Transcript PPT - Phase 2 - Canadian Light Source

Insert Title Here
SRI 2007: CLS Optical Metrology Facility- Overview
Brian
1
1
Yates ,
Dylan
1
Maxwell
Canadian Light Source Inc., 101 Perimeter Road, Saskatoon, SK S7N 0X4, Canada
Canadian Light Source Metrology Facility Overview:
The CLS Optical Metrology Facility performs beamline ray-tracing/modeling and determines the quality of the optical components that are installed in the CLS beamlines. This requires measurement of a
variety of mirror parameters, to ensure compliance by the vendor to CLS specifications. Typical mirror parameters are surface roughness, calibration of radius of curvature, and figure slope error.
Measurements from three distinct instruments are typically taken to characterize the quality of optical components.
The three instruments are the Ocean Optics Long Trace Profilometer (LTP), the Zygo Verifire AT Fizeau Interferometer (FI), and the Micromap 570 Interferometric Microscope (IM). The LTP directly
measures the slope along a single line on the optical surface. The radius of curvature and slope error is then calculated from this slope profile, which has a spatial period range of approximately 2 mm to
the length of the surface. Typically three tangential and three sagittal slope profiles are acquired for a complete measurement of an optical surface. The FI measures the height map of the optical surface
component. The radius of curvature and slope error can then be calculated from this surface height map, which has a spatial period range of approximately 1 mm to the length and width of the surface.
One measurement is sufficient to characterize the entire surface, but typically at least three measurements are acquired for error analysis purposes. The IM measures the height of the optical surface
over an approximately one millimetre square field-of-view. The surface roughness can then be calculated from this surface height map, which has a spatial period range of approximately 1µm to the
length and width of the field of view. For extremely smooth surfaces, three measurements are required at each sample location, and several locations must be sampled on the optical surface in order to
be fully characterized. Therefore, a single optical component that is fully characterized in the Optical Metrology Facility produces a significant number of data sets, in different formats, requiring analysis
and summarization.
Capabilities
Facility and Instrumentation
1. Class 10,000 Clean Room (± 0.1 degree Celsius temperature stability)
- All metrology instruments mounted on vibration isolation tables (TMC)
to minimize the effects of vibration.
2. 3D surface profiler (IM) from Micromap - Measures surface roughness.
3. Long Trace Profilometer (LTP) from Ocean Optics
- Measures figure “slope error” and radius of curvature.
4. Fizeau Interferometer (FI) from Zygo
- Measures figure “slope error” and radius of curvature using radius
bench measurement and distance measurement interferometer (DMI).
The role of the optical metrology facility is to determine the figure and finish of mirror, crystal and
diffraction grating surfaces, to ensure proper spot sizes at the beamline experimental stations. Figure,
finish and mirror coating determines the photon flux and photon energy range delivered to the
beamline experimental station.
The optical metrology facility instruments are used to determine actual height and slope profiles of
these optical mirror surfaces that are used in x-ray, visible light, and infrared synchrotron beamlines.
The instrument measurement ranges often overlap, giving greater confidence in each of the
measurements.
Removal of the figure curvature (e.g. spherical, cylindrical) from the surface height map or slope map
yields the slope error. A large slope error will result in an undesirable experimental focus, where
photon flux, brightness and spot size have been compromised.
Instrument
3D surface profiler (Micromap 570)
Long Trace Profilometer (Ocean Optics)
Fizeau Interferometer (Zygo VeriFire AT)
3D Surface Profiler (Micromap 570)
3D surface profile
of a smooth
mirror, taken using
a 10X objective
with the Micromap
570 in “smooth
phase” mode,
using filtered white
light (550 nm with
a 25 nm filter
bandwidth). The
image represents
approximately 0.5
x 0.5 mm2 of the
mirror.
Horizontal Spatial Period
Range
Vertical Dynamic
Range
Approximate Measurement Limit
3D Surface Profiler
(Micromap 570)
(0.0002 – 1.25) mm
(0.010 – 150000) nm
0.03 nm RMS Height
Long Trace Profilometer
(LTP) (Ocean Optics)
(2 – 1500) mm
(0.014 – 10000) µrad
10 nm RMS Height
0.45 µrad RMS Slope Error
Fizeau Interferometer
(Zygo Verifire AT)
(0.039 – 150) mm (normal)
(0.39 - 1500) mm (grazing)
(5 – 1000) nm
20 nm RMS Height
0.45 µrad RMS Slope Error
The Micromap 570 3D surface profiler (shown to the far left) is a microscope-based instrument that utilizes visible light
interferometry to measure the surface finish (i.e. surface roughness) of reflective optical surfaces. A CCD detector mounted
on the microscope is used to measure an area (on the order of ~ 1 x 1 mm2, which is dependent on the microscope objective
used) with a height resolution of approximately 0.03 nm.
Four different objectives are currently available: 5X, 10X, 20X, and 50X, which enable us to cover a wide range of spatial
periods (0.0002 - 1.25) mm or spatial frequencies (0.0004 - 2.5) µm-1.
Three measurement modes are available in software – smooth phase mode (interferometry phase measurement), wave
mode (white light interferometry phase scan) and focus mode (interferometry focus scan). Smooth phase mode is ideal for
very smooth surfaces (0.02 – 100) nm Sq roughness, but is limited to a maximum height measurement of 4 µm (focus depth).
Wave mode is ideal for smooth surfaces with heights and steps <5 µm. Focus mode is ideal for rough surface textures and
real surface heights >5 mm (maximum height measurement is 150 µm).
Long Trace Profilometer (LTP) from Ocean Optics (shown to the far left) measures the slope and curvature of
Long Trace Profilometer (LTP) (Ocean Optics) The
mirror surfaces along one dimension using a zero-path difference interferometer. A solid state HeNe laser source is
used to probe the mirror under test. The CLS LTP can measure optical surfaces up to 1.5 meters in length, covering
the range of spatial periods from (2 – 1500) mm or spatial frequencies of (0.00033 – 0.25) mm-1.
Surfaces of virtually any shape can be measured in situ, as long as the surface slope change is within the ± 5
milliradian acceptance angle of the LTP optical system, with a sensitivity of 0.1 µrad and 10 nm in height. Dr. Peter
Takacs and staff developed the LTP at Brookhaven National Laboratory (BNL).
Height profile of a “reference” spherical
mirror – the edges are raised ~20 µm.
The instrument is mounted on a vibration isolation table to minimize the effects from external vibrations, and
shrouded in a plastic curtain normally while measurements are taken (not shown in photograph to the far left), in
order to reduce air turbulence effects on the optical head. The bearing has recently been upgraded from a crossedroller bearing to a more precise linear air bearing. A further upgrade is planned in the near future to an even flatter
ceramic beam.
collimated HeNe laser beam (l=632.8 nm, 150 mm diameter)
The Zygo VeriFire AT Fizeau interferometer (shown to the far left) measures the
Return flat
Mirror under test
Fizeau Interferometer (Zygo Corp.)
height deviation of an optic under test from a reference surface. It accomplishes
this using phase shifting interferometry. The figure error of optical surfaces up to
150 mm in diameter at normal incidence, or larger optics at grazing incidence
angle (<1500 mm), can be measured with this instrument.
A picture of a grazing incidence measurement, in double pass mode is shown to
the left (middle position). A collimated HeNe laser beam (l=632.8 nm, 150 mm
diameter) is output from the Fizeau Interferometer through a reference
transmission flat (left), reflected off the mirror under test (right) and then reflected
back with a return flat (top right) to the detector in the interferometer. Double pass
refers to the fact that the collimated beam is reflected twice off the test mirror.
Transmission Flat
Fizeau Interferometer
Radius bench measurement/DMI Laser
Our Operating
Funding Partners
45° Grazing incidence measurement,
in so-called “double pass” mode.
3D height profile of a transmission flat,
measured in “single pass” normal incidence
mode.
www.lightsource.ca
SRI 2007: CLS Optical Metrology FacilityOptical Tip/Tilt Stage, Software Development
Insert Title Here
Brian Yates1, Dylan Maxwell1, Siyue Chen1, Bruce Truax2
1 Canadian Light Source Inc., 101 Perimeter Road, Saskatoon, SK S7N 0X4, Canada
2 Diffraction Limited Design LLC, 388 Wedgewood Road, Southington, CT 06489, U.S.A.
Canadian Light Source Metrology Facility – Software Development:
These metrology instruments are each provided with their own control and analysis software. In general this software provides
adequate functionality for standard metrology applications. However for the measurement of high quality precision synchrotron
optics, increased or improved functionality is often required. Furthermore, it is impossible to determine, verify, or modify the
algorithm that has been implemented in a closed-source commercial software package. The process of generating a detailed
report for each optical component can also be very tedious and time consuming. The data from each instrument is analyzed
using the corresponding commercial software packages. This analysis is recorded, typically with a screen capture, to be later
inserted into the report document. If analysis is needed that is beyond the capabilities of the provided software, then the data
must be exported to another custom software package. A summary is then created manually, typically by tabulating all of the
results. This process must be repeated for each optical component measured. It is prone to error and very difficult to
automate.
The CLS Optical Metrology Facility has begun developing software to address the above mentioned deficiencies in our
commercial software. The goal of this software is to provide advanced analysis functionality for all three metrology instruments,
pre-/post-processing, as well as support for automation and report generation. The core functionality to be implemented
includes error analysis, curve fitting, and power spectral density (PSD) for both profile (1D) and surface (2D) data. This
software has been developed in C/C++ using the ROOT data analysis framework [Brun et al.]. ROOT is an open source
package being developed at CERN and provides many advanced analytical and graphical features. Currently the software is in
the development and testing stages with a focus on curve fitting for the analysis of LTP and FI data.
Figure 1 (on the right) shows the results of a spherical fit (actually a polynomial approximation) to FI height data for a CLS
VESPERS beamline KB mirror. The surface height data was calculated by averaging three FI measurements acquired
successively. The surface height uncertainty was calculated from the variation in the original measurements. A data window
has been applied to select only the clear aperture of the mirror. The best fit piston (z = C), tilt (z = Ax + By +C),
power (z = D(x2 + y2) + Ax + By + C) and twist (z = Fxy + Ax + By + C + Dx2 + Ey2)) were then individually removed from both
the height data and height uncertainty data. The resulting surface height plot is shown in the left graph of Figure 1, with a
histogram of the surface height data below. The resulting surface height uncertainty is shown in the right plot, with a
corresponding histogram below. The histograms are useful to verify whether the distributions are approximately normal or not.
The root mean square (RMS) and mean values of the surface height plot are displayed in the upper left results box, along with
the respective errors calculated from the height uncertainty plot. The upper right results box displays the radius of curvature
and the fitted parameter results, all with their respective errors. This particular KB mirror has a height error of
0.0115 ± 0.0002 µm and a radius of curvature of 42591 ± 605 m. These results are formatted such that they can be easily
printed, or combined with others to generate a comprehensive report.
References:
Rene Brun and Fons Rademakers, ROOT - An Object Oriented Data Analysis Framework, Proceedings AIHENP'96 Workshop,
Lausanne, Sep. 1996, Nucl. Inst. & Meth. in Phys. Res. A 389 (1997) 81-86. See also http://root.cern.ch/.
Figure 1: Spherical fit of Fizeau Interferometer data for a KB mirror (VESPERS beamline).
Canadian Light Source Metrology Facility – Hardware Development:
An optical tip/tilt stage that can handle large synchrotron mirrors has been designed and machined (See Figure 2 below). Commercial stages were considered
but discarded later for a variety of issues - stability, high cost, insufficient mirror length capability, insufficient translation range, or deficiency in tip, tilt, pitch
adjustments. This flexible design can be used with all three metrological instruments in the CLS Optical Metrology Facility.
The conceptual design was developed by Bruce Truax of Diffraction Limited Design LLC. This preliminary design was further refined and finalized by Siyue Chen
within the CLS Engineering and Technical Services Division. The stage can handle optical elements up to 1400 mm in length and approximately 200 mm in
width, and weights up to 200 kg. A lateral motion of ± 25 mm in (x,z) can be achieved using micrometers. The height (y) can be finely adjusted ± 10 mm using
three finely threaded screws (750 mm per thread) that are mounted to 100 mm diameter air bearings (porous silicon carbide) from New Way. When these air
bearings are pressurized the stage can be easily moved while loaded with a heavy optic, even on a tapped optical table. The stage has sufficient rotational
degrees of freedom (tip, tilt, pitch) to allow for alignment of the optical element (See Figure 2). A retroreflector mount is provided to accommodate a distance
measurement interferometer (DMI), which is useful for radius bench measurements on short radius spherical or cylindrical optics (R < 2 m).
Figure 3 shows the underside of the optical table and gives a good view of the air bearings and ribbed support structure that provides rigidity. Figure 4 shows the
mounting base for the table – complete with the pivot post that holds the spherical bearing (Figure 5), micrometers to move the mounting base laterally (x,z), and
cut-outs to allow for rotation of the table through 180°. A symmetrical design allows for two possible configurations, which provides for either clockwise or
counter-clockwise rotation of the table. The lateral micrometers (x,z) can be removed to mount the radius bench measurement guide bar (2 m), which allows the
table and mount to be moved longitudinally (z). Figure 5 shows the rail release mechanism and cam followers (see also Figure 4) used for the radius bench
measurement.
Currently the stage is undergoing testing with the Fizeau Interferometer, and will be used for measuring the Phase II beamline optical components.
Retroreflector Mount (for DMI Laser + Radius Bench Measurement)
Retroreflector Mount
Cam follower (radius bench measurement)
Air Supply
Pitch
New Way air bearing (100 mm) - SiC
Micrometers for lateral (x) and
along beam (z) adjustments
y
Tilt
Rotational scale with pivot post below
(not visible, see Figure 4) –
pitch rotation of table used for
grazing measurements
z
x
Cut-out to allow rotation of table
(New Way air bearing)
Tip
Pivot Post for spherical bearing
Tension spring required for micrometer movement
Micrometers for lateral
(x,z) adjustments
Mounting Base securing screw with teflon spacer interface
Figure 4: Mounting Base for Table, showing pivot post, translation micrometers, retroreflector mount.
Removable handles
Figure 2: CLS optical tip/tilt stage (top-down view) to handle large synchrotron optics.
Cam Follower
Spherical Bearing (mounts to Pivot Post)
Rail Release – attaches to mounting
base for radius bench measurement
Fine Pitch adjustment for table
Compression Spring
New Way air bearing height adjustment (750 mm per thread)
Figure 5: Miscellaneous components
Figure 3: Side View showing ribbed support structure and New Way air bearings.
Our Operating
Funding Partners
www.lightsource.ca