Abstract
Interconnect materials for high operating temperature applications are becoming a limiting factor within the chain of materials. While materials such as capacitor dielectrics, semiconductor platforms (e.g. SiC), and baseplate materials (e.g. SiN composites) have paved a pathway to deploying electronics in high operating temperature applications, interconnect materials are a clearly identified weak link. As is often the case in advancements in technology, the materials technologies that were the bottlenecks to advancement give way to new solutions that create new bottlenecks in the material supply chain. Rather than a fluid march towards advancements in new frontiers in electronics, the high operating temperature sphere, like much of advanced electronic, suffers from a ‘slip-fault’ mode of development where advances occur in one segment while others lag behind creating drag on implementation.
For high operating temperature applications the available interconnect solutions are becoming the jarring stop to the smooth tectonic shift. Current solutions are diverse: high-lead, gold-based, and nano-sintering and its hybrids, but none are ideal. Even disregarding he toxicity of lead and the ongoing limbo of its regulatory status, the operating temperatures of the high-lead solders are on the low end of the requirements for future harsh environment electronics applications; whereas, the gold and nano-based alternatives have significant cost barriers - either at from the constituent materials perspective or the required investment in new processes. There is also the concern about the assessment of the action of nanomaterials in the waste stream due to their fundamentally different surface reactivity in a variety of situations.
Reliance on conventional, solder-type interconnection structures, regardless of composition, introduces the perennial problem of the growth of the interfacial phases due to the essentially unlimited volume of the bulk solder material. The changes in the interfacial structure with additional thermal work - as is provided by high operating temperature applications - creates an environment that is ripe for growth of a variety of failure mechanisms. These failure mechanisms are often related to the uncontrolled laminar growth of intermetallic phases at the interfaces and the mechanical characteristics of these intermetallic phases in comparison with the materials joined and the bulk constituent material of the solder.
An alternative class of interconnect materials, transient liquid phase sintering (TLPS) pastes, introduce a joint microstructure that is homogeneous throughout. The interfacial metallurgical reactions with the solderable surfaces are fundamentally similar to those that occur throughout the bulk of the joint. A reactant metal is included in the composition. This reactant metal, most often copper, reacts with and converts the bulk tin in the bulk of the solder interconnect to alloy structures with melting points well above the operating temperatures currently contemplated. At the conclusion of the joining process, which is generally a near drop-in for existing solder reflow processes, there is no large source of unreacted metal (e.g. Sn) that can continue to drive major microstructural changes with the continued thermal work provided by the application environment. For this reason the joints are homogeneous and do not have the free reactants necessary to drive substantial changes in joint morphology during cycling and use conditions.
In this paper, we will explore the differences between TLPS joints and solder-type joints with the anticipated thermal work that would be introduced in a high operating temperature environment.
High operating temperature InAs/GaSb type-II superlattice detectors on GaAs substrate for the long wavelength infrared
