The HIREX mission study is examining various optical configurations to determine a design approach with a high likelihood of mission success. As part of this process we are comparing a telescope designed around on-axis optics to one with off-axis optics. The trade off is a complex one with many pros and cons on either side. Because of this complexity, it has been decided that the comparison should be explicitly laid out.
The two design that are being compared are described.
The on-axis design is a standard gregorian with a field stop at the prime focus. The primary mirror is supported through its center hole with a TRACE-like mount. The mount seats consist of a similar epoxy/siltex casting held on invar clamping components. The details have been refined from the TRACE design because of the extreme figure requirements that the HIREX mission imposes.
The secondary is also held on axis. It is mounted on a 6 axis adjustment stage and the combination is suspended from a spider. The field stop is held either by another spider or a cantilever from the secondary assembly itself.
The off-axis design is far less mature at this point. It was proposed to alleviate the thermal problems that the secondary penumbra induces on the primary. The primary has no central holes and is supported by a set of tangential flexures. The flexures will be sized to balance the need for extremely low figure error with the modest mirror dynamic stability requirement.
The secondary is most likely on axis, mounted in the same manner as the on-axis system. However the secondary assembly will be supported on a solid plate mounted in one corner of the telescope truss. The field stop, again at the prime focus, would be supported in a way similar to that described above.
The effect of the nonuniform illumination of the on-axis system has to be compared to the thermal shape changes on an off-axis mirror. The essence of this distinction is that the non-uniform illumination, both in time and across the mirror may induce undesirable figure changes. On the other hand the off-axis mirror is supported about the periphery of the substrate, which is not symmetric with the surface figure. In Spite of the apparent symmetry of the support the shape change will not be symmetric with the true figure.
The mirror support is a large issue, second only to the ability to generate the mirrors themselves. The study has progressed to the point where we are able to predict a successful mount design for the on-axis case. Due to the symmetrical layout of the mirror and the mount there are straightforward approaches to reducing the surface effects of clamping and compensating for those that we can not remove. On the other hand the mount induced surface effects of an off-axis system are unknown but most likely non-compensatable. These must be determined, reduced where need be, and finally compensated for.
The on-axis mirror has a hole in it, the off-axis mirror does not. We must compare the relative difficulty of polishing around the central hole versus generating an off axis, non-spherical shape.
Metrology is a problem for this project in any event. We must compare the available metrology systems for on-axis and off-axis systems, and their chances of achieving the required resolution.
We have selected a standard of mirror performance that is being pursued by the electronics industry. We have tailored our specification to look as close to those proposed for XUV lithography as possible. The HIREX baseline design has a 0.6m mirror and a roughly 6 angstrom figure error tolerance to match the industrial goal. The HIREX mirror is going to have a much longer focal length, 35m vs 2-5m, and it may be either on-axis or off-axis versus the industrial goal of an off-axis mirror. We must determine if the fact that the baseline HIREX design becomes off-axis improves our argument that we are building a system similar to that being addressed by industry or is the focal length distinction such a large difference that all other considerations are minor.
There are real differences in the inherent stability between the secondary assembly support in the on-axis design and that in the off-axis case. These effects have to be examined.
The trade offs at the focal plane are unclear. By going with an off-axis design we open up considerable latitude in the axial position of the focal plane at the possible expense of the light rejection and thermal control of the focal plane. Aside from these possible issues the focals are nearly identical.