Development of Photoconductive Semiconductor Switch

Period of Performance: 07/18/2011 - 02/17/2012


Phase 1 SBIR

Recipient Firm

Ues, Inc.
4401 Dayton-Xenia Road Array
Dayton, OH 45432
Principal Investigator


OBJECTIVE: Advanced pulsed power switching technologies are needed to enable future nuclear weapon effects (NWE) experimentation capabilities and concepts for the active interrogation of special nuclear materials (SNM). The objective of this research is to advance the state of the art of high-gain optically-triggered switches by increasing the current density (to >1000 A/cm) and voltage hold-off (>67 kV/cm and >100 kV total) capabilities of complete switch assemblies that allow simple laser illumination, function in oil immersion, and have rise-times and timing jitter 100 kV) and high current (>100 kA) operations. Spark gap switches have limitations in terms of their triggering requirements, timing jitter and turn on time (both typically greater than a few nanoseconds). Photoconductive Semiconductor Switches (PCSS) are one method of switching high voltages without requiring direct electrically-connected trigger systems. PCSS have been demonstrated using silicon carbide (SiC), gallium nitride (GaN), and semi-insulating gallium arsenide (GaAs) for voltages over 100 kV with turn on times of 0.35 ns and timing jitter of ~0.1 ns. Unlike most semiconductors that only conduct as long as they are illuminated by enough light to generate current carriers, GaAs PCSS have the advantage of also being high-gain; once the device is turned on by a short laser pulse, they can remain conducting through a stable electron avalanche process. The primary issue with GaAs PCSS is that the current becomes filamentary with channel widths of ~50 micrometers and that the current per filament must be limited to 107 shots). This limit is set by the localized heating of the conducting channel and the need to keep the temperature below the melting point. If bulk GaAs is uniformly illuminated, the current tends to form a few, high-current channels that can damage the switch. Illuminating with narrow lines of laser light bridging the switching gap and spaced ~300 micrometers apart has been shown to allow multiple parallel channels to form and remain separate. However, this requirement limits the overall current density and makes the laser triggering optics more complex and/or inefficient. Research in this topic areas may address : (1) The development of techniques such as dead-bands between channels to prevent transverse current flow and the merging of neighboring channels. It may be possible to achieve this through ion implantation or other means. If the spacing of channels can be reduced to ~100 micrometers, the current density could be tripled. (2) The development of integrated focusing lens assemblies that work under transformer oil, do not reduce the voltage hold-off, and allow more efficient use of laser light. (3) The development of advanced bulk materials and contact designs that increase voltage hold-off, switch life-time (>100,000 shots) and enhance multi-channeling with uniform illumination. (4) The development of techniques to field large parallel arrays of switches that can uniformly carry high total currents with long mean-times between failures. Impact: The development of improved optically triggered switches will enable a new class of compact, rep-rated pulsed power systems for a number DoD and other applications. A immediate application would be for a short-pulsed (~10 ns) bremsstrahlung diode driver for high-fidelity NWE experimentation and potentially for active interrogation of SNM. Other DoD applications could include: ion beam drivers for gamma ray and neutron sources; microwave sources; electron beam sources for free-electron lasers, materials processing, and chem-bio remediation. There are many potential commercial applications. PHASE I: Develop a design concept for an improved PCSS and demonstrate a small-scale prototype of a single switch capable of holding off >100 kV. A final report detailing the design and prototype results will be delivered at the end of Phase 1. PHASE II: Develop a sub-scale prototype of a PCSS system capable of holding off >100 kV DC and then of triggering a discharge of over 10 kA with a conduction time of greater than 20 ns with a timing jitter of