focussing neutron optics - paul scherrer institute · focussing neutron optics erice school neutron...

J. Stahn solid state polarisers

Erice, 07. 2017 0

PAUL SCHERRER INSTITUT

Jochen Stahn

Laboratory for Neutron Scattering and Imaging

Solid State Polarisers

and

Focussing Neutron Optics

Erice School Neutron Science and Instrumentation, IV course

Neutron Precession Techniques

Erice, Sicily, Italy, 01. – 08. 07. 2017

outlineJ. Stahn solid state polarisers

Erice, 07. 2017 1

basics

◦ reflectometry

◦ supermirrors

◦ polarising coatings

polarisers

◦ overview

◦ reflective coatings

◦ comparison

focusing optics

◦ refractive

◦ reflective

reflectometryJ. Stahn solid state polarisers

Erice, 07. 2017 2

basics

◦ reflectometry

◦ supermirrors


polarisers

◦ overview


◦ comparison

focusing optics

◦ refractive

◦ reflective


Erice, 07. 2017 3

analogy to visible light

flat surfaces partly reflect light

→ image of the boot

some media also transmit light

→ ground below the water

parallel interfaces cause interference

→ colourful soap bubbles

reflectivity of a surface

function of index of refraction n

ω ω

k0 kR

k1

n0

n1

qz = 2|k0| sinω


Erice, 07. 2017 4

analogy to visible light

flat surfaces partly reflect light

→ image of the boot

some media also transmit light

→ ground below the water

parallel interfaces cause interference

→ colourful soap bubbles

reflectivity of plane parallel interfaces

interference pattern I(qz ,ni,di)

ω ω

k0 kR

k1

n0

n1

n2

n3

d1 thickness of layer 1

d2

d3

qz


Erice, 07. 2017 5

simulated reflectivity of a surface

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

critical edge ⇒ nsubstrate

R ∝

(q

qc

)4

for q ≫ qc


Erice, 07. 2017 6

simulated reflectivity of a thin layer

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)] amplitude ⇒ ∆nlayer,substrate

minimum ⇒ dlayer


Erice, 07. 2017 7

simulated reflectivity of a thick layer

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

amplitude ⇒ ∆nlayer,substrate

minimum ⇒ dlayer


Erice, 07. 2017 8

simulated reflectivity of a periodic stack of layers

= multilayer, ml

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

maxima ⇒ dbilayer

amplitudes ⇒ ∆nlayers

∆dlayers

minima ⇒ dfilm


Erice, 07. 2017 9

simulated reflectivity of a stack of mls

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

B

A

B A


Erice, 07. 2017 10

simulated reflectivity of a stack with thickness gradient

= supermirror, sm

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

critical edge of sm


Erice, 07. 2017 11

simulated reflectivity of a sm

and of Ni

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

qcNi critical edge of Ni

qcsm := mqc

Ni critical edge of sm


Erice, 07. 2017 12

index of refraction n :=|ki |

|k0|

≈ 1−V

2Ekin

V =2π~2

mnρb − µµµnB

︸︷︷︸

±µnB

:=2π~2

mn

(

ρb ± ρm)

ω ω

k0 kR

k1

n0

n1

qz = 2|k0| sinω

|k| = 2π/λ


Erice, 07. 2017 13

polarising sm

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

qz/A−1

log10[R(q

z)]

ρb − ρp ≈ ρsρb > ρs

ρb + ρp ≫ ρs


Erice, 07. 2017 14

polarising sm coatings

FM spacer substrate pro con

Fe89Co11 Si Si Co

Fe Si : N high transmission q|−〉

c

low activation

FeCoV Ti : N absorber q|−〉

c < 0 A−1 Co

Fe0.5Co0.5 Co

Co gets activated ⇒ avoid whenever possible!


Erice, 07. 2017 15

off-specular scattering

ρ = ρ(x , z)

⇒ R(qx) 6= 0!

⇒ dilution of phase space

background

losses

ρmagnetic(x , y) 6= 0

⇒ spin flip

Ni/Ti multilayer

k0in-plane off-specularspecular

out-of-plane off-specular

qxz

qzqxyz

qz

qx0


Erice, 07. 2017 16

keep in mind:R(qz) = R(n(z),B(z))

= 1 ∀ qz < qc

= 1 . . .0.55 for qz < qsm

∝ q−4z ∀ qz ≫ qc

• typical numbers:

ρb /10−6 A−2 qc / A−1 ωc @4A

Si, Fe|−〉

2.1 0.010 0.18◦

Fe|+〉

13.9 0.026 0.47◦

Ni 9.4 0.022 0.40◦

Ti -3.4

⇒ small angles ⇒ geometrical constraints

• roughness ⇒ off-specular scattering ⇒ background & depolarisation

-5

-4

-3

-2

-1

0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

polarisersJ. Stahn solid state polarisers

Erice, 07. 2017 17

basics

◦ reflectometry

◦ supermirrors


polarisers

◦ overview


◦ comparison

focusing optics

◦ refractive

◦ reflective


Erice, 07. 2017 18

overview

transmission throgh polycristalline Fe

6mm Fe: P = 33% at λ = 3.6 A

3He

→ talk by E. Babcock

Heusler alloy

crystal monochromator

thin film coatings


Erice, 07. 2017 19

Heusler alloy monochromator / analyser

Cu2MnAl single crystals with Fmagnetic(111) = ±Fnuclear(111)

⇒ F(111) reflex strong for µµµn ↑↑ B

weak for µµµn ↓↑ B

• λ ∈ [0.8, 6.5] A

• ∆λ/λ ≈ 1%

• P ≈ 95%P ≈ 95%P ≈ 95%

• used for triple-axis spectrometers

P. Courtois et al: N.I.M. A 529, 157 (2004)

polarisers - based on SMJ. Stahn solid state polarisers

Erice, 07. 2017 20

Using Reflected beam

• trajectory is inclined

• high polarisation

PR ≈ 96%− 99%PR ≈ 96%− 99%PR ≈ 96%− 99%

R|+〉

R|−〉

qzω1/λ

lost intensity

- transmitted

- absorbed

PR =R

|+〉−R

|−〉

R|+〉

+R|−〉

qzω1/λ

PR ≈ 1−R

|−〉

R|+〉

SwissNeutronics


Erice, 07. 2017 21

single, falt mirror switchable remanent polariser

HM ≈ 200Oe, Hg < 40Oe

-1

-0.5

0

0.5

1

-100 -50 0 50 100

magnetization

M

polarization field H/Oe

R|+〉

R|−〉

qz

R|+〉

R|−〉

qz

��

��

��

��

M

M

Hg

Hg

HM

HM


Erice, 07. 2017 22

benderlength

width≈ 150

|−〉 |+〉 |+〉

optimum parameters-1.6

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 20 40 60 80 100 120 140-1.6

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 20 40 60 80 100 120 140-3 -2 -1 0 1 2 3

-1.8

-1.6

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

-1

-0.5

0

0.5

1

1.5

2

y

θ

too large δθ-1.6

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 20 40 60 80 100 120 140

garland reflections of |−〉

too high λ-1.6

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 20 40 60 80 100 120 140


Erice, 07. 2017 23

S-benderlength

width≈ 250

• almost straight trajectory

• garland-problem is solved


Erice, 07. 2017 24

application: solid-state S-bender

Si wafers (150µm) used as channel

◦ thin and short channels

◦ q|−〉

c < 0 A−1

◦ no dark regiond due to substrate

◦ higher absorption

this principle also

applies to benders

T. Krist et al.: N.I.M. 698, 94 (2013)


Erice, 07. 2017 25

using Transmitted beam

• straight trajectory

• moderate polarisation

PT ≈ 60%− 80%PT ≈ 60%− 80%PT ≈ 60%− 80%

T|+〉

T|−〉

qz

PT =T

|−〉−T

|+〉

T|−〉

+T|+〉

qz

SwissNeutronics


Erice, 07. 2017 26

using Transmitted beam


• high polarisation

PT ≈ 96%− 99%PT ≈ 96%− 99%PT ≈ 96%− 99%

T|+〉

T|−〉

qz

PT =T

|−〉−T

|+〉

T|−〉

+T|+〉

qz

increase of efficiency by

multiple transmission:

• both sides of substrates coated

• several substrates in sequence

⇒ reduced intensitySwissNeutronics


Erice, 07. 2017 27

transmission bender + collimatorlength

width≈ 300


• dark areas due to substrates

T. Krist: solid-state transmission bender + collimator


Erice, 07. 2017 28

cavitylength

width≈ 50

|+〉 |−〉 |−〉

optimum parameters -1

-0.5

0

0.5

1

0 20 40 60 80 100

-1

-0.5

0

0.5

1

0 20 40 60 80 100-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

-1

-0.5

0

0.5

1

1.5

2

y

θ

too large δθ -1

-0.5

0

0.5

1

0 20 40 60 80 100

too high mchannel -1

-0.5

0

0.5

1

0 20 40 60 80 100


Erice, 07. 2017 29

cavitylength

width≈ 50

|+〉 |−〉 |−〉

optimum parameters -1

-0.5

0

0.5

1

0 20 40 60 80 100

-1

-0.5

0

0.5

1

0 20 40 60 80 100-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

-1

-0.5

0

0.5

1

1.5

2

y

θ

too low λ -1

-0.5

0

0.5

1

0 20 40 60 80 100

too high λ -1

-0.5

0

0.5

1

0 20 40 60 80 100

-1

-0.5

0

0.5

1

0 20 40 60 80 100-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

-1

-0.5

0

0.5

1

1.5

2


Erice, 07. 2017 30

V-cavitylength

width≈ 30

• straight beam geometry

F. Mezei et al.: Physica B 180-181, 1005 (1992)


Erice, 07. 2017 31

V-cavitylength

width≈ 30

• straight beam geometry phase space affected

• ∆λ/λmin ≈ 5

• P ≈ 99%P ≈ 99%P ≈ 99%

P. BoniF. Mezei et al.: Physica B 180-181, 1005 (1992)


Erice, 07. 2017 32

equiangular spiral

for beams

{emerging from

focused to

}

a narrow area

• same ω for all trajectories

→ flexibility for ω, m, λ

• phase space hardly affected

y

0 xmin xmax

xαδθ

θ

prototype at PSI

J. Stahn, A. Glavic: Journal of Physics: Conference Series 862, 012007 (2017)


Erice, 07. 2017 33

using Reflected and Transmitted beam

◦ split neutron guide for 2 polarised instruments (at HMI / HZB)

F. Mezei et al.: Physica B 213-214, 393 (1995)

◦ suggested analyser for Estia@ESS

A. Glavic


Erice, 07. 2017 34

wide-angle analysers

stack of cavities / benders / spirals pointing towards the sample

challanges:

◦ avoid / minimise black angles

◦ provide a high magnetisation field

◦ reduce losses

example: Hyspec analyser by PSI

60◦ coverage with 1000 benders

P ≈ 95%

transmission = 10% to 45% for λ ∈ [2, 5] A


Erice, 07. 2017 35

wide-angle analysers study for MIEZE@ESS

λ > 6 A

by P. Boni

optimisation of shape using an equiangular spiral: λ ∈ [6, 48] A

2.3◦

166mm

polarisers - comparisonJ. Stahn solid state polarisers

Erice, 07. 2017 36

comparison

Heusler

3He

−− − ± + ++

P ∆λ ∆λ/λpsl/d

polarisers - comparisonJ. Stahn solid state polarisers

Erice, 07. 2017 37

comparison

Heusler

3He

−− − ± + ++

P ∆λ ∆λ/λpsl/d

f ai l

ed

focusing opticsJ. Stahn solid state polarisers

Erice, 07. 2017 38

basics

◦ reflectometry

◦ supermirrors


polarisers

◦ overview


◦ comparison

focusing optics

◦ refractive

◦ reflective


Erice, 07. 2017 39

motivation

higher flux on small samples

no illumination of sample environment

selection of area on /within sample

control over phase space / trajectories

deal with small sources

remote footprint control


Erice, 07. 2017 40

focusing optics

reshapes the phase space of a n-beam (an ensemble of neutrons)

to a small spatial extent at a given position

shading optics

reshapes the phase space by restricting it in space (slit)

θ

x


Erice, 07. 2017 41

focusing optics vs. shading optics

high costs (needs high precision)

lower transmission

convenient beam manipulation

real focusing

aberration

robust

flexible

high transmission

high background

focusing optics – refractiveJ. Stahn solid state polarisers

Erice, 07. 2017 42

refractive optics

n ≈ 0.99999 . . .1 for all bulk materials

Snell’s law: n =sinαex

sinαin

⇒ αin ≈ nαex close to normal incidence

• used for SANS

αin

αex

M. R. Eskildsen et al. nature 391, 563 (1998)

focusing optics – reflectiveJ. Stahn solid state polarisers

Erice, 07. 2017 43

reflective optics

parabolic

parallel to convergent

hyperbolic

convergent to convergent

elliptic

divergent to convergent


Erice, 07. 2017 44

reflective focusing optics

elliptic

divergent to convergent


Erice, 07. 2017 45


early reflections suffer the most from coma aberration

⇒ multiple reflections

⇒ non-convergent beam behind guide exit

elliptic

divergent to convergent ?

L. Cusssen et al.: NIM A 705, 121 (2013)


Erice, 07. 2017 46



Erice, 07. 2017 47



Erice, 07. 2017 48

coma aberration


Erice, 07. 2017 49

coma aberration

. . . and its correction


Erice, 07. 2017 50

coma aberration

. . . and its correction

• I(xy) is restored

• I(θ) is not!


Erice, 07. 2017 51

Selene guide

point-to-point focusing

with

2 subsequent elliptical reflectors

for

horizontal and vertical direction

Selene

Selene picture:

ceiling painting in the Ny Carlsberg Glyptotek, København


Erice, 07. 2017 52

Selene guide

decoupling of

• spot-size

and

• divergence


Erice, 07. 2017 53

condenser: parabolic deflector to generate a parallel beam

parabola axis ⇒ beam direction

focal length ⇒ beam width

beam width& spot size

⇒ divergence

no collimator needed

tunable

adaptive parabola (convex)

focal spot with 170µm reached

(PSI, early version)


Erice, 07. 2017 54

astigmatic focusing: focusing to the detector by shifting the focal point

hyperbolic deflector


Erice, 07. 2017 55

solid-state neutron lense

T. Krist et al.: N.I.M. A 634, S104 (2011)

focusing optics - discussionJ. Stahn solid state polarisers

Erice, 07. 2017 56

focusing results in . . .

. . . no gain in brilliance

. . . a defined footprint

. . . a clean beam

homogeneous

uni-modal angular or spatial distribution

non-perfect optics

⇒ reduction of resolution / transmission

works best for small samples

weak aberration

thanks toJ. Stahn solid state polarisers

Erice, 07. 2017 57

Thomas Krist HZB

Peter Boni TUM

Uwe Filges PSI

Artur Glavic PSI

for discussions and for

contributing to these slides

PNCMI 2016 proceedings: Journal of Physics – Conference Series 862 (2017)

the future (or past)J. Stahn solid state polarisers

Erice, 07. 2017 58

sonic screwdriver used by the Doctor to

reverse the polarity of the neutron flow

There must be a similar device to polarise it!

focussing neutron optics - paul scherrer institute · focussing neutron optics erice school neutron...

Documents