3-D solid state battery architectures and fabrication

3-D structured all-solid-state batteries are a natural – and explicitly chosen – evolution of our group’s research. In seeking to fabricate 3-D structures, thin film technology is inherently centered on solid state materials. To maximize performance (e.g., simultaneous high power and energy), specific design ground rules are desired (e.g. relative material thicknesses, 3-D shapes). Interdigitated (interpenetrating) electrode architectures – representing a particularly efficient battery design – must be self-aligned, a requirement easily met by thin film technology. And using all solid state materials dramatically increases safety as compared to conventional batteries with flammable liquid electrolytes.

Our recent demonstration of a 3-D solid state battery array is instructive and promising. It follows onto prior regular energy storage arrays as electrostatic capacitor and liquid electrolyte nanobattery arrays formed from AAO nanopore. Here we patterned micropores in Si wafers and created ALD layers for anode, cathode, solid electrolyte, and current collectors as fully interdigitated electrodes, achieving proof of concept that such energy storage structures can markedly advance thin film solid state batteries and ultimately compete with conventional liquid electrolyte batteries.

Electronics applications

Growing research interest in solid state batteries parallels some of the emerging applications for microfabricating batteries. The need for small, high performance batteries is pervasive, from personal electronic devices to sensor/actuator systems and biomedical devices. Thus we anticipate increasing diversity of our research program with regard to application drivers.

Neuromorphic computing

A novel application for solid state batteries has been demonstrated by our collaborator A.A. Talin at Sandia, who showed that as charge/discharge of a battery changes the ion content in an electrode, it modifies properties of that electrode in a nonvolatile way. In the cases reported, the cathode resistance – measured by a revised contact structure – served as a nonvolatile analog memory signal.

Given that thin film technology already allows us to pattern small solid state battery devices with highly controlled thin layers and smooth interfaces, the projections for a nonvolatile analog memory technology are very promising. The potential impact is a new device paradigm for neuromorphic computing, just as the field is growing rapidly. We look forward to working in this area.