Before the end of this Century, the Great Observatories should be operational. In addition, the manned space station with its supporting infrastructure will be in place to support a variety of space astrophysics activities. This paper briefly describes some of the major Astrophysics Observatories that are awaiting launch, under development, planned, or being proposed for implementation for the next 25 years. Two classes of future observatories that could follow the Great Observatories are discussed in somewhat more detail, Large Optical/UV Observatories and an Advanced Gamma-ray Observatory that would utilize the Shuttle's External Tank. Studies of these concepts demonstrate that the historic approach for improving telescope performance has reached technology barriers and innovative telescope design approaches are required. Some of the key technology issues that have thus far been identified are discussed.

Two optical designs for an F/5 ground-based astronomical telescope, derived by modifying three-mirror designs, are described. The philosophy was to reduce the number of mirrors to two by combining the primary surface and tertiary surface on the same mirror. In the first design, the primary and tertiary mirror surfaces overlap at their vertices and share the same base curvature. Both mirrors are conics of revolution but differ in their conic constants. In the second design the deformation constants are identical. This design is an alternative for a prime focus corrector of a fast (Fp/1.5) primary, yielding sub-arcsecond images over a curved image plane.

The structural and thermal behavior of a ten-meter primary mirror for a space optical/near-IR telescope in geosyn-chronous orbit is studied. The glass-type lightweighted mirror is monolithic, of the double arch type, and is supported at only three points. The computer programs SSPTA (thermal), NASTRAN (finite element), and ACCOS V (optical) are used in sequence to determine the temperature, deformation, and optical performance of the mirror. A mirror temperature of 130°K or less appears to be obtainable by purely passive means. With a fused silica or standard Zerodur . blank, thermally induced deformation is unacceptable and cannot be fully corrected by an active secondary mirror over the desired field. Either active thermal control or a blank of lower thermal expansion coefficient would be required.

This paper concentrates on system trade-offs and their impact on the fabrication of quantities of large off-axis optical elements. Weight reduction techniques, substrate material selection, mirror surface quality (surface figure and reflectivity), and fabrication methods are discussed.

This paper discusses the space-based measurement of atmospheric winds from the point of view of the requirements of the optical system of a coherent CO₂ lidar. A brief description of the measurement technique is given and a discussion of previous study results provided. The telescope requirements for a space station based lidar are arrived at through discussions of the desired system sensitivity and the need for lag angle compensation.

The prospects for applying aperture-synthesis interferometry to the optical domain are reviewed. The radio examples such as the VLA provide a model, since the concepts are equally valid for radio and optical wavelengths. If scientific problems at the milliarc-second resolution level (or better) are to be addressed, a space-based optical array seems to be the only practical alternative, for the same reasons that dictated array development at radio wavelengths. One concept is examined, and speculations are offered concerning the prospects for developing real systems. Phase-coherence is strongly desired for a practical array, although self-calibration and phase-closure techniques allow one to relax the restriction on absolute phase stability. The design of an array must be guided by the scientific problems to be addressed.

Phased telescope arrays for broad-band imaging applications suffer from substantial sensitivity losses due to the large number of reflections required to phase and combine the images from the individual telescopes making up the array. They also suffer from severe field-of-view limitations (perhaps as small as a few tens of arc seconds) due to both pupil-mapping errors and the field curvature of the individual telescopes making up the array. In addition, thinned aperture optical systems (including phased telescope arrays) pose unique problems in specifying or characterizing image quality. Traditional image quality criteria such as "resolution" and encircled energy are woefully inadequate for many thinned aperture applications. Variations in the subaperture geometry which produce only subtle effects upon the core of the point spread function (PSF) may produce highly undersirable artifacts or spurious images as well as a modulation transfer function (MTF) which exhibits zero (or negligible) values over substantial regions within the cut-off spatial frequency of a filled aperture circumscribing the array. Clearly some minimum value of the MTF exists below which spatial information cannot be retrieved in the presence of noise. An MTF property called the "practical resolution limit" and defined as the reciprical of the maximum spatial frequency within which no zeros occur in the MTF thus becomes the image quality criteria of choice for those applications where fine detail is required from extended objects. This practical resolution limit and its effect upon mirror size and separation (subaperture configuration) will significantly impact the telescope mechanical architecture, stowage and deployment techniques, and perhaps even the booster vehicle selection for future large space telescopes.

We describe the conceptual design of a proposed first-generation optical in-terferometer in space, the Binary Star Explorer. The scientific objectives for this interferometer are to determine the fundamental astrophysical quantities of distance and mass for stars in binary systems. In particular, the interferometer will be able to make accurate distance measurements to an estimated 40 Cepheid binary systems in our Galaxy, and 28 supergiant binary systems in the LMC. The interferometer comprises two fixed telescopes on a 5 m baseline, beam-combining optics, and a visible/ultraviolet fringe detecting system. We determine the angular separation of binary systems made up of a cool giant star and hot dwarf companion by measuring the shift between the optical and ultraviolet fringes. In combination with knowledge of the physical size of the orbit (which must be obtained separately from radial velocity measurements on both stars), the distance to the binary is obtained as the ratio of the physical to angular sizes.

This paper will discuss how AOC R-Theta machine can produce superior aspheric surfaces in terms of tool marks and surface figure irregularities compared with other cascading (point to point) diamond turning surfaces. It will also discuss how easy and inexpensive it is to produce aspheric surfaces, which requires absolutely none, or little post polishing for lowscatter application.

The results of glancing incidence absorptance measurements performed on diamond-turned copper mirrors are presented. A photoacoustic calorimetry technique is used in which the output from a low power, chopped cw Nd:YAG laser (1.06 μm) is incident upon the mirror at angles of incidence from 0 to 87°, for both s and p-polarizations. Measurements are obtained as a function of the diamond turning groove orientation with respect to the plane of incidence. Minimum absorptance, at high angles of incidence, is achieved with s-polarized light and with the grooves aligned parallel to the plane of incidence. The affects on the absorptance of a large scratch at glancing incidence are also described.

Testing of large single point diamond turned mirrors and mirror systems requires the use of rapid, quantitative, optical interferometric measurements in the visible and IR ranges. Developments in this area and applications of these developments are discussed.

Diamond-turning applications for the opto-mechanical assembly of multimirror systems are given. These applications are categorized into three construction types: one-piece (monolithic) systems; assemblies using radial location surfaces; and assemblies employing non-radial location surfaces. A cross-sectional schematic is used to illustrate each application. The construction types are analyzed with regard to ease of manufacture, ability to hold assembly tolerances and limitations.

There has been considerable interest in the Electro-Optic community in utilizing diamond-turned optics for compact optical systems. The ability to generate complex, fast surfaces that far surpass the capability of glass substrate optics has proved invaluable in the design and development of automated test equipment. The current state-of-the-art diamond-turning process can introduce a variety of errors to the overall figure of the surface. The nature of these errors and the effect on surface quality are critical concerns for the system designer. To date, diamond-turned optics have been successfully used for both co-herent and incoherent applications in the mid to far Infrared (IR) spectrum where very good surface quality can be achieved. These applications range from laser resonator optics to telescopes. However, when the optics are used in the near IR to visible spectrum, surface quality becomes very critical. The same surface that may be exceptional for use in the mid to far IR spectrum may be totally unacceptable in the near IR to visible spectrum. In many applications, such as telescopes, involving the near IR to visible spectrum, system require-ments call for focusing incoherent radiation. However, there are also applications, such as laser performance in the far-field, where focusing coherent radiation is required and surface quality becomes an important system parameter. One method of determining surface quality of a diamond-turned optic is presented here, whereby the far-field diffraction pat-tern of a fast diamond-turned two mirror collimator, as an example, is observed for coherent radiation at 0.6328 microns, partially coherent radiation at 0.6328 microns, and coherent radiation at 1.0624 microns.

The family of afocal telescopes discussed here consists of two, three, or four confocal conic mirror sections. The two-mirror parabola/parabola combination (confocal parabolas, or Mersenne telescope) is the best known, but its performance is limited by curvature of field. The three-mirror Cassegrain/parabola combination (Offner afocal telescope) can be corrected for field curvature and is suitable for use at high magnifications. The four-mirror Cassegrain/Cassegrain combination, the least well-known design, is most effective as a unit power afocal relay. The properties of these designs will be reviewed briefly, with major emphasis given to the confocal Cassegrains design.

The test and evaluation of infrared optical systems is requiring collimators with rather stringent performance characteristics which must be simultaneously realized. Fields of view as large as 15 degrees, flat focal surfaces, clear apertures with uniform intensity across the focal plane and minimal spectral variations over passbands as wide as 7 pms are within the realm of common place. A centered, rotationally symmetric optical system with a tilted field of view and a decentered stop has been used to design just such an optical system. Evaluation shows that it meets performance requirements and that it is nearly diffraction limited over the central 5 of its 15-degree field of view.

A method of maintaining precise control of packaging constraints in the optimization of unobscured reflective systems is described. The method is illustrated by reference to an example of a four-mirror telescope objective. A command sequence for the optimization of this system is presented.