X-rays provide unique information for understanding the physical conditions and processes in virtually all astronomical systems. Their emissions are associated with hot plasmas, the creation of elements, and explosive events at the beginning and end of the life cycles of stars, galaxies, and clusters of galaxies. Most ordinary matter (baryons and electrons) in the universe is visible only in their light. They penetrate intergalactic space from cosmological distances. All these characteristics allow x-ray observations to place new and independent constraints on cosmology and the history of the universe.
X-ray astronomy faces great challenges to keep pace with planned advances in optical, IR, and radio astronomy. The reasons are that observations can only be made from satellites completely outside the Earth's atmosphere and because x-rays only reflect at very shallow grazing angles. Thus, x-ray telescopes must be very light weight to be launched into orbit, and the telescope must have approximately 100 times more area that is figured and polished than its effective collecting area. Figure 1 shows that many shells must be nested to build up a significant collecting area.
Figure 1. Schematic of the four shells of paraboloid and hyperboloid mirror pairs (MPs) comprising the Chandra X-ray Observatory. The shells were rotated about an optical axis in the figure's plane forming a complete cylindrical element. The mirror elements' full lengths must be precisely figured, although the effective area for collecting x-rays is only the small projected area for which gaps are left in the thermal precollimator. For the International X-ray Observatory and future observatories, each shell comprises distinct segments.
The Chandra X-ray Observatory, with 0.7 arcseconds full width at half maximum resolution and a collecting area equivalent to a 14inch-diameter telescope, represents the state of the art. Its high-resolution mirror is made from thick, rigid Zerodur glass and represents the end of the line for such technology because of the large mass. The International X-ray Observatory (IXO), which was highly recommended by the recent National Academy of Sciences' decadal review panel, incorporates a new technological approach. Its telescope could comprise some 14,000 thin sheets of glass, thermally formed on precisely figured mandrels, and co-aligned to build up a very large area. However, IXO will be challenged to achieve its required 5 arcsecond resolution, because its 0.4mm-thick glass is flimsy.
A new concept for an x-ray astronomy observatory1 could provide five times better angular resolution and 500 times the collecting area compared to present capability. Our innovation for the generations of telescopes to follow IXO would allow adjustment of the glass figure in situ. The approach uses thin-film piezoelectric actuators deposited directly on the back of reflecting elements (see Figure 2). Voltage across the narrow dimension of the piezoelectric material causes a strain parallel to the mirror's surface2 analogous to a bimetallic temperature sensor in a household thermostat. We have been depositing lead zirconate titanate crystals on 100mm-square flat sheets of Schott D263 glass.3 We use optical interferometry to measure the surface characteristics and deterministically adjust the actuators to control the figure. Previously, focusing adjustment was done in synchrotron x-ray beam lines4 in which the focusing is only in one dimension of a thick parabolic plate. In our application, we form a parabolic or hyperbolic shape along the optical-axis direction and a circular shape in a plane parallel to the focal plane using thin (0.1–0.4mm-thick) IXO-like segments.
Figure 2. Schematic of the piezoelectric material deposited on the nonreflecting side of a mirror. A voltage difference across the outer and inner electrodes causes a local strain in the mirror's plane. Proper application of strains along the mirror can correct figure errors on scales down to approximately twice the piezoelectric cell. The dielectric insulators define discrete cells in two dimensions.
The driving scientific requirement for the Generation-X Observatory is to study the first black holes, which formed when the universe was 300–500 million years old. This requires a 50m2 collecting area and 0.1 arcsecond angular resolution. We conceive that the observatory would be launched in a vehicle with capabilities similar to those studied for Ares V. The mirrors would unfold in orbit to form a partially filled 16m-diameter telescope, and a deployable optical bench would extend to a 60m focal length. An imaging detector positioned in front of the focal plane would intercept the converging x-ray beams where profiles of distinct rings could be measured.5
We expect that figure adjustment using actuators integral to the reflecting elements will be a standard feature of future x-ray telescopes. Missions with less demanding requirements might only calibrate the required adjustments on the ground and then apply them in orbit. Major observatory-class missions will use flight instrumentation to determine the best imaging capability in orbit and adjust for changing conditions as necessary. Such tweaking offers the capability, in principal, to optimize the imaging response over the field of view, trading poorer on-axis execution for better average performance to suit a variety of scientific objectives. The Generation-X observatory will offer future astrophysicists the ability to obtain x-ray data from the same objects seen by the most sensitive optical/IR and radio telescopes and with similar angular and spectral resolution.
This work has been supported in part by NASA contracts NNX08AT62G, NNX09AE87G, NAS8-39073, and by a grant from the Gordon and Betty Moore Foundation.
Roger Brissenden, Daniel Schwartz
Harvard-Smithsonian Center for Astrophysics
Roger Brissenden is manager of the Chandra X-ray Observatory and deputy director of the Harvard-Smithsonian Center for Astrophysics.
5. D. A. Schwartz, R. J. Brissenden, M. Elvis, G. Fabbiano, T. J. Gaetz, D. Jerius, M. et al., On-orbit adjustment calculation for the Generation-X x-ray mirror figure, Proc. SPIE 7011, pp. 70110W, 2008. doi:10.1117/12.790085