Search

Thin adjustable X-ray mirrors can correct deformations generated from fabrication, gravity release, mounting stresses, drifting stresses in the reflecting layer(s) and thermal variations while maintaining high angular resolution (< 0.5 arcsecond) and large effective area (< 2 m2) required for future X-ray missions. This work presents fabrication process developments for adjustable mirror segments with actuators for the Lynx X-ray observatory mission concept. Piezoelectric actuator arrays were fabricated on the convex side of precision slumped glass or curved silicon mirror segments using a 1.5 μm thick lead zirconate titanate (PZT) film. A two-layer metal routing scheme with a polymeric insulator was used to independently address 288 actuators on the mirror. The two-layer metal allows narrow kerfs between actuators and increased actuator density. Anisotropic conductive film was used to bond thin flexible copper cables to flat edges of the mirror to interface with external control electronics. This prototype mirror has eight cables with a total of 290 connections to access the array. To reduce the cabling complexity for future mirrors, thin film transistors have been fabricated on the curved mirror to function as access switches. To facilitate this, a mask aligner that allows precision alignment on curved mirror segments was developed and arrays of thin film transistors (TFT) on curved substrates have been tested. TFT and actuator integration on future mirrors will reduce external connections to just two cables with a total of 30 connections. Keywords: Lynx, adjustable optics, X-ray optics, thin film piezoelectric, curved substrate aligner

Exquisite angular resolution (< 0.5 arcsec) and high effective area (≥ 2 m^2 @ 1 keV) are requirements for a next-generation X-ray observatory capable of tackling outstanding problems in high energy astrophysics, including understanding how black holes grow over cosmic time and how hot baryonic material is distributed on the largest scales. However, realizing a telescope with this performance is challenging, as the thin optics required are susceptible to fabrication errors, thin film stress, and mounting deformations. One potential method of addressing these errors to fabricate adjustable X-ray optics – mirrors with actuators capable of correcting the optic’s figure following mounting. In this work, we present interferometric measurements of an adjustable X-ray optics prototype with lead titanate zirconate (PZT) actuators. We detail the realized actuator performance and correctability of the mirror prototype, and discuss the implications for the next-generation of adjustable mirrors.

Adjustable X-ray optics represent a potential mirror technology for the NASA Lynx X-ray observatory mission concept. Adjustable optics employ an integrated micron-thick piezoelectric film deposited on the convex side of silicon Wolter-type mirror segments. Discrete, independently addressable electrodes on the convex surface form individual actuators; the applied voltages are used to correct the shape of the mirror segments for figure errors resulting from a change in thermal environment, epoxy creep, or failure of an epoxy bond. On-orbit correction requires a metrology system to provide real-time feedback of mirror figure. We are examining the use of deposited semiconductor strain gauges to monitor mirror mechanical strains and surface temperatures. To establish requirements for monitoring we modeled a variety of thermal and mechanical disturbances to a mirror segment such as might occur on-orbit or from launch. Models are described and resulting requirements and performance discussed.

Arcus provides high-resolution soft X-ray spectroscopy in the 12-50 Å bandpass with unprecedented sensitivity, including spectral resolution < 2500 and effective area < 250 cm^2. The three top science goals for Arcus are (1) to measure the effects of structure formation imprinted upon the hot baryons that are predicted to lie in extended halos around galaxies, (2) to trace the propagation of outflowing mass, energy, and momentum from the vicinity of the black hole to extragalactic scales as a measure of their feedback, and (3) to explore how stars form and evolve. Arcus uses the same 12 m focal length grazing-incidence Silicon Pore X-ray Optics (SPOs) that ESA has developed for the Athena mission; the focal length is achieved on orbit via an extendable optical bench. The focused X-rays from these optics are diffracted by high-efficiency Critical-Angle Transmission (CAT) gratings, and the results are imaged with flight-proven CCD detectors and electronics. Combined with the high-heritage NGIS LEOStar-2 spacecraft and launched into 4:1 lunar resonant orbit, Arcus provides high sensitivity and high efficiency observing of a wide range of astrophysical sources.

Adjustable X-ray optics is the technology under study by SAO and PSU for the realization of the proposed X-ray telescope Lynx. The technology is based on thin films of lead-zirconate-titanate (PZT) deposited on the back of thermally formed thin substrates, and represents a potential solution to the challenging trade-off between high-surface quality and low mass, that limits the performance of current generation of X-ray telescopes. The technology enables the correction of mirror fabrication figure, mounting induced distortions, and on-orbit correction for variations in the mirror thermal environment. We describe the current state of development, presenting updated test data, anticipation of performances and expectations.

The combination of the hot slumping and the Ion Beam Figuring (IBF) technologies can be a very competitive solution for the realization of x-ray optics with excellent imaging capabilities and high throughput. While very thin mirrors segments can be realized by slumping with residual figure errors below few hundreds of nanometres, a non-contact and deterministic process (dependent on dwell time), like IBF, is a very effective post facto correction, as it avoids all the problems due to the handling and the supporting system. In the last years, the two processes were proven compatible with very thin sheet of Eagle XG glasses (0.4 mm thickness). Nevertheless, the fast convergence of the process is a key factor to limit the cost of the mirror plate production. A deeper characterization of removal function stability showed that its repeatability between each run has to be improved for a real enhancement of the process convergence factor. A new algorithm based on de-convolution has been implemented and tested, with important advantages in terms of calculation speed, minimum material removal and optimization possibilities. By analysing the metrological data of test slumped glasses, we showed how the IBF is effective in the correction of figure errors on scales above 8 - 10 mm. An overall figuring time of few hours is required with surface error around 100 nm rms. Thanks to the thickness measurement data, which are performed in transmission mode with an interferometric set-up, we demonstrated that it is possible to disentangle the effective amount of the material removed and the deformations introduced during the process.

The NASA Lynx mission concept is under study as a potential successor to the Chandra X-ray Observatory, for launch in the 2030s. Like Chandra, Lynx is to provide 0.5 arcsec half power diameter imaging at 1keV, but with 30 times the collecting area, and sub-arcsec imaging over a 10 arcmin (radius) field-of-view. Adjustable X-ray optics technology represents a potential approach to meet the challenging Lynx requirements by enabling the correction of mirror fabrication figure, mounting induced distortions, and on-orbit correction for variations in the mirror thermal environment. We describe the current state of development of the technology, including summarizing recent test data, development of mirror assembly error budgets, and discussion of the mirror assembly optical design and its anticipated performance.

The proposed Lynx telescope is an X-ray observatory with Chandra-like angular resolution and about 30 times larger effective area. The technology under development at SAO is based on the deposition of piezoelectric material on the back of glass substrates, used to correct longer wavelength figure errors. This requires a large number (about 8000) of figured segments with sufficient quality to be in the range of correctibility of the actuators. Thermal forming of thin glass offers a convenient approach, being based on intrinsically smooth surfaces (which doesn’t require polishing or machining), available in large quantity and at a low cost from flat display industry. Being a replica technique, this approach is particularly convenient both for development and for the realization of modular/segmented telescopes. In this paper we review the current status and the most recent advances in the thermal forming activities at SAO, and the perspectives for the employment of these substrates for the adjustable X-Ray optics.

The high-quality surface characteristics, both in terms of figure error and of micro-roughness, required on the mirrors of a high angular resolution x-ray telescope are challenging, but in principle well suited with a deterministic and non-contact process like the ion beam figuring. This process has been recently proven to be compatible even with very thin (thickness around 0.4mm) sheet of glasses (like D263 and Eagle). In the last decade, these types of glass have been investigated as substrates for hot slumping, with residual figure errors of hundreds of nanometres. In this view, the mirrors segments fabrication could be envisaged as a simple two phases process: a first replica step based on hot slumping (direct/indirect) followed by an ion beam figuring which can be considered as a post-fabrication correction method. The first ion beam figuring trials, realized on flat samples, showed that the micro-roughness is not damaged but a deeper analysis is necessary to characterize and eventually control/compensate the glass shape variations. In this paper, we present the advancements in the process definition, both on flat and slumped glass samples.

Piezoelectric adjustable optics are being developed for high throughput, high resolution, low mass Xray mirror assemblies. These optics require robust piezoelectric thin films and reproducible lithographic patterning on curved glass substrates. This work details the cleaning of Corning Eagle XG glass substrates for thin shell X-ray mirrors by a three stage acid and solvent cleaning procedure before a 0.02 μm Ti adhesion layer and a 0.1 μm Pt bottom electrode layer was deposited using DC magnetron sputtering. Piezoelectric Pb(Zr0.52Ti0.48)0.99Nb0.01O3 thin films with a thickness of 1.5 μm were then deposited by radio frequency magnetron sputtering in three 0.5 µm layers with intermittent annealing steps in a rapid thermal annealing furnace at 650°C for 60 seconds. Defects observed in the piezoelectric thin films were linked to residue remaining on the glass after cleaning. 112 piezoelectric cells and 100 μm wide conductive Pt traces were patterned using bilayer photolithography. The photoresist layers were deposited using spin coating at 2000 and 4000 RPM to achieve uniform 1 μm thick layers, resulting in reproducibly resolved features with limiting resolutions of approximately >25 μm. The resulting mirror pieces achieved a 100% yield, with average relative permittivity of 1270, dielectric loss 0.047, coercive field 30 kV/cm and remanent polarization of 20 μC/cm2 . While the defects observed in the films appeared to have not influence on the electrical properties, additional cleaning steps using DI water were proposed to further reduce their presence.

Thin x-ray optics with high angular resolution (≤ 0.5 arcsec) over a wide field of view enable the study of a number of astrophysically important topics and feature prominently in Lynx, a next-generation x-ray observatory concept currently under NASA study. In an effort to address this technology need, piezoelectrically adjustable, thin mirror segments capable of figure correction after mounting and on-orbit are under development. We report on the fabrication and characterization of an adjustable cylindrical slumped glass optic. This optic has realized 100% piezoelectric cell yield and employs lithographically patterned traces and anisotropic conductive film connections to address the piezoelectric cells. In addition, the measured responses of the piezoelectric cells are found to be in good agreement with finite-element analysis models. While the optic as manufactured is outside the range of absolute figure correction, simulated corrections using the measured responses of the piezoelectric cells are found to improve 5 to 10 arcsec mirrors to 1 to 3 arcsec [half-power diameter (HPD), single reflection at 1 keV]. Moreover, a measured relative figure change which would correct the figure of a representative slumped glass piece from 6.7 to 1.2 arcsec HPD is empirically demonstrated. We employ finite-element analysis-modeled influence functions to understand the current frequency limitations of the correction algorithm employed and identify a path toward achieving subarcsecond corrections.

The adaptive optics system for the SMARTX telescope consists of piezoelectric PbZr 0.52 Ti 0.48 O 3 (PZT) thin films deposited on the backside of the mirror. To achieve sufficient strain response from the piezoelectric films, the substrates chosen are thin, flexible glass pieces that can be slumped into the desired curvature. Preliminary testing has been performed using flat pieces and parts that were slumped into a cylindrical geometry. Commercially available boro-aluminosilicate glass segments (100 × 100 mm) were slumped along one axis to a radius of curvature of 220 mm. 2-μm-thick PZT films were deposited via RF magnetron sputtering on flat glass substrates to demonstrate the viability of the processing approach. The deposited PZT showed high yield (>95%) on 1-cm 2 electrodes. The films exhibited relative permittivity values near 1500 and loss tangents below 0.05. In addition, the remanent polarization was 20 μC/cm 2 with coercive fields near 30 kV/cm. 1-μm-thick films with comparable electrical parameters were then deposited onto the slumped substrates. To understand the piezoelectric response of the films and characterize the device performance, influence function measurements were performed. The typical cell response using a 4 V (1.3V c ) drive voltage corresponded to a 0.5 μm out-of-plane displacement, which relates to an in-plane strain value larger than 150 ppm. Both of these parameters meet benchmark requirements for reaching the targeted 0.5 arcsecond resolution goal of the SMART-X telescope. These results demonstrate a viable route to fabricate highly functional mirror segments for the SMART-X telescope.

Lynx, one of the four strategic mission concepts under study for the 2020 Astrophysics Decadal Survey, provides leaps in capability over previous and planned x-ray missions and provides synergistic observations in the 2030s to a multitude of space- and ground-based observatories across all wavelengths. Lynx provides orders of magnitude improvement in sensitivity, on-axis subarcsecond imaging with arcsecond angular resolution over a large field of view, and high-resolution spectroscopy for point-like and extended sources in the 0.2- to 10-keV range. The Lynx architecture enables a broad range of unique and compelling science to be carried out mainly through a General Observer Program. This program is envisioned to include detecting the very first seed black holes, revealing the high-energy drivers of galaxy formation and evolution, and characterizing the mechanisms that govern stellar evolution and stellar ecosystems. The Lynx optics and science instruments are carefully designed to optimize the science capability and, when combined, form an exciting architecture that utilizes relatively mature technologies for a cost that is compatible with the projected NASA Astrophysics budget.

Piezoelectric adjustable x-ray optics use magnetron sputtered thin film coatings on both sides of a thin curved glass substrate. To produce an optic suitable for a mission requiring high-angular resolution like “Lynx,” the integrated stresses (stress×thickness) of films on both sides of the optic must be approximately equal. Thus, understanding how sputtered film thickness distributions change for convex and concave curved substrates is necessary. To address this, thickness distributions of piezoelectric Pb0.995(Zr0.52Ti0.48)0.99Nb0.01O3 films are studied on flat, convex, and concave cylindrical substrates with a 220-mm radius of curvature. A mathematical model of the film thickness distribution is derived based on the geometric properties of the sputter tool and the substrate, and film thicknesses deposited with a commercially available sputtering tool are measured with spectroscopic ellipsometry. Experiment and modeled results for flat and convex curved substrates demonstrate good agreement, with average relative thickness distribution difference of 0.19% and −0.10% respectively, and a higher average difference of 1.4% for concave substrates. The calculated relative thickness distributions are applied to the convex and concave sides of a finite-element analysis (FEA) model of an adjustable x-ray optic prototype. The FEA model shows that, left uncorrected, the relative film thickness variation will yield an optic with an optical performance of 2.6 arc sec half power diameter (HPD) at 1 keV. However, the mirror figure can be corrected to diffraction-limited performance (0.3 arc sec HPD) using the piezoelectric adjusters, suggesting that the tolerances for applying a balanced integrated stress on both sides of a mirror are alleviated for adjustable x-ray optics as compared to traditional static x-ray mirrors. Furthermore, the piezoelectric adjusters will also allow changes in mirror figure over the telescope lifetime due to drift in the stress states of the x-ray surfaces to be corrected on orbit.

In this paper we review the progress and current status of thermal forming activities at SAO, highlighting the most relevant technical problems and the way to solve them. These activities are devoted to the realization of mirror substrates for the X-ray surveyor mission concept, an observatory with Chandra-like angular resolution and 30 times more effective area or larger. The technology under development at SAO is based on the deposition of piezoelectric material on the back of the substrates. About 8000 mirror segments, with initial quality of 10 arcseconds or better are required for the telescope.

The X-ray Surveyor is a mission concept for a next generation X-ray observatory. This mission will feature roughly 30 times the effective area of the Chandra Observatory while matching its sub-arcsecond angular resolution. The key to meeting these requirements is lightweight, segmented optics. To ensure these optics achieve and maintain sub-arcsecond performance, we propose to use piezoelectric coatings for post-bonding and on-orbit figure correction. We have fabricated a cylindrical prototype optic with piezoelectric adjusters and measured its performance using optical metrology. We present the results of this experiment and discuss their implications for an observatory featuring adjustable X-ray optics.

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....

Lynx, a next generation X-ray observatory concept currently under study, requires lightweight, high spatial resolution X-ray mirrors. Here we detail the development and fabrication of one of the candidate technologies for Lynx, piezoelectric adjustable X-ray optics. These X-ray mirrors are thin glass shell mirrors with Cr/Ir X-ray reflective coatings on the mirror side and piezoelectric thin film actuators on the actuator side. Magnetron sputtering was used to deposit metal electrodes and metal-oxide piezoelectric layers. The piezoelectric (Pb0.995(Zr0.52Ti0.48)0.99Nb0.01O3) was divided into 112 independent piezoelectric actuators, with 100% yield achieved. We discuss the fabrication procedure, residual thermal stresses and tuning of the Cr/Ir coating stress for the purposes of stress balancing.

Adjustable X-ray optics represent a potential enabling technology for simultaneously achieving large effective area and high angular resolution for future X-ray Astronomy missions. The adjustable optics employ a bimorph mirror composed of a thin (1.5 μm) film of piezoelectric material deposited on the back of a 0.4 mm thick conical mirror segment. The application of localized electric fields in the piezoelectric material, normal to the mirror surface, result in localized deformations in mirror shape. Thus, mirror fabrication and mounting induced figure errors can be corrected, without the need for a massive reaction structure. With this approach, though, film stresses in the piezoelectric layer, resulting from deposition, crystallization, and differences in coefficient of thermal expansion, can distort the mirror. The large relative thickness of the piezoelectric material compared to the glass means that even 100MPa stresses can result in significant distortions. We have examined compensating for the piezoelectric processing related distortions by the deposition of controlled stress chromium/iridium films on the front surface of the mirror. We describe our experiments with tuning the product of the chromium/iridium film stress and film thickness to balance that resulting from the piezoelectric layer. We also evaluated the repeatability of this deposition process, and the robustness of the iridium coating....

The alignment of thin-shell, segmented mirrors for Wolter-I telescopes frequently involves the use of a Hartmann test. In order to get optical throughput in the Hartmann test, the mirrors must first be coarsely aligned to one another and to the metrology system. In the past, the coarse alignment of these mirrors at the Smithsonian Astrophysical Observatory has largely relied upon component machine tolerances and contact measurements with a coordinate measurement machine (CMM). This process takes time and does not produce reliable nor repeatable results. Thus, methods were developed to allow for the quick and reliable coarse alignment of thin- shell, segmented mirrors at their final locations in the mirror assembly. We present the coarse alignment system developed at the Smithsonian Astrophysical Observatory and its use in the alignment of thin-shell, segmented mirrors for the adjustable X-ray optics program.

NASA's Chandra X-ray Observatory continues to provide an unparalleled means for exploring the high-energy universe. With its half-arcsecond angular resolution, Chandra studies have deepened our understanding of galaxy clusters, active galactic nuclei, galaxies, supernova remnants, neutron stars, black holes, and solar system objects. As we look beyond Chandra, it is clear that comparable or even better angular resolution with greatly increased photon throughput is essential to address ever more demanding science questions—such as the formation and growth of black hole seeds at very high redshifts; the emergence of the first galaxy groups; and details of feedback over a large range of scales from galaxies to galaxy clusters. Recently, we initiated a concept study for such a mission, dubbed X-ray Surveyor. The X-ray Surveyor strawman payload is comprised of a high-resolution mirror assembly and an instrument set, which may include an X-ray microcalorimeter, a high-definition imager, and a dispersive grating spectrometer and its readout. The mirror assembly will consist of highly nested, thin, grazing-incidence mirrors, for which a number of technical approaches are currently under development—including adjustable X-ray optics, differential deposition, and new polishing techniques applied to a variety of substrates. This study benefits from previous studies of large missions carried out over the past two decades and, in most areas, points to mission requirements no more stringent than those of Chandra.

We report on improvements in our efforts to control and characterize piezoelectrically adjustable, thin glass optics. In the past, an optical profilometer and a Shack-Hartmann wavefront sensor have been used to measure influence functions for a at adjustable mirror. An electronics system has since been developed to control > 100 actuator cells and has been used in a full calibration of a high-yield at adjustable mirror. The calibrated influence functions have been used to induce a pre-determined figure change to the mirror, representing our first attempt at figure control of a full mirror. Furthermore, we have adapted our metrology systems for cylindrical optics, allowing characterization of Wolter-type mirrors. We plan to use this metrology to perform the first piezoelectric figure correction of a cylindrical mirror over the next year....

The new cosmological frontier will be the study of the very first stars, galaxies and black holes in the early Universe. These objects are invisible to the current generation of X-ray telescopes, such as Chandra. In response, the Generation-X (“Gen-X”) Vision Mission has been proposed as a future X-ray observatory which will be capable of detecting the earliest objects. X-ray imaging and spectroscopy of such faint objects demands a large collecting area and high angular resolution. The Gen-X mission plans 100 m2 collecting area at 1 keV (1000× that of Chandra), and with an angular resolution of 0.1″. The Gen-X mission will operate at Sun–Earth L2, and might involve four 8 m diameter telescopes or even a single 20 m diameter telescope. To achieve the required effective area with reasonable mass, very lightweight grazing incidence X-ray optics must be developed, having an areal density 100× lower than in Chandra, with mirrors as thin as 0.1 mm requiring active on-orbit figure control. The suite of available detectors for Gen-X should include a large-area high resolution imager, a cryogenic imaging spectrometer, and a grating spectrometer. We discuss use of Gen-X to observe the birth of the first black holes, stars and galaxies, and trace their cosmic evolution.

We describe progress in the development of adjustable grazing incidence X-ray optics for 0.5 arcsec resolution cosmic X-ray imaging. To date, no optics technology is available to blend high resolution imaging like the Chandra X-ray Observatory, with square meter collecting area. Our approach to achieve these goals simultaneously is to directly deposit thin film piezoelectric actuators on the back surface of thin, lightweight Wolter-I or Wolter- Schwarschild mirror segments. The actuators are used to correct mirror figure errors due to fabrication, mounting and alignment, using calibration and a one-time figure adjustment on the ground. If necessary, it will also be possible to correct for residual gravity release and thermal effects on-orbit. In this paper we discuss our most recent results measuring influence functions of the piezoelectric actuators using a Shack-Hartmann wavefront sensor. We describe accelerated and real-time lifetime testing of the piezoelectric material, and we also discuss changes to, and recent results of, our simulations of mirror correction.

The future of x-ray astronomy depends upon development of x-ray telescopes with larger aperture areas (approx. = 3 square meters) and fine angular resolution (approx. = 1 inch). 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 (approx. = 600 square meters) of lightweight (approx. = 1 kilogram/square meter areal density) high-quality mirrors at an acceptable cost (approx. = 1 million dollars/square meter 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.

Addressing the astrophysical problems of the 2020's requires sub-arcsecond x-ray imaging with square meter effective area. Such requirements can be derived, for example, by considering deep x-ray surveys to find the young black holes in the early universe (large redshifts) which will grow into the first supermassive black holes. We have envisioned a mission based on adjustable x-ray optics technology, in order to achieve the required reduction of mass to collecting area for the mirrors. We are pursuing technology which effects this adjustment via thin film piezoelectric "cells" deposited directly on the non-reflecting sides of thin, slumped glass. While SMARTX will also incorporate state-of-the-art x-ray cameras, the remaining spacecraft systems have no more stringent requirements than those which are well understood and proven on the current Chandra X-ray Observatory.

The proposed SMART-X telescope consists of a pixelated array of a piezoelectric lead zirconate titanate (PZT) thin film deposited on flexible glass substrates. These cells or pixels are used to actively control the overall shape of the mirror surface. It is anticipated that the telescope will consist of 8,000 mirror panels with 400-800 cells on each panel. This creates an enormous number (6.4 million) of traces and contacts needed to address the PZT. In order to simplify the design, a row/column addressing scheme using ZnO thin film transistors (TFTs) is proposed. In addition, connection of the gate and drain lines on the mirror segment to an external supply via a flexible cable was investigated through use of an anisotropic conductive film (ACF). This paper outlines the design of the ZnO TFTs, use of ACF for bonding, and describes a specially designed electronics box with associated software to address the desired cells.

Post-mounting figure correction is a promising avenue to produce low-mass, high-resolution X-ray telescopes. We have demonstrated the feasibility of this approach using piezoelectrically adjustable glass mirrors. Influence functions for various piezoelectric cells have previously been measured with an optical profilometer, but with significant noise. We have improved on both the speed and accuracy of these measurements using a Shack- Hartmann wavefront sensing system. Additionally, we have altered our wavefront sensing system to investigate the mid frequency roughness of our slumped glass mirrors. We report on initial results for measurements of both influence functions and mid frequency roughness and describe our path forward.

Piezoelectric PbZr(0.52)Ti(0.48)O(3) (PZT) thin films deposited on thin glass substrates have been proposed for adjustable optics in future x-ray telescopes. The light weight of these x-ray optics enables large collecting areas, while the capability to correct mirror figure errors with the PZT thin film will allow much higher imaging resolution than possible with conventional lightweight optics. However, the low strain temperature and flexible nature of the thin glass complicate the use of chemical-solution deposition due to warping of the substrate at typical crystallization temperatures for the PZT. RF magnetron sputtering enabled preparation of PZT films with thicknesses up to 3 μm on Schott D263 glass substrates with much less deformation. X-ray diffraction analysis indicated that the films crystallized with the perovskite phase and showed no indication of secondary phases. Films with 1 cm(2) electrodes exhibited relative permittivity values near 1100 and loss tangents below 0.05. In addition, the remanent polarization was 26 μC/cm(2) with coercive fields of 33 kV/cm. The transverse piezoelectric coefficient was as high as -6.1±0.6 C/m(2). To assess influence functions for the x-ray optics application, the piezoelectrically induced deflection of individual cells was measured and compared with finite-element-analysis calculations. The good agreement between the results suggests that actuation of PZT thin films can control mirror figure errors to a precision of about 5 nm, allowing sub-arcsecond imaging.

The direct deposition of piezoelectric thin films on thin substrates offers an appealing technology for the realization of lightweight adjustable mirrors capable of sub-arcsecond resolution. This solution will make it possible to realize X-ray telescopes with both large effective area and exceptional angular resolution and, in particular, it will enable the realization of the adjustable optics for the proposed mission Square Meter Arcsecond Resolution X-ray Telescope (SMART-X). In the past years we demonstrated for the first time the possibility of depositing a working piezoelectric thin film (1-5 um) made of lead-zirconate-titanate (PZT) on glass. Here we review the recent progress in film deposition and influence function characterization and comparison with finite element models. The suitability of the deposited films is analyzed and some constrains on the piezoelectric film performances are derived. The future steps in the development of the technology are described.

In the half century since the initial discovery of an astronomical (non-solar) x-ray source, the sensitivity for detection of cosmic x-ray sources has improved by ten orders of magnitude. Largely responsible for this dramatic progress has been the refinement of the (grazing-incidence) focusing x-ray telescope. The future of x-ray astronomy relies upon the development of x-ray telescopes with larger aperture areas (greater than 1 m2) and finer angular resolution (less than 1.). Combined with the special requirements of grazing-incidence optics, the mass and envelope constraints of space-borne telescopes render such advances technologically challenging.requiring precision fabrication, alignment, and assembly of large areas (greater than 100 m2) of lightweight (approximately 1 kg m2 areal density) mirrors. Achieving precise and stable alignment and figure control may entail active (in-space adjustable) x-ray optics. This paper discusses relevant programmatic and technological issues and summarizes progress toward active x-ray telescopes.

The proposed SMART-X telescope includes adaptive optics systems that use piezoelectric lead zirconate titanate (PZT) films deposited on flexible glass substrates. Several processing constraints are imposed by current designs: the crystallization temperature must be kept below 550 °C, the total stress in the film must be minimized, and the yield on 1 cm 2 actuator elements should be < 90%. For this work, RF magnetron sputtering was used to deposit films since chemical solution deposition (CSD) led to warping of large area flexible glass substrates. A PZT 52/48 film that wasdeposited at 4 mTorr and annealed at 550 °C for 24 hours showed no detectable levels of either PbO or pyrochlore second phases. Large area electrodes (1cm x 1 cm) were deposited on 4” glass substrates. Initially, the yield of the devices was low, however, two methods were employed to increase the yield to near 100 %. The first method included a more rigorous cleaning to improve the continuity of the Pt bottom electrode. The second method was to apply 3 V DC across the capacitor structure to burn out regions of defective PZT. The result of this latter method essentially removed conducting filaments in the PZT but left the bulk of the material undamaged. By combining these two methods, the yield on the large area electrodes improved from < 10% to nearly 100%.

The next generation of large X-ray telescopes with sub-arcsecond resolution will require very thin, highly nested grazing incidence optics. To correct the low order figure errors resulting from initial manufacture, the mounting process, and the effects of going from 1 g during ground alignment to zero g on-orbit, we plan to adjust the shapes via piezoelectric “cells” deposited on the backs of the reflecting surfaces. This presentation investigates how well the corrections might be made. We take a benchmark conical glass element, 410×205 mm, with a 20×20 array of piezoelectric cells 19×9 mm in size. We use finite element analysis to calculate the influence function of each cell. We then simulate the correction via pseudo matrix inversion to calculate the stress to be applied by each cell, considering distortion due to gravity as calculated by finite element analysis, and by putative low order manufacturing distortions described by Legendre polynomials. We describe our algorithm and its performance, and the implications for the sensitivity of the resulting slope errors to the optimization strategy.

We report on technical progress made over the past year developing thin film piezoelectric adjustable grazing incidence optics. We believe such mirror technology represents a solution to the problem of developing lightweight, sub-arc second imaging resolution X-ray optics. Such optics will be critical to the development next decade of astronomical X-ray observatories such as SMART-X, the Square Meter Arc Second Resolution X-ray Telescope. SMART-X is the logical heir to Chandra, with 30 times the collecting area and Chandra-like imaging resolution, and will greatly expand the discovery space opened by Chandra’s exquisite imaging resolution. In this paper we discuss deposition of thin film piezoelectric material on flat glass mirrors. For the first time, we measured the local figure change produced by energizing a piezo cell – the influence function, and showed it is in good agreement with finite element modeled predictions. We determined that at least one mirror substrate material is suitably resistant to piezoelectric deposition processing temperatures, meaning the amplitude of the deformations introduced is significantly smaller than the adjuster correction dynamic range. Also, using modeled influence functions and IXO-based mirror figure errors, the residual figure error was predicted post-correction. The impact of the residual figure error on imaging performance, including any mid-frequency ripple introduced by the corrections, was modeled. These, and other, results are discussed, as well as future technology development plans.

To extend the effective collecting area for future X-ray astronomy observatories, it is necessary to use highly nested, very thin glass shells. The capability to adjust the figure of these shells on-orbit is essential in order to achieve the present X-ray imaging capability, of order 1/2 arcsec resolution, which is provided to astronomers world-wide by the Chandra X-ray Observatory. We are pursuing concepts to carry out this adjustment using piezoelectric crystals deposited directly on the back sides of the reflectors, and divided into a pattern of discrete actuators by individually controlled electrodes. In this paper we carry out a simulation of how well gravity induced errors might be corrected by this process. We consider a conical glass piece 205 mm axially by 410 mm azimuthally, and with 1 m radius of curvature. We do a finite element analysis to calculate the influence function of each of a set of 20 × 20 piezo pixels. Then we apply a 1g force axially, using various constraint conditions, and calculate the coefficients of the set of influence functions such that the adjusted shape minimizes the slope errors.

Generation-X is required to be an X-ray observatory with 50 m2 effective collecting area and 0.1 arcsec half-power diameter (HPD) angular resolution at 1 keV. It is conceived that a launch vehicle such as that studied for the Ares V will carry a monolithic 16-m-diameter mirror to the earth-sun L2 point. Even with such a vehicle, the reflectors comprising the ≈ 250 nested shells must be extremely light-weight. Therefore their figure and alignment cannot be achieved on the ground, and likely could not be maintained through the launch environment. We will present a conceptual solution to those constraints: adjustable X-ray optics, as a case of "adaptive" optics where the stability once in orbit should require adjustments no more frequently than yearly. The figure would be adjusted via thin-film actuators deposited directly to the back (non-reflecting) side of each element. This bi-morph configuration would impart in-plane strains via the piezoelectric or electrostrictive effect. Requirements of the adjustment are to the order of a few nanometer precision. Each shell, and each module, must also be aligned, to tolerances of about 0.1 micrometer. We conceive that on-orbit data would be acquired by a built-in Hartmann system for the alignment adjustments and low-order figure, and by ring profile measurements of a very bright celestial X-ray source to correct figure errors up to the mid-frequency range of several hundredths cycles mm −1 .

Adjustable x-ray optics offer the promise of much higher imaging resolution with lightweight optics, providing the key technology for the development of the next generation of astronomical x-ray telescopes such as Generation-X. These adjustable grazing incidence optics might be adjusted only once, on-orbit. To produce theses optics will require the development of several component technologies along with their integration into a new mirror concept. In this paper we define a number of the key technologies necessary for adjustable x-ray optics for astronomy, give a brief description of the issues involved, and some status of these activities being developed as part of our adjustable optics development program at the Smithsonian Astrophysical Observatory.

The Chandra X-ray Observatory , with its sub-arc second resolution, has revolutionized X-ray astronomy by revealing an extremely complex X-ray sky and demonstrating the power of the X-ray window in exploring fundamental astrophysical problems. Larger area telescopes of still higher angular resolution promise further advances. We are engaged in the development of a mission concept, Generation-X, a 0.1 arc second resolution x-ray telescope with tens of square meters of collecting area, 500 times that of Chandra . To achieve these two requirements of imaging and area, we are developing a grazing incidence telescope comprised of many mirror segments. Each segment is an adjustable mirror that is a section of a paraboloid or hyperboloid, aligned and figure corrected in situ on-orbit. To that end, finite element analyses of thin glass mirrors are performed to determine influence functions for each actuator on the mirrors, in order to develop algorithms for correction of mirror deformations. The effects of several mirror mounting schemes are also studied. The finite element analysis results, combined with measurements made on prototype mirrors, will be used to further refine the correction algorithms.

The proposed adaptive optics system for the Gen-X telescope uses piezoelectric lead zirconate titanate (PZT) films deposited on flexible glass substrates. The low softening transition of the glass substrates imposes several processing challenges that require the development of new approaches to deposit high quality PZT thin films. Synthesis and optimization of chemical solution deposited 1 μm thick films of PbZr 0.52 Ti 0.48 O 3 on small area (1 in 2 ) and large area (16 in 2 ) Pt/Ti/glass substrates has been performed. In order to avoid warping of the glass at temperatures typically used to crystallize PZT films (~700°C), a lower temperature, two-step crystallization process was employed. An ~80 nm thick seed layer of PbZr 0.30 Ti 0.70 O 3 was deposited to promote the growth of the perovskite phase. After the deposition of the seed layer, the films were annealed in a rapid thermal annealing (RTA) furnace at 550°C for 3 minutes to nucleate the perovskite phase. This was followed by isothermal annealing at 550°C for 1 hour to complete crystallization. For the subsequent PbZr 0.52 Ti 0.48 O 3 layers, the same RTA protocol was performed, with the isothermal crystallization implemented following the deposition of three PbZr 0.52 Ti 0.48 O 3 spin-coated layers. Over the frequency range of 1 kHz to 100 kHz, films exhibit relative permittivity values near 800 with loss tangents below 0.07. Hysteresis loops show low levels of imprint with coercive fields of 40-50 kV/cm in the forward direction and 50-70 kV/cm in the reverse direction. The remanent polarization varied from 25-35 μC/cm 2 and e 31,f values were approximately -5.0 C/m 2 . In scaling up the growth procedure to large area films, where warping becomes more pronounced due to the increased size of the substrate, the pyrolysis and crystallization conditions were performed in a box furnace to improve the temperature uniformity. By depositing films on both sides of the glass substrate, the tensile stresses are balanced, providing a sufficiently flat surface to continue PZT deposition. The properties of the large area film are comparable to those obtained on small substrates. While sol-gel processing is a viable approach to the deposition of high quality PZT thin films on glass substrates, preliminary results using RF magnetron sputter deposition demonstrate comparable properties with a significantly simpler process that offers a superior route for large scale production.

The next large x-ray astrophysics mission launched will likely include soft x-ray spectroscopy as a primary capability. A requirement to fulfill the science goals of such a mission is a large-area x-ray telescope focusing sufficient x-ray flux to perform high-resolution spectroscopy with reasonable observing times. The IXO soft x-ray telescope effort in the US is focused on a tightly nested, thin glass, segmented mirror design. Fabrication of the glass segments with the required surface accuracy is a fundamental challenge; equally challenging will be the alignment of the ~7000 secondary mirror segments with their corresponding primary mirrors, and co-alignment of the mirror pairs. We have developed a system to perform this alignment using a combination of a coordinate measuring machine (CMM) and a double-pass Hartmann test alignment system. We discuss the technique, its ability to correct low-order mirror errors, and results of a recent pair alignment including progress toward the required alignment accuracy of < 2 arcseconds, and discuss the influence of the alignment process on mirror figure. We then look forward toward its scalability to the task of building the IXO telescope.

Platinum is being explored as an alternative to the sprayed boron nitride mandrel release coating under study at GSFC for the International X-ray Observatory (IXO). Two and three inch diameter, polished (PFS) and superpolished (SPFS) fused silica flat mandrels, were used for these tests. Pt was applied to the mandrels by DC magnetron sputtering. The substrate material was 400 micron thick D263 glass, the material which has been proposed for the IXO segmented optics. These substrates were placed on the mandrels and thermally cycled with the same thermal profile being used at GSFC in the development of the BN slumping for IXO. After the thermal cycle was complete, the D263 substrates were removed; new D263 substrates were placed on the mandrels and the process was repeated. Four thermal cycles have been completed to date. After initially coating the mandrels with Pt, no further conditioning was applied to the mandrels before or during the thermal cycles. The microroughness of the mandrels and of the D263 substrates was measured before and after thermal cycling. Atomic force microscopy (AFM) and 8 keV X-ray reflectivity data are presented.

High-energy astrophysics is a relatively young scientific field, made possible by space-borne telescopes. During the half-century history of x-ray astronomy, the sensitivity of focusing x-ray telescopes-through finer angular resolution and increased effective area-has improved by a factor of a 100 million. This technological advance has enabled numerous exciting discoveries and increasingly detailed study of the high-energy universe-including accreting (stellar-mass and super-massive) black holes, accreting and isolated neutron stars, pulsar-wind nebulae, shocked plasma in supernova remnants, and hot thermal plasma in clusters of galaxies. As the largest structures in the universe, galaxy clusters constitute a unique laboratory for measuring the gravitational effects of dark matter and of dark energy. Here, we review the history of high-resolution x-ray telescopes and highlight some of the scientific results enabled by these telescopes. Next, we describe the planned next-generation x-ray-astronomy facility-the International X-ray Observatory (IXO). We conclude with an overview of a concept for the next next-generation facility-Generation X. The scientific objectives of such a mission will require very large areas (about 10000 m2) of highly-nested lightweight grazing-incidence mirrors with exceptional (about 0.1-arcsecond) angular resolution. Achieving this angular resolution with lightweight mirrors will likely require on-orbit adjustment of alignment and figure., Comment: 19 pages, 11 figures, SPIE Conference 7803 "Adaptive X-ray Optics", part of SPIE Optics+Photonics 2010, San Diego CA, 2010 August 2-5

The next large x-ray astrophysics mission launched will likely include soft x-ray spectroscopy as a primary capability. A requirement to fulfill the science goals of such a mission is a large-area x-ray telescope focusing sufficient x-ray flux to perform high-resolution spectroscopy with reasonable observing times. The IXO soft x-ray telescope effort in the US is focused on a tightly nested, thin glass, segmented mirror design. Fabrication of the glass segments with the required surface accuracy is a fundamental challenge; equally challenging will be the alignment of the ~7000 secondary mirror segments with their corresponding primary mirrors, and co-alignment of the mirror pairs. We have developed a system to perform this alignment using a combination of a coordinate measuring machine (CMM) and a double-pass Hartmann test alignment system. We discuss the technique, its ability to correct low-order mirror errors, and results of a recent pair alignment including progress toward the required alignment accuracy of < 2 arcseconds. We then look forward toward its scalability to the task of building the IXO telescope.

Generation-X is being studied as an extremely high resolution, very large area grazing incidence x-ray telescope. Under a NASA Advanced Mission Concepts Study, we have developed a technology plan designed to lead to the 0.1 arcsec (HPD) resolution adjustable optics with 50 square meters of effective area necessary to meet Generation-X requirements. We describe our plan in detail. In addition, we report on our development activities of adjustable grazing incidence optics via the fabrication of bimorph mirrors. We have successfully deposited thin-film piezo-electric material on the back surface of thin glass mirrors. We report on the electrical and mechanical properties of the bimorph mirrors. We also report on initial finite element modeling of adjustable grazing incidence mirrors; in particular, we examine the impact of how the mirrors are supported - the boundary conditions - on the deformations which can be achieved.

Generation-X will be an X-ray observatory with 50 m 2 collecting area at 1 keV and 0.1" angular resolution. A key concept to enable such a dramatic improvement in angular resolution is that the mirror figure will be adjusted on-orbit; e.g., via piezo-electric actuators deposited on the back side of very thin glass and imparting strains in a bi-morph configuration. To make local adjustments to the individual mirror shells we must employ an imaging detector far forward of the focal surface, so that rays from the individual shells can be measured as distinct rings. We simulate this process on a few representative shells via ray-traces of perfect optics, perturbed axially by low order Legendre polynomial terms. This elucidates some of the requirements for the on-orbit measurements, and on possible algorithms to perform the on-orbit adjustment with acceptably rapid convergence.

The Constellation-X Spectroscopy X-ray Telescope (SXT) is a segmented, tightly nested Wolter-I telescope with a requirement of approximately 12.5 Ž half-power diameter (HPD) for the mirro r system. The individual mirror segments are 0.4-mm thick, formed glass, making the task of mounting, alignment and bonding extremely challenging. Over the past year we have developed a series of tools to meet these challenges, the latest of which is an upgrade to the 600-meter x-ray beam line at GSFC. The new facilities allow us to perform full-aperture and sub-aperture imaging tests of mirror segment pairs to locate th e source of deformations and correlate them with our optical metrology. We present the optical metrology of the axial figure and Hartmann focus, x-ray imaging performance predictions based on analysis of the optical metrology, and both full-aperture and sub-aperture x-ray imaging performance of test mirror segment pairs at 8.05 keV. Keywords: X-ray optics, x-ray imaging, metrology, performance prediction, integration and test

Each of the four Spectroscopy X-ray Telescopes (SXT) on Constellation-X contain a mirror assembly comprised of 2600 primary and secondary mirror segments. Critical to the performance of the mirror assemblies is the alignment of secondary to primary, and alignment of mirror pairs to one another. Focus errors must be corrected in order to meet imaging error budgets. The use of segmented mirrors enables unique alignment strategies not feasible with mirror shells of a full revolution. We discuss the relative advantages and disadvantages of two Con-X alignment strategies to minimize focus errors between shells. In the first approach, the mirrors are bent azimuthally to adjust the focal length of the mirror pair. In the second approach, coma is used to compensate for the transverse focus error. We examine the limits of applicability of the two approaches, and also discuss alignment error budgets.