#
2009 APS March Meeting

## Volume 54, Number 1

##
Monday–Friday, March 16–20, 2009;
Pittsburgh, Pennsylvania

### Session T3: Keithley Award Session (GIMS)

2:30 PM–5:30 PM,
Wednesday, March 18, 2009

Room: 301/302

Sponsoring
Unit:
GIMS

Chair: James Matey, US Naval Academy

Abstract ID: BAPS.2009.MAR.T3.2

### Abstract: T3.00002 : Near-Quantum-Limited SQUID Amplifier

3:06 PM–3:42 PM

Preview Abstract
Abstract

####
Author:

John Clarke

(University of California, Berkeley and Lawrence Berkeley National Laboratory)

The SET (Single-Electron Transistor), which detects charge, is
the dual of
the SQUID (Superconducting QUantum Interference Device), which
detects flux.
In 1998, Schoelkopf and co-workers introduced the RFSET, which
uses a
resonance circuit to increase the frequency response to the
100-MHz range.
The same year saw the introduction of the Microstrip SQUID
Amplifier$^{1}$
(MSA) in which the input coil forms a microstrip with the SQUID
washer,
thereby extending the operating frequency to the gigahertz range.
I briefly
describe the theory of SQUID amplifiers involving a tuned input
circuit with
resonant frequency f. For an optimized SQUID at temperature T,
the power
gain and noise temperature are approximately G = f$_{p}$/$\pi $f
and T$_{N}$
= 20T(f/f$_{p})$, respectively; f$_{p}$ is the plasma frequency
of one of
the Josephson junctions. Because the SQUID voltage and current
noise are
correlated, however, the optimum noise temperature is at a
frequency below
resonance. For a phase-preserving amplifier, T$_{N}$ =
($\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/
\kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $ + A)hf/k$_{B}$,
where Caves'
added noise number A = $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/
\kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $ at the quantum
limit.
Simulations based on the quantum Langevin equation (QLE) suggest
that the
SQUID amplifier should attain A = $\raise.5ex\hbox{$\scriptstyle
1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $.
We have
measured the gain and noise of an MSA in which the resistive
shunts of the
junctions are coupled to cooling fins to reduce hot electron
effects. The
minimum value A = 1.1 $\pm $ 0.2 occurs at a frequency below
resonance. On
resonance, the value A = 1.5 $\pm $ 0.3 is close to the
predictions of the
QLE, suggesting that this model may fail to predict the
cross-correlated
noise term correctly. Indeed, recent work suggests that a fully
quantum
mechanical theory is required to account properly for this
term$^{2}$. This
work is in collaboration with D. Kinion and supported by DOE BES.
$^{1}$M. Mueck, \textit{et al}., \textit{Appl. Phys. Lett.}
\textbf{72}, 2885 (1998).
$^{2}$A. Clerk, \textit{et al.,} http://arxiv.org/abs/0810.4729.

To cite this abstract, use the following reference: http://meetings.aps.org/link/BAPS.2009.MAR.T3.2