Proceedings Article | 10 July 2018
KEYWORDS: Mirrors, Segmented mirrors, Telescopes, Surface finishing, Glasses, Large telescopes, Polishing, Manufacturing, Ceramics, Distortion
Segmented mirror technology is essential for large telescopes. Actually, the world largest telescopes such as Keck I and II, and GTC employ segmented mirrors for the primary mirrors, and all the planning extremely large telescopes (TMT, E-ELT, and GMT) also plan to use the segmented mirrors although the size of one mirror is far from each other. Needless to say, the technology of segmented mirrors realizes not only telescopes with the large apertures, which cannot be fabricated with a single dish, but also lightens the total weight of the telescopes. In addition, it enables the manufacture of the mirrors with a relatively small equipment and can simplify the mechanical structure of the telescope, resulting in saving the total cost for development.
However, there are native difficulties and issues for the segmented mirrors technology. First, the segmented mirrors must be aligned to each other in the positions and angles with an accuracy of several tens of nanometers and arcsec, and it must be stabile for a long time by the active control as for TMT, the very complicated technique and the human resources are needed for precise control of as many as 492 mirrors. Secondly, individual polishing process of segmented mirrors produces, unavoidable sagging at the outer circumference of the mirrors. Because these local surface deformation cannot be compensated by the alignment of mirrors, the PSF by the telescope is strongly affected by the remaining wave front error, which is the higher order.
Finally, the ambient thermal radiation leaked t from the gaps between segment mirrors , which becomes a background noise source, significantly degrades the sensitivity of the instrument, especially near-IR or mid-IR instruments, compared with a single dish telescope.
These problems arises from the fact that each segmented mirrors are independently supported with the gaps. Therefore, we have been developing Alignment-Free Gapless Segmented Mirror (AFGSM), which is mechanically fixed to each other without any gaps between mirrors, as if it were a monolithic type of mirror. Specifically, after fan-shaped or similar segment mirrors are manufactured individually, they are mechanically fixed to each other by fastening, fusing or bonding. Since the Orthogonality or parallelism of the mirror surface from its reference surface is finished with a machine precision of 1 um or less, the assembly can be mechanically carried out with a well accuracy. Polishing to obtain a highly precise surface for use as a telescope, can be to the extent of corrected (re-)polishing.
^
In order to realize such a segmented mirror, we selected ceramics cordierite CO720 as a material, which is a type of fine ceramic developed by Kyocera. The cordierite CO720 has a coefficient of thermal expansion (CTE): <0±0.02ppmK - 1 for 23 deg., which is comparable to that of low thermal expansion glass such as Clearceram, Zerodur, or ULE. Since it can be easily polished, it is suitable for the base material of a mirror for a telescope application. Indeed, as a material comparable to low thermal expansion glass such as Zerodur, cordierite CO720 is widely used in cutting-edge technology of semiconductor manufacturing equipment which requires extremely high accuracy.
Cordierite has a stiffness (E / ρ) of about 1.5 times that of general low thermal expansion glass material. With AFGSM, it is necessary to fasten / bond / join segments to each other. Cordierite CO720 is more suitable for gapless mirrors because the mirror surface is less likely to be deformed compared to glass against the stress generated during assembly, thanks to its high stiffness characteristics.
Also, the high flexibility of formability of cordierite is also important in order to realize a gapless segmented mirror. In order to minimize the alignment error, the two contacting surfaces must ideally be processed as perpendicularity angles to the mirror reference. In addition, if there is a gap on the contacting surface, deformation will occur due to the tightening force at the time of fastening, so flatness also becomes important. Further, its high flexibility of formability of cordierite is an advantage also for providing a honeycomb structure for weight reduction. In ordinary glass materials, a honeycomb structure is realized by grinding from the bulk state, so that much processing time is required. On the other hand, in the case of cordierite ceramics, machining can be carried out at the stage of a soft molded body such as chalk after raw material forming and before firing. Therefore, costly grinding can be performed quickly and easily, which contribute overall cost reduction. Finally, the long term dimensional stability of cordierite is also suitable for general telescope mirrors. Low thermal expansion glass is cooled for a very long time in order to release the internal stress in the material during molding process. Therefore, the molded material takes a long time to deform. For example, in the case of Zerodur, deformation value 129 nm/m/1 year is expected as according to our research with PTB. In case of cordierite, deformation after firing is extremely small (< 3.8nm/ m/ year) compared to the low thermal expansion glass and it is possible to maintain stable dimensions over a long period of time.
In order to demonstrate the feasibility of Gapless mirrors, in this study we made a prototype of Dia.300mm parabolic mirror sample. As a result of the structural analysis for fixing the mirrors, the shape of the segmented mirror was a fan shape with an internal angle of 60 degrees and it was divided into six parts. Triangle parts (three parts) was one of the concepts for this trial, but distortion at mirror surface coming from assembling each individual mirrors tend to be created easily, and therefore, we rejected this three parts option.
Regarding the honeycomb structure, there are many cases where honeycomb designs of the same shape are used over the entire surface such as triangle and hexagon in the case of a monolithic mirror. Segmented mirror used for this evaluation, however, has a relatively small diameter of 300 mm, and it was confirmed by simulation that sufficient rigidity of cordierite material can be obtained without providing a complicated honeycomb structure. For this reason, we prioritize the workability for bolting in the horizontal direction on the contacting surface of the mirror, and an assembling structure was selected that can easily perform the assembly process. In addition, the thickness of the rib is made thicker than the other parts so that the contacting surface does not deform even by tightening with a bolt. It is confirmed that this structure does not lose practicality even if the mirror up to 1 m in diameter is scaled up.
Individual segmented mirrors are manufactured independently from forming, green machining, firing and grinding, which is a standard ceramic manufacturing process, and then six segment mirrors are fastened with bolts with precision. After aligning the height of all mirror surfaces by grinding process, the polishing process is performed at the end in order to remove any height gap at the mirror surface. As a result, it was confirmed that sagging hardly occurred at the joint area, and ideal mirror surface condition was achieved measured by interferometer. Furthermore, in order to quantitatively investigate the occurrence of deviation of each segment after assembly, a flat surface sample was also prepared. After checking the surface accuracy using the laser interferometer for the polished flat mirror, we confirmed that there was no significant segment mirror misalignment by thermal cycle test and vibration test.
In this presentation, in addition to the design results of the prototype AFGSM, the optical performance of the parabolic mirror obtained as a result of the polishing process, and the results of the environmental test are reported. We also discuss applicability of ceramic segment mirrors for space telescopes and some other observation devices.
