Automation of MKID Simulations for Array Building with AEM (Automated Electromagnetic MKID Simulations)

TL;DR

This paper presents AEM, a software tool that automates electromagnetic simulations of MKID arrays, improving pixel yield and accuracy over manual or interpolated methods, demonstrated on a 100-pixel array.

Contribution

Introduction of AEM, an automated simulation software that constructs and simulates MKID arrays, enhancing efficiency and accuracy in resonant frequency and Qc predictions.

Findings

Automated designs increase pixel yield by avoiding interpolation losses.

Resonance frequency and Qc values are more accurate with AEM.

Simulation of 100 MKIDs completed in under 11 hours with high precision.

Abstract

Microwave Kinetic Inductance Detectors (MKIDs) are photon detectors comprised of superconducting LC resonators with unique resonant frequencies corresponding to their geometrical structure. As each pixel has its own geometry, electromagnetic simulations by hand of every pixel in a kilo-pixel array are impractical. Simulating fewer pixels and interpolating in between risks reduced pixel yield in arrays due to overlapping resonant frequencies. We introduce a new software called AEM (Automated Electromagnetic MKID simulations) that automates the constructions and simulations of every simulated MKID pixel in an array according to specified resonant frequencies and a Qc range. We show automated designs to have an increased pixel yield (avoiding loses due to interpolation completely), increased accuracy in resonance frequency and Qc values when compared to interpolated structures. We also demonstrate a simulated trial of AEM for 100 MKIDs between 4 & 8 GHz to produce MKIDs with accuracies of +-0.2 MHz with a runtime of 10 hrs 45 mins.