There are two main challenges in the manufacturing of Wolter optics: to achieve a perfect figure and to keep the micro-roughness of the optical surface low. The accuracy of the figure described in terms of roundness and slope errors governs the overall point spread function and thus the angular resolution of the optics. A particular difficulty is the proper figuring of the break between the parabolic and hyperbolic part or the alignment of separate parboloids and hyperboloids. Surface errors range from long-wave deformations to high spatial frequencies where interference effects dominate the imaging quality rather than geometric optics. In the regime of interference, the surface errors are called micro-roughness. As X-rays have wavelengths about a factor of hundred shorter than optical light, the requirement on the micro-roughness is correspondingly high. There is a distinct relationship between the X-ray scattering and the surface topography described by vector electromagnetic scattering theory [1,5]. The reducing of micro-roughness required extreme efforts in the early days of X-ray optics, whereas super-polishing down to a few tenth of a nanometer micro-roughness is now standard.
Most of the first Wolter telescopes have been manufactured in the classical way adopted from optical mirror production. They were made from thick glass blanks, which were ground, polished, and finally coated. For example, fused quartz glass with nickel coating was used for Einstein and Zerodur glass with gold coating for ROSAT (Fig. 6.8). Chandra, the most precise Wolter telescope so far, is made of iridium-coated Zerodur and has an angular resolution of better than 1'' half power diameter (HPD). High-Z corrosion-resistant metals like gold, iridium, and nickel are well suited for the coating of the reflecting mirror sufaces. Multilayer coatings combined of different materials can enhance the reflectance beyond the limit of total external reflection by using constructive interference from the thin layers. Multilayer Wolter optics are under study for future X-ray missions.
After the pioneering astronomical X-ray missions the demand for high throughput X-ray optics initiated the development of light-weight, thin-walled X-ray mirrors. One result of this development, mainly driven by the XMM-Newton mission, was the technique of coated electroformed nickel shells replicated from super-polished mandrels (Fig. 6.8). Another result, e.g., applied for ASCA, was the technique of replicated foil mirror segments (aluminum foil plus epoxy) . Both techniques allow a high degree of nesting. The nickel mirrors are heavier but are suited for an angular resolution up to 16'' HPD (XMM-Newton) whereas the very light foil mirrors achieve a resolution in the arcminute range only. Replication techniques have the advantage that once a mandrel is polished several integration-ready replications can be drawn from it. Even the coating is part of the replication process.
Scientific goals of today require X-ray telescopes with even larger collecting areas on the order of several square meters and angular resolution in the range of arcseconds to avoid source confusion. At the same time, however, limited launch masses are a severe constraint driving the development of new technical solutions for extremely light weight and rigid Wolter optics. Two approaches are currently under study: thin glass mirror segments can be formed in appropriate moulds when the glass has a certain viscosity through heating [6,14]. Still lighter are pore optics made from stacked silicon sheets with small grooves as "light channels" . These techniques are favored for the Wolter optics of the planned X-ray missions: "slumped glass" for Constellation-X and pore optics for XEUS.
Was this article helpful?