Folie 1

1
The UCN double plate experiment
S. Heule Paul Scherrer Institut, Villigen
and Physik-Institut der Universität Zürich
2
Ultracold neutrons

• Ultracold neutrons (UCN) are free neutrons which
  are reflected on certain materials under any
  angle of incidence
• Purpose measurement of fundamental properties of
  the neutron

Properties Have kinetic energies in the 100 neV
range Move with velocities of few m/s
3
Interactions

• The four fundamental interactions
• Gravitational interaction VG mngh
• 100 neV m-1
• Magnetic interaction VM -µ B
• 60 neV T-1
• Weak interaction
• beta decay of the neutron
• Strong interaction with nuclei
• Fermi potential up to 300 neV
• Description potential step
• ?reflection on surfaces / walls ? UCN storage

4
Fermi potential

• Reflection on walls

vnvC
vn
vn
VFermimvC2/2
5
UCN wall materials
UCN source
diffuse reflection
UCN experiment
specular reflection
5-6 m

• High Fermi potential
• Low loss probability per wall collision
• For UCN guides Specular Reflectivity
• Diffuse reflection leads to storage effect /
  decreased transmission

6
Wall materials

• Good wall materials have a Fermi potential above
  250 neV, e.g.
• Beryllium
• Nickel (natural, 58Ni)
• Non-hydrogenated diamond-like carbon (DLC)

7
Testing surfaces

• UCN guides Highly specular reflection required
• Idea Reflectometry with many reflections
• Test under real conditions use UCN
• The solution
• The double plate experiment

8
The parallel plates experiment

• of reflections determined by gap of the plates

S2
S3
S1
S4
D2
D1
9
Setup at ILL
10
Sample plates

• Length 500 mm
• RMS-roughness determined by atomic force
  microscopy (AFM, 1µm x 1µm)
• Floatglass
• Untreated (RRMS0.9 nm)
• HF etched (RRMS1.2 nm)
• Sand-blasted
• Nickel
• Coating on Floatglass
• PSI (RRMS1.6 nm)
• S-DH Heidelberg (RRMS0.25 nm)
• Diamond-like carbon
• From Fraunhofer IWS Dresden (by Ion beam sputter
  deposition, RRMS0.6 nm)
• Stainless steel (RRMS1.6 nm)

11
first analysis steps

• Subtraction of separately measured background for
  both detectors
• Normalized rates D2/D1
• Compensate for fluctuations in the neutron flux
• n(d) calculated from the plate distance
• Angular acceptance determined by apertures S1 and
  S2

12
Simple model

• Loss factor per wall collision
• Non-specularly reflected UCN are lost
• I(d)I(n)I0 Rn
• Incident rate I0 should be the same for all
  samples

I00.87
I00.24
13
Micro roughness

• Textbook model developed by Steyerl and
  Ignatovich
• Small surface irregularities lead to constructive
  interference of scattering from different depths
  of the surface layer
• Leads to conditions which are satisfied for the
  double plate experiment if RRMS 5 nm
• Two characteristic numbers RMS-roughness and
  correlation length

14
Probability diffuse scattering

• RRMS1 nm, Lcorr25 nm, ?I50, En200 neV,
  VFermi252 neV

z
?
y
F
x
According to A. Steyerl, Z. Physik 254 (1972)
169-188
15
Integral probability

• RRMS1 nm, Lcorr25 nm, VFermi252 neV

vnvC
According to A. Steyerl, Z. Physik 254 (1972)
169-188
16
Test 1 converging plates

• Smallest angle (36) is transformed to 90 at
  0.24 converging angle of upper plate(gap 5 mm)
• For all samples
• Rates above0.24 aresignificantlyabovebackgrou
  nd

17
Test 2 Variation of S4

• Double slit aperture S4 is replaced by a bar or
  removed completely

NiHD Ni 5/10
18
Converging angle

• Systematic variation of the RMS roughness shows
  agreement with measurement at R1 nm

Sample Ni HD
19
S4 variation

• Systematic variation of the RMS roughness shows
  agreement with measurement at R1.4 nm

Sample Ni HD
20
Conclusions

• UCN double plates experiment performed at ILL
  Grenoble
• Preliminary analysis shows significant fractions
  of diffuse reflection for various samples
• Diffuse reflection can be explained qualitatively
  by the Micro roughness model

21
Acknowledgement

• The work was performed at the Paul Scherrer Instit
  ut, Villigen, Switzerland and at the Institut
  Laue-Langevin, Grenoble, France
• Thanks to all who contributed to the double plate
  experiment
• This work is supported by the Swiss National
  Science Foundation (SNF, grant 200021-105400)
• Information about UCN_at_PSI onlinehttp//ucn.web.p
  si.ch

22
The PSI UCN source
rexpgt 1000 cm-3
Diamond-LikeCarbon coatedUCN guides

• Requirements
• High Fermi potential
• Low losses on wall collisions
• Specular reflection

rUCN6000cm-3
2 mA,600 MeVproton beam1 duty cycle
3 m3 D2O