Prototype Module of a Robust 18-channel Magnetometer System · 2016. 7. 31. · §History of...

Prototype Module of a Robust 18-channel Magnetometer System J. Storm, M. Burghoff, D. Drung, R. Körber

IEEE/CSC & ESAS SUPERCONDUCTIVITY NEWS FORUM (global edition), July 2016.Paper based on this presentation: J.-H. Storm et al., Supercond. Sci. Technol. 29, 094001(2016).

§ Introduction§ The prototype module§ Module and system design

§ The magnetometer

§ Field distortion due to Niobium shields

§ Noise performance of the prototype

§ Proof experiments§ Magnetoencephalography

§ ULF nuclear magnetic resonance

§ Summary

Introduction

06.07.2015 2

Outline:


§ History of biomagnetism at PTB Berlin:§ 1980 Berlin magnetically shielded room (BMSR)

§ 1991: 37-SQUID Multichannel System

§ 1994: 83-SQUID Multichannel System

§ 2000 Berlin magnetically shielded room (BMSR-2)

§ 2003: 304-SQUID Vector magnetometer system

Introduction

3

§ Main applications for the “old” 304 SQUIDvectormagnetometer:§ Magnetoencephalography

§ Material properties characterisation


Presenter

Presentation Notes

This presentation gives an overview on the development of a new SQUID magnetometer system intended for high-precision in biomagnetic measurements and nuclear spin precession experiments. The following slides are a collection of several internal and external talks.

Motivation

4

§ New applications for the “new”126 SQUID vector magnetometer:§ Ultra-low-field Nuclear Magnetic Resonance

§ Quantitative imaging of magnetic nanoparticles viamagnetorelaxometry

§ Ultra-sensitive spin precession measurements fordetermination of fundamental constants of naturesuch as the electric dipole moments of 129Xenucleus

304

126 Key features of the new system: § A scalable and modular system design

§ Vector magnetometer with different field sensitivities

§ Robust against pulsed magnetic fields up to 50 mT


Presenter

Presentation Notes

To extend the measurement capabilities at PTB Berlin, the 304-channel system will be replaced by a new SQUID vector magnetometer with up to 126 channels.


§ The magnetometer





§ Summary

The Prototype Module

06.07.2015 5

Outline:


Module and System Design – Overview

§ Top plane:1 x-y-z triplet d=17.1 mm1 hexagon d=74.5 mm

§ Bottom plane:7 z-loop d=17.1 mm1 hexagon3 x-y duplet d=17.1 mm

§ System:z-loop’s hexagonal gridx-y duplet hexagonalgrid rotated by 10.89°

§ SQUID capsule:niobium shield d=5 mmdetachable contact forthe flux antenna

6


Presenter

Presentation Notes

Overview of the multichannel module and the magnetometer arrangement. On the right hand, the intended 7-module arrangement containing 126 magnetometer channels.

bottom small z-loops

bottom large z-loop

top small x,y,z-loops

bottom small x-loop bottom small y-loop

Materials: • Magnetometer coils:Niobium wire d=100 µm

• Support structure: fiber reinforced plastic (G10)

SQUID chip

niobium foil for superconducting wire bonds

detachable contact for the pick-up coil

niobium shield d=5mm

SQUID bias pins

35 m

m

z1

z2 z3

z4

z5 z6

z7

Module and System Design – Details

7


Presenter

Presentation Notes

Detail view on the realization of SQUID package and the different pick-up coils.

The Magnetometer control line current limiter

outp

ut v

olta

ge/

SQU

ID b

ias feedback line

16x unshunted SQ

UID

‘s

SQUID chip PTB C70-M1 PTB C7-L1

Input inductance 150 nH 400 nH

Input coupling 2.5 nH 4.1 nH

Current limiter Ioff=20 µA, Ion=1 µA

Flux antenna r=8.5×10-3 m raq=37.3×10-3 m

Loop area 2.3×10-4 m² 4.4 ×10-3 m²

loop inductance 58.4 nH 325.5 nH

tp inductance »20-50 nH

Field sensitivity 822-930 pT/F0 86-90 pT/F0

SQU

ID c

hip

flux

ante

nna

rloop

twisted pair

small magnetometer

large magnetometer

8


Presenter

Presentation Notes

Schematic illustration of the magnetometer and the base parameters of the SQUID-current sensors used for the 17.1 mm and the 74.5 mm diameter pick-up coils respectively.

Field Distortion Due to Niobium Shields

0 -50 -100 -150 -200 -250 -300-2,0x10-3

-1,5x10-3

-1,0x10-3

-5,0x10-4

0,0

measured data @ B0= 65µT FEM solution

Bz(r=

0,z)

-B0 /

B0

z (mm)

0 10 20 30 40 50-50

-40

-30

-20

-10

0

z (m

m)

-0.10-0.08-0.06-0.04-0.020.000.020.040.060.080.10

B0

DBz/B0

r (mm)

single shield multichannel arrangement xy-plane yz-plane

B0 ey

9


Presenter

Presentation Notes

To investigate the influence of the superconducting shields to a surrounding magnetic field, FEM simulations were carried out. On the left side a detailed illustration of the field distribution for a single shield and a comparison with measured values is given. Simulation data for an arrangement of 126 shields is shown on the right side.

Field Distortion Due to Niobium Shields

17.11.2015 10

-280 -260 -240 -220 -200 -180

-1500

-1000

-500

0

Helmholtz coil d=1m shield distortion z-direction

DB/

B 0 (pp

m)

z (mm)-100 -50 0 50 100

-4000

-3500

-3000

-2500

-2000

-1500

-1000

-500

0

Helmholtz coil (d=1m) shield distortion x-direction shield distortion y-direction

DB/

B 0 (pp

m)

s (mm)

→ The distortion of one module is one order of magnitude smaller

z = -175 mm


Presenter

Presentation Notes

For exemplification of the results, cut lines from the simulated field data were plotted together with the normalised field profile of a 1 m in diameter Helmholtz coil.

Noise Performance of the Prototype

Æ17.1mm loops: 1.28 fT/√Hz Æ74.5mm loop: 0.56 fT/√Hz → dominated by Dewar noise

Æ17.1mm loop: 0.52 fT/√Hz → intrinsic SQUID noise Æ74.5mm loop: 0.16 fT/√Hz → dominated by ambient noise

10-1 100 101 102 103 104 10510-1

100

101

102

z3

z6

17.1 mm pickup coils

74.5 mm pickup coil (s7)

ÖSB

(fT/Ö

Hz)

f (Hz)

Lower z-loops (software gradiometers)

10-1 100 101 102 103 104 105

10-1

100

101

ÖSB

(fT/Ö

Hz)

f (Hz)

Upper z-loops (magnetometers)

11

Dotted lines: intrinsic SQUID noise


Presenter

Presentation Notes

Presented is the measured field noise for the z-channels of the prototype module. In the bottom plot, software gradiometers were implemented with the upper 74.5 mm magnetometer as reference. Due to the early state of development, strong electromagnetic disturbances that arise from the readout setup are visible.


§ The magnetometer





§ Summary

Proof Experiments

06.07.2015 12

Outline:


Magnetoencephalography

m e d i a n n e r v e

s e n s o r y c o r t e x

s t i m u l a t o r

13


Presenter

Presentation Notes

In order to demonstrate the suitability of the new system, two proof experiments were carried out. At first, magnetoencephalography with somatosensory evoked brain activity. The experimental setup is shown schematically.


N20

s-band 450-750 Hz

k-band 850-1200 Hz

§ Ultra-low-noise EEG/MEG systems enable bimodal non-invasive detection of spike-likehuman somatosensory evoked responses at 1 kHz (T. Fedele et al. Physiol. Meas. 2015)

12 000 averages

14


Presenter

Presentation Notes

As a comparison, here are recent results of ultra-low-noise EEG and MEG measurements. The MEG data was measured using a single channel system which has a white noise level of 0.5 fT/rtHz and a warm-cold distance of 8.8 cm at room temperature. With permission form Physiological Measurement: doi:10.1088/0967-3334/36/2/357


• electric stimulation at median nerve at t=0 s• N20 visible at t»20 ms after stimulation• 16200 averages

Somatosensory evoked brain activity, Prototype module:

0.01 0.02 0.03 0.04 0.05-1

-0.5

0

0.5

1x 10

-13

Time [sec]

Fiel

d [T

]

x1x2x3xo

0.01 0.02 0.03 0.04 0.05-1

-0.5

0

0.5

1x 10

-13

Time [sec]

Fiel

d [T

]

y1y2y3yo

0.01 0.02 0.03 0.04 0.05-1

-0.5

0

0.5

1x 10

-13

Time [sec]Fi

eld

[T]

z1z2z3z4z5z6z7ZuzoZo

N20

15

x1 x3 x5 x9

y1 y3 y5 y9

s7 z9 s9


Presenter

Presentation Notes

MEG results for N20 measurement with the prototype module



0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]

Fiel

d [T

]

x1x2x3xo

0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]

Fiel

d [T

]y1y2y3yo

0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]Fi

eld

[T]



s-range (450-750 Hz)

16

x1 x3 x5 x9

y1 y3 y5 y9

s7 z9 s9


Presenter

Presentation Notes

Frequency domain filtered N20 signal (sigma-range 450-750Hz).



0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]

Fiel

d [T

]

x1x2x3xo

0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]

Fiel

d [T

]y1y2y3yo

0.01 0.02 0.03 0.04 0.05

-4

-2

0

2

4

x 10-15

Time [sec]

Fiel

d [T

]



k-range (850-1200 Hz)

17

x1 x3 x5 x9

y1 y3 y5 y9

s7 z9 s9


Presenter

Presentation Notes

Frequency domain filtered N20 signal (kappa-range 850-1200Hz).

Bpol To boost magnetisation of sample use prepolarisation, usually mT-range.

Bdet Expose sample to detection field, ususally µT-range.

SQUID-Sensor

Basics ULF nuclear magnetic resonance

x

z

y

t

Mz

During polarisation magnetisation Mz grows with T1.

18


Presenter

Presentation Notes

Second proof experiment: ultra low field nuclear magnetic resonace (ULF-NMR). A step-by step description of the experimental setup is given on the next two slides. In the first step, the sample (distilled water) is polarized by a strong magnetic field.

Bdet Expose sample to detection field, ususally µT-range.

SQUID-Sensor

Basics ULF nuclear magnetic resonance

x

z

y

t

Bz at SQUID

During detection magnetisation M precesses around Bdetwith Larmor frequency w=gB and decays with T2

*.

T2* describes

dephasing of the M (spins). intrinsic + instrumental contributions.

19


Presenter

Presentation Notes

In the second step, the polarisation field has been quickly turned off and the free precession decay of the sample is being measured in the detection field. The decay time T2* depends on intrinsic effects in the sample and on the homogeneity of the detection field over the sample volume.

Nuclear magnetic resonance of protons

Experimental setup: Sample inside polarising coil

Detection field coil

0 1 2 3 4 51,0x102

1,5x102

2,0x102

2,5x102

3,0x102

mag

. flu

x de

nsity

(pT)

t (s)

• baseline correction,• high-pass filter (f0=60Hz)

Sample: distilled water detection field: 2.56 µT Polarization field: 35 mT (centre sample) Polarization time: 5 s SQUID reset time : 50 µs

Raw B-field data Filtered FID

0 1 2 3 4 5

-1,5x103

-1,0x103

-5,0x102

0,0

5,0x102

1,0x103

1,5x103

mag

. flu

x de

nsity

(fT)

t (s)

20


Presenter

Presentation Notes

Specifications and measured time domain data of the ULF-NMR experiment.

Nuclear magnetic resonance of protons

Experimental setup: Sample inside polarising coil

Detection field coil

Sample: distilled water detection field: 2.56 µT Polarization field: 35 mT (centre sample) Polarization time: 5 s SQUID reset time : 50 µs

Amplitude spectra with the respective fits

Points are data Fit: Lorentzian to data real and imaginary parts ® Resonance frequency: 108.97 Hz ® T2* (for bottom z-magnetometer): 1.75 s

Parameters are in accordance with expected values

107,0 107,5 108,0 108,5 109,0 109,5 110,0 110,50

20

40

60

80

100

120

x1 x2 x3 xo y1 y2 y3 yo z1 z2 z3 z4 z5 z6 z7 Zu zo Zo

ÖSB (

fT/Ö

Hz)

f (Hz)

21

x3 x5 x9

y3 y5 y9

s7 z9 s9


Presenter

Presentation Notes

Presented is frequency domain data of the ULF-NMR experiment. The results are in reasonable accordance with previous work and confirm the suitability of the new system for measurement methods involving pulsed magnetic fields. (Nuclear magnetic relaxation in water revisited Hartwig at al, The Journal of Chemical Physics, 135, 054201 (2011), DOI:http://dx.doi.org/10.1063/1.3623024)

Summary • 18-channel SQUID magnetometer module wasdesigned and constructed

• Different coil sizes allow maximum SNRs for differentsource depths and configurations

• The designed module forms the basis for a scalablemulti-module system (we plan a 126 channels configuration)

• Magnetic simulations of the magnetic distortions ofthe niobium shields were estimated and geometry ofthe shields optimised

• Sensitive MEG and pulsed ULF NMR experiments were performed

• Ultra-low noise performance enabled multi-channel detection of high- frequency components at around 1 kHz of somatosensory evoked activity

22

® Accepted paper: A modular, extendible and field-tolerant multichannel vector magnetometer based on current sensor SQUIDs, 2016 Supercond. Sci. Technol.


Prototype Module of a Robust 18-channel Magnetometer System · 2016. 7. 31. · §History of...

Documents

Transcript of Prototype Module of a Robust 18-channel Magnetometer System · 2016. 7. 31. · §History of...