You are here

Fully Automated Quantum Cascade Laser Design Aided by Machine Learning with up to 100X Design Cycle Time Reduction

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N68936-20-C-0083
Agency Tracking Number: N20A-T003-0280
Amount: $139,890.00
Phase: Phase I
Program: STTR
Solicitation Topic Code: N20A-T003
Solicitation Number: 20.A
Solicitation Year: 2020
Award Year: 2020
Award Start Date (Proposal Award Date): 2020-06-24
Award End Date (Contract End Date): 2020-12-28
Small Business Information
ORLANDO, FL 32826-1111
United States
DUNS: 052971564
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Hong Shu
 (407) 733-9776
Business Contact
 Elena Lyakh
Phone: (310) 720-3286
Research Institution
 University of Central Florida
 Arkadiy Lyakh
4000 Central Florida Blvd
Orlando, FL 32816-1111
United States

 (407) 823-0699
 Nonprofit College or University

A Quantum Cascade Laser's (QCL) core material is a series of nanometer scale layers of conduction band barrier and well materials designed to induce lasing electron energy levels. The key design feature of a QCL is the ability to repeat the laser core superlattice design to cascade electrons through the superlattice repetitions by repeated stimulated emission. Such designs are generally carefully created manually by well trained scientists with a strong intuition for quantum behavior and quantum well simulation programs. A method is proposed to design and evaluate superlattice designs by training a neural network to replicate the intuition of the superlattice designer and the quantitative results of the superlattice simulation. Training such a neural network with thousands of example designs and evaluations will allow the network to produce usable QCL superlattice designs without human assistance.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government