Gradient-Index Research Group Metrology Instruments
Instrument
Wavelength
(μm)
0.4 – 1.1
Index
Range
Any
Absolute
/Relative
Relative
Material
Type
GRIN
Schmidt
Immersion
0.4 – 1.1
< 1.7
Absolute
GRIN
MWIR MachZehnder
Interferometer
3.0 – 5.0
Any
Relative
GRIN
*LWIR MachZehnder
Interferometer
8.0 – 12.0
Any
Relative
GRIN
Abbe and
Pulfrich
Refractometers
0.4 – 1.7
< 1.7
Absolute
Homogeneous
Metricon
Refractometer
0.4 – 1.7
< 2.0
Absolute
Homogeneous
Visible Sagnac
Interferometer
0.4 – 1.1
Any
Absolute
Homogeneous
Instrument
Wavelength
(μm)
3.0 – 5.0
Index
Range
Any
Absolute
or Relative
Absolute
Material Type
Visible MachZehnder
Interferometer
MWIR Sagnac
Interferometer
Homogeneous
Comments
Sample Geometry
Gives two-dimensional relative
index. Accuracy limited by
sample preparation and
mechanical properties.
Same as Mach-Zehnder but uses
index fluid that matches the
sample at some point. Limited by
available fluids.
Gives two-dimensional relative
index. Accuracy limited by
sample preparation and
mechanical properties. Cannot
use index fluid.
Gives two-dimensional relative
index. Accuracy limited by
sample preparation and
mechanical properties. Cannot
use index fluid.
Requires index fluid and wellcharacterized reference prism.
Sample must have lower index
than prism and fluid.
Critical angle measurement with
reference prism.
Thin (~1mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Thin (~1mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Thin (~2mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Figure error < 1λ.
Wedge error < 5μm over
region of interest.
Thin (~3mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Figure error < 1λ.
Wedge error < 5μm over
region of interest.
Figure error < 3λ
Accuracy limited by sample
thickness measurement and
sample preparation.
Comments
Small sample blank (5x5x5
mm3, up to (20x10xN mm3).
One polished surface and one
edge that is at least ground
Small sample blank (5x5x5
mm3, up to (20x10xN mm3).
One polished surface
required
Plane parallel window.
Diameter > 15mm
Thickness > 5mm
Sample Geometry
Accuracy limited by sample
thickness measurement and
sample preparation.
Plane parallel window.
Diameter > 15mm
Thickness > 5mm
Wedge < 5 arcmin
Thicker samples reduce
measurement uncertainty
*Instrument under construction or upgrade, expected completion by spring of 2015
www.optics.rochester.edu/workgroups/moore/downloads
Sample Prep Requirements
Figure error < 1λ.
Wedge error < 5μm over
region of interest.
Figure error < 1λ.
Wedge error < 5μm over
region of interest.
Figure error < 3λ
Wedge < 5 arcmin
Thicker samples reduce
measurement uncertainty
Sample Prep Requirements
7/14/2014
Gradient-Index Research Group Metrology Instruments
*LWIR Sagnac
Interferometer
8.0 – 12.0
Any
Absolute
Homogeneous
*Low Coherence
Distance
Measuring
Interferometer
0.650
N/A
Absolute
Thickness
GRIN or
Homogeneous
0.4 – 0.7
Any
Relative
GRIN
*Thermal
Interferometer
0.6328
Any
dn/dT and
CTE
GRIN or
Homogeneous
*Mini MachZehnder
Interferometer
0.6328
Any
Relative
GRIN
Beam
Deflection
Accuracy limited by sample
thickness measurement and
sample preparation.
Measures thickness of plane
parallel samples. Expected
accuracy around 100nm.
Plane parallel window.
Diameter > 15mm
Thickness > 5mm
Nominally plane parallel
sample with polished
surfaces
Wedge < 5 arcmin
Thicker samples reduce
measurement uncertainty
TBD
Calculate relative GRIN profile
by measuring laser deflection
through GRIN material.
Gives two-dimensional
measurement of index and
thickness change as a function of
temperature. Limited by phase
measurement error and
environmental stability and
monitoring.
Microscope configuration for
small samples or large fringe
densities.
One-dimensional gradient.
Plane parallel window or
cylinder geometry
Thin (~1mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Thin (~1mm), plane, parallel
slice. Index is constant
through sample thickness
(exposed gradient).
Advantages
Figure error < 1λ.
Wedge error < 3μm over
region of interest.
Reflective coating on one
half of one surface
Figure error < 1λ.
Wedge error < 3μm over
region of interest.
Instrument
Best Case Error Bar
Mach-Zehnder
Interferometer
2x10-5
- Very versatile and robust measurement method
- Schmidt immersion method provides absolute index
- Sample prep problems can limit measurement accuracy
- Sample prep is typically destructive
Abbe and Pulfrich
Refractometers
Sagnac Interferometer
1x10-5
- Robust measurement method
1x10-4
Limited by reference prism and available index fluids
Surface measurement only
Limited by sample thickness measurement
Sensitive to scatter and surface shape
Beam Deflection
1x10-3
Thermal Interferometer
CTE and dn/dT to 2
or 3 digits
-
-
Limited accuracy
Requires constant index in propagation direction
Double pass (reduces effective fringe resolution)
Requires environmentally controlled chamber
Average index measurement
Used for any index or wavelength range
Uses plane parallel sample rather than prism
Nondestructive test for certain geometries
Can be implemented in-process
Gives CTE and dn/dT as a function of (x,y)
Useful for GRIN and homogeneous materials
Disadvantages
*Instrument under construction or upgrade, expected completion by spring of 2015
www.optics.rochester.edu/workgroups/moore/downloads
7/14/2014
Gradient-Index Research Group Metrology Instruments
*Instrument under construction or upgrade, expected completion by spring of 2015
www.optics.rochester.edu/workgroups/moore/downloads
7/14/2014