Non-Uniform Time-Stepping For Fast Simulation of Photodetectors Under High-Peak-Power, Ultra-Short Optical Pulses

Abstract

County (UMBC)ScholarWorks@UMBC digital repository on the Maryland Shared Open Access (MD-SOAR) platform.

Designing silicon-germanium photodetectors with numerical optimization: the tradeoffs and limits

“…This approach dramatically reduces the number of time steps used for the dynamic analysis. 5 To verify the accuracy of this solver, we first study an experimentally characterized Si-Ge photodetector, 1 shown in Fig. 1(a).…”

“…1 However, designing a high-performance photodetector or even improving the performance of an already existing design is a challenging task due to the required computation time, difficulties in estimating the sensitivity of the device to the design parameters, and the existence of design constraints. [2][3][4] To overcome this challenge, we recently developed an efficient drift-diffusion equations solver that uses a non-uniform time-stepping 5 and both single-frequency and broadband excitations 6 to study photodetectors that have several layers of semiconducting materials with varying thicknesses and doping concentrations. With this numerical solver, we can calculate photodetectors' phase noise, quantum efficiency, and response time to estimate their stability, efficiency, and speed.…”

Enhancing the Resolution of Local Near-Field Probing Measurements With Machine Learning

“…For our numerical study, we selected photodetectors [12], [13], [14], [15] as the device under investigation. By solving the drift-diffusion equations [12] on nonuniform spatial and temporal meshes [13], using monochromatic or broadband excitations [14], we can accurately and efficiently calculate both the field and current distributions along the photodetector. Particularly, with the nonuniform time-stepping capability, we can analyze thousands of photodetectors in a few hours using regular personal computers [14], [15].…”

“…Other than some peaks occurring at the interfaces, the field strength is typically high only in the intrinsic (i) region. We calculate the electric field profiles with the aforementioned drift-diffusion equations solver [12], [13], [14], [15]. Throughout this work, we will refer to these field profiles as the "true" field profiles.…”

Designing Silicon-Germanium Photodetectors with Numerical Optimization: The Tradeoff Between Quantum Efficiency & Phase Noise

“…To determine the stability, efficiency, and speed of a photodetector, one must measure or calculate the phase noise, quantum efficiency, and response time of the photodetector. We recently developed an efficient drift-diffusion equations solver that uses a non-uniform time-stepping [1] and both single-frequency and broadband excitations to calculate the phase noise [2], quantum efficiency [2], and bandwidth [3] of photodetectors that have several layers of semiconducting materials with varying thicknesses and doping concentrations.…”