Download Optics for next generation light sources Anton Barty

Optics for next
generation light sources
Anton Barty
Centre for Free Electron Laser Science
Hamburg, Germany
Tuesday, March 2, 2010
Key issues
• Optical specifications
• Metrology (mirror surfaces)
• Metrology (wavefront, focal spot)
• Coatings and damage
• Novel optical structures
Tuesday, March 2, 2010
Overview of the LCLS front-end optics
Source
Beamline
Offset mirrors
Endstation
- ~30 µm Gaussian electron
beam near undulator exit
- Can easily incorporate
a variety of source models
- Apertures
- Slits
- etc.
- Measured figure
- Measured PSD
- Known geometry
- Intensity at endstation
- Intensity at focus
- Total wavefront error
Tuesday, March 2, 2010
LCLS front end mirrors direct the beam into
different end stations
l
Near experiment hall
Front end enclosure
LCLS X-ray beam
M1S
e
-
Soft X-ray offset mirrors
(SOMS, 0.8 to 2 keV)
Tuesday, March 2, 2010
M2S
0.8 - 2 keV
0.8 - 2 keV
M3S2
M3S1
2 - 24 keV
M1H
M2H
Hard X-ray offset mirrors
(HOMS, 2 to 24 keV)
Hutch 1
(AMO)
Hutch 2
(SXR)
Hutch 3
(XPP)
Tunnel
2-
Hu
(X
A wave optics formalism is must be used for
coherent X-ray beams
M1S
M2S
M3S/M4S
Source
Beamline
+
(Gaussian profile)
Tuesday, March 2, 2010
Offset mirrors
+
(Apertures)
+
(Measured 1D figure)
Endstation
=
(Statistical roughness)
????
(Intensity at focus)
Complete mirror PSD is obtained from
interferometry, surface microscopy and AFM data
Interferometer
Si test substrate by Vendor 1
Si test substrate by Vendor 2
~ 0.16 µrad rms!
~ 0.3 µrad rms!
! = 0.19 nm rms!
! = 0.19 nm rms!
! = 0.16 nm rms!
! = 0.36 nm rms!
Tuesday, March 2, 2010
Zygo 2x
Zygo 20x
AFM
AFM
2.95 mm
0.37 mm
10 µm
2 µm
Regina Soufli, LLNL
[email protected]
Tuesday, March 2, 2010
Interferometry measurements are used to create
composite mirror surfaces
Interferometer data
Zygo
Figure
Model surface
(Measured 1D figure)
PSD
Model surface
AFM
(Statistical roughness)
Tuesday, March 2, 2010
Grazing incidence optics lead to a dilation of the
mirror surface power spectrum
Incident wavefront
Reflected wavefront
(x’, y’)
W=2h sin(!)
y’ = y
x’=x sin(!)
!
Mirror surface h(x,y)
fy
ymirror
Mirror surface height
(viewed normal to surface)
Tuesday, March 2, 2010
Mirror surface PSD
(isotropic distribution)
fy
ybeam
fx
xmirror
z
fx
xbeam
Wavefront error
(mirror reflecting at angle !)
Wavefront PSD
(elliptical distribution)
Coherent X-rays demand wavefront propagation
to model the optical system
At M1S
At M2S
At M3S
Intensity
Intensity
Intensity
(0.0139 radian from grazing)
(0.0417 radian from grazing)
(0.0417 radian from grazing)
SOMS #4 Figure (phase)
SOMS #4 Figure (phase)
SOMS #4 Figure (phase)
Source
(Not to scale)
(colour = phase)
~1cm
Note: For testing purposes the same figure and PSD was used for all SOMS mirrors. This is trivial to change
when we have new metrology data - the use of beamline scripting simplifies changes (although some format
checking is still required).
Tuesday, March 2, 2010
Mid spatial frequency range gives rise to
additional beam structure
At AMO endstation no MSFR
At AMO endstation with measured MSFR
~5 mm at enstation
Tuesday, March 2, 2010
Beam structure is affected by the
mid spatial frequency components
Figure only (no MSFR)
Tuesday, March 2, 2010
With MSFR included
We expect measurable intensity variation
at the entrance to the AMO endstation
1.0
0.8 keV
0.8
0.6
0.4
0.2
0.0
0
100 200 300 400 500 600
1.0
2.0 keV
0.8
0.6
0.4
0.2
Without MSFR
0.0
0
With measured MSFR
~5 mm at endstation
Tuesday, March 2, 2010
100 200 300 400 500 600
Specifications for SOMS mirrors
Table 1. Surface specifications within the clear aperture (CA) of the SOMS mirror substrates.
§
Spatial frequency
range
-3
-1
CA to 10 µm
MSFR
HSFR
10 µm ! 0.5 µm
-1
-1
0.5 µm ! 50 µm
-3
Mirror #
SN1
SN2
SN3
SN4
SN5
Tuesday, March 2, 2010
-1
-1
Spatial wavelength
range
CA to 1 mm
2 µm ! 1 mm
20 nm ! 2 µm
Figure (nm RMS)
1.8
1.3
1.2
0.64
1.4
Specification
<2 nm RMS and
<0.25 µrad RMS
< 0.25 nm RMS
< 0.4 nm RMS
Slope error (µrad RMS)
0.19
0.2
0.37
0.14
0.37
Fig. 3. Figure (left) measured at LLNL for the five SOMS mirror substrates available to be installed
at LCLS. Measured PSD curve (right) for SOMS sn4 mirror, with the PSD in the HSFR range shown
Tuesday, March 2, 2010
SOMS system (dashed line). Because metrology data are not yet available for the AMO
focusing optics this calculation likely places an upper bound on the focusing spot properties
given the predicted SOMS front-end performance. One point that is immediately apparent is
that the focal spot is likely to vary in shape with energy due to the dependence of wavefront
error and propagation properties on the photon energy, for different optical path errors and
footprints on the mirrors across the SOMS energy range.
Tuesday, March 2, 2010
Fig. 4. Calculated intensity distribution of the FEL beam at the last pop-in intensity monitor before the
endstation in hutch 1 (at z=120 m from the undulator exit). Calculation takes into account the measured
SOMS figure and roughness of all three SOMS mirrors in the beam path. Vertical banding is due to the
Intensity at K-B focus is less severely degraded
Tuesday, March 2, 2010
Fig. 5. Focal plane intensity structure in the plane of highest intensity, calculated based on the
predicted SOMS front-end performance, figure 3. Solid lines represent the predicted intensity
profile and dashed lines represent the case of perfect optics. Strehl ratios for the above two
discussed in this manuscript. For these reasons, the measured results for surface figure and
roughness on the HOMS Si substrates shown in Fig. 6 were used in the wavefront
propagation calculations and are expected to be fairly representative of the wavefront results
from the final, SiC-coated HOMS mirrors. For the purposes of this study we use the figure
and finish data for HOMS substrates sn2 and sn3 respectively for the M1H and M2H optics.
A total of 4 HOMS mirror substrates have been fabricated, with the intention to ultimately
install 2 and use the remaining 2 as spares.
Specifications for HOMS mirrors
Mirror #
SN1
SN2
SN3
SN4
Figure (nm RMS)
2.4
1.0
2.0
1.5
Slope error (µrad RMS)
0.27
0.27
0.22
0.23
Fig. 6. Figure error (left) measured at LLNL for the four HOMS mirror substrates, and measured
PSD (Right) for HOMS #2 Si substrate. . Bottom: rms values for the figure of each HOMS
substrate within the central 420 mm-length, derived after subtraction of the best-fit sphere.
Tuesday, March 2, 2010
HOMS performance at CXI endstation
Tuesday, March 2, 2010
Fig. 7. Calculated intensity distribution of the FEL beam at the last pop-in intensity monitor
before the CXI endstation in hutch 5 (at z=383 m from the undulator exit). Intensity images at
at 0.8 and 2.0 keV are shown on the top and horizontal intensity profiles through the maximum
are plotted on the bottom Solid lines represent the predicted intensity profile and dashed lines
Predicted focal spot of the CXI instrument
Tuesday, March 2, 2010
Fig. 8. Calculated focal plane intensity distribution at the focus of the CXI endstation using the
1m focal length K-B optics. Solid lines are the predicted intensity profile and dashed lines
represent the case of perfect optics. Strehl ratios for the above two calculations (actual peak
The LCLS mirrors are coated with B4C
J.Bozek, R.Soufli
Tuesday, March 2, 2010
A compromise must be drawn between
surface roughness and stress
J.Bozek, R.Soufli
Tuesday, March 2, 2010
Optical mounts must be precision engineered for
ultra high stability
~300m to FEH
1 µrad => 0.3 mm
1 nrad => 0.3 µm
Tuesday, March 2, 2010
Offset mirrors
- Measured figure
- Measured PSD
- Known geometry
J. Krzywinski
Tuesday, March 2, 2010
Focal spot size is inferred from
crater size and damage models
J. Krzywinski
Tuesday, March 2, 2010
and now for something
completely different....
Tuesday, March 2, 2010
Sasa Bajt, PXRMS 2010
Tuesday, March 2, 2010
Pump-probe geometry with small windows allows us to
measure only the delayed diffraction signal
i) Drive the sample
ii) Probe the sample some time later
iii) Collect diffraction from probe
Tuesday, March 2, 2010
We demonstrated that a tamper is effective in containing
nanoparticle expansion
Tampered
Untampered
Low fluence
Tuesday, March 2, 2010
High fluence
Aluminium dot
Substrate (silicon nitride)
Silicon tamper
Aluminium dot
Substrate (silicon nitride)
Multilayer optics for two-colour pump-probe
experiments
Andy Aquila, PXRMS 2010
Tuesday, March 2, 2010
Multilayers for X-ray two-colour
time delay experiments
Andy Aquila, PXRMS 2010
Tuesday, March 2, 2010
Disposable zone plates for extreme focussing
Anne Sakdinawat, LBNL
Tuesday, March 2, 2010
Chirped multilayers for X-ray pulse compression
Andy Aquila, PXRMS 2010
Tuesday, March 2, 2010
•
Coherent beams demand wavefront propagation
for optics analysis and optic specification
•
•
Precision metrology is essential
•
Some optics are designed to be blown up!
Anything in the beam will diffract
(apertures, solid attenuators - everything!)
• •Credits:
•
•
Tuesday, March 2, 2010
LLNL X-ray optics group
Sasa Bajt, Andy Aquila (CFEL)
Jacek Kryzwinski (SLAC)
Measured HOMS figure error is used to determine the
optical surface profiles
10.00
8.00
Figure (nm)
6.00
4.00
2.00
HOMS #1
HOMS #2
0.00
HOMS #3
HOMS #4
-2.00
-4.00
-6.00
-8.00
-250.00
-200.00
-150.00
-100.00
-50.00
0.00
50.00
Position (mm)
Tuesday, March 2, 2010
100.00
150.00
200.00
250.00
We expect overfilling of HOMS mirrors to
degrade performance in the far experimental hall
Intensity at 8.0 keV and z=383 m
1.5
1.5
1.0
1.0
Position (mm)
Position (mm)
Intensity at 2.0 keV and z=383 m
0.5
0.0
0.5
0.0
-0.5
-0.5
-1.0
-1.0
-1.5
-1.5 -1.0 -0.5 0.0 0.5
Position (mm)
-1.5
-1.5 -1.0 -0.5 0.0 0.5
Position (mm)
Tuesday, March 2, 2010
1.0
1.5
1.0
1.5
Tuesday, March 2, 2010
Tuesday, March 2, 2010
Tuesday, March 2, 2010