squids then and now: from slugs to axions · • tangent magnetometer • magnet and mirror...

SQUIDs Then and Now: From

SLUGs to Axions

Josephson 50th Anniversary

Applied Superconductivity Conference

Portland, OR

• SQUIDs Then

• SQUIDs Now

• Cold Dark Matter:

The Hunt for the Axion

9 October 2012

SQUIDs Then…

1 October 1964

• Royal Society Mond Laboratory

Cambridge University

• Thesis advisor: Brian Pippard

• Thesis research required measuring

voltages of 10−12 to10−13 volts

Eric Gill 1933

• Tangent magnetometer

• Magnet and mirror suspended from a long

quartz fiber at the centre of a Helmholtz pair

• Current in the coil caused a deflection of a

reflected light beam

• Voltage resolution 1 pV in 12 s (≈ 3 pVHz-1/2)

A Superconducting GalvanometerA. B. Pippard and G. T. Pullan 1951

Courtesy Brian Josephson

Brian Josephson Explains Tunneling

November 1964

1961 to 1964

I

Superconductor 1 Superconductor 2

~ 20 Å

Insulatingbarrier

I

V

Φ = n Φ0

J

1961 1962

1963 1964 Φ

Φ

Φ0I0

Deaver and FairbankDoll and Näbauer

Josephson

Anderson and

Rowell

Jaklevic, Lambe, Silver and Mercereau

Φ0 ≡ h/2e

Brian Pippard:

“John, How would you like a

voltmeter with a resolution of

2 × 10−15 V in 1 second?”

The Very Next Day—November 1964

Courtesy Kelvin Fagan

Brian’s Idea

I

VoutVin

R

L L

M

M = L

τ = L/R

Digital voltmeter: IΦ0 = Φ0/M = Φ0/L

Voltage resolution: Vin = IΦ0R = Φ0/τ = 2 × 10-15 V

for τ = 1 s

At Tea

• Paul Wraight pointed out that Nb has a surface oxide

layer and PbSn solder is a superconductor (February 1965)

5 mm

At Tea

Copper

Niobium

SnPb

solder

I

I

V

• Paul Wraight pointed out that Nb has a surface oxide

layer and PbSn solder is a superconductor (February 1965)

Before dinner

5 mm

Copper

Niobium

SnPb

solder

The Very Next Day

I

I

V

Brian Pippard:

It looks as though a slug

crawled through the

window last night and

expired on your desk!”

5 mm

Copper

Niobium

SnPb

solder

The SLUG (Superconducting Low-Inductance Undulatory Galvanometer)

Current IB in niobium wire

Vo

ltag

e

February 1965

Nb wire

and solder

I

I

V

IB

IB

5 mm

Copper

Niobium

Solder

The SLUG as a Voltmeter

Analog voltmeter

Voltage noise at 4.2 K

10 fVHz-1/2

(10-14 VHz-1/2)

V

1 µΩ

0 1 2Φo

Φ

δV

δΦ

V

I(mA)

V(mV)

Vd(nV)

Vd

V

Observation of Charge Imbalance

• Current injected through the Al-AlOx-Sn

junction creates a charge imbalance between

the electron-like and hole-like quasiparticle

branches in the superconducting Sn film.

• This imbalance establishes a potential

difference across the Sn-SnOx-Cu junction.

• Detailed investigation of nonequilibrium

superconductivity.

(1)

(2)

(3)

(4)

SLUG

JC 1972

Beasley and Webb 1967

0.001′′ Nb foil

0.001′′ mylar

0.03′′ Nb wire

Adjustable Niobium SQUID

Nb wire

and foil

…and Now

Thin-Film, Square Washer DC SQUID

SQUID with input coil

Josephson junctions

500 µµµµm 20 µµµµm

• Wafer scale process

• Photolithographic patterning

• Nb-AlOx-Nb Josephson junctions (John Rowell)

Ketchen and Jaycox (1981)

At 4.2 K

Flux noise: 1 – 2 µΦ0/Hz−1/2

~ Φ0/1,000,000 in 1 second

Superconducting Flux Transformer:Magnetometer

SQUIDMagnetic field noise

~ 10-15 THz-1/2

B J

Closedsuperconducting

circuit

Roomtemperatureelectronics

Magnetic Fields

1 femtotesla

tesla

10-16

10-10

10-8

10-6

10-4

10-12

10-14

10-2

1

Earth’s field

Urban noise

Car at 50 m

Human heart

Fetal heart

Human brain response

SQUID magnetometer

Conventional MRI

Quantum Design "Evercool"

UC Berkeley Flux Qubits

35 µµµµm

Qubit 1

Qubit 2

SQUID

UC Berkeley

High-Tc SQUIDs Prospecting for Mineral Deposits

Courtesy Cathy Foley, CSIRO

Gravity Probe-BTests of General Relativity

Courtesy Stanford University and NASA

• Geodetic effect—

curved space-time due

to the presence of the

Earth

• Lense-Thirring effect—

dragging of the local

space-time frame due to

rotation

• Gyroscopes:

Spinning, niobium-coated

sapphire sphere develops

magnetic dipole moment,

read out by a SQUID

South Pole Telescope: Searching for galaxy clusters

• Antarctica 9,500 feet

• 960 transition edge sensors with

multiplexed SQUID readout

• 12,000 SQUIDs

The “Bullet Cluster”

UC Berkeley + Many other groups

300-Channel SQUID Systems for Magnetoencephalography (MEG)

Magnetic Resonance Imaging at 132 µµµµT

UC Berkeley

Cold Dark Matter: The Hunt for the Axion

S.J. AsztalosG. CarosiC. HagmannD. Kinion,

K. van BibberM. Hotz

L.J. RosenbergG. Rybka, J. HoskinsJ. HwangP. SikivieD. B. TannerR. BradleyJC

Our Bizarre Universe

Neutrinos 0.6%

Baryons (ordinary matter) 4.6%

Dark Energy (DE) 73%

Cold Dark Matter (CDM) 22%

• A candidate particle for CDM is the axion, proposed in 1978 to explain

the absence of a measurable electric dipole moment on the neutron

• Roberto Peccei and Helen Quinn 2013 J.J. Sakurai Prize

• An alternative candidate particle is the WIMP—Weakly Interacting

Massive Particle

• Blas Cabrera and Bernard Sadoulet 2013 W.K.H. Panofsky Prize

0 100 200 300 400 500 6000

100

200

300

400

2.73K

Inte

nsi

ty (

MJy

sr-1

)

Frequency (GHz)

Po

wer

Frequency

610~ −−−−∆∆∆∆

νννν

νννν

Resonant Conversion of Axions into PhotonsPierre Sikivie (1983)

Primakoff Conversion

Expected Signal

HEMT* Amplifier

*High Electron Mobility Transistor

Magnet

Cavity

• Predicted axion mass:

ma ≈ 1 µeV – 1 meV (0.24 - 240 GHz)

• Need to scan frequency

Axion Dark Matter eXperiment (ADMX)

• Cooled to 1.5K

• 7 tesla magnet

Originally: Lawrence Livermore

National Laboratory

Now: University of Washington,

Seattle

Amplifier Noise Temperature

-AR

Ti

( )[ ]RRTT4kA(f)S NB

20

V +⋅=

V0

Quantum limit: TQ = hf/kB

i

LLNL Axion Detector

• Original system noise temperature: TS = T + TN ≈ 3.2 K

Cavity temperature: T ≈ 1.5 K

Amplifier noise temperature: TN ≈ 1.7 K

• Time to scan the frequency range from f1 = 0.24 to f2 = 0.48 GHz:

τ(f1, f2) ≈ 4 x 1017(TS/1 K)2(1/f1 – 1/f2) sec

≈ 270 years

Microstrip SQUID Amplifier: Principle

Conventional SQUID Amplifier Microstrip SQUID Amplifier (MSA)

• Source connected to both ends of coil • Source connected to one end of the

coil and SQUID washer; the other end

of the coil is left open25

20

15

10

5

Gai

n (

dB

)

20015010050Frequency (MHz)

MSA Gain Measurements

600

400

200νre

s (M

Hz)

604020Coil Length (mm)

30

25

20

15

10

Gai

n (

dB

)

700600500400300200100

Frequency (MHz)

71 mm

33 mm

15 mm

7 mmGai

n (

dB

) νre

s (M

Hz)

Gain for four coil length

M. Mück, J. Gail, C. Heiden†, M-O André , JC

MSA Noise Temperature vs. Bath Temperature

Gain = 20 dB

Quantum limit TQ = 30 mK

TQ

Tbath = 50 mK

Noise temperature:

TNopt = 48 ± 5 mK

Darin Kinion, JC

MSA:Impact on Axion Detector

≈≈≈≈ 100 days

• Original LLNL axion detector: TS ≈ 3.2 K

• Microstrip SQUID amplifier: TN ≈ 50 mK

• Next generation: Cool system in a dilution refrigerator to 50 mK

• For T ≈ TN ≈ 50 mK: TS ≈ T + TN ≈ 0.1 K

• Scan time ∝ Ts2 :

τ(0.24 GHz, 0.48 GHz) ≈ 270 years x (0.1/3.2)2

• ADMX II now funded at the University of Washington by DOE

Epilogue

• SQUIDs are remarkably broadband: 10−4 Hz (geophysics)

to 109 Hz (axion detectors).

• At low temperatures, the resolution of SQUID amplifiers is essentially limited by Heisenberg’s Uncertainty Principle.

• SQUIDs are amazingly diverse, with applications in

physics, chemistry, biology, medicine, materials science,

geophysics, cosmology, quantum information,……..

Brian: Thanks and

Congratulations!

Then…

…and Now