Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Applied Physics Letters

Low-noise tunnel junction dc SQUID’s

Michael W. Cromar, P. Carelli

Published May 1, 1981• 87 citations
Publisher

Abstract

We present the results of an investigation of thin-film dc super conducting quantum interference device (SQUID’s) with very high energy resolution. The SQUID’s were fabricated using integrated-circuit techniques appropriate for superconducting devices. A weakly coupled SQUID with very low inductance (1 pH) had an intrinsic energy sensitivity 6×10−34 J/Hz = 0.9 h. A well-coupled SQUID (coupling constant α = 0.43) had an energy sensitivity referred to the input coil of 4.7×10−32 J/Hz = 71 h.

DOI
10.1063/1.92472

Related papers

An ultra-low-noise tunnel junction dc SQUID

M. B. Ketchen, Richard F. Voss

The design, fabrication, and noise properties of an ultra-low-noise tunnel junction dc SQUID (superconducting quantum interference device) are presented. The two resistively shunted Pb-alloy tunnel junctions are each 3.6 μm diam. The SQUID has an inductance L of ≈11 pH, a critical current per junction, I0 of ≈100 μA, and a resistance per shunt R of ≈2 Ω. At 2.3 K, eI0R≈kBT, and simple theory predicts that the minimum intrinsic energy sensitivity should be ≈2h. The actual minimum value of the measured high-frequency (70 kHz) flux noise at 4.2 K is φn=1.3×10−7φ0 Hz−1/2. This corresponds to an intrinsic energy sensitivity, φ2n/2L, of 3.2×10−33 J Hz−1 or 5h. Lowering the bath temperature to 1.8 K gave no improvement in performance.

Applied Physics Letters•Nov 15, 1979•61 citations

Very low noise, tightly coupled, dc SQUID amplifiers

B. Muhlfelder, J. A. Beall, Michael W. Cromar +1 more

We have fabricated and tested thin film, niobium edge junction, double transformer, dc superconducting quantum interference devices (SQUID’s) that were stable under room-temperature storage and thermal cycling and that had very good noise performance. The input inductance, approximately 1.7 μH, was large enough to facilitate good matching to many experiments. When the SQUID was operated as a small-signal amplifier, the minimum detectable energy per unit bandwidth (Se) was 5×10−33 J/Hz at 100 kHz, referred to the SQUID loop (uncoupled). The minimum detectable energy per unit bandwidth was 1.8×10−31 J/Hz at 100 kHz, referred to the input coil. The SQUID’s had good characteristics for flux-locked operation since the small signal Se was low over a substantial range of bias current and magnetic flux. For operation in a flux-locked feedback circuit, Se was 6×10−32 J/Hz at 1 kHz.

Applied Physics Letters•Oct 27, 1986•19 citations

Ultra-low-noise tunnel junction dc SQUID with a tightly coupled planar input coil

M. B. Ketchen, J. M. Jaycox

We present a new superconducting quantum interference device (SQUID) that combines ultra-low-noise performance with tight coupling to a useable input coil. The inductive loop of the SQUID consists of a superconducting square washer that is slit in a low inductance fashion from its central hole to an outside edge where two Josephson tunnel junctions are incorporated. The inductance of the SQUID is about 90 pH, and each junction has a critical current of 15 μA and a resistive shunt of 4.5 Ω. The 190-nH input coil is a planar spiral of 50 turns situated immediately above the washer and coupled to the SQUID with a coupling constant k2 of 0.86. The SQUID has a minimum coupled energy sensitivity of 19h at 4.2 K and 8h at 1.5 K, in good agreement with theory. The crossover between white and low frequency noise occurs at ∼250 Hz.

Applied Physics Letters•Apr 15, 1982•248 citations

Low noise niobium dc SQUID with a planar input coil

V. J. de Waal, P. van den Hamer, T. M. Klapwijk

A practical all-niobium dc superconducting quantum interference device (SQUID) with a niobium spiral input coil has been developed. The SQUID utilizes submicron Josephson junctions. The best intrinsic energy resolution obtained with a 1-nH SQUID is 4×10−32 J/Hz. A 20-turn 1.2-μH input coil is coupled to a 2.3-nH SQUID with an efficiency of 0.5. The energy resolution with respect to the coil is 1×10−30 J/Hz.

Applied Physics Letters•Feb 15, 1983•12 citations

Use of low-noise dc-SQUIDs in fundamental physics experiments

P. Carelli, Maria Gabriella Castellano, Roberto Leoni +1 more

The noise performance of dc-SQUIDs (Superconducting Quantum Interference Devices), which nowadays can reach values close to the quantum limit, makes them very attractive as amplifiers for experiments using low temperatures and demanding for extremely high sensitivity. After reviewing the working principle of the dc-SQUID, we describe a thin-film device with a new structure, which allowed us to reach an energy sensitivity of 5.5 h- at 0.9 K and a good coupling with a large input coil. At last we discuss some fundamental physics experiments which take advantage from the SQUID use.

Journal de Physique IV (Proceedings)•Jun 1, 1998•0 citations

A coupled DC SQUID with low 1/f noise

Heikki Seppä, M. Kiviranta, Alexandre Satrapinski +3 more

A low-noise coupled DC superconducting quantum interference device (SQUID) especially optimized for low frequency is discussed. Using large Josephson junctions and a low loop inductance the contribution of the critical current fluctuation to 1/f noise can be minimized. To minimize the inductance the Josephson junctions are placed in the center of the washer close to the SQUID loop. A scaled-up copper model of the SQUID and the signal coil is used to analyze the impedance of the SQUID loop affected by the signal coil. An equivalent circuit model describing the effective inductance of the SQUID loop as a function of the frequency is used to design appropriate damping of the resonances. The DC SQUID characteristics are smooth and the noise performance of the SQUID does not markedly suffer from resonances. The contribution of the 1/f noise at 1 Hz is found to be about 5*10/sup -7/ Phi /sub 0// square root Hz.>

IEEE Transactions on Applied Superconductivity•Mar 1, 1993•25 citations
We present the results of an investigation of thin-film dc super conducting quantum interference device (SQUID’s) with very high energy resolution. The SQUID’s were fabricated using integrated-circuit techniques appropriate for superconducting devices. A weakly coupled SQUID with very low inductance (1 pH) had an intrinsic energy sensitivity 6×10−34 J/Hz = 0.9 h. A well-coupled SQUID (coupling constant α = 0.43) had an energy sensitivity referred to the input coil of 4.7×10−32 J/Hz = 71 h.
The design, fabrication, and noise properties of an ultra-low-noise tunnel junction dc SQUID (superconducting quantum interference device) are presented. The two resistively shunted Pb-alloy tunnel junctions are each 3.6 μm diam. The SQUID has an inductance L of ≈11 pH, a critical current per junction, I0 of ≈100 μA, and a resistance per shunt R of ≈2 Ω. At 2.3 K, eI0R≈kBT, and simple theory predicts that the minimum intrinsic energy sensitivity should be ≈2h. The actual minimum value of the measured high-frequency (70 kHz) flux noise at 4.2 K is φn=1.3×10−7φ0 Hz−1/2. This corresponds to an intrinsic energy sensitivity, φ2n/2L, of 3.2×10−33 J Hz−1 or 5h. Lowering the bath temperature to 1.8 K gave no improvement in performance.
We have fabricated and tested thin film, niobium edge junction, double transformer, dc superconducting quantum interference devices (SQUID’s) that were stable under room-temperature storage and thermal cycling and that had very good noise performance. The input inductance, approximately 1.7 μH, was large enough to facilitate good matching to many experiments. When the SQUID was operated as a small-signal amplifier, the minimum detectable energy per unit bandwidth (Se) was 5×10−33 J/Hz at 100 kHz, referred to the SQUID loop (uncoupled). The minimum detectable energy per unit bandwidth was 1.8×10−31 J/Hz at 100 kHz, referred to the input coil. The SQUID’s had good characteristics for flux-locked operation since the small signal Se was low over a substantial range of bias current and magnetic flux. For operation in a flux-locked feedback circuit, Se was 6×10−32 J/Hz at 1 kHz.
We present a new superconducting quantum interference device (SQUID) that combines ultra-low-noise performance with tight coupling to a useable input coil. The inductive loop of the SQUID consists of a superconducting square washer that is slit in a low inductance fashion from its central hole to an outside edge where two Josephson tunnel junctions are incorporated. The inductance of the SQUID is about 90 pH, and each junction has a critical current of 15 μA and a resistive shunt of 4.5 Ω. The 190-nH input coil is a planar spiral of 50 turns situated immediately above the washer and coupled to the SQUID with a coupling constant k2 of 0.86. The SQUID has a minimum coupled energy sensitivity of 19h at 4.2 K and 8h at 1.5 K, in good agreement with theory. The crossover between white and low frequency noise occurs at ∼250 Hz.
The noise performance of dc-SQUIDs (Superconducting Quantum Interference Devices), which nowadays can reach values close to the quantum limit, makes them very attractive as amplifiers for experiments using low temperatures and demanding for extremely high sensitivity. After reviewing the working principle of the dc-SQUID, we describe a thin-film device with a new structure, which allowed us to reach an energy sensitivity of 5.5 h- at 0.9 K and a good coupling with a large input coil. At last we discuss some fundamental physics experiments which take advantage from the SQUID use.
A low-noise coupled DC superconducting quantum interference device (SQUID) especially optimized for low frequency is discussed. Using large Josephson junctions and a low loop inductance the contribution of the critical current fluctuation to 1/f noise can be minimized. To minimize the inductance the Josephson junctions are placed in the center of the washer close to the SQUID loop. A scaled-up copper model of the SQUID and the signal coil is used to analyze the impedance of the SQUID loop affected by the signal coil. An equivalent circuit model describing the effective inductance of the SQUID loop as a function of the frequency is used to design appropriate damping of the resonances. The DC SQUID characteristics are smooth and the noise performance of the SQUID does not markedly suffer from resonances. The contribution of the 1/f noise at 1 Hz is found to be about 5*10/sup -7/ Phi /sub 0// square root Hz.>