Efficient Mid-Infrared Single-Photon Counting Kinetic Inductance Detectors

Abstract

The search for extraterrestrial life is currently one of the main topics of astronomical research. One of the methods for this is to look for the spectral lines of molecules needed for life, called biosignatures, in the atmosphere. The Large Interferometer for Exoplanets (LIFE) mission will explore the mid-IR in the search for biosignatures. This mission is a spectrometer that uses nulling interferometry to cancel the signal from the nearby star. A spectrometer then splits the remaining planet signal to find biosignatures. The wavelength band of LIFE is 4-18.5 µm. LIFE places a strict requirement on the dark current: ID ≤ 10−2 counts/sec. Kinetic Inductance Detectors (KIDs) are a type of superconducting detectors that are single-photon counting, with inherent energy-resolving power, for wavelengths down to 25 µm. KIDs do not have a dark current and, therefore, meet the detector requirements for LIFE. This absence of dark current makes KIDs ideal for other low-light level applications in the mid-IR.

In this work, we present the design of lens-absorber coupled single-photon counting KIDs for two wavelengths within the band of LIFE: 10 µm and 18.5 µm. The absorber consists of two meandering 23 nm thick Al lines (ρ = 2.4 × 10−8 Ω m) with widths of 250 nm and 200 nm for 18.5 µm and 10 µm, respectively. The absorber is simulated using the Floquet theorem in CST. The absorber is divided into unit cells simulated with periodic boundary conditions. These unit cells must be smaller than 0.8λd to avoid crosstalk between absorbers. The simulation is performed for all angles from broadside incidence up to the subtended angle of the lens. These simulations show absorption efficiencies for the TE and TM fundamental Floquet modes of up to 81% and 65% for the 18.5 µm design and the 10 µm design, respectively. The absorber acts dominantly as an inductor and is coupled to a coplanar waveguide (CPW) to tune the resonance frequency of the KIDs.

We measured one of the designed chips: a chip with 18.5 µm lens-absorber coupled KIDs on a substrate. We performed power-integrating measurements to estimate the optical coupling of these devices. For these KIDs, we expect a total optical coupling of 1.77. This coupling is calculated with respect to a single mode and 1 polarization. Our absorber can couple to 2 polarizations and multiple modes, resulting in a coupling of larger than 1. The measurements show a reasonable match between the expected and measured optical coupling.

We also performed single-photon counting experiments. The KIDs showed a low resolving power, making distinguishing the source’s photons from noise and cosmic rays impossible. Phonon losses due to the substrate were assumed to cause this. Therefore, another chip was measured with 25 µm lens-absorber coupled KIDs on a membrane. The two chips share identical design and fabrication processes, differing only in foundation: one is built on a membrane, the other on a substrate. The KIDs on the membrane showed a relatively high resolving power of approximately 4. This confirmed the hypothesis that phonon losses were the cause of the low resolving power for the KIDs on a substrate.

Future work should focus on redesigning the absorbers and KIDs for use on a membrane and adding a quarter-wavelength backing reflector and matching layer to increase the optical efficiencies of the devices.