The state-of-the-art in near-net-shape Hot Isostatically Pressed (HIPed) ultralight weight beryllium mirror substrates is demonstrated. Design and fabrication of two identical 5.75-inch diameter, bare beryllium mirrors weighing less than 100 grams are described. Dimensional accuracy is presented and dimensional stability issues are discussed. The mirror is representative of the class of bare beryllium optics required for light weight space-borne telescopes and satellites. The technology is generally applicable to other classes of optics including electroless nickel plated optics. It appears feasible to fabricate mirror substrates with enough accuracy to be electroless nickel-plated and optically finished without machining the substrate. Initial results on an enhanced method to make net-shape HIPed beryllium substrates directly from powder with reusable mandrels are presented. The cost of substrates made by this process would be 50% less than a current near-net-shape HIP method.

Scattering sites in diamond-turned electroless nickel (EN) coatings are typically the result of surface defects originating from the plating process. This paper presents a description of these defects, their origin, and how they can be minimized.

This paper compares the manufacturing effort and estimated relative cost of traditionally finished surfaces and single point, precision machined beryllium surfaces which have been subsequently polished by loose abrasive polishing techniques. The study includes O50 optical grade beryllium and a preliminary look at a new sputter deposited beryllium cladding showing both of their characteristics. The study was done in a manner which has allowed optimization of the polishing process for any specific desired, final mirror quality allowing determination of the minimum cost process. The objective of this study has been to establish both a 'baseline' of quality and costs while assisting optimization. The study co-addresses quality, effectiveness, and effort when producing low scatter optical surfaces from the single point turned beryllium surface. The results are useful in selecting minimum cost beryllium mirror manufacturing processes while mapping the latitude that is available. Optimized polishing is shown to reduce polishing times by 70 to 90 percent below that normally needed.

A method has been derived that can image radiation scattered from particles in the near field to the focal plane of an imaging sensor, using only two system level parameters. The parameters needed are the effective focal length and back focal length of the imaging system. The inherent advantages of this method are that imaging of scattered radiation and near field objects can be modelled with very little precise knowledge of the system's design, and cumbersome ray tracing techniques are not used. The method derived uses a mathematical transfer matrix to describe the system as a whole, so individual elements don't have to be modelled. Results from a sample case are presented.

The knife-edges of baffle vanes can specularly reflect out-of-field optical radiation in such a manner that it can scatter from the primary mirror into the field-of-view thereby degrading the performance of the optical sensor. Decreasing the radius of curvature of the knife-edges is an obvious technique for minimizing this effect. However, the x-ray hardness of the baffle vanes is negatively impacted by decreasing the radius of curvature of the knife-edges. Thus, there are tradeoff considerations on the radius of curvature of the baffle knife-edges. This results in a need for an accurate requirement on the radius of curvature of the baffle knife-edges. The Physitron-developed computer simulation KNIFEDGE is used to compute the required baffle knife-edge radius for exoatmospheric interceptors and surveillance sensors. A scaling expression is then derived that allows the results to be scaled to sensor designs other than those considered in the investigation.

The manner in which diffraction can increase the reflectance measured from very rough surfaces is shown by observations of diffracted light within the instrument profile of a far-IR reflectometer system. A correction to the calibration signal based on numerical integration of the diffracted part of the instrument profile is described. Diffraction correction factors as large as 2.94 have been found with small optics at long wavelength (630 microns). The effect of diffraction on diffuse reflectance measurements of a very rough perfect reflector is shown at wavelengths from 56 to 200 microns.

The Stratosphere Observatory For Infrared Astronomy, or SOFIA, is being designed at NASA's Ames Research Center as a replacement for the Kuiper Airborne Observatory (KAO). A 2.5 meter Nasmyth telescope will be mounted in a Boeing 747 SP and flown at 41,000 feet, above most of the H₂O in the Earth's atmosphere. In the original SOFIA design, the telescope is located in front of the wings, as it is in the KAO. An alternative design with the telescope placed behind the wings is being studied as part of an effort to reduce cost and weight. In this location, the emission from the engines and the hot H₂O molecules in the exhaust become significant straylight sources. The engines and exhaust radiate into the telescope cavity, and illuminate the primary and tertiary mirrors at low telescope elevation angles. The APART/PADE program was used to analyze the straylight at the SOFIA focal plane as a function of wavelength and telescope elevation angle. The emission from the engines and exhaust gas is compared to that from the Earth and the telescope itself. The secondary mirror will be chopped during most observations, allowing subtraction of a constant background signal. The importance of fluctuations in the exhaust emission is that they limit the amount of straylight that can be reduced by chopping. For the aft position of the SOFIA telescope to be acceptable from a scientific point of view, the fluctuations in the exhaust straylight must be lower than the shot noise in the telescope mirror emission. Based on the results of this analysis, the SOFIA telescope has been moved behind the wings. The degradation in performance at low elevation angles reported here has been deemed acceptable.

Baffles are placed on the front of telescopes to shade the telescope interior from solar and earth radiation that is outside the instrument's field of view. Benefits and drawbacks of specular baffles, which are being investigated as an alternative to the standard diffuse black baffle, are summarized. An overview of four specular baffles designs, the elliptical baffles, Linlor baffle, bielliptical baffle, and Lockheed-Stavroudis baffle, is given. A stray light analysis of the Lockheed-Stavroudis baffle is reported.

Bidirectional reflectance distribution function has been measured on highly polished uncoated and silver coated samples of CVD SiC in the wavelength range 0.325-10.6 microns to determine the dependence of scatter as a function of wavelength. From these data, total integrated scatter, the power spectral density as function of spatial frequency, and the root mean square surface roughness were calculated. The results indicate that the uncoated CVD-SiC scatter topographically (i.e., follow the lambda exp -4 scaling law) in the wavelength region, 0.325-1.06 micron but not in the region, 1.06-10.6 microns. At 10.6 microns, CVD-SiC exhibits unusually large surface scatter which can be significantly improved by coating CVD-SiC with a thin layer of silver.

Generally, there are two sources of diffraction to consider when modeling optical systems: diffraction due to apertures (edge diffraction) and slit diffraction (spatial filtering). Most optical design and analysis programs can handle aperture diffraction, but until recently, none could easily handle slit diffraction. This type of analysis is needed in many areas of optical instrumentation such as spectroscopy or laser systems that utilize spatial filtering. In order to compute slit diffraction, an incoming optical field must first be truncated by a very small aperture (slit, pinhole, etc.). A Fourier decomposition is then performed on this truncated field to produce new direction vectors that then become the diffracted field. At the request of Ball Aerospace Systems group (BASG), Breault Research Organization (BRO), Inc. has developed a new subroutine called DECOMPOSE for their Advanced Systems Analysis Package (ASAP) software. This paper will describe how BASG verified the performance of this subroutine and evaluated it against textbook theory to prove its viability. This work was performed in an effort to analyze a portion of the Infrared Spectrograph (IRS) for the Space Infrared Telescope Facility (SIRTF).

A technique for calculating scatter to within a few thousandths of a degree from specular is discussed. Results are given for several mirrors.

Measurement of integrated scatter near the specular beam is an excellent way to qualify optics and optical coatings that are used in imaging systems. This paper reviews an instrument design to measure Total Integrated Scatter (TIS) over the range from 0.05 degree(s) to 3.0 degree(s) from specular. The measurement scheme utilizes an integrating plate instead of an integrating sphere.

One of the major hardships with using APART is its lack of graphics that clearly picture the optical-mechanical system input to the program. The APART MOdel Development (APMOD) program was designed to overcome this problem, and at the same time make the development of complex models simple and efficient. APMOD combines a screen editor, a syntax checker, a help system, and a graphics package into one easy-to-use program. One of the unique aspects of APMOD is that it is an outstanding tool for teaching new users how to develop optical-mechanical models for stray light analysis by APART.

A method has been developed which allows optical system designers to determine the effects at the focal plane from noise generated due to contamination in a sensor's near field-of-view and deposited on system mirrors. This method is embodied in the PEARLSS code, which allows an 'end to end' simulation of contamination generation, transport, deposition, and the resulting performance degradation for spaceborne optical systems. The code is constructed in such a way as to allow trade studies over parameters such as system materials, dimensions, operating temperatures and wavebands, pointing directions, orbital locations, and ground-processing cleanliness levels. PEARLSS outputs include a 2-D map of the scattered/emitted noise at the first mirror, the BRDF there due to particle deposition, and a map of the structured noise on the focal plane of the sensor system. All of these outputs are generated as functions of time. A simple test case is run through the code to demonstrate its various capabilities.

A design review of a broadband, full reflection hemispherical Bidirectional Scatter Distribution Function (BSDF) instrument is given. Monochromator-based and Fourier Transform Spectrometer-based (FTS) sources are described. The goniometer design shown allows the measurement of full reflection hemisphere scatter with transmission measurements possible.

The GEMINI 8-meter Telescopes Project is designing twin 8-meter telescopes to be located in Hawaii and Chile. The GEMINI telescopes will have interchangeable secondary mirrors for use in visible and infrared. The APART/PADE program is being used to evaluate the effective infrared emissivity of the IR configuration plus enclosure as a function of mirror contamination at three infrared wavelengths. The goal is to design a telescope whose effective infrared emissivity is no more than 2% when the mirrors are clean.

WINDII is an imaging, field widened Michelson interferometer built by Canada and France for flight on NASA's Upper Atmosphere Research Satellite, which was launched September 12, 1991. Its primary purpose is to measure winds in the 80-300 km region of the atmosphere by measuring the Doppler shift of the airglow emissions. This paper discusses the design, testing and performance of the baffle system used for daytime observations.

Three phases of a stray light analysis are delineated which are typical of the stages that can be used to evaluate and significantly improve the stray light performance of a sensor. The application is described of the three-phase analysis to the stray light analysis of the Cassini Narrow Field of View and Wide Field of View telescopes which will be sent to Saturn. The resulting telescope redesign analyses are described.

For all of its excellent merits as a stray light analysis program, APART has not been designed to function as a traditional optical raytracing code, such as ACCOS-V, CODE-V or Synopsis. Raytracing with APART gets tricky, and difficult, when the optical model at the heart of the analysis no longer consists exclusively of conic surfaces of revolution or no longer consists of optics whose vertices are on the optical centerline. A technique is presented here that makes use of non-optical objects, which can be freely ROTATEd and SHIFTed within the last image space in the APART program 1 model. These non-optical objects are free to have curvature, to be second order aspheres (as set by the conic constant) and, unlike a true optical object, to be located anywhere with any orientation.

The cone visibility from the CCD detector array plane in the calibration channel of wide field planetary camera II (WFPC II) is analyzed, using APART, for three representative wavelengths as characterized by indices of refraction. The light pipe walls are visible from the corners of the equivalent CCD detector array when imaging with the smallest index of refraction, n = 1.375. Painting the inside of the light pipe walls will result in a decrease in their visibility.

Specular and diffuse reflectance measurements made near normal incidence of two very rough, solid aluminum surfaces are presented for the wavelength range from 2.2 to 512 microns. The diffuse measurements made at nonspecular angles by two different detectors indicate that between 33 and 201 microns the reflectance of one surface is nearly Lambertian (isotropic) with a bidirectional reflectance distribution function (BRDF) value within 32 percent of the theoretical value of (1/pi)/sr for a perfectly diffuse, perfect reflector. Photometric reflectance spectra at the specular angle show that between 6.9 and 100 microns the specular BRDF of these surfaces is within 5 percent of the theoretical value of (1/pi)/sr. At longer wavelengths of 235, 320, and 512 microns the specular reflectance rapidly departs from that of a perfectly diffuse, perfect reflector. The two samples studied have rms surface roughnesses of 44 and 60 microns. A durable metal surface with these near perfect reflectance characteristics can be advantageously used in the FIR as a black-body source, the interior surface of an integrating sphere, and most especially as an absolute calibration standard. BRDF measurements at 40 deg incidence, though still highly diffuse, show a significant departure from Lambertian reflectance.

This paper reviews a scatterometer which is capable of measuring scatter throughout most of the sphere surrounding the sample. The instrument can be configured to operate at many different laser wavelengths, or with a broadband source, at virtually any angle of incidence. Automated polarization control of both source and receiver has been accomplished, which allows calculation of incident and scattered Stokes vectors as well as the Mueller matrix associated with either reflective or transmissive samples. This paper forms the background material for the two papers that follow it in this proceeding.

This paper includes a very brief review of Stokes vectors and Mueller matrices. Matrix characteristics are reviewed for a number of different samples--i.e.: isotropic, anisotropic, diffuse, etc. Measurements are given for a number of different samples at both 633 and 1060 nm. This paper relies on hardware descriptions presented in the previous paper in this document.

This paper reviews a simple (economical) technique for measuring retro-scatter. Noise floors below 10^-5 sr^-1 have been achieved. The technique relies on a unique Stokes/Mueller method of characterizing sample polarization characteristics. Data is presented for several samples. This paper relies on material presented in the preceding two papers of this document.

Scatter measurements were performed on one sample each of ZnSe and ZnS (sometimes called 'cleartrans') in 5.0 degree intervals from 5.0 degrees to 180 degrees (an additional point, 3.7 degrees from specular was also obtained). Two detectors were available: A battery powered Si detector, used in the voltage mode and chosen to match potential IR sources and a photomultiplier tube (PMT, used in the current mode), chosen to match a visible source such as a Xe arc lamp. Two sources were available. The first was an HeNe laser operating at 6328 Angstroms with a nominal power of 15 mW. The second was a Xe arc lamp with an input of 1000 watts. The lamp was used, alone, as a broadband source or it was used with an interference filter having peak transmission of 5000 Angstroms and bandwidth of 100 Angstroms. The resulting matrix of 12 tests characterizes the relative scatter from these two materials over a significant part of the visible spectrum and demonstrates the potential difficulties in comparing tests with differently convolving source-detection systems.

Many analyses are found to be simpler in direction cosine coordinates. The analyses occasionally concern data obtained at angles far from specular. In order to avoid design error, the magnitude of the difference between transformed value and measured value must be known. Three Si discs (two uncoated and one coated with Al) were tested. All tests began with normal incidence. First, the sample was fixed and the detector was moved in 5.0 degree intervals about a vertical axis to measure reflective scatter. In the second test, the sample was simultaneously rotated about the same axis but in the opposite direction in order to yield 0.1 intervals in direction cosine coordinates. The data from the first test were transformed into direction cosine coordinates and compared. The transformed data began diverging significantly from the direct data at about 15.0 degrees. This report investigates the region of divergence.

The design of laser radars and the determination of the detectability of various vehicles require accurate data for the monostatic reflectivity of surface materials to far-field laser illumination at many different wavelengths. Monostatic bidirectional reflectance can be measured with an interferometric reflectometer using a homodyne arrangement in which the signal is developed from the Doppler shift of light reflected from a slowly moving target. Such an instrument has been successfully developed for far-field measurements of target materials. In this paper we will present recent theoretical analysis of the speckle statistics and the signal over noise ratio of the monostatic bidirectional reflectometer. Calculation is also compared with experimental results.

Commonly used methods of developing paints and evaluating their performance involve calculating the signatures of vehicles and backgrounds, this requires experimental determination of the directional and bidirectional reflectance of the surfaces involved. This paper describes the measurements required, the instruments used to make such measurements, and computer codes and techniques used for paint development and signature evaluation. Examples of bidirectional reflectance data obtained using full experimental mapping are presented. Applications of BRDF data in IR paint development are demonstrated with emphasis on validation and confirmation of computer modeling codes. Calculations of signatures using BRDF data are given using bidirectional reflectance data for two different coatings.

A general procedure is described for arriving at a correct choice of a black baffle surface in optical systems. It is argued that the selection process is invoked too late in the design of many optical systems or is given only cursory early attention. The advantages of a systems approach to black baffle surface design are briefly discussed. High-level selection issues including wavelengths and wavebands of interest, position of black surfaces in optical systems, general robustness needed for surfaces, environment of the optical system, cost, and schedule are addressed.

A stray light model of the Apache Point 2.5 meter telescope has been constructed to predict its performance from a stray light perspective. We conducted extensive modelling of the baffle and vane surfaces and of the surface scatter characteristics of reflective and refractive optical elements in order to characterize the overall scattering characteristics of the telescope system. Baffle and vane surfaces were modelled two ways. The first model used the specular surface properties and the second used the Bidirectional Reflectance Distribution Function (BRDF) of the surfaces. The lens and mirror surfaces were modelled using scatter characteristics of surfaces that are cleaned on a regular basis. A scatter analysis yielded the irradiances at the focal plane for off-axis sources from in-field to 65 degrees. An analysis of the scatter paths contributing to irradiance levels has identified the most important scatter paths. Most of the scatter paths which involve baffle surfaces can be blocked or eliminated. These options are not discussed here, but will be discussed in a second paper.

Properties of black surfaces for space and ground-based optical systems are reviewed. Tables are presented of available black anodized surfaces and black paints for optical instrumentation. Published reviews are listed which describe the properties of many of the surfaces.

A new approach to darkening surfaces for suppressing stray light in optical systems is to microscopically texture a metal surface directly with light-trapping pores and cones. This leaves a sturdy but light-absorbing surface that is ideal for controlling stray light in space-based telescopes. Enhanced raggedness is particularly needed to survive challenging conditions without generating debris or contaminants that can disable sensor components. Several ion bombardment configurations, including seeded ion beams and plasma immersion extraction, can produce these microscopic textures without the fragile interfaces characteristic of coatings and anodization. Experiments have identified key process parameters controlling feature size, spacing, and their resulting optical effects. Performance has been evaluated at arbitrary wavelengths by reflection and scattering. Both broad and narrow absorption bands can be engineered. Textured metal light absorbers are quite versatile and easily tuned to each application by simply adjusting a process parameter (typically substrate temperature). Required hardware can fit on a table top, for small telescope components, anyway. Thus, it could easily be incorporated into production lines.

Optically black baffle surfaces are essential in determining the signal-to-noise level for many optical systems. Requirements for new, advanced optical systems indicate that new, advanced materials or improvements in current industrially available materials are needed. Broadband optically black beryllium baffle surfaces have been developed and performance of black beryllium knife-edged baffle vanes demonstrated. In this paper an overview of the requirements for baffles, optical characteristics of the developed optically black beryllium baffle surfaces, and the performance of black beryllium knife-edged baffle assembly will be discussed.