Lightweight, aspheric, reflective optical designs commonly are designed and built for demanding space-based remote sensing, targeting systems, and aerial reconnaissance. Traditional designs utilizing low expansion optical glasses steadily are giving way to metals such as aluminum, beryllium, and AlBeMet, and ceramics such as silicon carbide. These materials can be produced in extremely lightweight, yet robust and athermalized, designs by virtue of their superior tensile strength, fracture toughness, and the ability to compose support structures and mirrors from identical materials.

Figure 1: Illustration of symmetric verses off-axis design at left, with the three-mirror off-axis design at right.

In addition, these materials can often best exploit advancing freeform generating and computer polishing technologies. This discussion focuses on the performance tradeoffs among aluminum, AlBeMet, beryllium, and silicon carbide in the context of modern substrate manufacturing, optical generating, and computer polishing technologies. The goal is to provide the opto-mechanical engineer an overview of the confluence of material science and the state of the art in manufacturing technologies as they relate to ever more demanding optical performance objectives.

Material Choices

Reflective optical systems typically are chosen over refractive systems to achieve wavelength independence and/or lighter, more robust, and shorter packaging for a given aperture and effective focal length (EFL). Reflective designs become attractive for apertures greater than approximately 100-200 mm and where comparatively large fields of view are not required. Because surface figure irregularities of reflecting surfaces cause four times as much transmitted wavefront error (WFE) compared to refractive surfaces, and non-circular apertures and aspheric prescriptions are common, reflective surface manufacturing tolerances and corroborating metrology can be very challenging.

Types of reflecting surfaces range in difficulty from simple round and spherical progressing to irregular shaped flats, symmetric and off-axis aspheric, and finally “free form” surfaces that lack an axis of symmetry. These types of surfaces are commonly combined into symmetric and off-axis assemblies having two or more elements, as shown in Figure 1.

Reflective Optics Materials

A quick reference to the relative properties of various materials suited for manufacture of reflecting optical systems is included in the table on page 6a.

Beryllium exhibits low density, high stiffness, has attractive thermal properties, and can be machined into very thin sections. It is also very expensive, both as bulk material and post-processing. Most typically, beryllium is plated with electroless nickel plating to enable the manufacture of a high-quality optical surface. Beryllium may be bare polished, although at considerable added effort and cost. Aggressively light-weighted, bare polished beryllium was selected for the 18 mirror segments that will comprise the 8.0-meter aperture James Webb Space Telescope (JWST). Beryllium structures also may be brazed with aluminum alloy to create optical bench structures.

Aluminum beryllium such as Brush Wellman (Warren, MI) AlBeMet 162H is lower in bulk material and manufacturing cost, less hazardous, and has advantages in ductility and fracture toughness over beryllium. Laser and electron beam welding of AlBeMet recently has been introduced to form intricate lightweight structures to integrate AlBeMet mirrors, thus minimizing material utilization costs while achieving an athermalized, stiff, strong, and very lightweight design.

There is currently considerable interest in silicon carbide as a challenger to beryllium in many contexts. Many producers of SiC recently have emerged, reducing SiC costs compared to beryllium. Formulations of SiC, such as those produced by Poco Graphite (Decatur, TX), are listed in Table 1. The Poco process consists of manufacturing very intricate mirrors and matching structural elements from inexpensive and easily machined graphite prior to conversion to SiC.

Table 1: Relative properties of common mirror materials

SiC can be cladded with both CVD SiC and silicon. Both of these surfaces can be ground and polished to optical finishes approaching those on glass surfaces. Since CVD SiC is very hard and difficult to optically process, silicon cladding is much more favorable. Silicon cladding may be both ground and diamond point machined, with a well-matched coefficient of expansion to SiC.

Aluminum 6061-T6 holds reign at the low end of the reflective optic cost spectrum. This material is light, inexpensive, strong, thermally isotropic, and may be readily diamond point machined. It is also easy and practical to build all of the mirrors and structural elements from the same material. Its principal disadvantages are relatively low stiffness and high thermal expansion.

Recent renewed interest and success in post-polishing aluminum to achieve improved surface figure and finish compared to diamond machining rapidly is expanding the applications for aluminum mirrors. Many cryogenic programs have demonstrated that Aluminum 6061-T6 can expand and contract very uniformly, provided thermal gradients and resulting stresses are avoided. Like beryllium, aluminum mirrors also are commonly plated with nickel to enable refined figure and finish by diamond machining and post-polishing. If appropriate attention is paid to the design and processing of nickel-plated aluminum mirrors, bimetallic thermal distortion effects can be minimized.

Manufacturing Techniques

Recent introduction of “slow tool servo” and raster-generating-type freeform technologies applied to both diamond machining and grinding technologies is launching a revolution in optical manufacturing. Freeform optical surfaces, previously unimaginable, now can be produced. Lightweight optics significantly distorted by centrifugal or “fling” effects during diamond machining can be measured and re-cut to correct for reproducible error patterns.

Once the exclusive realm of the highly skilled and often secretive master optician, finishing of aspheric optical surfaces now is quickly entering a computerized- and technology-driven era. Many optical manufacturing companies now have developed their own machinery, while several companies now offer generic turnkey, small tool, and freeform computer polishing machines for commercial sale. These technologies enable surfaces and molds generated by freeform diamond machining and grinding to be further improved for figure and finish to the limits of measurement technology. As a result, much lighter and higher accuracy surfaces can be produced at a lower cost.

Measurement technologies also are advancing in locked step with manufacturing as an increasing number of companies are offering an assortment of measurement technologies based on computerized slope integration, 2D, 3D profilometry, phase measuring interferometry, and micro-lithographic such as computer-generated holograms (CGH). These technologies also can be used in the creation of replication molds for high-volume production in a wide spectrum of industries.

Researchers in astronomy, microlithography, aerospace and defense, and industrial replication and molding for commercial products will benefit from metal optics and the opportunities to produce surfaces and systems that are lighter, better, faster, and cheaper thanks to new tools, new processes, advancing materials, and greater competition.

This article was written by Michael N. Sweeney of Axsys Technologies Imaging Systems (ATIS, Rochester, MI). For more information, contact Mr. Sweeny at This email address is being protected from spambots. You need JavaScript enabled to view it..


Photonics Tech Briefs Magazine

This article first appeared in the October, 2006 issue of Photonics Tech Briefs Magazine.

Read more articles from the archives here.