Advanced numerical modeling and characterization of infrared focal plane arrays

Mercury cadmium telluride (HgCdTe) is the detector material of choice for high performance infrared (IR) focal plane arrays (FPAs). Unique among semiconductor materials, HgCdTe features a widely tunable, direct bandgap from the ultraviolet to very long wavelength IR, lattice matched substrates, excellent quality epitaxial layers, and long minority carrier lifetimes. The requirements for better HgCdTe device performance are driving advancements in numerical modeling and characterization. In particular, numerical models and physical device simulations have become indispensable tools to understand the physics and optimize the operation of complex HgCdTe pixel arrays. Existing limitations in current numerical models prevent the study of new, higher performance pixel designs with much higher resolutions necessitating smaller pixels and novel structures that take advantage of the wave nature of light. Additionally, the costs associated with larger FPAs are driving the development of non-destructive testing methods during production and assembly. This thesis describes the contributions I have made to both the development of novel models for physical device simulation and to innovative characterization techniques for FPAs with read-out integrated circuits (ROICs). I have developed an integrated model for simulating the electromagnetic and electrical response of pixel arrays. I describe the results I have obtained applying the newly developed simulation approach to study planar detectors, small pixel mesa-type detectors, and photon trapping (PT) structures. I have examined the effect that reducing pixel pitch has on performance and the benefits of detectors with PT structures. The outcome of the work on the PT structured detector has led to the development of a new detector by BAE Systems, Inc. I will also present work towards the development of a temporary hybridization method. HgCdTe-based detectors are hybrid structures where HgCdTe FPAs are indium bump bonded to silicon ROICs. FPAs are tested by permanently hybridizing an array to a ROIC and both must be discarded if an FPA fails. The latest generation arrays are larger than 7 cm square and the corresponding ROICs are increasing in cost and complexity. Using the temporary hybridization technique I have evaluated, the array and ROIC can be nondestructively separated in case of a test failure for reuse.

Grantor dc:publisher

Boston University

Year dc:date.issued

2012

Author dc:creator

• Keasler, Craig Alan

Language dc:language.iso

en_US

Dc Identifier Other

b38909479

OAI identifier oai:identifier

oai:open.bu.edu:2144/31575