The work reported here examines the roughness evolution of several thin films including gold, copper, aluminum and silicon; all applied to fused silica substrates. The films were deposited by an ion beam sputtering process and ranged in thickness from a few hundred nanometers to several micrometers. Roughness evolution during ion milling was evaluated by exposing the thin films to an argon ion beam and measuring roughness at regular intervals until the thin film was completely removed. These results suggest that roughness evolution depends on initial film roughness as well as depth of material removed.

An approach to applying design of experiments to multi-layer optical coating processes was explored. Two small designed experiments were first performed to determine the effect of process parameters on optical properties of silicon nitride and silicon dioxide single layer films. These results were used to predict the optimum operating region for the production of multi-layer dielectric mirror. A third experiment was then performed to test the prediction. The results of these experiments showed that while a quantitative optimization of the multi- layer process based on the single layer experiments was not possible; it was possible to qualitatively determine a region of parameter space for optimum mirror performance based on the single layer experiments.

Silicon carbide (SiC) is emerging as an important ceramic material for optical applications requiring stiff, lightweight structures with good thermal conductivity. This report discusses the application of ion milling in the fabrication of SiC optical components. Ion beam milling combined with either ductile grinding or polishing provides an excellent approach to deterministic fabrication of SiC optical components. Results of recent roughness evolution studies for SiC samples prepared by several pre-ion milling fabrication processes suggest that ductile grinding and some polishing processes can be used to produce low-subsurface-damage components suitable for ion milling. Typical improvements with optical figures after ion milling have convergences on the order of 2 or 3. Overall, these experiments indicate that ion milling combined with other fabrication processes represents a viable, highly deterministic approach to producing high-precision SiC optical components.

Conventional spot tools are precise, monolithic hard tools made to manufacture specific optics. A modular spot tool is described which consists of a base sphere and individual lathe-turned pucks functioning as lens seats. It can be reconfigured for a variety of radii, thicknesses and wedges, and it may be fabricated and qualified as fit for use with equipment available to the small optics shop.

Ductile regime grinding or shear-mode grinding is an enabling technology that is finding application in the optics manufacturing industry. By definition, ductile removal occurs when the scale of machining is conducted below a material-dependent critical dimension or chip size. Much of the interest in ductile grinding resides in manufacturing glass optics, which experience brittle-ductile transition at on the order of magnitude 10 nm. Silicon carbide has a critical dimension that is on the order of 100 - 200 nm, making it a more attractive candidate for the technology. Preliminary results from an on-axis chemical vapor deposited (CVD) SiC sphere (f/14) indicate 317 angstrom surface roughness and 0.96 wave P-V figure were achieved with this manufacturing method. Surface finish, interferometric measured figure and BRDF scatter results are presented from ductile regime grinding of 40 and 75 mm flats of CVD silicon carbide, bare reaction bonded SiC (RBSiC) and uniaxial hot-pressed SiC (HPSiC).

The beryllium cryogenic off-axis telescope (BeCOAT) uses a two-mirror, non re-imaging, off- axis, Ritchey Chretian design with all-beryllium optics, structures and baffles. The purpose of this telescope is the system level demonstration of advanced manufacturing technologies for optics, optical benches, and baffle assemblies. The key issues that are addressed are single point diamond turning of beryllium optics, survivable fastening techniques, minimum beryllium utilization, and technologies leading to self-aligning, all-beryllium optical systems.

Siiij1e point diamond turning technique for optical crystals is reported in this paper. The main factors influencing the machined surface quality are discovered and the regularities of machining process are drawn. A new paraieter Rs is introduced to evaluate the surface quality of diamond turned optical crystals. Under the optiiized machining conditions, airror surface of roughness value Ra 0.006 jii is obtained on single crystal gerianiui.

An automated inspection station is a specifically designed for quality assurance inspection of multifocal intraocular lenses in the manufacturing environment. The station provides automatic measurements for all required optical parameters of the AMO ARRAY^TM multifocal IOL, which includes labeling parameters such as effective focal length, aqueous diopter power (distance power), add power (near power), and depth-of-focus (DOF) around the distance power, and optical quality parameters such as resolution efficiency, image contrast modulation (MTF), through focus response (TFR), and astigmatism.

We present here the results obtained from a Shack-Hartmann analysis of the optical quality of the primary mirror of the Telescopio Nazionale Galileo. We compare our results with those obtained from Zeiss from interferometry, finding a good agreement. The 80% encircled energy is found to be within a diameter of 0.14 arc sec. Calculations of the MTF, EE and PSF and comparison with diffraction theory show a remarkable agreement, indicating that the mirror quality is very near the diffraction limit.

This paper concerns a new, machine mounted aspheric metrology device designed to measure a broad range of figures without the use of auxiliary optics. A prototype device, based on the classical Hartmann test, called a Hartmann Optical Surface Tester (HOST) was evaluated on a single point diamond turning machine. Design, initial testing, and validation data from reference spheres, and two types of aspheres are discussed. Results of a simulation model for estimating acceptable alignment errors for the HOST on the diamond turning machine also are presented. Peak-to-valley measurement uncertainty on the test optics was found to be better than 0.08 micrometers .

The Ronchi test is performed by placing a periodic grating at or near the focus of an optical system. Like most geometric tests, the Ronchi test measures wavefront slope instead of the contour of the wavefront like conventional interferometry. A two-channel Ronchi test with a horizontal and a vertical grating can be used to reconstruct a non-symmetric test surface. This paper investigates the errors introduced by different rotational and translation misalignments of the gratings in the two-channel Ronchi test relative to one another. The Ronchigrams are analyzed with a modal technique based on Zernike polynomials. The severity of each misalignment is dependent upon the frequency of the gratings used and on the f/# of the system.

The stray light rejection properties of the Lockheed-Stavroudis reflective baffle geometry as implemented in the BeCOAT telescope are described in terms of PST and baffle efficiency. PST is dominated by scatter from diffuse baffle surfaces (aperture stop, etc.) and mirror scatter and reflective baffle edge specular reflections. Manufacturing and alignment tolerances for this system are discussed.

Results of highly sensitive phase retardance and optical extinction symmetry measurements using a precision ellipsometry apparatus are described. Quartz wave plate retardance measurement accuracy was enhanced through the use of three laser wavelengths at 1064, 632.8, and 532 nm. Optical extinction symmetry phenomena in crystal quartz, KNbO₃, Nd:YVO₄ and TiO₂ was also investigated in order to correlate variations in retardance measurements to crystal quality and orientation accuracy.

An existing 10.6 micrometers CO₂ laser bi-directional reflectance distribution function (BRDF) station was converted into a small scale off-axis rejection (OAR) facility for testing the point source transmittance (PST) of an infrared (IR) sensor cold shield at room temperature with a room temperature detector. A cold shield with a defective coating was identified; the coating was refurbished and performance remeasured. An improved performance was observed. A simple Gaussian integration technique for total transmitted energy was applied to assess performance in the sensor. Construction of a combined BRDF/OAR facility for component and telescope testing is discussed.

Aspherical surface quality testing by using a null lens in an interferometer is described. The wavefront variation of the testing system due to the manufacture change in radius of curvature of the null lens and the wavefront modification of the testing system by optimization are discussed by example. The fringe pattern obtained from the testing was shown and analyzed by using Wyko phase II software.

An interferometric system for aspheric surface testing is described. The system consists of a lateral shearing interferometer with a polarization phase shifter, CCD camera, a video frame grabber and a computer. A key element of the interferometer is a plane parallel birefringent plate which can give any amount of lateral shear. Phase shifting interferometry is used to obtain the derivative information of the wavefront under test. By integrating the derivative data, we have the wavefront shape. The system has measured surfaces with rms repeatability of better than (lambda) /20.

An interferometric optical profiler using Nomarski microscopy has been developed for measuring surface roughness and 3-D profiles. A commercial Nomarski differential interference contrast (DIC) microscope is modified to add a quarter-wave plate for phase- shifting. Quantitative slope data can be obtained by utilizing phase shifting techniques that rotate the analyzer to change the phase difference between the two orthogonally polarized beams. The slope data is integrated to generate surface height. The instrument is insensitive to vibration. It has measured surface features with a repeatability of better than 1 nm in an ordinary laboratory environment without vibration isolation.

In this paper, photosensitive polymer, AEPS resin, is used in experiment with two light sources. For each type of light source, we obtained the relationship between the light intensity and the thickness of the solidified layer, and the exposed time and the thickness of the solidified layer. From these relationships, we look for the optimal relationship among the type of light source, light intensity, exposed time and the thickness of each solidified layer for the photosensitive polymer.

An automatic spectrophotometer has been developed for spectral measurements of transmittance, reflectance and total optical losses of thin film specimens. The general system design and automation is described. The measured results of narrowband filters, ZnS films, mirrors, etc., are presented. In the visible region, the overall photometric accuracy is verified to be 0.1% and 0.2% for transmittance and reflectance; respectively. The wavelength scale is accurate to within 0.5 nm with a reproducibility of 0.1 nm.

The recording of multiplexing hologram on one holo film has been studied theoretically and verified experimentally in this paper, and it has been put forward that when multiframe holograms are recorded on one holo film, superposition and modulation of the interference fringes is the main cause to reduce the diffraction efficiency of some holograms having disperse fringes in the multiframe holograms.

User requirements and economics demand that large optics components will have to be produced to more stringent tolerances and lower costs. While manufacturing and testing technology have been improving rapidly, the challenge will continue to be how to integrate technology improvements into industrial processes. This paper describes Itek's approach to continuous process improvement in optical manufacturing. Specific examples of process innovations show how evolutionary upgrading of the process has resulted in improved performance and reduced costs.

Mirror blanks used in high-reliability optical systems for airborne and spaceborne applications have many requirements in terms of weight, stiffness and moment of inertia, as well as mounting and gravitational influences. Lightweight and ultra-lightweight mirror blank design techniques have been enhanced by recent technological developments in mirror blank fabrication and optical figuring. This paper briefly reviews traditional mirror blank design considerations in light of new fabrication technologies such as abrasive water jet machining of mirror cores and ion figuring of optical surfaces. The impact of these new technologies on mirror blank design is also discussed, as well as new design and analytical techniques using NASTRAN. Actual production data using these techniques are presented.

The continuing development of rapid fabrication methods for large optics at the Steward Observatory Mirror Lab has resulted in the completion of the WIYN 3.5 m primary mirror in only five months. The use of these methods, though rapid, also resulted in one of the best surface figures we have produced (16 nm rms), excellent microroughness (8 angstroms rms), and very smooth small-scale figure error as determined by the structure function of the surface errors. In this paper, we review the important techniques in the grinding, polishing, and testing of the mirror used to achieve these results.

Contraves Inc. is currently under contract to develop the facility and tooling for processing the Subaru 8.3-meter primary mirror. The optical polishing of an 8-meter class astronomical mirror presents several challenges. Simply the size of the substrate makes all handling and transportation difficult. The mirror substrate is presently being manufactured by Corning Incorporated in Canton, New York. The project includes the transportation from Canton to Wampum, Pennsylvania, the optical processing, and the final delivery to Mauna Kea, Hawaii. This program is based on existing technology. The polishing machine, optical test equipment, and support fixtures are proven equipment. The processes, equipment, and facilities required for the production of the Subaru 8.3-meter monolithic, concave, hyperbolic, thin meniscus mirror are presented here.

The purpose of this paper is to illustrate the rapid response of lightweight mirror blanks to small thermal perturbations and also to show how a Zernike polynomial representation of an optical surface can be helpful in analyzing the time varying deformations of the surface. In this work, a localized thermal load was introduced into the back of a lightweight mirror blank while the front surface was monitored interferometrically. Interferograms were taken every minute for the 14 minute application of the load and then continued for another 24 minutes to show the thermal relaxation of the blank.

Computer-controlled methods have been applied for figuring two telescope mirrors of 1.25 m in diameter to a very high optical quality. Figuring and testing are performed in a closed loop scheme. The polishing tool comprises several non-rotating subtools in a linear array. The subtools are equipped with force actuators to regulate dynamically the polishing forces on the basis of the mirror metrology results. Fast and continuous convergence of the mirror surface accuracy has been achieved. A discussion is presented of the applicability of the developed methods for manufacturing very large mirrors.

Using a small polisher with a specific movement to figure the optical surface under the control of a personal computer not only can one locally figure the surface precisely to improve the image quality of optical elements but also can figure the aspherical surface, beginning with the closest spherical surface, if the departure between these two surfaces is not large.

An optical test for measuring null correctors has been developed that uses a rotationally symmetric computer-generated hologram (CGH) to synthesize the wavefront that would be reflected by a perfect primary mirror. The test of a null lens is performed by measuring the CGH through the null corrector. Agreement between the null lens and the CGH indicates a high probability that both the null lens and the CGH are correct. This paper presents results that confirm the ease and accuracy of the test. Three null correctors for 3.5-m primary mirrors were measured at the Steward Observatory Mirror Lab using this test. Two of the null correctors were shown to have the correct conic constants within the measurement uncertainty of +/- 78 ppm. The test of the third null corrector was limited by gross flaws in the holograms caused by a fabrication error. A proposed solution to the fabrication problem is presented.

Testing of large aspheric concave optics to a surface quality of 1/50(lambda) rms or better requires knowledge of the test configuration error sources as the overall test configuration consists of a chain of test optics used to measure the actual optic figure. The intent of this paper is to discuss the testing and calibration techniques used on the Kodak segmented primary mirror with a reflective null compensator. Many of the calibration techniques discussed are applicable to a refractive test configuration (or for that matter, any optical test set).

The best surface testing of large optical components frequently requires large auxiliary test optics which are expensive. One example is the testing of a large primary mirror in autocollimation. A method has been developed which allows the testing of large optics using subaperture interferometric testing. Sufficient subaperture profiles are obtained to cover the entire surface. Then these subaperture profiles are combined into a unified profile of the mirror's surface.

Multiple reflections between two surfaces of a window introduce a fixed pattern error in the transmitted wavefront. In a Fizeau or Twyman-Green interferometer, this wavefront is reflected by a return flat and transmits through the window. The fixed pattern error is carried in the measurement result. This error is negligible, only if the wedge angle is so large that the interference fringes formed by the two surfaces are too dense for the detector to resolve. However, if the wedge angle is small (e.g., several arc-seconds), the phase error (pv) could be up to 0.025 fringes for most glass (n equals 1.5). By tilting both the window and the return flat properly, it is possible to cancel the effect of multiple reflections of a window.