Comparing Transition-Edge Sensor Response Times in a Modified Contact Scheme with Different Support Beams

Abstract

We present measurements of the thermal conductance, G, and effective time constants, $\tau$, of three transition-edge sensors (TESs) populated in arrays operated from 80–87 mK with T$_\mathrm{C}$ $\sim$ 120 mK. Our TES arrays include several variations of thermal architecture enabling determination of the architecture that demonstrates the minimum noise equivalent power, the lowest $\tau$, and the trade-offs among designs. The three TESs we report here have identical Mo/Cu bilayer thermistors and wiring structures, while the thermal architectures are: (1) a TES with straight support beams of 1 mm length, (2) a TES with meander support beams of total length 2 mm and with two phonon-filter blocks per beam, and (3) a TES with meander support beams of total length 2 mm and with six phonon-filter blocks per beam. Our wiring scheme aims to lower the thermistor normal state resistance R$_{N}$ and increase the sharpness of the transition $\alpha =$ dlogR/dlogT at the transition temperature T$_\mathrm{C}$. We find an upper limit of $\alpha$ given by ($25\pm 10$), and G values of 200 fW/K for (1), 15 fW/K for (2), and 10 fW/K for (3). The value of $\alpha$ can be improved by slightly increasing the length of our thermistors.