Space and military applications of optical systems require proof of performance of components at acceptance of contract and an active QA system in use throughout manufacture. To support these and other requirements the National Physical Laboratory provides a range of IR spectrophotometric reference standards that are individually calibrated to a known uncertainty of absolute measurement under conditions that match the customer's needs. These standards of transmittance, of regular reflectance, of hemispherical reflectance and of wavenumber are necessary because of the significant errors that may occur in IR spectrophotometers. The nature and causes of the errors are discussed, and why they are worse in IR instruments than in UV/visible ones. The NPL techniques, equipment and special modifications are described. Stable samples submitted by customers are also calibrated, and in some cases these have the advantage of giving an advantageous like-with-like measuring situation for the customer.

Choosing tolerances for parameters that affect optical performance presents difficulty. Cost of the finished product is proposed as the only selection criterion and evaluation of cost requires an estimate of yield. A methodology is outlined for interpreting tolerances statistically for use in a monte carlo ray tracing program in order to estimate performance variability and hence yield. The method has been used to achieve high production yield but the loop through cost has not yet been completed. Feedback from production has been used with the aim of refining both the tolerance modeling techniques and the production processes.

In designing and building an optical system for use in space, it is essential to have a high degree of confidence that the final instrument will perform within specification. Projects such as lens systems for micro-satellites may have relatively tight schedules for design and construction, and similarly restricted budgets. These factors will not permit the construction of prototype and development models and so the only way to have confidence in the final performance of the system is to construct a thorough tolerance budget at an appropriate level of detail. The system performance has to be assessed for sensitivity to both tolerances relating to manufacture of optical components and tolerances relating to the mechanical mounting methods. Practical considerations and cost then drive how these tolerances are distributed to create a budget that will deliver the specified performance. Testing methods are chosen to confirm that this performance has been achieved.

In recent years we have used several kinds of aspherical lenses in our minute optical system. To assure the lens quality, establishing definite evaluation technique of the aspherical lenses has been required as one of the most important matters. For that purpose, we developed a technique using the Mach-Zehnder interferometer. Mach- Zehnder interferometry has the capability of estimating the total characteristic of an aspherical lens in almost the same situation as the lens is under practical use. However it has difficulty in finding out that what kinds of manufacturing errors exist and how much the errors are. Our method concerns these problems and solves by the following: (1) obtain differential coefficients of Zernike coefficients for the aspherical test sample lens by the computer simulation, giving small manufacturing errors; and (2) separate quantitatively each manufacturing error by solving the first-order-simultaneous-equation constructed with the above differential coefficients and the measured values of the Zernike coefficients.

We present a new technique that provides a functional response similar--but not identical--to the point spread function which can be used for (i) characterizing the apodization conditions of a high numerical aperture (NA) lends, or (ii) imaging with an improved resolution over confocal. This measurement technique employs a spatial autocorrelation of the focal field, absorption and subsequent fluorescence in a sample containing a suitable fluorophore. The experimental results presented show the sensitivity of the new technique to different apodization conditions of the lens. Theoretically the experiments are modeled from the first Rayleigh-Sommerfeld integral of scalar diffraction theory in the Kirchhoff approximation. Without any fitting parameters, the calculated curves show excellent agreement with the experimental results. Based on the same principle, the technique can be used to obtain improved resolution over that obtainable in confocal microscopy. The imaging method is presented and the imaging resolution is analyzed for two model objects. The method employs only one high NA lens for the potentially robust common path delivery of the autocorrelated excitation beams.

There are several problems is using MTF to measure the performance of visual afocal sights. Earlier research by one of the authors has shown that a good correlation exists between a subjective assessment of performance and the Strehl intensity ratio of visual sights. The paper outlines some of the advantages in using this as a criterion of image quality and describes equipment for measuring the line Strehl ratio of binoculars both on and off axis.

This paper describes a method for 'stitching' multiple overlapping interferometric measurements of the equator of a high quality sphere to produce a single profile representing the roundness of the ball. The resulting optical profile measurement is compared with one obtained using a mechanical measurement and it is demonstrated that very good agreement exists between the two independent measurements.

Realization of diffraction limited multi-element optical system to yield the best image quality, calls for tight fabricational tolerance on vital optical parameters such as radius of curvature, wedge, optomechanical centering etc., and state of art technologies in terms of fabrication, assembly and characterization at element as well as at system level. It was seen that, though the vital parameters are measured to accuracies of +/- 5 micrometers, < 10 arc sec, < arc sec respectively at integrated level, the MTF, the nodal parameter for assuring the system performance used to degrade even up to 20 percent. A systematic analysis at coupling the optical parameters to performance indicates that there exists an optimized plane specific to a system which can be arrived at only through global optimization of axial separation, relative elemental planes and tilts etc. In this paper, the details of a novel method to arrive at the global optimization of optical systems experimented is presented. It was seen that through this, realization of MTF close to the diffraction limit could be achieved. The methodology has successfully been adopted for various lens assemblies of f/6 and f/4.5 working in visible, near infrared and short wave infrared regions and covering a field of 26.0 degrees. Besides optimizing the MTF, it helps in relative matching of image format of different lens assemblies.

The requirements on the homogeneity of optical glasses have increased significantly. Especially microlithography lens applications of extreme resolution and imaging quality have driven this development. In the past global specification of the glass homogeneity by P-V-limits was common. Now there are more specific and higher requirements on definite terms of the Zernike wavefront expansion. This new challenge initiated significant improvements at the Schott 20 inch aperture Fizeau type interferometer. While exploiting the advantages of the Carl Zeiss direct measuring type interferometer we emphasized on the reduction of background noise and systematic error constributions. Together with the increase in specifications the demand for qualified glass discs has risen sharply. This led us to the introduction of a semi-automated systems for the sample inspection at the interferometer. We have performed a series of reproduction measurements and detailed time-resolved analyses for the interesting homogeneity characteristics. The results show significant advances which have been reached both in increasing the measurement accuracy and in the production process for optical glasses of extreme homogeneity.

The thermo-optic coefficient (delta) n divided by (delta) T of CaF₂, ZnS and Ge single crystals have been measured in the infrared from 20 degrees C to 100 degrees C. The laser interferometric method employed allows a determination of (delta) n divided by (delta) T with an accuracy close to 10^-6K^-1 in the case of nonabsorbing materials. For Ge the uncertainty is increased by a factor of 3 and is mainly due to its increasing absorption coefficient with temperature. The behavior of ZnS was examined at 1.06 micrometers and 10.6 micrometers laser radiations; CaF₂ and Ge were investigated respectively at 1.06 micrometers and 10.6 micrometers.

The main results obtained with optical devices using coherence multiplexing are described. The demodulation methods for interferometers illuminated by low coherence sources are studied. The applications of coherence multiplexing in the field of fiber-optics sensors and telecommunications together with the future developments are also presented.

Absolute optical flat testing using the classical three flat method poses repositioning problems and basically only yields one absolute diameter. Although not giving access tot he mean curvature, the method we propose only requires relative displacements and yields a 2D map of the departure of the flat to the best sphere. Simulations and experimental data have been achieved up to a 301 by 301 sampling, showing a repeatability in the neighborhood of 1 nm P-V.

An engineering of the Z-scan technique is proposed for a direct measurement of the nonlinear variation of the refractive index of optical materials. The proposed set-up is fast and it gives a real-time control of the experiment. The configuration is based on a fast oscillation of the sample across the focal point of a lens by which the whole experimental signal can be visualized at once on an oscilloscope. Direct digitalization of the oscillograms could increase the experimental sensitivity and gives an on- line control of the noise. In this way both optical and electrical noises can be identified and reduced by suitable filtering.

Synthetic quartz glass surfaces have been examined after consecutive stages of the manufacturing process: cutting, flat honing, and polishing. Results are presented of microtopography investigations performed using several optical and non-optical methods such as optical and mechanical profilometry, atomic force microscopy, and integrated light scattering. Rms-roughness values have been determined for each step of the surface finishing process. Subsurface damage effects have been imaged by scanning electron microscopy.

We report on the development of a laser calorimeter with sub ppm sensitivity. Besides the adaptation of the sample holder for low loss measurements, different ways of evaluating calorimetric data and methods for the detection and compensation of scatter irradiation effects as well as other signal distortions are discussed. As laser sources, an AR- laser and a Nd-YAG laser with frequency doubling option are employed for the measurements. The system is used for the characterization of different types of KTP crystals for frequency conversion. The results show a distinct polarization dependency of the absorption, throughout all crystals investigated.

The pyroelectric effect exhibited by two nonlinear optical materials is used to measure the residual absorption in their transparency window. The polarization behavior of potassium titanyl phosphate KTiOPO₄ and thallium arsenic selenide Tl₃AlSe₃ are first described to determine the value of their pyroelectric coefficient. The pyrospectroscopic technique is then applied for absorption coefficient measurements at discrete laser wavelengths with a relative uncertainty estimated around 10 percent.

There are various modifications of conventional schemes of interferometers for the different measurements. We present some modification of known types of interferometers for testing small angular deviations during the geodetic works.

An optical method for measuring carrier concentration and mobility in n-type semiconductors is presented. The method depends on an accurate knowledge of the free electron effective mass,(M), as a function of temperature and dopant concentration. These values are not generally available in the literature. Measured values for M in HgCdTe, for example, are rare and, according to several authors, significantly higher than calculated values in the literature. M for n-type HgCdTe, InSb, GaAs, and Si were measured using Faraday rotation and van der Pauw testing over the range 77 K < T < 296 K. Good agreement was obtained with available published values of M in InSb and GaAs. M in HgCdTe was found to be on the order of 90 percent higher than calculated values at 296 K. The specific results have ben sent to the open literature for publication and consequently cannot be included for republication here. Faraday rotation and absorption can be combined with a constant, C, to yield carrier mobility. C was measured in the above materials at selected wavelengths. With M and C for a material known, carrier concentration and mobility can be determined from Faraday rotation and absorption measurements alone.

Low frequency microechography was a nondestructive technique, which could open a perspective to study photoimplementing optical properties in polymerizable materials. Indeed, this technique is particularly suitable for the characterization of crosslinking gradients in photopolymers and can visualize the local variation of mechanical properties. Two sets of samples were studied. The former which had been polymerized through a two level amplitude mask allowed to determine the resolution of the apparatus; the latter which were insulated through a apodizing filter with a radial density variation (in order to fabricate GRIN material) was an example of the study of mask pattern transfer into a polymer.

Basic properties of multiple-beam reflecting interferometers are described. The design principles of interferometer's mirrors are described, affecting the shape of response function. Special attention is paid to obtaining transmission-like fringes in reflected light and to minimizing frequency dispersion.

The power spectral density (PSD) is the most appropriate way to characterize the microroughness of a surface with a spatial wavelength range from several centimeters to a few hundred nanometers. Novel concepts to specify optical surfaces using the PSD are discussed. The results are applied to measurements obtained by a n optical heterodyne profilometer, by a mechanical profiler nanostep, and by an atomic force microscope. It is demonstrated that this method could yield a bandwidth independent model for the description of surface parameters.

Optical scatterometry, defined as the characterization of surfaces via diffracted light analysis, has been shown to be an attractive tool for the metrology of microlithographic structures. To tackle the inverse scattering problem, advanced data analysis schemes have been developed. This paper illustrates the application of light scattering to characterize developed resist lines in terms of multi- parameter measurements. Additionally, the depth prediction and the width prediction of special silicon concentration profiles, embedded in a plane resist layer, are reported. Substantial accuracy gains have been achieved by using partial least squares (PLS) regression along with quasi- nonlinear data preparation techniques, including range splitting or enhanced quadratic and cubic approaches. Moreover, a combination of PLS and minimum mean square error methods enables rapid and nearly arbitrarily accurate measurements.

The surface profile and the corresponding power spectral density function achieved by optical or mechanical profilometry operating in xz domain are always affected by probe dimensions and data digitization of the instruments. For mechanical profilometers there are typical changes in surface topography caused by system parameters like stylus radius, data digitalization and filtering. In this presentation the dependence of surface topography and statistical surface parameters on tip radius and data digitization is investigated by using a new numerical simulation model. In contrast to existing models the simulation algorithm presented in this paper determines the surface profile convoluted with the probe profile for any selection of surface and probe. For a sinusoidal and a fractal-like surface the distortion of the real profile by the probe geometry and the data digitalization is investigated and statistical surface parameters like root- mean-square-roughness R_q and correlation length l_c are calculated. The statistical surface data achieved from the simulation procedure are compared with experimental data measured by a mechanical profile using different tips.

Stray light in optical systems is usually assessed by techniques which are essentially similar to those used in the design synthesis of the system. In general however the dimensionality of the actual build of the system is far higher than that of the design and the traditional 'ray tracing' assessment techniques can become unmanageable. Modern computer graphics software developed for CAD and the entertainment industry has been applied to this problem at Pilkington Optronics as part of the process of integrated optomechanical design. This paper describes the relevant features of the new software packages as applied to the analysis of stray light and presents some results from particular systems. A useful feature of the technique is the use of animation to reduce the dimensionality; these examples cannot be presented here in written form but some simplified examples are available on request on 3.5 inch diskette format.

A new stray light test facility developed by SODERN for space instruments characterization is described. The main purpose of this instrumentation is to measure the point source transmittance (PST) and non-uniformity for attenuations up to 10 degrees but the instrumentation design, capacities and performances have been sized to perform tests on a broad variety of instruments or sub- assemblies. The test and checking process to achieve a perfect instrument reliability will be discussed. The limiting factors of the bench performances, including cleanliness level, air, source and light trap scattering have been estimated and measured to determine the level of unwanted stray-light.

We investigated thin MgF₂/LaF₃- and LaF₃/MgF₂-layers on opaque glass with angle resolved light scattering to analyze the different scattering contributions. We already showed the possibility of separating the volume scattering from the interface contribution for a single MgF₂-layer on BK7. In this work we concentrated on layers with thicknesses common for multilayer mirrors in the UV. The roughness of the interfaces was included in the calculations by the power spectral density measured with the atomic force microscope. This has proven to be better than a Gaussian or exponential correlation function ansatz. In order to obtain as much information as possible we also investigated the uncoated substrates and the corresponding single layer systems. We found again that the investigated layers, which have columnar structure, showed generally non-negligible volume scattering. Additionally crosscorrelation between the different interfaces must be taken into consideration.

A total integrated scattering measurement set-up is described which enables the detection of scattered light within a spatial frequency range from 0.0041 micrometers^-1 to 4 micrometers^-1. The apparatus is based on a Coblentz sphere and is equipped with light sources in the wavelength range from the UV to IR. Examples are presented of measurements on samples with rms- roughnesses from angstroms to microns.

Heterodyne interferometry is a powerful technique for measuring very small phase variations of modulated signals which has been widely used for measuring the roughness of optical surfaces. This paper will begin with a general recall of the principles of this technique. We then propose a method based on the use of the statistical moments of the height distribution of roughness. In order to quantify the stationary of the surface roughness, methodology based on the study surface roughness of BK7, heavy glass and silicon dioxide.

In working on optical specifications for Megajoule, CEA's laser for inertial confinement fusion (ICF), we came to look closely at optical beam propagation in non-linear material. All materials are non-linear, in that their refractive index at each point is a function of the intensity of the propagating beam at that point. However, only at the high intensities encountered, for example, in ICF lasers does this behavior become noticeable. We show in this paper the effect of non-linearity on beam propagation, in a basic and graphical manner, using easily understood first principles. To this effect, we look at plane waves modulated by sinusoidal ripples. It is possible, with this simplified model, to compute the spatial frequencies of ripples that suffer high amplitude amplification, leading to optical component damage, and those that do not suffer such large amplification. We also highlight the close analogy between a sequence of laser slabs in a laser amplifier, and a sequence of layers in an optical thin film coating. This analogy helps us understand the small ripple gain function of a Megajoule amplifier.

This paper describes the use of Fourier techniques to characterize the transmitted and reflected wavefront of optical components. Specifically, a power spectral density (PSD), approach is used. High power solid-state lasers exhibit non-linear amplification of specific spatial frequencies. Thus, specifications that limit the amplitude of these spatial frequencies are necessary in the design of these systems. Further, NIF optical components have square, rectangular or irregularly shaped apertures with major dimensions up to 800 nm. Components with non-circular apertures can not be analyzed correctly with Zernicke polynomials since these functions are an orthogonal set for circular apertures only. A more complete and powerful representation of the optical wavefront can be obtained by Fourier analysis in 1 or 2 dimensions. The PSD is obtained from the amplitude of frequency components present in the Fourier spectrum. The shape of a resultant wavefront or the focal spot of a complex multi-component laser system can be calculated and optimize using PSDs of the individual optical components which comprise the system. Surface roughness can be calculated over a range of spatial scale-lengths by integrating the PSD. FInally, since the optical transfer function of the instruments used to measure the wavefront degrades at high spatial frequencies, the PSD of an optical component is underestimated. We can correct for this error by modifying the PSD function to restore high spatial frequency information. The strengths of PSD analysis are leading us to develop optical specifications incorporating this function for the planned National Ignition Facility.

The damage threshold specifications for the National Ignition Facility will include a mixture of standard small- area tests and new large-area tests. During our studies of laser damage and conditioning processes of various materials we have found that some damage morphologies are fairly small and this damage does not grow with further illumination. This type of damage might not be detrimental to the laser performance. We should therefore assume that some damage can be allowed on the optics, but decide on a maximum damage morphology limit. A new specification of damage threshold termed 'functional damage threshold' was derived. Further correlation of damage size and type to system performance must be determined in order to use this measurement, but it is clear that it will be a large factor in the optics performance specifications. Large-area tests have verified that small area testing is not always sufficient when the optic in question has defect-initiated damage. This was evident for example on sputtered polarizer and mirror coatings where the defect density was low enough that the features could be missed by standard small-area testing. For some materials, the scale-length at which damage non- uniformities occur will effect the comparison of small-area and large-area tests. An example of this was the sub- aperture tests on KD*P crystals on the Beamlet test station. The tests verified the large-area damage threshold to be similar to that found when testing a small-area. Implying that for this KD*P material, the dominate damage mechanism is of sufficiently small scale-length that small-area testing is capable of determining the threshold. The Beamlet test station experiments also demonstrated the use of on- line laser conditioning to increase the crystals damage threshold.

A theoretical discussion on the resolution of buried air defects by using the photothermal deflection technique is presented. The thermal wave interferometry theory is described for the general case of layered materials. The principle of measurement of thermal parameter is reviewed. The analysis of buried air gap is made considering the defect like a mirror which is able to reflect partially the thermal wave giving rise to an interference effect at the solid surface. The condition of this interference is studied together with the limit to the capability to detect defects. The surprising theoretical result is that an air buried layer of thickness less than 1 percent of the air thermal wavelength is still detectable. The photothermal deflection technique is introduced as a suitable technique to investigate the thermal interference effects.

We demonstrate that part of the absorption of coated glass substrates is due to the photoinduced absorption of the substrate during the evaporation process. This phenomenon, called solarization, results from short wavelength radiation due to the interaction between electrons emitted by the electron gun and the material to be evaporated. We develop a method of front and back illumination of samples in order to separate each part of the in-depth absorption using photothermal measurements: bulk and interface absorption of the thin film, substrate absorption. This way, we can present the balance of absorption losses in different single layer films.

The development of coatings which show a good thermal conductivity plays an important role in the realization of new laser schemes like the Yb:YAG disk laser. We present results on investigations on the thermal conductivity of different thin film materials produced by conventional thermal evaporation and ion beam sputtering. With the optimized IBS-process high reflecting coatings are produced on the end faces of Yb:YAG disk lasers. A significant reduction of the thermal gradient in the laser disk was observed, which is an essential requirement for high power lasers besides the excellent optical quality of the coating.

Thermal guiding effect plays a dominant role in beam formation of diode end-pumped lasers. Thermally induced aberrations occur for high pump density resulting in beam quality degradation. Thermal distortions in such lasers were investigated in this paper both theoretically and experimentally. It was found that severe wavefront distortions are generated outside the pump volume whereas for beam confined to pump volume the aberration effects are negligible. For high pump density these distortions results in atypical caustics shape. This effect was examined in two cases: fundamental mode laser pumped by 2-W cw diode and multimode laser pumped by 10-W fiber coupled bar. Two crystals: Nd:YAG and Nd:YVO₄ were used as active media. Caustics shape and beam quality measurements, as well as an interferometric inspection of wavefront distortion were performed. Satisfactory agreement between theory and experiments was achieved.

Precise determination of the main refractive indices of biaxial anisotropic coatings is essential for the calculation of their optical performance. Measurements based on wave guiding methods have proved to yield excellent results with these layers. A comprehensive study on coatings with relatively low anisotropy, obliquely deposited under angles of 32 degrees or less, was done before, but coatings of higher structural anisotropy have not been concerned yet. With higher deposition angels, new problems arise because the microstructure deviates more and more from the model, and the column inclination angle has an increasing influence on the mathematical separation of the main refractive indices. In a theoretical study, this effect is shown to limit the precision and reliability of the method. Experimental results on HfO₂ thin films, deposited under 0 degrees, 30 degrees, 45 degrees, and 60 degrees, will be given. They were achieved using additional information about the coatings from transmission measurements between crossed polarizers and compared to results from ellipsometric and spectral photometric measurements. All measurements were performed at a wavelength of 632.8 nm.

Aspheres with no rotational symmetry offer new perspectives for the development of high power laser optics. Confronted with the dynamic range of the manufacturing process recently opened by scientists in ultra-precision engineering, interferometric form testing is having to meet the challenge posed by complex mirror shapes with high local slopes and/or stepheights exceeding the range of unambiguity which conventional interferometry can handle. For this reason, special interferometric techniques such as two wavefront interferometry, sub-Nyquist- and sub-aperture-approaches are to be combined, to overcome the above mentioned problems. But even if the refined experimental setup is capable of resolving the interferometric fringe pattern, the question as to how the data collected should be interpreted, is extremely time-consuming. The use of ray-tracing software in experimental setups with chromatic aberrations or distortions is, therefore inevitable, building a theoretical background to distinguish between system and manufacturing errors.

A set of orthonormal functions is used to expand the field of the fundamental mode of a weakly-guiding optical fiber. The parameters of the expansion are evaluated by minimizing the norm of the difference between the actual field within the fiber and the approximating one. The simple elementary form of the functions involved into the expansion allows an easy evaluation of parameters which need to correctly design an optical fiber based transmission system. For fibers with both triangular and parabolic refractive index profile, a satisfactory field approximation is obtained by retaining only a few terms into the expansion. In particular, by using only the first three even order terms, the accuracy is greater than that of the classical gaussian field approximation and is comparable with the accuracy of other approximations based on more complicate functions.

In this paper a method of shearing interferometry enabling us to perform a characterization of the basic statistical properties of non-gaussian randomly rough surfaces of solids will be described. This method is based on a statistical analysis of the form of dark fringe centers taking place in interferograms obtained by using a suitable interference microscope. By means of the method the root-mean-square values of heights and slopes, the values of autocorrelation lengths, autocorrelation coefficients of the heights and autocorrelation coefficient of the slopes of surface irregularities of glass diffusers will be determined.

This paper presents an algorithm of phase-shifting interferometry with precise phase steps. In the algorithm, the phase steps between the interferograms captured by a CCD camera are taken as unknowns and they are directly determined by fringe pattern matching with subpixel resolution. The algorithm is therefore inherently free from the phase step errors caused by air turbulence, thermal drift, mechanical vibration, miscalibration and nonlinearity of the phase shifter. The computer simulation in the 3D surface measurement has shown that fringe pattern matching is effective in phase-shifting interferometry to find the phase steps from the interferograms with equi-spaced fringe patterns. In addition, this algorithm is potentially useful in the measurement of the physical quantities that are related to optical phase changes such as length and displacement.

Diffractive optics five a means of decreasing the number of components in infrared devices, while meeting the current performances of conventional designs. The testing of such lenses needs two steps: the first one deals with MTF evaluations. The corresponding measurement tool will be briefly presented. An optical bench dedicated to diffraction efficiencies assessments for hybrid lenses is described: the main principles are addressed including the tolerance budget of such a means and experimental results are sketched out.

To evaluate the MTF performance of optical systems, the TOR option of the CODEV software from Optical Research Associates Inc. (ORA) is used for assessing MTF changes due to such tolerance aspects as decenters and tilts. In this evaluation, CODEV assumes internal coupling of the tolerances and a Rayleigh function for this tolerances' probability distribution, which assumption is well suited to serial modes. However, where one or more models are manufactured on a one-off basis, e.g. models of space instruments, this probability distribution is not realistic, in view of the independent nature and Gaussian distribution of the decenter generated on space instruments. To study this case and present the first results, AEROSPATIALE uses a new feature of the TOR option, which allows the user to modify the probability density functions.

This paper describes the design of test equipment constructed to enable the Marine Corps Logistics Base the capability to provide the required optical/IR boresight and electronic tests for rebuild and TOSH (TOW optical sight hardening) modification for the TOW (tube launched, optically tracked, wire guided) guided missile launcher optical/infrared sight. Optical modification and calibration are performed to improve system accuracy and reliability against failure, especially in a battlefield environment. Although principle concern was for the day sight, provision for the thermal night sight was also included. This design includes a large diameter optical beam, provision for thermal targets and night sight boresight alignment capability with the day sight. Primarily, a specially constructed collimator test set is discussed, implemented as the TOW 150 test and alignment station. In addition, a precision TOSH prism test system is briefly described, which is used to facilitate internal optical component boresight alignment utilizing a HeNe laser, alignment telescope and video camera. Radiometric theory is applied to the collimator geometry, which includes the modulation factor, Planck radiation function and appropriate bandpass integrated to yield effective rms irradiance values for calibration of the day sight.

This paper reports the development and production activities performed during the last six years ending with the technical acceptance of the first VLT 8.2m primary mirror.

Stability of beryllium mirrors is said to be unpredictable. Three recent mirrors demonstrate excellent stability. JPL produced a plano-concave, 0.5 m solid test mirror that was machined from a HIP'ed billet of special I-70 Be powder, polished bare and tested to 4K. It was thermally stable and had no hysteresis. The JPL ITTT 0.85 m primary mirror used similar material and processes, but with more stress relied treatments. Tests of this bare-polished, very lightweight single arch hyperboloidal mirror to 5K showed similar excellent results. The ESO VLT chopping secondary is a 1.12 m, machined lightweight, nickel plated, convex paraboloid. Similar fabrication processes are being used but with higher strength I-220 Be. In-processes testing indicates a stable mirror. The results show beryllium to be a stable and predictable mirror material.

This paper outlines recent progress in a novel active fabrication technique for severely aspheric optics. The underlying technology has been initially developed at the Optical Science Laboratory (OSL), at University College London, and has been incorporated in the OSL full-size active lap, used for polishing large astronomical optics. Optical Generics Limited (OGL) and OSL are currently collaborating to further develop the technology. A prototype OGLP-400 computer controlled polishing machine has been developed which exploits the technology for use with smaller aspheric optics up to 600mm diameter. The recent experimental results on both machines are presented, and their implications discussed.

The Centre d'Etudes de Limeil-Valenton is currently involved in a project which consists of the construction of a 2 KJ/500TW pulsed Nd:glass laser and which will be devoted to inertial confinement fusion research in France. With 240 laser beams and almost 10 000 m² in coated area required, the proposed megajoule-class laser will be the largest laser system ever built in the world. The proposed laser conceptual design necessitates 44-cm by 44-cm by 6-cm cavity-end mirrors representing more than 50-m² of coated area. These dielectric mirrors are made of quarterwave stacks of SiO₂ and ZrO₂-PVP and are prepared from colloidal suspensions. After a sustained search effort, we have prepared mirrored coatings with up to 99 percent damage thresholds ranging 8-10 J/cm² at 1053- nm wavelength and with 3-ns pulse duration were achieved. Mirrors with good coating uniformity and weak edge-effect could be produced by dip-coating and laminar-flow coating at room temperature and atmospheric pressure. To satisfy the coating demand, coating equipments and a 1000-m² sol-gel coating production plant are planned to be built at REOSC- Group SFIM company in order to start producing within schedule and cost goal.

This paper describes the results of a study carried out to measure the actual deformation of a Zerodur telescope mirror due to electron irradiation simulating the dose expected on a satellite in geostationary orbit (GEO). The purpose of the work was to provide a sound experimental basis for the predicted performance of the Silex telescope on the ARTEMIS GEO satellite. Silex is one of the data-relay payloads onboard ARTEMIS, which is an experimental satellite developed for the European Space Agency to test and operate new telecommunications services. The Silex telescope is a 250 mm diameter Cassegrain afocal telescope with an F/1.5 primary mirror which has been specified to meet very stringent long term focusing stability requirements in order to ensure a transmitted wavefront radius of curvature of 30 km or better over the ten year lifetime of the experiment. Communication will primarily be conducted between the GEO terminal and an identical one to be flown on the French SPOT 4 satellite in low earth orbit. The results presented here show that it is not possible in general to predict the behavior of a relatively large and complex Zerodur mirror by applying the results form small samples. Previously published results show that Zerodur obeys a square root lay for compaction versus electron irradiation dose. We find that the change in radius of curvature of a 250 mm diameter Zerodur mirror has a linear dependency on the dose, and that the compaction behavior is not dependent on the dose rate, but is confirmed to vary as the square root of the dose.

Nd-doped phosphate glasses are the preferred gain medium for high-peak-power lasers used for inertial confinement fusion research because they have excellent energy storage and extraction characteristics. In addition, these glasses can be manufactured defect-free in large sizes and a t relatively low cost. To meet the requirements of the future megajoule size lasers, advanced laser glass manufacturing methods are being developed that would enable laser glass to be continuously produced at the rate of several thousand large plates of glass per year. This represents more than a 10 to 100-fold improvement in the scale of the present manufacturing technology.

SFIM ODS is involved in the French MEGAJOULE Project managed by the Commissariat a l'Energie Atomique. It is a high power UV-laser using 240 square beams. SFIM ODS achieved a technical and economical analysis concerning the production of flat-parallel debris-shields in silica using large and specific dual-side fine-grinding and polishing machines. This analysis was supported by trials using half-scale windows showing the ability to reach the specifications with low production costs.

In the framework of the megajoule laser project driven by the french atomic energy board, one of the most valuable optical programs in terms of material volume as well as in terms of component size, the actual glasses and production means appear to be inconsistent with the economical objectives. Corning proposed an alternative, based on the use of a low cost glass, together with an evolution of the production process. Combining its experience in quality optical glasses manufacturing and its mastery of forming processes, Corning conducted a production cost reduction program; the objective of this program was to validate the concept of large slab melting, where blocks are cut off, versus the conventional single block melting. Economical improvements are based on a reduction of lost time and production lead-time by increasing the feeding yield, on a better glass utilization, and a reduced number of molds. The technical issues were: increase the feeding yield maintaining a given glass quality level, reduce the glass allowance, improve the materials of the molds, reinforce the thermal process control, automatism of critical operations, especially at the start-up and at the end of the mold feeding.

A fiber handling system (FHS) is described in this paper to minimize fiber damages during handling in manufacturing processes of optical components. The manufacture of an undersea telecommunication component is used as an example to demonstrate the FHS concept. However, this FHS concept can be used in any manufacturing process that involves optical fibers.

The manufacturing of beam shaping optics consisting of two or more microoptical components in a hybrid microsystem places high demands on the assembling process for positioning accuracy making a manual alignment process inadequate. The basis for the micro-assembly forms a robotic system. This robot makes it possible to handle, to grip and to join the components with very small position deviations and to apply very small drops of adhesives. The dispensing of adhesive drops of approximately 1 nl volume is an essential part of the task.

The growing demand for complex components for integrated- optic requires fabrication methods allowing good reproducibility and a minimized number of processing steps. The ion-exchange technique is attractive because it can be used to make inexpensive and versatile waveguides in glass. To ensure reliable operation, the waveguides have to be buried beneath the surface. Various methods for making buried waveguides are based on field assisted ion exchange with one or more steps in the procedure. We propose a new assembling method in which a buried optical waveguide is obtained by direct bonding of two separate waveguides. We found that direct bonding of preprocessed wafers is a useful and versatile step in the fabrication of integrated optical components. Wafer direct bonding is compatible with VLSI batch technology and device miniaturization therefore low cost combined with high performance can be achieved. Furthermore, a large variety of symmetrical or asymmetrical index profiles is possible and the method allows simultaneous fabrication of many identical optical components. Optical properties of the component are studied and the advantages of this new process are summarized and compared with other techniques.

In the conception of a planar optical backplane interconnecting comprising of a uniform power distribution, we proposed to use cascaded high frequency diffractive optical elements, each with a different diffraction efficiency in the + 1 transmitted order. We introduced a new type of binary grating: the side walls are inclined angle. The gratings were optimized with a rigorous coupled wave algorithm, Here, we investigate various etching techniques in order to realize such slanted structures in fused silica. These methods include both ion beam milling and reactive sputter etching and their variants.

Severely aspheric lenses have surfaces which depart from spherical by millimeters rather than microns. Optical testing of these lenses can be achieved using aspheric null mirrors. Although such mirrors can give full null compensation at diffraction limited precision most severely aspheric lenses are made by molding for applications which allow many fringes of form error. Diamond turned aspheric null mirrors verified by Form Taysurf methods give a test lens plus mirror systems which can be tested using either interference or a less sensitive autocollimation or encircled energy method more economical and more appropriate to the quality required.

Binary diffractive optical elements for beam diagnostics have been calculated and fabricated as reflective phase elements in copper. Our calculation combines electromagnetic theory and Fourier optics for coding phase and amplitude in the diagnostic beam. First the binary structure is fabricated in photoresist. Then a galvanic process is used to transfer these structures in copper. By including spatial filters in the design of the element a compact sensor for beam position is achieved.

A profile grinding equipped with a new mode of NC command is characteristic of being provided with a feedback sensor for detecting a displacement of table to control plunge feed and a sensor for detecting grinding load to decide selectively a sparkout time or a traverse feed rate. The new model gives a plunge feed command for motion copying and also a selective size reduction or sparkout time command similar to pressure copying. As a result, profile area with large error can be fast removed under high grinding load and for the other area, profile feed rate can be increased by shortening sparkout time selectively. Applying the new mode NC command, a high profile accuracy and also high productivity in grinding can be expected.

In the 3 to 5 micron waveband the material that gives the best ratio of optical power to density is silicon. Optics using this material would tend to be the lightest possible. Unfortunately it has high dispersion so that optical systems need achromatizing. A diffractive surface can achromatize a lens without having the disadvantages of an extra element that will have both extra mass and will lead to tight tolerances between the two halves of the doublet. Silicon is a notoriously difficult material to a machine, however recent trials have enabled this to be carried out satisfactorily on a lens of 125 mm diameter.

A software tool for the simulation of ion beam sputter deposition is presented. All calculations are performed in 3D space using the original deposition chamber geometry. Starting at the extraction grid of the ion gun the distribution of the ion beam at the target is determined. A special version of transport of ions in matter code is applied to simulate the sputtering process. Using the energy and angular distribution of all sputtered and reflected atoms, the energy deposition as well as growth rates and particle compositions at any location and orientation in the deposition chamber can be calculated. Some experimental results of growth rates and optical absorption losses of SiO₂ and TiO₂ layers are presented and are compared to the results of simulation.

SiO₂ films doped with the organic dye DO3, exhibiting high electron delocalization have been prepared by the sol- gel method and deposited on soda-lime glass substrates. On these films an electric field of 5-7 kV/cm has been successively applied in order to induce orientation of the DO3 molecules and freeze them in the condition by heating and reticulation of the inorganic matrix. The thermal stability of the films and the variation of the molecules' orientation as a function of time have been studied. The films have been characterized by means of differential thermal analysis, thermogravimetry, mass spectrometry, ellipsometry and atomic force microscopy.

The development of new optical glass polishing methods that improve the production efficiency of standard components is of key interest for the optical industry. Laser polishing is a very attractive method since it requires no mechanical abrasives or surface-adapted polishing tools. Moreover, it can be applied to surfaces of very complicated topography that are difficult or even impossible to polish using more traditional techniques and provides the possibility of automatic processing. We demonstrate rapid generation of large area polished optical surfaces by high power CO₂ laser irradiation. Results focus on glasses with high expansion coefficients conventionally used in the optical industry such as TRC-33. The technique is applied to preheated glass samples with initial rms roughness up to 500 nm. To find out the conditions for successful and reliable use of the proposed laser polishing method, the laser-driven heating process is monitored by means of the surface and depth temperature distributions. Whereas the former is determined in-situ from the IR radiation emitted by the glass surface, the latter is obtained by comparison of the IR radiation emitted by the bulk sample with a theoretical model. Laser polishing of 5000 mm² glass surfaces is reported and processes involved in the surface modification of irradiated samples are described and discussed.

Seen in the context of manufacturing machining process, the ability to rough-grind glass, an essentially brittle material, firstly to form, incurring considerable sub- surface damage, and then to free-abrasively polish it to nanometric roughness levels and micrometric form, is remarkable. Single-point studies of the fracture mechanics and machinability of glasses show that, at incursion levels of sub-micrometer order and forces of the order of grams, plastic deformation dominates material deformation processes. The implications of brittleness being scale- dependent ar considerable embracing the possibilities for single-point machining. Whereas it is difficult to experimentally observe the action of a single-particle among the many under a lap, observation and analysis of single- point machining may be usefully employed to explain lap-grit actions. Surface roughness values, damage-depths, interface presentation, limitations on inherent form-convergence and the role of statistical averaging in abrasion may be combined to provide a mechanical model for certain particulate actions.

The demands of higher precision, higher speed, higher stiffness, and new files of application determine the improvements and new developments of precision bearings. In many applications air bearing systems are often the only alternative, especially when precise movements free of vibration, are necessary. The performance of high precision bearings for scanning is primarily dependent on the design and on the individual components. Main components are the bearing, the drive system, the controller and the optical mirror unit. Runout errors and vibration excitation lead to planar and roundness errors and to an increase in image errors. It is difficult to compensate the highly dynamic errors of a bearing system. Desirable is a very stiff, reproducible and very smooth running bearing system as a basis for good scanning results. Through suitable technology and choice of materials it is possible to optimize an air bearing for a special field of use. With the application of glass and glassceramic new possibilities for the production of precision bearings are present. One such example is a precision glass bearing made of these materials with an integrated polygonal mirror for scanning based upon the well-known design principle of double spherical air bearings.

Lithium niobate substrates were processed in plasma of radio-frequency glow discharge in hydrogen and subsequently in oxygen. The modified surface layers were examined using the infrared spectroscopy and ⁶Li depth profiling. It was found that hydrogen plasma treatment enhances the OH band absorption which is explained as hydrogenation of Li- reduced samples. At the same time the sample surface is modified so that the reflectivity increases and the extinction coefficient of a wafer is substantially lowered. The Li-reduction is not higher than 50 percent, but the depth of the modified surface layer extends to about 20 micrometers. It is attributed to a relatively high sample heating caused by plasma which facilitates the Li mobility. A possibility of this method as an alternative technique to the annealed proton exchange is mentioned. An interesting reversed ⁶Li depth profile was observed, i.e. decreasing with depth. The reason of this phenomenon might be a negative surface biasing in plasma, which makes the Li⁺ ions to drift to the surface. It suggests an idea of thus produced 'buried' waveguide.

An optimized aspherical lens with large numerical aperture is demanded in many optical systems. In addressing the requirements, we have obtained analytical solutions of such a desired two-side aspherical lens by solving differential equations. Therefore geometrical parameters and index of refraction of the lens are explicitly expressed in the formulas. Lots of aspects of the aspherical lens can be seen by analyzing the obtained formulas rather than conventionally analyzing discrete data. Criteria of manufacturing the optimized lenses in different precision form uniquely depend on the refractive index of the plastics or glass material. On the basis of summarizing the work in this paper, we further derive formulas of power transmission for the optimized aspherical lens in several cases. The power transmission analysis is associated with typical radiation distribution of a source such as Lambertian or Gaussian distribution. Two polarized states of optical waves are also taken into account. Another criterion of manufacturing aspherical lenses with a specific transmittance is given according to the numerical aperture. It can be used to evaluate the power transmittance of plastic aspherical lenses without anti-reflection coating. A design according to the proper criteria allows more tolerances in the process of plastic lens manufacturing and assembling, while the analytically optimized results are more adaptive to compensation design for correcting deformation of materials.

Low temperature doping procedure is employed to make erbium doped lithium niobate substrates. The doping process is studied for X-, Y- and Z-cuts in both virgin and proton exchanged wafers. The amount as high as 9 weight percent of erbium was found in X-cuts when the doping was performed from a mixture of molten nitrates containing 10 weight percent of erbium slat after 5 hours diffusion at 400 degrees C. The content of erbium in Z- and Y-cuts as well as in the proton exchanged X-cuts was found to be much lower, below 0.5 percent. In-diffused erbium ions concentration is localized in a very thin layer, which can be extended by a subsequent annealing. Annealed proton exchanged waveguides were fabricated in the erbium doped wafers without any deterioration of the samples surfaces.

There is a trend in the optical industry to automate the edge-blackening of optical components. The goal of edge- blackening is to coat the areas outside the functional region of optical components, namely the peripheral areas including the surface edge or 'diaphragm' and apertures, with light absorbing, index-matched material to minimize scattering light. The common method today is the manual application with brush or ink-writer of quick-drying liquid pigmented synthetic resin. The lenses are fitted at a rotary table with a vacuum-pump or with considerable disadvantage inhibiting the automation. Another problem is the use of a high number of lacquers with different or partly unknown properties, due to varying chemical bases.

Surface characterization has a vital importance in many fields of research and technology. In the recent years, increasing efforts have been made in order to use light scattering rather than classical optical and mechanical tools for the purpose of characterization. After a brief recall on the classical optical means, a presentation of recent achievements in the field of characterization from surface scattering analysis will be made. Two kinds of surfaces and two kinds of characterizations will be distinguished.