neutron optics - cern

59
INSTITUT MAX VON LAUE - PAUL LANGEVIN 1 1 THE EUROPEAN NEUTRON SOURCE INSTITUT LAUE LANGEVIN Neutron Optics P. Courtois Service for Neutron Optics EIROforum School on Instrumentation 13-17 May 2019

Upload: others

Post on 16-Oct-2021

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 11T H E E U R O P E A N N E U T R O N S O U R C E

I N S T I T U T L A U E L A N G E V I N

Neutron Optics

P. CourtoisService for Neutron Optics

EIROforum School on Instrumentation 13-17 May 2019

Page 2: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 2

The neutron

Why we need neutron optics

Basic concepts & Optical components

Reflection Mirrors & Super-Mirrors

Diffraction Mosaic crystals

Filters Crystals, polycrystalline materials

Polarization Super-Mirrors, crystals, 3He spin filters

Some examples of applications at I.L.L.

Neutron OpticsNeutron Optics are key components for neutron instrumentation

Page 3: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 3

Particle Mass m = 1.675 10-27 Kg → Kinetic Energy E = ½ m v2 ~ meV

No Charge Spin = 1/2 Magnetic moment m = -1.913 mN

Beta life time 886 s

neutrons E (meV) l (Å)

Hot 900 - 80 0.3 - 1

Thermal 80 - 5 1 - 4

cold 5 - 0.03 4 - 50

Wave Wavelength l = h/mv Wave vector k = 2p/l

Moment p = ћk Energy E = ћ2k2/2m

E(meV) = 81.81 / l2 (Å)

Neutrons have wave-like properties…

The Neutron

Page 4: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 4

Interaction with the individual nuclei via short range forces

Coherent scattering : depends on the direction of scattering vector Q = ki-kf

→ Coherent scattering length bcoh (~10-12 cm)

→ Diffraction, reflection, refraction …

Incoherent scattering : uniform scattering→ Incoherent scattering length binc (~10-12 cm)

→ Background , beam attenuation

Coherent scattering

atomnki

kf

Neutron Scattering

Incoherent scattering

natom

Page 5: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 5

Interaction with the individual nuclei via short range forces Neutron capture

Absorption cross section : sabs (~10-24 cm2)

Activation of materials - Emission of gamma rays

→ Radioprotection - shielding

Neutron Absorption

neutron capture

Atom*n

g

59Co

60Co*

n

g

E = 1.17 MeVg

E = 1.33 MeV

60Ni

Page 6: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 6

Interaction with unpaired electrons via a dipole interaction Magnetic scattering

Magnetic scattering length p ≈ 0.269 mat (10-12 cm/mB)

p ≈ bcoh (e.g. mat (Fe)= 2.2mB , p = 0.59 10-12cm, bcoh = 0.94 10-12 cm )

→ study of magnetic structures, production of polarized neutron beams

Magnetic scattering

atomn

e-

Magnetic Scattering

Page 7: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 7

Element bcoh (10-12 cm) binc (10-12 cm) sabs (10-24 cm2)

H -3.74 25.2 0.33

D 6.67 4.04 0.0005

BGdCd

- -767

488902520

C 6.64 ~ 0 0.0035

Ni 10.3 0.64 4.49

Si 4.14 0.015 0.17

Cu 7.72 0.20 3.78

Background

Neutron absorber

Optics !

Moderator

Page 8: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 8

Neutron sources are large compared to sample size ( 100 cm2 vs 1 cm2)

Neutrons sources are weak (max neutron flux on sample is 1010 neutrons/cm2/s, while x-ray fluxcan be 1020 photons/cm2/s)

Neutron beams are divergent (beam divergence in the range of 0.2° to 1°)

Neutron beams are ”polychromatic” (wide energy range, e.g. thermal range: 5-80 meV)

Neutron beams are not polarized

Basic Properties of neutron beamsNeutrons are extracted by beam tubes from the reactor core and transported by neutron guides to the experimental area

a= 2 qcSampleS ~1 cm2Neutron source

(guide, beam tube)

Beam size S ~ 100 cm2Divergent beam

flux ~ 1010 n /cm2/sReactor core

flux ~ 1015 n /cm2/s

Page 9: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 9

What do we need to perform neutron scattering experiments ?

We need to transport neutrons to the experimental area (guides)

Mirror & Super-Mirror : Flux, divergence, neutrons with wavelengths in the right range

We need to determine incident and scattered neutron directions

Collimator : Angular resolution

We need to select out unwanted neutrons - Filter

We need to determine incident and scattered wavelengths (Energies)

Crystal Monochromator : Wavelength resolution (Energy resolution)

We need also to polarize neutron beams

Crystal/ Super-Mirror/ 3He spin filter : Orientation of the neutron spin

Instrument Performances are then determined by Neutron Optics !

Page 10: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 10

Page 11: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 11

Neutron MirrorsReflection/Refraction at Surfaces

plq coh

c

Nb=

incident reflected

refracted

Nbcoh = Scattering Length DensityN = atoms/cm3

bcoh = coherent scattering length

for natural Ni: qc(°) = 0.1 x l (Å)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.40.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

l = 5 Å

refle

ctiv

ity

q [°]

qc

Ni mirror

θ

MaterialNb

(x 1038 /m2)qc (mrad)

58Nickel 13.31 2.03

Nickel 9.41 1.7

Iron 8.2 1.62

Copper 6.7 1.39

Silicon 2.08 0.81

Aluminium 2.08 0.81

Total reflection for q < qc

Page 12: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 12

Periodic MultilayersReflection/Refraction at Surfaces

Total reflection for q < qc

+

additional Bragg peak at qB ≈ l / 2d

Rµ 4N 2d4 N1b1 - N2b2( )

2

π 2n4

Total reflection

Bragg peak

Total reflection

qc

Neutron Reflectivity

Wavelength band

qB l 2d

Dl

l=

2d2 N1b1-N2b2

π

d

θ

substrate (glass, Si…)

Nbcoh = Scattering Length Density

N : number of bilayers

n : order of the reflection

Page 13: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 13

Super-Mirrors

substrate (glass, Si…)

d3

d2

d1

θ

Sequence of bi-layers of variable thicknesses d

Total reflection for q < qc + additional Bragg peaks

m =qc

SM

qc

Ni

G=qc

sm

qcNi

æ

èçç

ö

ø÷÷

2

µm2

0.00 0.25 0.50 0.75 1.00 1.250.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.5 1.0 1.5 2.0 2.5

m - value

l = 5 Å

refle

ctiv

ity

q [°]

regime ofsupermirror

regime oftotal

reflection

How to increase angular acceptance of mirrors

→ significant increase in critical angle (qc = l 2dmin)

m Super-Mirror

Gain in neutron flux

Page 14: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 14

Ni/Ti Super-MirrorsProperties

Neutron Reflectivity R (N1b1-N2b2)2 High contrast → high reflectivity !

• Ni Nb = 9.40 (10-6Å-2)

• Ti Nb = -1.95 (10-6Å-2)

Reflectivity profiles of Ni/Ti super-mirrors for 4 ≤m ≤ 7(sources : www.swissneutronics.ch)

m=4

Performances

R > 80% for m = 4 Ni/Ti

Gain factor / Ni mirror = 16 (2D)

but transmission T Rn !!

eg : for a 100 m long guide , at least ten

reflections → Transmission < 10% ...

Page 15: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 15

Ni/Ti Super-Mirrors Deposition : Reactive DC Magnetron Sputtering

Sputtering machine (ILL) - Production 0.8 m2 / day

Production m= 4 : 1600 layers ! Substrate : 0.5 cm thick Si wafers or 0.2 cm thick Glass/Si/Sapphire

Page 16: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 16

1614/05/2019

Neutron Guide

m=2 Ni/Ti Super-Mirrors

Length > 50 m

Section ~ 20 x 10 cm2

4 reflective internal surfaces

Curvature R ~ 1-10 km

A neutron guide can transport neutrons over a long distance with “no loss”

Ni/Ti Super-Mirrors

Page 17: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 17

Bragg MirrorsLarge wavelength band Monochromator

(Laue diffractometer for Protein crystallography)

guide

to produce a « monochromatic » beam at 3.5 Å with a bandwidth Dll = 20%

Q

Stacked Multilayer monochromator Beam width = 20 mm qB = 1.25 ° -> d0 = 80 Å Dll = 20 % -> graded d-spacing 74-90 Å Beam size >> Sample size (1 mm)

-> Focusing device (fan geometry)

•mirror length = 25 mm • thickness 0.5 mm• 40 mirrors Ni/Ti• Si substrate

(low attenuation factor)

monochromatorsample

L mono-sample = 2135 mm

Page 18: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 18

Bragg Mirrors

guide

Q

monochromator

sample

monochromator

neutrons

direct beam

reflected beam l0 ≈ 3.5 Å

Dll ≈ 20%

Reflectivity ≈ 75%

TOF spectra on LADI

Large wavelength band Monochromator

(Laue diffractometer for Protein crystallography)

Page 19: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 19

Page 20: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 20

Use of single crystals to select a given wavelength band according to the Bragg’s Law

2 dhkl sinqB = nl0d= distance of the lattice (hkl) planes qB = Bragg angle

Determining the wavelength (Energy)Crystal Monochromator

polychromatic beam

monochromatic beam

l0

crystal

If a ~b , the resolution and intensity are said to be optimised

Mosaic Crystal, i.e. crystal with defects such as dislocations,must be used to match the divergence a of primary beam which is typically 0.2°- 0.8°

The relative bandwidth Dll is given for Gaussian distributions by Dll = (a 2 + b 2)1/2cotqB

a : divergence of the primary beamb : full width at half maximum of the neutron “rocking curve” or neutron mosaic spread

n/s

b q

Page 21: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 21

A mosaic crystal consists of a distribution W(q) of small crystallites (perfect crystals) making small angles relative to the nominal plane. W(q) is usually taken to be a Gaussian distribution.

Defect free crystal = perfect crystal (Silicon) Crystal with defects such as dislocations = mosaic crystal

)2/exp(2

1)( 22

pq D=w

The mosaic crystalPerfect crystal vs Mosaic crystal

Perfect crystalDarwin width < 0.001°

w

Mosaic crystal width > 0.05 °

q

Page 22: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 22

Mosaic Crystal

The reflection range of mosaic crystals is of the order of 0.3 °even more....

Mosaic crystals can reflect efficiently neutrons from the source of divergence a, if b a.

High intensity and reasonable resolution

Suitable for neutron beams !

fwhm < 10-4°

Dll =a cotqB ; flux → 0 !

Perfect crystal

<hkl> aa

Dll = (a 2 + b 2)1/2cotqB

Mosaic crystal

fwhm > 0.1 °

a+2bb

a

Perfect crystal

The reflection range of perfect crystals is in the order of 0.001°.

Perfect crystals do not reflect efficiently neutrons from the source of divergence a.

Very high resolution but no intensity

Not suitable for neutron beams !

Perfect crystal vs mosaic crystal

Page 23: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 23

Mosaic CrystalsGood Materials should have

High neutron reflectivity R scattering power Q

Large structure factor Fhkl

High coherent scattering length bcoh (Fhkl bcoh) Small unit cell Volume V : Cubic, Diamond d-spacing optimized for a given wavelength (d ~ l )

Low incoherent scattering length → low background

Low absorption cross sections → small attenuation

Availability of large single crystals with suitable width and uniform mosaic block distribution !

I0 IR

crystal

R(l) = I0(l) / IR(l)

Bql /sin2/V)e(F=Q 32-W

hkl

Page 24: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 24

Common mosaic crystals for neutron monochromators

Crystal orientationCrystal Mosaic

NeutronEnergy

Application Supplier

HOPG (C)d002= 3.35

ÅHOPG(002) 0.5°- 3°

ColdThermal

High fluxPanasonic, Optigraph, Momentive

Cu d111= 2.08

Å

(111)(220)

(200)…0.01° - 3°

Hot Thermal

Highresolutionor high flux

I.L.L. (or ?)

Sid111= 3.13

Å

(111)(113)…

Perfect Cold

ThermalHigh

resolutionmany !

Ged111= 3.26

Å

(111)(113)…

< 0.4°Cold

ThermalHigh

resolutionI.L.L.(or ?)

HeuslerCu2MnAl

(111) 0.2 – 0.6 °Hot

ThermalPolarized neutrons

I.L.L.

Page 25: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 25

Crystal MonochromatorHighly Oriented Pyrolitic Graphite (HOPG)

Hexagonal unit cell - a = 2.461 Å; c= 6.707 Å ; space group P63mc

Structure FactorF00l = 4 bC l = 2N

F002 = 4 x 6.64 10-12 cm

Large Graphite single crystals not available

Production of mosaic HOPG crystals micro-crystallites oriented along the c axis random distribution in the (a,b) planes

Suppliers : Panasonic, Momentive, Optigraph

Only (0 0 2l) Reflections available HOPG(002) mosaic FWHM > 0.4°

Suitable for neutron applications !

Page 26: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 26

Diamond structure ; space group Fd3m ; aSi = 5.43 Å and aGe = 5.65Å

Structure Factors Fhkl = 0 h,k,l mixed even or oddFhkl = 0 h,k,l all even and h+k+l=4n+2Fhkl = 8 b h,k,l, all even and h+k+l=4nFhkl = 4√2 b h,k,l all odd

Large perfect crystals available from industry

For h,k,l “all odd” reflections: Fhkl ≠ 0 and F2h,2k,2l = 0

→ No reflection of the second harmonic→ No l/2 contamination

but the mosaic distribution (FWHM) of a perfect crystal is too narrow for neutron applications !!

Crystal MonochromatorSi and Ge perfect crystals

Page 27: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 27

Face-centred cubic structure aCu = 3.615Å; space group Fm3m

Structure Factors Fhkl = 0 h,k,l mixed even or oddFhkl = 4 bCu h,k,l all even or all odd

Fhkl = 4 x 7.72 10-12 cm

Crystal Growth at ILL Bridgman Technique large crystals of high quality

but the mosaic distribution (FWHM) of the as-grown crystal is too narrow for neutron applications !!

m = 8 kg-4 -2 0 2 4

Refle

ctiv

ity (a

rb.u

nits

)

diffraction angle (minutes)

as-grown Cu crystal

fwhm ~ 30 arc sec

Crystal MonochromatorCu mosaic crystal

Page 28: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 28

crystal

Anvil

Heating element

FWHM=0.15°

FWHM=0.45°FWHM=1°

Mosaic Crystals Control of the mosaic distribution by plastic deformation (I.L.L.)

FWHM=0.4°Perfect crystal

Ge111

Cu200

Peak reflectivity at l = 3.4 ÅRexp = 50-55% (FWHM=0.4°)

Rexp ≈ 70 % of Rth

Peak reflectivity at l = 1.1 ÅRexp = 40-45% (FWHM=0.3°)

Rexp ≈ 80-90 % of Rth

T = 800 °C

P = 1 MPa

as-growncrystal

Cu200 P,T

Page 29: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 29

Mosaic CrystalsBent perfect Silicon crystals (I.L.L.)

Effective mosaic d (rad) d = cot(qB) t / Rt = total crystal thicknessR = radius of curvatureqB = Bragg angle

<hkl>R

qB

Perfect Si crystal

neutrons

Stack of thin wafers to allow bending wafer thickness = 1 mm 10 wafers to get t = 10 mm (or more) Curvature : flat to RH ≈ 2 m

Page 30: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 30

Monochromators

HOPGD19

• Large effective area to cover the direct beam

Assembly of mosaic crystals

• FWHM ~ beam divergence 0.2 < FWHM < 0.6 °

• Crystal size ~ Sample size (1 to 10 cm2)

• Crystals fixed onto specific mechanics

• 10B4C is used to reduce background and activation

• Orientation accuracy ± 0.03°

• Vertical Focusing

• Horizontal Focusing (Triple axis spectrometers)

crystal

AlB4C

crystal + support

Page 31: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 31

Focusing devicesDivergent beamflux ~ 1010 n /cm2/s

a= 2 qcSampleS ~1 cm2

Neutron source (guide, beam tube)

Beam size S ~100 cm2

lost neutrons !

Focusing devices are used to increase the neutron flux at the sample position. However, theincrease of neutron flux implies a degradation of the angular resolution. (Liouville’s theorem!)

H1H2H1 a1 = H2 a2Divergence a1

a2

Page 32: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 32

Focusing DevicesPrinciples

Vertical Focusing

Image size ~ crystal size (height) Gain in flux compression factor Increase of vertical angular divergence Rv l

Horizontal Focusing

Horizontal focusing affects Q resolution Rh 1/l

Sample

1

Rv

=1

2sinq

1

L1

+1

L2

æ

èç

ö

ø÷

1

RH

=sinq

2

1

L1

+1

L2

æ

èç

ö

ø÷

Rv

Source

Sample

L2

L1

θ

L2

L1Source

Sample

RHθ

Page 33: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 33

Focusing DevicesMonochromators for neutron Diffractometers

Production of a fixed neutron wavelength at a given take-off angle

Ge (335) - l = 1.6 Åcrystal mosaic = 0.2 °

High Resolution

D2B

D20

Ge (111)- l = 3.15 Åcrystal mosaic = 0.4 °

High Resolution

D20

Ge monochromator is well suited

for high resolution neutron diffractometers

D20 (high intensity difffratometer with variableresolution) - D2B & D1B (powder diffractometers)

Large dimensions Heigth 300 mm

Width 50-120 mm

no l/2 contamination (odd reflections)

Fixed vertical focusing Rv = 2 LMS sinqB

D20 : Rv = 3200 mmD2B : Rv = 4600 mm

Page 34: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 34

D20

Single crystal diffractometerHOPG (002) - l = 2.4 Åcrystal mosaic = 0.5 °

High Flux

D19

Single crystal diffractometerCu(200) - l = 1.2 Å

crystal mosaic = 0.2 °High Resolution

D10

HOPG monochromator

• Provides high neutron flux (thermal & coldneutrons)

l/2 contamination (002 reflections)

Use of HOPG Filter !

Fixed vertical focusingCu monochromator

provides high neutron flux or high resolution(hot & thermal neutrons)

l/2 contamination

use of HOPG Filter !

thermal neutron guide cut-off

Focusing DevicesMonochromators for neutron Diffractometers

Page 35: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 35Auteur - Date

Optimization of instrument performances for a wide energy range

Variable horizontal and vertical curvature

Focusing monochromator is used in combination with focusing guide (virtual source)

Focusing DevicesMonochromators for Triple Axis Spectrometers

Page 36: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 36

D20

HOPG(002) monochromatorcrystal mosaic = 0.5 °

cold neutrons

Cu(200) monochromatorcrystal mosaic = 0.3 °

High Flux – Thermal neutrons

Si(113) monochromatorbent perfect crystals

Hot neutrons

Thales IN8C IN1

Focusing DevicesMonochromators for Triple Axis Spectrometers

Page 37: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 37

Page 38: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 38

Neutron Filtersm super-mirrors can be used to select out unwanted neutrons

Neutron long wavelength cut-off filter

qm super-mirror

White beam

Transmittedl <lC

Ni

c

cmq

ql =

reflectedl >lC

sample

Page 39: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 39

m Super-Mirrors Long wavelength cut-off filter (SANS instrument)

White beamTransmitted

l <lC

reflected l >lC

Q = 2.2

lc = 20 Å

to select out unwanted neutrons of wavelengths > lc

m=1.1 SM NiV/Ti on 0.5 cm thick Si substrate mounted at an angle q in a neutron guide

Borofloat glass m=0.65

0.84 m

Ni

c

cmq

ql =

Page 40: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 40

to select out unwanted neutrons of wavelengths > lc

m=1.1 SM NiV/Ti on 0.5 cm thick Si substrate mounted at an angle q in a neutron guide

m Super-Mirrors Long wavelength cut-off filter (SANS instrument)

White beamTransmitted

l <lC

reflected l >lC

Q = 2.2lc = 20 Å

Q = 1.32

lc = 12 Å

Borofloat glass m=0.65

0.84 m

Transmittedl <lC

Ni

c

cmq

ql =

Page 41: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 41

m Super-Mirrors Long wavelength cut-off filter (SANS instrument)

White beamTransmitted

l <lC

reflected l >lC

Q = 2.2lc = 20 Å

Q = 1.32

lc = 12 Å

Borofloat glass m=0.65

0.84 m

Transmittedl <lC

to select out unwanted neutrons of wavelengths > lc

m=1.1 SM NiV/Ti on 0.5 cm thick Si substrate mounted at an angle q in a neutron guide

Ni

c

cmq

ql =

Page 42: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 42

m Super-Mirrors Long wavelength cut-off filters at I.L.L. (SANS instrument)

Neutron Flux at the sample position on D33

C. D. Dewhurst et al. , J. Appl. Cryst. (2016). 49, 1-14

no filter

lc = 20 Å

lc = 12 Å

Page 43: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 43

Neutron Filterspolycrystalline Be & Saphirre crystal

θ

Polycrystalline Beryllium at 77Kis used to select out neutrons of l< 4 Å

(Cut-off at l = 2dBe max )

Perfect Saphirre crystal is used to select out fast neutrons of l < 0.5 Å

Reduction of background

T ~ 5%

T ~ 80%

n l > lcut

l < lcut

Cut-off

Page 44: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 44

Neutron FiltersHOPG crystal

θ

l= 1.2 Å

l= 2.4 Å

Wavelength l (Å)

tran

smis

sio

n

l0 2

HOPG crystal (Highly Oriented PyroliticGraphite) is commonly used for eliminatinghigher-order contamination of amonochromated neutron beam

PG filter has strong attenuation at 1.2 Å butpasses 2.4 Å

• T > 80 % at 2.4 Å

• T < 1% at 1.2 Å

l= 0.8 Å

l0 = 2.4 Ål0 = 2.4 Å+ l0 2 = 1.2 Å

HOPG filter , 4 cm thick

HOPG filter

Page 45: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 45

Production of Polarized neutron beams

at I.L.L.

Page 46: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 46

Magnetic Field B (quantization axis) → sneutron= ± ½

Polarization of the neutron beam +

+

+

=

NN

NNP

B Principal methods

Fe/Si and Co/Ti super-mirrors (reflection, transmission)

Heusler Cu2MnAl crystal (diffraction)

3He3 spin filters (absorption by polarized 3He nuclei)

polarizer

P =1

N+ neutrons with >N- neutrons with >

Page 47: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 47

Refraction index

for a magnetic layer π

pbNλn

2

)(1

2 =

Magnetic contribution

Reflectivity R of a bilayer system

> : R+ [N1b1-N2(b2+p2)]2

> : R- [N1b1-N2(b2-p2)]2

Polarization N1b1= N2(b2±p2)

→ Reflection / Transmissionof one spin state 1: non magnetic layer

2: ferromagnetic layer

q

>

>

> + >

B

Q

1

2

Page 48: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 48

Co/Ti Super-Mirrors

> : N(b+p)Co = 6.55;

> : N(b-p)Co = -2.00 ≈ Nb Ti = -1.95

m= 3.2 Super-Mirrors

but Activation of Cobalt ! (lifetime 5 Years)

can be used only for polarization analysis

Fe/Si Super-mirrors

> : N(b+p)Fe = 13.04;

> : N(b-p)Fe = 3.08 ≈ Nb Si = 2.08

m = 4 super-Mirrors

→ Polarizer in reflection / Transmission geometry

Spin up Fe/Siq

>

>

P = 94%

R+

R-

Page 49: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 49

Polarizing Fe/Si Super-Mirrors

to produce a polarized cold neutron beam

S bender Polarizer for neutron reflectometer

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5 3

0 0.5 1 1.5 2 2.5 3 3.5 4

Re

fle

ctiv

ity

Omega

m=q / qc(Ni)

m = 3.8 Fe/SiR+ > 80%

R+

R-

P

m=3.8 Fe/Si SMson 0.2 cm thick Si substrate

Neutrons propagate into Si Spin > transmitted Stack of 60 blades(l 50 mm; h 160 mm; w 10 mm)

Pneutrons > 95% (5.5 Å) T ~ 40% of good neutrons

200 mm

Si channel

B

Polarizing supermirrors

>

neutrons

Page 50: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 50

Polarizing Co/Ti Super-MirrorsAnalyzer for wide angle spin echo spectrometer (WASP)

Co/Ti m= 2.8

Teta (°)

R+

R-

P

C bender - optimized for l = [4 -12 Å]

Neutrons propagate into air

Curvature R = 7m to avoid direct line of sight : at least on reflection !

Double sided mirrors (h 141 x w 254 mm2)

C bender

substrate

absorbing layer

Co/Ti SM

0.2 cm thickBorated glass

substrate

Page 51: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 51

Bragg reflection

from a ferromagnetic crystal

> : I+ [FN(Q) + FM(Q)]2

> : I- [FN(Q) - FM(Q)]2

FN(Q) nuclear structure factor

FM(Q) magnetic structure factor

Polarization FN(Q) = ± FM(Q)

>

BUnpolarized

White beam

Polarized

Monochromatic beam

Q

Crystal planes

M

>

>

>

Page 52: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 52

Single crystal

• (111) reflection F111N = - F111M

• mosaic 0.2° < fwhm < 0.6°

• Reflectivity R experimental ≈ Rtheoretical

• Polarization P > 92 % 0

10

20

30

40

50

-0.4 -0.2 0 0.2 0.4

— calculated — experimental

Re

fle

ctiv

ity

(%)

Rocking angle (degree)

● Mn

● Cu

●AlCubic structure L21

Structures Factors(111)

Nuclear: F111N = 4 (bMn-bAl ) (F111N = -2.8 10-12 cm)

Magnetic: F111M= 4 pMn

(F111M = 2.78 10-12

cm)

Heusler Cu2MnAl single crystal

Page 53: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 53

Neutron Filters – Polarized Neutrons3He spin filters

B

3He spin filter cell

P

For fully polarized gaseous 3He (PHe = 1), one spin state goes through the filter withzero absorption. The other spin state is almost fully absorbed since σ = 6000 barns→ polarized neutron beam

Absorption cross section of 3Helium nuclei If the nuclear spin of He and the neutron spin are parallel, sa ≈ 0 If the nuclear spin of He and the neutron spin are anti-parallel, sa ≈ 6000 barns

Page 54: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 54

Neutron Filters – Polarized Neutrons3He spin filters

Polarization of 3He is time dependant : P(t) = exp(-t/T1)

]b[

750]h[

PTd =

h 400200 wT

Tm[h] =P[b]

7000

¶B /¶r[cm]

B

æ

èç

ö

ø÷

-2

1

T1

=1

Td

+1

Tw

+1

Tm

Tw : Relaxation due to interaction with paramagnetic impurities

Tm : Relaxation due to field gradients

Td : Dipolar relaxation among 3He nuclear spins

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

He-3

Fit

Hours

Heliu

m-3

Pola

risation

T1 = 96 hrs

Page 55: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 55

The total relaxation time T1 must be in the order of 100 hours to perform high quality neutron experiments

High quality 3He spin filters (Polarization, Tw)

Homogeneous magnetic field

Production of polarized gas 3He !

Neutron Filters – Polarized Neutrons3He spin filters

¶B /¶r

B

æ

èç

ö

ø÷ <10-3cm-1

Page 56: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 56Auteur - Date

Production of 3He spin filters at I.L.L.Metastability Exchange Optical Pumping (MEOP)

3He filling station at I.L.L.

polarization of 3He nuclei

Page 57: Neutron Optics - CERN

I N S T I T U T M A X V O N L A U E - P A U L L A N G E V I N 57Auteur - Date

Production of 3He spin filters at I.L.L.Metastability Exchange Optical Pumping (MEOP)

Production rate > 1 bar-litre/h , Final pressure up to 4 bar, Polarization of 3He ≈ 80 %

Page 58: Neutron Optics - CERN

58T H E E U R O P E A N N E U T R O N S O U R C E

Conclusion

Neutron Optics define beam properties

- Direction, Divergence, Wavelength, Energy, Polarization

- Angular Resolution, Wavelength resolution, Energy resolution

Vertical focusing devices allow the optimization of the neutron flux at the sample position

Double variable focusing devices allow the optimization of instrument performances for a wide energy range

Since the power of the source is low, neutron optical components must be of high quality and properly designed

Neutron Optics obey to Liouville’s Theorem : It costs flux to increase resolution and

it costs resolution to increase flux

The optimization of instrument performances is always

a compromise between flux and resolution

Page 59: Neutron Optics - CERN

59T H E E U R O P E A N N E U T R O N S O U R C E

Thank You for your attention