Thermoelectric materials provide a means for converting heat into electrical power using a fully solid-state device. Power-generating devices (which include individual couples as well as multi-couple modules) require the use of n-type and p-type thermoelectric materials, typically comprising highly doped narrow band-gap semiconductors which are connected to a heat collector and electrodes.
Despite the favorable attributes of the bulk YMS material, it must ultimately be incorporated into a power-generating device using a suitable joining technology. Typically, processes such as diffusion bonding and/or brazing are used to join thermoelectric materials to the heat collector and electrodes, with the goal of providing a stable, ohmic contact with high thermal conductivity at the required operating temperature.
Since YMS is an inorganic compound featuring chemical bonds with a mixture of covalent and ionic character, simple metallurgical diffusion bonding is difficult to implement. Furthermore, the Sb within YMS readily reacts with most metals to form antimonide compounds with a wide range of stoichiometries. Although choosing metals that react to form high-melting-point antimonides could be employed to form a stable reaction bond, it is difficult to limit the reactivity of Sb in YMS such that the electrode is not completely consumed at an operating temperature of 1,000 °C. Previous attempts to form suitable metallization layers resulted in poor bonding, complete consumption of the metallization layer or fracture within the YMS thermoelement (or leg).
An approach to forming a stable metallization layer to consolidated YMS parts or legs has been developed, however, using thin molybdenum foils. The foil is diffusion bonded to the YMS part at temperatures above 900 °C, under the application of adequate pressure for several hours. Use of a thin foil eliminates the fracture typically observed within YMS parts when thicker foils are used (induced by thermomechanical stresses) as seen in the figure. The metal can be bonded prior to dicing of the legs from a hotpressed/consolidated YMS form, or it can be bonded to a pre-cut leg of the correct geometry. The contact is thermally stable with respect to YMS, exhibiting no de-bonding or increase in contact resistance after a 1,000 °C vacuum heat treatment for 1,500 hours. This metal lization layer may then be bonded or brazed to other components, such as heat collectors or current-carrying electrodes. To enable successful bonding of the molybdenum metallization requires the preparation of YMS legs of sufficient strength, which is facilitated by the hot pressing of YMS powders above 900 °C (using proper cooling rates to minimize residual stress formation and avoid fracture of the parts).
This metallization layer may then be bonded or brazed to other components, such as heat collectors or current-carrying electrodes. Furthermore, it can be implemented at both the hot and cold sides of the leg. It can also be applied as a metallization layer to other similar compositions of thermoelectric materials. The specific nature of the interaction between Mo and YMS is still under investigation; however, it is clear the molybdenum reacts sufficiently to form an adequate bond, without the extensive reaction observed with similar metals such as nickel, niobium, or titanium. The process may be similar to that involved in the formation of other metal/ceramic bonds, wherein the metal has limited solubility in the ceramic material (for example, in diffusion bonded metal/alumina joints).
This work was done by Samad Firdosy, Billy Chun-Yip Li, Vilupanur Ravi, Jeffrey Sakamoto, Thierry Caillat, Richard C. Ewell, and Erik J. Brandon of Caltech for NASA’s Jet Propulsion Laboratory. NPO-46670
This Brief includes a Technical Support Package (TSP).
Metallization for Yb14MnSb11-Based Thermoelectric Materials
(reference NPO-46670) is currently available for download from the TSP library.
Don't have an account?
Overview
The document titled "Metallization for Yb14MnSb11-Based Thermoelectric Materials" presents research conducted at NASA's Jet Propulsion Laboratory (JPL) on advanced thermoelectric materials, specifically focusing on the compound p-type semiconductor Yb14MnSb11 (YMS). Thermoelectric materials are essential for converting heat into electrical power in solid-state devices, and they typically require two different materials—n-type and p-type semiconductors—connected in a junction to function effectively.
YMS has been identified as a promising candidate for thermoelectric applications, particularly in space-based radioisotope thermoelectric generators (RTGs), due to its high figure of merit (zT value) at elevated temperatures, specifically around 1000°C. The document discusses the challenges associated with integrating YMS into power-generating thermocouples, emphasizing the need for effective metallization techniques to ensure stable and efficient electrical contacts at high temperatures.
The metallization process involves using metallurgical or diffusion bonding techniques to create a reliable interface between the thermoelectric material and the electrodes. The document details a specific metallization scheme that employs molybdenum (Mo) foils, which are bonded to the YMS material. This approach has demonstrated a low contact resistance of ≤25 μΩ-cm² and has shown thermal stability, with no significant increase in contact resistance after extensive heat treatment (1500 hours at 1000°C).
The research also explores the interaction between Mo and YMS, noting that the bonding process likely involves the diffusion of molybdenum atoms into the YMS material. Importantly, no evidence of compound formation between Mo and YMS has been detected, indicating a stable bonding mechanism. The document includes microscopic images and resistance measurements that illustrate the effectiveness of the metallization layer before and after thermal treatment.
Overall, this research highlights the potential of Yb14MnSb11 as a high-performance thermoelectric material and the importance of developing suitable metallization techniques to enhance the efficiency and reliability of thermoelectric devices. The findings contribute to the broader field of thermoelectric technology, with implications for aerospace applications and beyond. The work was conducted under NASA's Commercial Technology Program, aiming to make aerospace-related developments accessible for wider technological and commercial applications.
