The use of 3D structures has been shown to have the high potential for solving several problems that are currently preventing the performance improvements in lithium batteries. The high-surface area structures can provide elimination of dendrites and elimination of safety concerns, improved energy density and specific energy, high power density, low cost per kWh and improvement in cycle life. Compared with traditionally designed battery electrodes three-dimensional structures also enable short diffusion path in the active mass, better current collection and lower impedance.

A perfectly engineered three-dimensional structure is demonstrated using porous silicon. Both anode and cathode can be fabricated on the porous substrate, leading to numerous improvements in performance. A lithium metal anode in porous silicon is shown to be dendrite-free for over 200 cycles as a result geometrical effects in suppressing dendrite growth and low current density. The suppression of dendrites enables the use of lithium metal anodes and eliminating the safety concerns. Both anode and cathode also show improved specific capacity as a result of lower overall impedance and better utilization of the active material; as well as higher power density due to larger three-phase boundary and perfectly arranged active material inside uniformly spaced pores.

A complete cell based on the porous silicon structure can use any anode and cathode material; and can therefore be optimized for a particular application or requirements. It is also amenable to use of both liquid and solid electrolytes. A design is demonstrated that enables lithium metal air battery using a solid electrolyte. Another design shows a monolithic full cell within a single porous silicon wafer, eliminating the need for a separator or additional packaging.

Results of a half-cell, full cell and symmetrical cell testing are presented. The half-cell testing demonstrates the basic behavior of the porous structure and shows current densities in the excess of 5 mA/cm² (geometrical or projected surface area), but with a low current density per true surface area. The symmetrical cell testing shows stable anode cycling and absence of dendrites for over 200 cycles, while the full cell testing proves the feasibility of pairing the porous silicon metal lithium anode with several conventional cathode materials.