Ion implantation is a method of correcting figure errors in thin silicon or glass substrates. For future high-resolution, highthroughput x-ray observatories, such figure correction may be critical for thin mirror substrates. Ion implantation into both glass and silicon results in surface stress, which bends the substrate. We demonstrate that this stress may be used to improve the surface figure of flat glass wafers. We then describe three effects of ion implantation in glass and silicon. The first effect is the stress resulting from the implanted ions, and the implications for figure correction with each material. Second, each material studied also shows some relaxation after the ion beam is removed; we report on the magnitude of this relaxation and its implications. Finally, the surface stress may affect the strength of implanted materials. We report on ring-on-ring strength tests conducted on implanted glass samples.
The X-ray optics community has been developing technology for high angular resolution, large collecting area X-ray telescopes such as the Lynx X-ray telescope concept. To meet the high collecting area requirements of such telescope concepts, research is being conducted on thin, segmented optics. The mounts that fixture and align segmented optics must be the correct length to sub-micron accuracy to satisfy the angular resolution goals of such a concept. Set-andforget adjustable length optical mounting posts have been developed to meet this need. The actuator consists of a cylinder made of metal. Halfway up the height of the metal cylinder, a reduced diameter cylindrical neck is cut. To change the length of this actuator, an axial compressive or tensile force is applied to the actuator. A high-current electrical pulse is sent through the actuator, and this electrical current resistively heats the neck of the actuator. This heating temporarily reduces the yield strength of the neck, so that the applied force plastically deforms the neck. Once the current stops and the neck cools, the neck will regain yield strength, and the plastic deformation will stop. All of the plastic deformation that occurred during heating is now permanent. Both compression and expansion of these actuators has been demonstrated in steps ranging from 6 nanometers to several microns. This paper will explain the concept of ThermoYield actuation, explore X-ray telescope applications, describe an experimental setup, show and discuss data, and propose future ideas.
Slumping (or thermal-shaping) of thin glass sheets onto high precision mandrels was used successfully by NASA Goddard Space Flight Center to fabricate the NuSTAR telescope. But this process requires long thermal cycles and produces mid-range spatial frequency errors due to the anti-stick mandrel coatings. Over the last few years, we have designed and tested non-contact horizontal slumping of round flat glass sheets floating on thin layers of nitrogen between porous air-bearings using fast position control algorithms and precise fiber sensing techniques during short thermal cycles. We recently built a finite element model with ADINA to simulate the viscoelastic behavior of glass during the slumping process. The model utilizes fluid-structure interaction (FSI) to understand the deformation and motion of glass under the influence of air flow. We showed that for the 2D axisymmetric model, experimental and numerical approaches have comparable results. We also investigated the impact of bearing permeability on the resulting shape of the wafers. A novel vertical slumping set-up is also under development to eliminate the undesirable influence of gravity. Progress towards generating mirrors for good angular resolution and low mid-range spatial frequency errors is reported.
In order to advance significantly scientific objectives, future x-ray astronomy missions will likely call for x-ray telescopes
with large aperture areas (≈ 3 m2) and fine angular resolution (≈ 12). Achieving such performance is programmatically
and technologically challenging due to the mass and envelope constraints of space-borne telescopes and to the need for
densely nested grazing-incidence optics. Such an x-ray telescope will require precision fabrication, alignment, mounting,
and assembly of large areas (≈ 600 m2) of lightweight (≈ 2 kg/m2 areal density) high-quality mirrors, at an acceptable cost
(≈ 1 M$/m2 of mirror surface area). This paper reviews relevant programmatic and technological issues, as well as possible
approaches for addressing these issues-including direct fabrication of monocrystalline silicon mirrors, active (in-space
adjustable) figure correction of replicated mirrors, static post-fabrication correction using ion implantation, differential
erosion or deposition, and coating-stress manipulation of thin substrates.
Figure correction of thin x-ray telescope mirrors may be critical for future missions that require high angular resolution and large collecting areas. One promising method of providing figure correction is to use stress generated via ion implantation. Since stress-based figure correction strategies cannot correct high spatial frequency errors, it is critical to obtain glass with only low spatial frequency error. One method is thermal gas bearing slumping, where glass is softened while floating on thin films of gas. This method avoids introducing mid- or high- spatial frequency errors by eliminating contact between the glass and mandrel. Together, these two methods form a promising approach to fabricating mirrors for a high angular resolution, large-area x-ray observatory. In this paper we report on progress in understanding gas bearing slumping, and advancing the technology to curved geometry. We also report on continued progress on advancing the ion implantation technology toward correcting flight-sized mirror substrates.
Figure correction of X-ray telescope mirrors will be critical for future missions that require high angular resolution and large collecting areas. In this paper, we show that ion implantation offers a method of correcting figure errors by imparting sub-surface in-plane stress in a controllable magnitude and location in Schott D-263 glass, Corning Eagle XG glass, and crystalline silicon substrates. In addition, we can in theory achieve nearly exact corrections in Schott D-263 glass, by controlling the direction of the stress. We show that sufficient stress may be applied to Schott D-263 glass to achieve figure correction in mirrors with simulated initial figure errors. We also report on progress of a system that will be capable of correcting conical shell mirror substrates.
The successful NuSTAR telescope was fabricated with thin glass mirrors formed into conic shapes by thermal slumping of thin glass sheets onto high precision mandrels. While mirrors generated by this process have very good figure, the best mirrors to date have a resolution limited to ~7 arc sec, due primarily to mid-range scale spatial frequency errors. These mid-range errors are believed to be due to clumping and particulates in the anti-stick coatings used to prevent sticking between mandrel and mirrors. We have developed a new slumping process which avoids sticking and surface-induced mid-range error by floating hot glass substrates between a pair of porous air bearing mandrels through which compressed nitrogen is forced. We report on the design and testing of an improved air bearing slumping tool and show results of short and long slumping cycles.
An ideal bonding agent for thin-shell x-ray mirrors could be quickly applied to joints and set with deterministic and stable properties. Unfortunately, mirror assembly methods have typically utilized various epoxy formulations which are messy, slow to apply and cure, and far from deterministic or stable. Problems include shrinkage, creep and high thermal and humidity sensitivity. Once the bond is set errors are frozen in and cannot be corrected. We are developing a new method for bonding thin-foil mirrors that has the potential to solve these problems. Our process to bond mirrors to housing reference points is achieved via small beads of a low-melting-point bonding agent (such as solder or thermoset). The mirror is bonded to small contact surface points under real-time metrology. If the position of the mirror needs to be adjusted after bonding, a small force is applied normal or parallel to the contact surface and a pulsed fiber laser is used to melt an ultrathin layer of the solder for a very short time. The joint is then compressed, stretched or sheared while molten before refreezing in a new position, enabling repeatable and stable mirror position adjustments along the direction of the force in nm-level steps with minimal heat input. We present results from our prototype apparatus demonstrating proof of principle. The initial experiment includes developing a technique to bond D263 glass to Kovar, designing and building a one-dimensional stage to precisely apply force, and using an infrared laser pulse to heat the joint while measuring position and force.
The future of x-ray astronomy depends upon development of x-ray telescopes with larger aperture areas (≈ 3 m2) and
fine angular resolution (≈ 1″). Combined with the special requirements of nested grazing-incidence optics, the mass and
envelope constraints of space-borne telescopes render such advances technologically and programmatically challenging.
Achieving this goal will require precision fabrication, alignment, mounting, and assembly of large areas (≈ 600 m2) of
lightweight (≈ 1 kg/m2 areal density) high-quality mirrors at an acceptable cost (≈ 1 M$/m2 of mirror surface area). This
paper reviews relevant technological and programmatic issues, as well as possible approaches for addressing these
issues—including active (in-space adjustable) alignment and figure correction.
Air bearing glass slumping followed by ion implantation for fine figure correction constitutes a promising process for fabricating thin glass segmented mirrors for future high-resolution x-ray telescopes. We have previously demonstrated the feasibility of both air bearing slumping and ion implantation figure correction to produce mirrors with good figure and without introducing mid spatial-frequency errors or roughness. In this work, we describe a mechanically-robust slumping tool design that can be adapted to Wolter I mirror shapes; and we describe progress on understanding ion implantation for use as a figure correction process, by using in-situ curvature measurements in a tandem ion accelerator.
Molding glass by using air bearings is a promising procedure for inexpensive and high precision glass shaping. Thin glass sheets are sandwiched between air bearings and pushed flat while being thermally cycled. In this study, a novel device for shaping glass is created and tested using 0.5 mm thick, 100 mm round, Schott D263 wafers. Numerous samples were shaped with varying values for bearing-to-glass gap and maximum temperature, and were measured with a Shack Hartmann metrology tool. Glass was shaped with bearing-to-glass gaps of >50 μm, 36±2.5 μm, and 30.5±2.5 μm. The best peak-to-valley (P-V) flatness achieved is 6.7/3.6±0.5 μm for front/back of the glass sheet, using a gap of 36±2.5 μm. The average steady-state P-V achieved is 12 μm. Using the same device parameters, the best repeatability achieved over the whole 100 mm wafer is 2.7±0.5 μm P-V and 9.5 arcseconds RMS slope error. When looking at 60 mm sections, the repeatability improves to <1 μm P-V and 5±0.5 arcsec.
Achieving both high resolution and large collection area in the next generation of x-ray telescopes requires highly accurate shaping of thin mirrors, which is not achievable with current technology. Ion implantation offers a promising method of modifying the shape of mirrors by imparting internal stresses in a substrate, which are a function of the ion species and dose. This technique has the potential for highly deterministic substrate shape correction using a rapid, low cost process. Wafers of silicon and glass (D-263 and BK-7) have been implanted with Si+ ions at 150 keV, and the changes in shape have been measured using a Shack-Hartmann metrology system. We show that a uniform dose over the surface repeatably changes the spherical curvature of the substrates, and we show correction of spherical curvature in wafers. Modeling based on experiments with spherical curvature correction shows that ion implantation could be used to eliminate higher-order shape errors, such as astigmatism and coma, by using a spatially-varying implant dose. We will report on progress in modelling and experimental tests to eliminate higher-order shape errors. In addition, the results of experiments to determine the thermal and temporal stability of implanted substrates will be reported.