with the TE leg, causing dimensional swelling of the TE legs. The reaction of the solder with the TE leg occurs as the lack of a nickel diffusion barrier on the side of the TE leg where the displaced solder and/or the preexisting solder beads is directly contact the TE material. This study concludes that the present TE modules are not suitable for long-term use at temperatures above 160°C due to the reactivity between the Sn-solder and the (Bi,Sb)2(Te,Se)3 TE alloys. In order to deploy a reliable TE power generator for use at or below 200 ºC, alternate interconnect materials must be used and/or a modified module fabrication technique must be developed.
Acknowledgements The authors would like to thank Miles Clift, Andy Gardea, Jeff Chames, Ryan Nishimoto, April Nissen and Dr. Neville Moody for their assistance with experiments. We also would like to thank Scott Whalen and Dr. Per Van Blarigan, Dr. Joshua Sugar, Dr. Bryan Wong and Dr. Richard Karnesky for their valuable technical inputs to this report. Special thanks also to Dr. Peter Van Blarigan for supplying the commercial modules for the project. The programmatic support and guidance from Davina Kwon, Bill Even, Tim Shepodd and Paul Spence are greatly appreciated.
2. Experimental Methods
2.2 Sample preparation
2.3 Analytical techniques
2.3.1 Materials characterization
2.3.2 Thermal stability evaluations
3. Experimental Results
3.1 As-received module dimensions and construction
3.2 Metallurgy of as-received module
3.2.1 Chemical composition
3.2.2 Microstructure and Texture
3.2.3 Fracture resistance and crack propagation
3.2.4 Vickers microhardness
3.3 Thermal stability and failure mode
3.3.1 Module configuration and dimensions
3.3.2 Metallurgical change of the base alloy for P and N legs
4. Summary and Discussion
6. Future Work
Figures Figure 1. A net electromotive force is generated by a temperature gradient across the TE module legs through the Seebeck effect
Figure 2. As-received commercial module.
Figure 3. Schematic outlining the procedure used to prepare metallographic cross sections of TE modules.
Figure 4. Typical data output of EBSP.
The color-coded map and dot inverse pole figure show the preferred crystallographic orientation relative to the three sample axes
Figure 5. The principle of a fracture toughness measurement method by G.
Anstis in 1981...... 15 Figure 6. TMA schematic and typical data output showing dimension change in the TE module upon heating.
Figure 7. Optical images of the metallographic polished cross section reveal the overall dimension and construction of the TE module
Figure 8. High-magnification optical images showing the detailed construction of the asreceived module.
Figure 9. Optical images show the solder beads in direct contact with the side of the P leg on the sideway
Figure 10. Summary of the overall TE module construction revealed by SEM /BEI images
Figure 11. Schematic of the module construction specified by the manufacturer
Figure 12. EDS results show pure Ni and Cu surface coatings.
Figure 13. EDS identified alloy composition of the materials used for each interconnect layered component.
Figure 14. SEM/BEI image and EDS show Sn-Ni-containing reaction layers at the Sn solder-Ni barrier interfaces
Figure 15. EDS shows the qualitative chemical composition of the P and N leg
Figure 16. Summary of the material design and construction of the TE module determined by SEM/ EDS analyses.
Figure 17. Schematic of the material and construction of the TE module based on the SEM/EDS analyses.
Figure 18. WDS results show alloy composition profiles across the interconnect assembly.
..... 27 Figure 19. Reaction layer revealed by a) Sn and Ni composition profile, b) SEM/BEI image and Sn and Ni x-ray intensity maps
Figure 20. Typical microstructure of P and N legs imaging by SEM/SEI (upper) and polarized optical imaging (lower).
Figure 21. SEM/BEI (upper) and SEM/SEI (lower) image pairs show pores at interparticle boundaries.
The cracks seen in these images are artifacts from the intended Vickers indentation testing
Figure 22. Sample geometry (Row I) and texture measurements (Row II for P leg and Row III for N leg)
Figure 23. Cracks initiated at the tips of the diamond indents and subsequently traveled along the longitudinal direction in both the P and N legs.
Figure 24. Zigzag propagation path corresponds to alternating intergranular and Tran granular fractures
Figure 25. Typical area size of Vickers hardness measurements.
Figure 26. TMA plot shows a big drop in dimensional change of the module as a whole at 200°C
Figure 27. Sn-solder was either missing initially or disappeared after heating TMA to 200°C and then holding for 2 hours.
Figure 28. Optical images of the modules taken before and after annealing show deterioration of interconnect assemblies at 160°C and 200°C for two hours.
........ 36 Figure 29. Optical images reveal the missing and displaced solder as well as the swollen legs at 160ºC and 200ºC for two hours.
Figure 30. Failure of the interconnect assembly seen on the cross section from the side constructed with excess Sn-solder beads.
Figure 31. Microstructure of the swollen P and N legs.
Compositional variation is seen in the P and N legs near the interconnect assemblies
Figure 32. Close-up view of the icrostructure of the swollen P and N legs.
Compositional variation is seen in the P and N legs near the interconnect assemblies
Figure 33. X-ray intensity maps show non-uniform elemental distribution adjacent to the interconnect assembly and Sn diffusion into swollen legs
Figure 34. Non-uniform composition across the swollen P and N legs, annealed at 200°C, by elemental composition profile using WDS.
Figure 35. BEI images show the locations of the WDS analyses.
Figure 36. Phase diagram of Sn-Te binary alloy.
Figure 37. Chemical composition profiles by WDS show alloy composition of the P and N legs remain unchanged between ambient temperature and 200°C for 2 hours.
...... 45 Figure 38. SEM/SEI shows minimal changes in defects and pores in the P leg between (a) as-received and (b) 200°C for 2 hours.
Figure 39. Polarized optical images show grain structure at ambient temperature, 160°C, and 200°C, holding 2 hours
Figure 40. EBSP color-coded maps show the stable texture of the P leg.
Figure 41. Construction of the commercial module.
Figure 42. Illustration of Level I and Level II failures.
Figure 43. Level II failure seen in the module constructed with Sn-Au solder.
Tables Table 1. Chemical composition (wt%) by EPMA/WDS
Table 2. Vickers microhardness at 5-gram load
Table 3. Chemical composition (at %)
Table 4. Vickers micro-hardness
Table 5. Modulus and nano-hardness
1. Introduction The construction and operation of thermoelectric (TE) power generating modules are well understood. A typical Bi-Te module has three major components. On the outside of the module one finds electrically insulating cover plates. Adhered to the inside of each cover plate are electrical interconnects that connect the TE legs in series. The interconnect assembly is a layered structure that contains a solder sandwiched between diffusion barriers with Cu traces on one side and the Bi-Te leg on the other side. Attached to the interconnects one finds of course the P-doped or N-doped TE legs.
The P and N TE legs are grouped into P-N couples connected electrically in series and thermally in parallel. A net electromotive force is generated by a temperature gradient across the TE module legs through the Seebeck effect, as shown in Figure 1 below. The Seebeck effect can be thought of as the thermally induced diffusion of charge carriers from the hot side to the cold side.
Figure 1. A net electromotive force is generated by a temperature gradient across the TE module legs through the Seebeck effect.
The efficiency of the TE module is determined by the figure-of-merit ZT=(S2σ/κ)T, where α is the Seebeck coefficient; σ is the electrical conductivity; κ is the thermal conductivity; and T is the average temperature between THot and TCold (Ref 1-3). We are interested in TE modules that will operate between 25°C and 200°C and for that temperature range, (Bi,Sb)2(Te,Se)3 alloys have the highest known ZT (Ref. 4, 5, 12).
Our ultimate goal is to develop design and construction specifications for a reliable TE module based on well-understood materials science. It is well known that small changes in the metallurgical characteristics of TE materials and of interconnect components can profoundly affect the performance of TE modules (Ref. 4, 5, 9, 12). Thus, we examined the metallurgy and thermal stability of (Bi,Sb)2(Te,Se)3 TE compounds and of other materials used in currently available commercial TE modules in the temperature range between 25°C and 200°C in an effort to determine if these modules will meet our technical requirements. The details of the scientific experiments are outlined in the following sections.
This page intentionally left blank.
2. Experimental Methods