30 +44 (0) 1904 788600 | Edmund Optics®
Section 10:
Superpolished Optics
10-1
Spatial Frequency (mm-1)
Relative Intensity
4” Conventional Interferometer
(Figure) White Light
Interferometer
(Waviness) Atomic Force
Interferometer
(Roughness)
100 101 102 103 104 105
Figure 10.1: Measurable spatial frequency ranges of several metrology
technologies, revealing overlapping capabilities1
The desire to achieve higher throughput and lower loss in laser systems
drives the need for optical components with minimal scatter, particularly
in applications involving short wavelengths or high-power lasers. Optics
that minimize scatter using ultra-low surface roughness are commonly
referred to as “superpolished”. While no industry standard exists for
what roughness qualifi es an optic as superpolished, Edmund Optics® established
a process for polishing optical surfaces to a root mean square
(RMS) surface roughness of less than one angstrom (10-10 m) for partsper
million-level scattering. These sub-angstrom, low-loss surfaces are
ideal for precise laser applications including cavity ring-down systems
and laser gyroscopes.
Superpolished optics also compliment low-loss coating technology such
as ion beam sputtering (IBS). The spectral performance of these coatings,
if skillfully deposited, is often limited by the roughness of their
substrates.
Section 10.1: Measuring
Sub-Angstrom Surface Roughness
Every tool used for metrology has its own unique measurable spatial
frequency range. Figure 10.1 displays the overlapping spatial frequency
ranges of three diff erent devices commonly used to measure surface errors:
conventional interferometry, white light interferometry (WLI), and
atomic force microscopy (AFM).
Diff erent spatial frequency ranges are classifi ed as diff erent categories of
surface errors. These groups are not clearly defi ned frequency boundaries
but are widely-accepted to cover certain general ranges. A conventional
interferometer using a HeNe laser is well-suited for measuring
fi gure error, which is low spatial frequency error associated with typical
Zernike polynomials. The spatial frequency range of conventional interferometers
slightly overlaps WLI’s mid-spatial frequency range, but WLI
is still a better option for measuring waviness, which is a fi ner level of
surface errors. Waviness begins to contribute to performance degradation
from scatter. Both WLI and AFM can measure roughness, or higher
spatial frequency errors, but application requirements determine the optimal
instrument to use. Applications in the visible or IR spectra are generally
measured at frequencies less than 2,000 cycles/mm, in which case
WLI is an ideal metrology method. AFM is optimal for taking a closer
look at an optic’s surface and can be necessary for UV applications, as
higher spatial frequencies may need to be measured.
Figure 10.2: White light interferometry data of a superpolished surface
manufactured by Edmund Optics®, showing a sub-angstrom RMS surface
roughness