Two-Stage Dc-Squid Amplifier Achieves Low Noise with 100 SQUID Cells

Why It Matters

The ultra‑low noise readout dramatically improves the sensitivity of TES‑based CMB instruments, enabling more precise polarization measurements and deeper cosmological insights. Its scalable design can elevate performance across diverse low‑temperature detector platforms.

Key Takeaways

• Two‑stage dc‑SQUID achieves ~1 µΦ₀/√Hz noise.
• Input SQUID uses four cells; SSA contains 100 cells.
• Noise meets requirements for CMB TES detectors.
• Voltage swing reaches 3 mV; VΦ ≈10 mV/Φ₀.
• Enables lower NEI (≈4 pA/√Hz) than typical TES readouts.

Pulse Analysis

Superconducting quantum interference devices (SQUIDs) have long been the workhorse for ultra‑sensitive magnetic detection, especially when paired with transition‑edge sensor (TES) arrays. As astronomical experiments push toward ever‑fainter signals—such as the polarized imprint of the cosmic microwave background—readout electronics must deliver noise floors well below the intrinsic detector noise. The two‑stage dc‑SQUID presented by Li and colleagues addresses this challenge by integrating a compact four‑cell input stage with a 100‑cell series array, leveraging Nb/AlAlOₓ/Nb Josephson junctions to achieve a flux noise near 1 µΦ₀/√Hz, a benchmark that rivals the best single‑stage designs.

The technical innovation lies in the cascaded architecture and careful biasing strategy. The input SQUID’s double‑transformer layout, complemented by dummy cells, optimizes gradiometric coupling while the series array amplifies the signal with a voltage swing of roughly 3 mV and a flux‑conversion coefficient VΦ of about 10 mV/Φ₀. These parameters translate into a system‑noise‑equivalent current of ~4 pA/√Hz, an order of magnitude lower than conventional TES readouts that typically hover around 100 pA/√Hz. Such low noise directly improves the signal‑to‑noise ratio for CMB polarization measurements, where detecting micro‑kelvin variations is essential.

Beyond cosmology, the amplifier’s performance opens doors for a spectrum of low‑temperature detector applications, from millimeter‑wave astronomy to X‑ray spectroscopy and even dark‑matter searches that rely on minute energy deposits. The demonstrated scalability suggests that larger TES arrays can be read out without compromising sensitivity, potentially reducing system complexity and cost. As next‑generation observatories like AliCPT‑40G prepare for deployment, this two‑stage SQUID architecture positions itself as a critical enabling technology, promising richer data sets and deeper scientific returns across the quantum‑sensing landscape.