With modern semiconductor detector arrays approaching near-ideal performance, the throughput of telescopes and their associated instrumentation is becoming limited by the coatings applied to the optical components. Mirror coatings must have the highest reflectivity attainable, be able to withstand extremes in temperature and humidity, be cleanable, and be as durable as possible. Transmissive optics (lenses), by contrast, require extremely low-reflectivity surfaces, yet the coatings to achieve these low light losses must also be essentially impervious to contaminants and periodic cleaning. Today's best coatings leave much to be desired in these regards. Aluminum is still the standard telescope mirror coating, despite losses as high as 10% in the optical red spectral region, while the best antireflection coating is soluble in water and therefore difficult to maintain.
Astronomy is not the only discipline that would gain from improvements to optical coatings. Solar arrays and solar concentrators, for example, also benefit directly from better throughput, and the potential economic impact of a 10% gain in electrical conversion efficiency would be astounding, when considered on the worldwide market.
Improvements in optical coatings require not only new and better formulations but also more uniform and reliable application techniques. The coatings lab at the University of California Observatories (Santa Cruz), headed by Dr. Andrew Phillips, has been engaged in efforts to improve both reflective and anti-reflection coatings for astronomical optics for a number of years. Starting with improvements to the infrastructure of their coating facilities, Dr. Phillips seeks to pursue two goals: a definitive comparison of the efficiency and uniformity of depositing coatings through the relatively new technique called "ion-assisted deposition" vs. the more traditional approach of "sputtering". The improvements will also extend their coating capabilities to include the reactive deposition of nitrides, which are critical to the highly reflective coating considered to be the current state of the art. A novel moving stage inside the vacuum chamber will also allow application onto large substrate areas with improved thickness and process uniformity. This same new equipment will be used to develop both high-performance protected-silver coatings for mirrors as well as multi-layer sol-gel anti-reflection coatings for large lenses. Funding for this work is being provided by NSF's Division of Astronomical Sciences through its Advanced Technologies and Instrumentation program.
Observational astronomy depends on the collection of light for analysis. In modern telescopes plus spectrograph, typically half the light is lost due to slight absorption in mirrors, or slight reflections at lens surfaces. Telescope mirrors must be recoated every three years or so. By increasing the reflectivity of mirrors (for example, by using silver-based coatings rather than aluminum), and improving anti-reflection coatings on lenses, we may gather more light without building larger telescopes. Similarly, recoating mirrors causes a loss of observing time, entails risk to the optics, and costs manpower, so the less frequently it must be done, the better. A silver-based mirror coating with good blue reflectivity that lasted 5-to-10 years without recoating would be a phenomenal improvement that could be used on any reflecting telescope with immediate gains of 20-30% in efficiency while lowering operating costs. The primary merit/impact of the project is improving infrastructure for observational astronomical research. The project involved upgrading an existing coating chamber to allow coating development with state-of-the-art equipment. It also demonstrated a unique design that allows coating large optics with relatively modest-sized equipment. Finally, the upgrades included installing both equipment for e-beam deposition and magnetrons for sputtering so that the two processes could be compared head-to-head. The first major component was building a stage that carried the deposition equipment from the center of the chamber to near the edge on a "swing-arm" (see figures). Coupled with rotation of the mirror substrate, this allows large surfaces to be "painted" using a single e-gun or small magnetron. The advantages are that the geometry of deposition sources and substrate being coated is fixed for any radius, so film structure is constant, and by controlling the speed of the swing-arm, coating thickness can be made uniform. Furthermore, all service lines (gas, water, electrical) to the equipment can be "hard-plumbed" in the vacuum so that leaks and electrical shorts will not develop. We demonstrated that this scheme works, coating mirrors up to 30-inches diameter with good thickness uniformity. The next upgrades were to install a large cryopump for the vacuum system, and a six-pocket "e-gun" (a device that uses an electron beam to spot-heat materials for evaporation). The e-gun means we can evaporate up to 6 materials in any pumpdown, and about that many materials are typical for silver-based coatings, where both adhesive layers below the silver and protective barrier layers above the silver are essential. Combined with an existing ion source, we are able to deposit oxides using reactive "ion-assisted deposition" (IAD). With the improved vacuum system, we can produce nitrides as well as oxides; some nitrides appear to be superior to oxides as barrier layers for protecting silver. The final modification was the installation of three magnetrons, which can also be used for deposition. In magnetrons, a trapped plasma is used to knock atoms out of the target (as opposed to heating the target material to evaporation temperature as with the e-gun). Magnetron sputtering has been used in the most successful protected-silver mirror coatings to date, but it is possible that IAD e-beam coatings might be as good (or better). With both magnetrons and e-gun in the chamber, we can make direct comparisons of the two processes for any layer in the coating. While the award funded just the chamber upgrades, research has been on-going while the upgrades were in progress. Major findings include: Sputtering and ion-assisted e-beam deposition appear comparable in the quality of oxide and nitride layers. Silver deposited by e-beam is slightly more reflective and makes a more durable coating than silver deposited by magnetron sputtering. This is likely because the lower pressures with e-beam means less entrapped gas in the silver. Yttrium fluoride and ytterbium fluoride layers near to the silver layer increase the lifetime of the coating. However, they do not yet reach the target lifetime of 5-10 years. In field-testing at observatories, samples placed looking upward degraded after about 2 years, while identical coatings facing downward did not degrade. This implies something settling on the surface reacts chemically with the coating to degrade it, and suggests new avenues of attack on the problem of durability. Condensed moisture on silver-based mirrors causes rapid degradation and/or corrosion. In addition to research on coatings for telescope mirrors, we developed a number of silver-based coatings for use inside certain instruments. These coatings can be optimized for a narrower wavelength range, allowing more flexibility for making the coatings both highly-reflective and durable. We expect these coatings to maintain their high performance for 10 years or longer in the protected environment inside the instruments. See http://coatings.ucolick.org for more information on the upgraded coating chamber and the research being carried out with it.