Optics in Orbit

The success of the Hubble Space Telescope and the Chandra X-ray Observatory, along with the availability of improved technology, has encouraged astronomers to look beyond traditional mirror designs for the telescopes of tomorrow. The high-precision optics in the next generation of space telescopes will be an order of magnitude lighter in weight than those in the Hubble or Chandra. Sponsored by agencies like NASA, the European Space Agency (ESA), and the U.S. Department of Defense (DoD), developers are using a host of materials and designs to build lightweight mirrors a meter or more in diameter.

As designers push toward larger telescopes, the weight-per-unit area of collecting aperture becomes the driving factor. The Hubble primary mirror was 2.4 m in diameter and had an areal density (weight/collecting aperture area) of approximately 180 kg/m². The areal density goal for the Next Generation Space Telescope (NGST) is 15 kg/m², and the plans and desires of the scientific community for the coming decades will require areal densities pushing below 1 kg/m² (see sidebar on page 24). Similarly, recent developments in x-ray mirror technology have resulted in a 0.5-m-diameter, 0.6-m-long electroformed nickel alloy mirror that weighs only 1.2 kg (see Figure 1). A similarly sized mirror in the Chandra telescope weighs more than 100 kg.

Figure 1. An electroformed nickel mirror for x-ray detection developed by NASA's Marshall Space Flight Center weighs only about 1/100th as much as a similar mirror in the Chandra X-ray.

Many NASA and ESA missions use infrared optics, which require operation at cryogenic temperatures. Mirror makers pursue various approaches to manage the distortion introduced during cooling, which range from active correction via actuators that operate at cryogenic temperatures to cryo-figuring: figuring the optic at room temperature in a manner that accounts for shape changes that occur under cryogenic conditions.

A metal with a high strength-to-weight ratio, beryllium is attractive for space applications. The Space Infra-Red Telescope Facility (SIRTF), due for launch in July 2002, is essentially an all-beryllium telescope. It was fabricated by a team from Ball Aerospace (Boulder, CO) using cryo-figuring. Developed to operate over a wavelength range from 3 to 180 µm, the telescope is a two-mirror Ritchey-Chretien design with an 850-mm clear aperture. Structurally, it is a single arch configuration that allows the primary mirror to be supported off a central ring. The primary mirror has an areal density of approximately 28 kg/m².

The cryo-figuring process was extremely successful and resulted in a final surface figure of approximately 0.067 µm rms. The Ball Aerospace team also is developing a beryllium mirror as part of the NGST Advanced Mirror System Demonstrator (AMSD) program. It is a 1.39-m hexagonal mirror attached to a composite back plane structure by seven actuators. This design permits adjustment of the radius of curvature and correction of some low-frequency error terms. The mirror system will have a total areal density of less than 15 kg/m², including the mirror, back plane, and actuators.

This mirror is still in the early stages of fabrication. In preparation for this effort, however, Ball Aerospace developed a technology demonstration mirror to determine the extent to which a beryllium mirror could be lightweighted. The team manufactured a 0.53-m-diameter mirror, using a spherical beryllium powder and achieving an areal density of less than 9.8 kg/m². This is the lightest-weight beryllium mirror ever manufactured. It has been cryogenically tested to less than 25 K, demonstrating a surface figure shift on the order of 0.08 µm rms, a factor of three less than was seen on the SIRTF mirror.

Silicon carbide has long been a promising material with highly desirable properties for robust lightweight mirrors. Unfortunately, producing mirrors of any appreciable size has been difficult. Two groups have made mirror blanks larger than a meter in diameter, however.

Industrieanlagen-Betriebsgesellschaft GmbH (IABG; Ottobrunn, Germany) has made a 1.04-m-diameter blank for a NASA program called Solar Lite. The substrate has an areal density of 56 kg/m², although it is not designed to be a lightweight blank. IABG has also produced a 0.5-m carbon-silicon-carbide blank with an areal density of 7.8 kg/m², however. The blank is made by creating a block of chopped carbon fibers embedded in a phenolic resin. The makers machine the block to shape and then fire the block until the resin turns to carbon. At even higher temperatures, they add silicon, which infiltrates the block to produce silicon carbide.

The resulting mirror blank shrinks very little and can be machined into near net shape. The bare material can be polished to approximately 10 nm. IABG has developed a silicon-carbide slurry that can be applied to the blank in a subsequent step. This surface has been polished to 3 nm. It is possible to obtain smoother surface finishes by depositing a silicon-carbide layer over the blank using chemical or plasma vapor deposition, and then polishing. This latter process yields surface finishes in the 0.5-nm regime.

Meanwhile, a French collaboration between Astrium (Toulouse) and Boostec (Bazet) fabricated a 1.35-m-diameter optic as a demonstration for the Far InfraRed and Submillimeter Telescope (FIRST; also known as the Herschel Space Observatory) program. This telescope is due for launch in 2007 and is designed to cover the wavelength range from 60 to 670 µm. The mirror for the demonstration telescope is made from sintered silicon carbide. The developers pressed a finely ground silicon-carbide powder mixed with organic binders and additives at high isostatic pressure. They machined the compacted material to shape and then heated it, causing the particles to agglomerate. Astrium-Boostec fabricated the mirror in nine segments, then brazed them together to produce a unified blank, which they ground and polished to produce the final figure and finish. The bare material can be polished to around 3 nm and can be ion-figured. Adding vapor-deposited silicon carbide allows even smoother surfaces.

The demonstrator mirror has an areal density of approximately 26 kg/m². The lightweighting was done to meet FIRST requirements and does not represent the limits to which the mirror weight can be minimized. At 110 K, the developers observed a distortion of approximately 1.31 µm rms, but they did not detect any stress release or hysteresis effects (which would introduce unpredictable shape changes).

The diameter of the primary mirror for the full-sized telescope is 3.5 m. Astrium/Boostec has fabricated one full-size segment, but it has not yet been sintered.

Two glass mirror systems—the mirror and any equipment necessary to make the mirror perform as designed—are being built under the NGST AMSD program. The University of Arizona (Tucson, AZ) is fabricating a 2-m-diameter mirror system that consists of a 3-mm-thick, 2-m-diameter-thin meniscus attached to a composite back plane structure by an array of actuators every 7 cm (see figure 2). In the completed mirror system, the actuators will compensate for gravitational distortion and cryogenic distortion. The university is completing fabrication of the thin meniscus and plans to test the mirror system at cryogenic temperatures this winter.

Figure 2. The 2-m-diameter mirror will be made from this 3-mm-thick thin meniscus of glass. Actuators spread across the back of the glass will compensate for temperature and gravity changes in space. (University of Arizona).

A team from Composite Optics (San Diego, CA) fabricated a 1.6-m-diameter glass/composite mirror that consists of a composite sandwich with a Zerodur (Schott Glass; Duryea, PA) face sheet. The areal density of the primary mirror is 11 kg/m², and that of the assembled system, including actuators and back-plane structure, is 15 kg/m². The actuators attaching the primary mirror to the back plane provide tip, tilt, and piston control. A fourth actuator adjusts the radius of curvature.

Because the design of the system only permits radius-of-curvature correction, the mirror must be cryo-figured. The mirror system has been cooled to 25 K with optical tests performed at 40 K. Figuring to correct for cryogenic distortion is now underway at SAGEM (St. Pierre du Peray, France). SAGEM expects to perform final cryogenic testing in the winter of 2001.

The NGST AMSD program also is responsible for two mirrors in the early stages of fabrication. Eastman Kodak Co. (Rochester, NY) is developing a mirror system that consists of a 1.5-m-diameter hexagonal mirror attached to a composite back-plane structure by 16 actuators. The actuators will allow gravity off-loading for 1-g testing and also will provide radius-of-curvature correction and some low-frequency error correction. The mirror consists of a two-face honeycomb ULE (Corning Inc.; Corning, NY) face sheet fabricated by fusion-bonding a face sheet and a back sheet to a honeycomb core.

The BF Goodrich Co. (Danbury, CT) mirror is a 1.3-m hexagonal fused silica meniscus attached to a composite reaction structure by 37 actuators. These actuators provide 1-g gravity correction and correct radius of curvature and low- and mid-spatial-frequency errors.

In the past decade, mirror builders have been able to fabricate high-precision, lightweight optics for space-based systems at comparatively low costs, driven by DoD-funded research in the 1980s and advances in microelectronics, optoelectronics, and computers. These advances led to more sophisticated and sensitive metrology systems that are capable of measuring surface finishes at angstrom levels and geometrical figures to hundredths of a wave. The fruit of these advances will be the large telescopes that will be launched in space over the next decade. oe

Acknowledgments

Thanks to Gary Mathews, Kodak; Norbert Pailer, Astrium-Germany; Bob Brown, Ball Aerospace; Dave Crowe, Composite Optics Inc.; Emanuel Sein, Astrium-France; Phil Stahl, Marshall Space Flight Center; Eri Cohen and Dave Galagher, Jet Propulsion Laboratory; Enricque Garcia, BFG Aerospace; and Bernie Seery, Goddard Space Flight Center for providing information for this article.

Space wasn't always in James Bilbro's career plan. Bilbro, special assistant for optics to the director of NASA's Marshall Space Center (Huntsville, AL), started out as a cowboy. "After working a ranch, at temperatures of 110 °F in the shade, and riding horseback all day, I figured there must be a better way to make a living," he says. After working on the Santa Fe Railroad and having a short stint in a sawmill, he saw the potential for a career in engineering.

Bilbro spent his first 15 years at NASA as an electrical engineer, helping develop coherent lidar, but when it came time for him to get his doctorate, he changed direction. "I was about to go for my Ph.D. in electrical engineering, when a friend mentioned he was going to the University of Arizona for optics," Bilbro says. "I thought, 'Now that sounds interesting.'" So he switched.

Several years ago, Bilbro started the Space Optics Manufacturing technology center at Marshall. The mission of this 60-person organization is to develop the advanced optics for future missions of NASA, a challenge he welcomes. "When I was a kid I used to keep a scrapbook of space stuff," says Bilbro. "I had no idea that someday I would be working here." Today, the organization is pushing the envelope on lightweight mirrors.

But at NASA, pushing the envelope is just part of the culture. Daniel Goldin, NASA's administrator, saw Bilbro's display of 15 kg/m² lightweight mirrors at an American Astronomical Society conference. "He was impressed for about 30 seconds," says Bilbro. "Then he looked at me and asked, 'Can you make mirrors that are 5 kg/m²?' I said, 'It will take some work, but I believe it is in the realm of possibility.'"

Then Goldin asked, "How about 1 kg?"

"I don't think these technologies lend themselves to that," Bilbro answered.

"Good," Goldin replied. "The next time I see you, tell me how to make 1 kg/m²."

"I avoided him for the next two years," Bilbro says with a laugh. Today, Bilbro has an inflatable membrane mirror sitting on his desk that weighs a tenth of kilogram. "It isn't imaging quality," he says, "but it is good enough to burn holes in Coke cans."

—Laurie Ann Toupin