Ultracompact Inductorless Noise-Canceling LNAs in 40-nm CMOS Achieving 2.2-K Noise Temperature for Qubit Readout

This article presents two ultracompact inductorless low-noise amplifiers (LNAs) in 40-nm CMOS for cryogenic qubit readout. Both LNAs utilize an inverter-based input stage as the main amplifier, and LNA-I and LNA-II employ a complementary and common-source (CS) stage, respectively, as the auxiliary amplifier for achieving noise canceling (NC). LNA-I also incorporates self-forward body bias (SFBB) to counter V <tex-math notation="LaTeX">ₜₕ</tex-math> reduction and improve r <tex-math notation="LaTeX">ₒᵤₜ</tex-math> during cryogenic operation, while LNA-II uses source-follower feedback(SFFB) to enhance gain and noise figures (NFs) without compromising input impedance matching. At room temperature (RT), LNA-I achieves a measured gain ( S <tex-math notation="LaTeX">_21</tex-math> ) of 25.6 dB and a minimum NF of 0.63 dB at 2 GHz. At 4 K, it demonstrates a measured gain of 29 dB and a minimum NF of 0.033 dB (corresponding to a noise temperature T <tex-math notation="LaTeX">_N</tex-math> of 2.2 K), along with a 3-dB bandwidth f <tex-math notation="LaTeX">3\,dB</tex-math> from 10 MHz to 3 GHz, while consuming 19.4 mW. LNA-II achieves a measured gain of 27 dB and a minimum NF of 1.16 dB at 0.5 GHz at RT. At 4 K, it demonstrates a measured gain of 30.2 dB and a minimum NF of 0.29 dB ( T <tex-math notation="LaTeX">_N</tex-math> of 20 K), accompanied by a f <tex-math notation="LaTeX">3\,dB</tex-math> from 20 MHz to 4 GHz and with a total power consumption of 10.8 mW. The circuit only occupies an active area of 0.018 and 0.0072 mm <tex-math notation="LaTeX">^2</tex-math> for LNA-I and LNA-II, respectively.