Trey Turner, REO Inc.
A review of superpolished surfaces, including fabrication and measurement, as well as some of the leading applications that depend on these optics.
Superpolished optics – that is, components with subangstrom surface roughness – represent a critical enabling technology in some of the most sensitive and high-precision lasers and laser-based systems.
They are some of the most demanding components to make, coat and measure, pushing optical fabrication and metrology technology to their very limits. However, the extraordinary performance delivered by these optics enables some of the most cutting-edge and exciting applications for lasers today.
Precision optics virtually always carry a specification for surface shape, or maximum wavefront distortion on transmission. Typically, shape is specified as power and irregularity. This is the acceptable departure from the desired radius of curvature (power) and flat or spherical shape (irregularity).
These specifications allow the system designer to calculate the amount of incident light energy that can be concentrated into a focused spot or propagated in a given beam diameter over a particular distance. But they do not address one important practical characteristic of all real-world optical surfaces: scatter. Scatter is dependent upon small-scale surface microroughness and causes light to be deflected out of the desired direction of propagation, reducing system efficiency or signal-to-noise ratio.
Superpolishing is an optical fabrication technique developed specifically to achieve minimal defects and micro-roughness values. Whereas traditionally polished glass components might have a surface roughness of about 3 Å, superpolished optics are characterized by roughness values below 1 Å, which can reduce scattered light below the 1-ppm level. One of the most common techniques used for superpolishing is submerged polishing.
In traditional polishing, optics are mounted on a spindle, which is rotated while a lap – a metal-grinding tool – moves back and forth over the component surfaces to polish them. A slurry, consisting of small abrasive particles in a fluid, flows over the optical surfaces. The sizes of the abrasive particles are reduced over time to achieve finer degrees of polishing.
Most superpolishing techniques modify this method by submerging the entire spindle/lap assembly in the polishing fluid. This provides two main benefits: First, the surface tension of the top (exposed) surface forms a barrier that helps protect the lap and workpieces from outside contaminants that would scratch the optical surface. Second, submersion increases thermal conductivity, which causes the lap and substrate to be at virtually the same temperature. This results in improved shape consistency of the tool and workpiece, which is also a factor in achieving a smoother polished surface.
Every manufacturer has variants on this basic method as well as proprietary techniques to achieve specific goals. For example, at REO, we pay close attention to the particle size distribution in the slurry throughout the polishing process; also, we monitor the actual chemistry of the slurry.
This is because polishing particles tend to have functional groups on their outsides that can hold a positive or negative charge – which, depending upon the pH of the overall slurry, can cause them to stick to each other or to the glass. Thus, by controlling slurry pH, we can control where polishing particles tend to migrate, avoiding agglomeration. As a result, we can consistently achieve microroughness levels lower than 0.5 Å on materials such as fused silica, optical glass (for example, BK 7) and Zerodur.
Making superpolished optics requires the ability to quantify surface roughness or its consequences, and this requires a different set of tools than is traditionally used for optical surface metrology. One common technique for examining uncoated superpolished substrates is optical profilometry.
An optical profilometer looks similar to a conventional optical microscope with a halogen light source. Internally, it uses interference between the surface under test and an extremely flat internal reference surface to create fringes that are a direct indicator of the profile of the test surface.
An optical profilometer is a microscope-based instrument that can measure surface height to the subangstrom level. Suitable for production environments, optical profilometers are a sensitive, accurate way of quantifying surface roughness.
Commercially available optical profilometers can routinely achieve a vertical resolution (measuring the height of features) of less than 0.1 nm, and can even achieve 0.01-nm resolution under certain circumstances. Thus, they are an accurate and sensitive way of quantifying surface roughness and are suitable for production environments.
However, it is important to understand that optical profilometers are limited in terms of the range of spatial frequencies in the X-Y plane over which they can measure surface undulations. For example, variations that occur over either very large (tens of millimeters) or small (atomic scale) distances usually cannot be measured.
The differential interference contrast – or Nomarski – microscope is another method frequently used for evaluating the surface roughness of uncoated surfaces. This technique again uses interference; however, instead of comparing the surface under test to a flat internal reference, the Nomarski microscope compares two views of the surface under test that are slightly shifted relative to one another, in a manner similar to shearography. Thus, small local changes in surface slope are converted into contrast differences in the image, making this a particularly good technique for identifying small defects. But this method generally is not as quantitative as optical profilometry.
For coated optical surfaces, it is generally more useful to measure the net effects of both surface roughness and coating imperfections than to attempt to quantify the precise surface profile. The most direct method for accomplishing this is to expose the surface under test to laser illumination and then look at scattered light at one or more off-axis positions. Another very sensitive technique for gauging the total loss for an optic (including absorption and scatter) is to place it in a cavity and measure the ringdown time.
CEAS: Trace gas measurement
Superpolished mirrors are vital to several important applications that depend on the ultralow scatter (<5 ppm) and extremely high reflectivity (>99.999 percent) that these optics can uniquely provide. Instruments based on cavity-enhanced absorption spectroscopy (CEAS) are standout examples. In particular, CEAS is being used in a new generation of instruments for monitoring greenhouse gases and other atmospheric pollutants, industrial process monitoring, leak detection in natural gas pipelines, agricultural monitoring and surveying, and water cycle and rainfall studies.
CEAS is a form of absorption spectroscopy. Every small molecule has a unique absorption spectrum consisting of a well-known pattern of sharp absorption lines. So, in principle, this absorption spectrum can be used to make real-time composition measurements on any gas mixture. In practice, however, the absorption signal from trace gases is usually orders of magnitude less than the instrument shot noise and cannot be observed.
In CEAS instruments, a sample of the gas under test flows into an optical cavity with superpolished end mirrors. Laser light is introduced into this cavity, where the high reflectivity and low scatter allow the beam to make many back-and-forth trips before leaking out through one of the mirrors.
This yields an effective path length more than 10,000 times longer than the cell itself, resulting in a highly improved signal-to-noise ratio. This precision enables CEAS instruments to routinely detect trace gases with parts-per-billion sensitivity; even parts-per-trillion performance is possible. Furthermore, the technique is accurate enough to distinguish between molecules of a given species that have different isotopic constituents.
Cavity ringdown spectroscopy is a form of cavity-enhanced absorption spectroscopy. A laser beam is sent through an optical cavity containing the gas mixture to be measured; then the laser is shut off, and the amount of time it takes for the laser intensity to decay to 0 (called the ringdown time) indicates the concentration of a given species in the cavity.
Common CEAS techniques include various types of cavity ringdown spectroscopy using pulsed or CW lasers, and off-axis integrated cavity absorption spectroscopy. More exotic variants include mode-locked CEAS, based on broadband femtosecond lasers and high-resolution grating monochromators. Mirrors for the latter are some of the most demanding optics we produce at REO because they must offer high reflectivity and low scatter over a wide (>75 nm) bandwidth while avoiding stretching the femtosecond pulses with unwanted group delay dispersion.
Ring laser gyroscopes
The ring laser gyroscope (RLG) is a commonly used component in navigational systems for spacecraft, and commercial and military aircraft. Specifically, the RLG is used to sense rotation in a single axis, and it delivers several advantages over traditional mechanical gyroscopes. For example, it does not suffer from wear because it has no moving parts, and the lack of internal friction eliminates the drift inherent to mechanical gyros. Also, RLGs are typically smaller, lighter and more rugged than mechanical gyros, which is particularly advantageous in airborne and spaceborne systems.
A ring laser gyroscope is a HeNe laser cavity configured as a ring. Rotation in the ring plane causes a shift in the interference pattern between the two counterpropagating cavity modes; this shift is sensed by a detector.
Most RLGs are based on helium-neon (HeNe) laser cavities. However, the cavity is constructed with three or more mirrors, rather than just the usual two, so that it has a ring shape. It is configured so that only two counterpropagating longitudinal modes are supported (clockwise and counterclockwise). One cavity mirror is constructed to allow a slight amount of incident light to be transmitted. The two transmitted beams are recombined and sent to a photodetector.
When the system is at rest, the frequencies of the two longitudinal modes are the same, and the interference pattern created by the superposition of the two beams remains fixed. However, rotation of the system in the axis perpendicular to the beam plane creates an apparent path length difference for the modes. Namely, in the inertial frame of reference, the beam going in the direction of the rotation appears to have farther to travel to reach its starting point, while the opposite is true for the beam going in the other direction. In the RLG frame of reference, this manifests itself as a slight frequency difference, causing the interference pattern of the two modes to move from its original position. Both the rate and direction of movement of the interference pattern are sensed by the photodetector and used to accurately measure gyro motion.
RLGs require superpolished mirrors so as to prevent any light from being scattered from one mode into the other. Any modal crosstalk reduces the contrast of the interference fringes upon which the measurement technique critically depends.
Green HeNe lasers
A more standard type of HeNe laser also requires superpolished mirrors. Specifically, these are HeNe lasers that output in the green (543 nm), rather than in the more commonly available red line (633 nm).
Probably the most important applications for these green HeNe lasers are in various bioimaging techniques, such as confocal microscopy and flow cytometry. The latter is a technique widely used to count and/or sort specific cell types from a large population of living cells. It is used extensively in research and clinical diagnostics; e.g., for diagnosing blood cancers and for measuring various genetic- and protein-related cell characteristics.
In flow cytometry, fluorescence-tagged cells are forced to flow in single file through one or more tightly focused laser beams; a detector array collects fluorescence excited by the laser as well as any scattered light. The green HeNe is a compact, reliable and economical source for this process.
Flow cytometry is most commonly performed on a whole blood sample that has been mixed with one or more fluorescent antibodies that bind to specific cell types. The prepared cells are then forced to flow in single file through one or more tightly focused laser beams. Detector arrays collect fluorescence excited by the laser and may also sense scattered light (which provides information about cell size, shape and structure). Current instruments can count several thousand cells per second.
The larger the number of different excitation wavelengths used, the greater the range of fluorescent probes that can be employed. This, in turn, enables a higher number of distinct cell parameters to be measured. Several popular fluorescent probes can be excited by 543-nm light, and the green HeNe is a compact, reliable and economical source.
The challenge in making green HeNe lasers is that the medium has low gain at the 543-nm transition. Thus, mirror reflectivity must be maximized and cavity losses minimized to successfully extract any significant power at this wavelength. Scatter is particularly problematic because it increases rapidly with decreasing wavelength. Thus, ultralow-scatter, superpolished mirrors are key to the successful production of green HeNe lasers.
Meet the author
Trey Turner is chief technology officer at REO Inc. in Boulder, Colo.; email: firstname.lastname@example.org