Pierre J. Roche

Refractive index and inhomogeneity of thin films

The fact that the optical characteristics of thin-film materials are generally different from those of the same materials in bulk form is well known. The differences depend very much on the conditions in which the deposition has been carried out. A good understanding of these differences, their causes, and the influence of deposition parameters is vital if we are to be able to improve coating quality. We have developed two complementary methods with the objective of deriving information on the index of refraction and its variation throughout the thickness of the film. Perceptible optical inhomogeneity is normally present and appreciable inhomogeneity is frequently present in thin films. Such inhomogeneity is usually associated with layer microstructure. The first is a postdeposition technique that makes use of measurements in air of the transmittance and reflectance of the layer under study over a wide wavelength region. The second, in contrast, makes use of in situ measurements, that is measurements made under vacuum and during the actual deposition of the layer. We shall show with the help of several examples that the two methods lead to results that are consistent and demonstrate the existence in deposited materials of an inherent variation of the index of refraction normal to the surface. The thermal sensitivity of the layer properties and their tendency to adsorb atmospheric moisture must be taken into account before the residual differences between the two techniques can be explained.

Characterizations of optical surfaces by measurement of scattering distribution

The micropolish of good quality optical surfaces can be characterized by measuring the scattered light distribution. Very often the surface defects are not isotropic but display preferred orientations that are translated into an anisotropy of the scattered light distribution. The total amount of light scattered by very high quality surfaces, coated or uncoated, scarcely exceeds a few hundred parts per million. Precise measurement of the distribution of the scattered light is always a task requiring great care and attention to detail. The apparatus is described. All the necessary degrees of freedom have been included so that the scattering may be completely analyzed. It is possible to make measurements out of the plane of incidence so that the complete spatial distribution of the scattered light can be obtained, whatever the angle of incidence of the primary beam. Thus to characterize the geometry of the system we use four fundamental parameters: the angle of incidence i, the two angles θ and ϕ that define the scattering direction, and the angle α that defines the orientation of the scattering surface in its own plane. Only two free parameters need exist because the surface roughness itself, which is the source of the scattered light, only depends on two variables. We have verified experimentally the validity of the relationships linking i, θ, ϕ, and α. In these relationships the expression for the intensity scattered in a particular direction (θ,ϕ) for an uncoated surface at angle of incidence i can be written in the form of the product of a coefficient, depending only on illumination and observation conditions, and of the 2-D Fourier transform of the autocorrelation functions of the surface roughness. Experimental measurements with uncoated surfaces of black glass have accorded with the theory. When the surfaces are coated with one or several layers the problem is more complicated, but it should be possible to derive information on the autocorrelation functions of each of the interfaces and the degree of correlation between them.

Theory and application of antiscattering single layers: antiscattering antireflection coatings

We recall the analytical expression that gives, for a rough surface illuminated at normal incidence, the light scattered in the half-space containing the specular reflection direction. Two cases are studied: the bare substrate and the substrate coated with one transparent layer. It is shown, for this specular direction, that the light scattered from a single layer can be equal to zero (perfect antiscattering) in certain conditions relative to the roughnesses of the two layer interfaces. Data calculation proves that this antiscattering effect occurs in all directions of the half-space of the reflected light. The experimental results are in good agreement with this theoretical analysis for five different dielectric materials. This study brings out most information about the grain of the material, which is responsible for the residual roughness.

Description of a scattering apparatus: application to the problems of characterization of opaque surfaces

We show how the complexity of a micropolished optical surface can be investigated in detail by measurement of the distribution of scattered light. We deal with problems of roughness anisotropy and uniformity together with cleaning problems. Experimental results concern numerous black glasses from different polishing shops and allow a determination of the polish inhomogeneity in a same glass set. After that, we present a detailed study of the apparatus function of the scatterometer, and we determine the limits of validity of our optical characterization method.

Simulation of the degradation of optical glass substrates caused by UV irradiation while coating

The radiation induced degradation of optical glass substrates during thin film deposition, which has been shown previously, is studied using simulation. Influence of UV irradiation by spectral lamps and excimer laser on fused silica and borosilicate glasses is investigated. Photothermal deflection is used to measure the increase in BK7 absorption versus time during UV irradiation. We also study means permitting a curing of UV induced absorption: first thermal annealing, and second focusing a visible laser beam onto the irradiated area. The results show that UV radiation can be responsible for an important part of substrate absorption increase while coating.

Overlapping of roughness spectra measured in macroscopic (optical) and microscopic (AFM) bandwIDths

Light scattering and Atomic Force Microscopy (AFM) are used together to analyze surface roughness in a very wide frequency bandwidth, extending from macroscopic (optical) to microscopic (AFM) scales. The two techniques are shown to be in large agreement since the roughness spectra overlap at intersection of bandwidths. A particular behavior of roughness is emphasized that permits to predict scattering at very short wavelengths. Thin film materials obtained by different techniques (IAD, Ion Plating, EB) are also investigated via a comparison of roughness spectra measured before and after coating in all bandwidths.

Front and back illumination of coated substrates for in-depth localization of absorption

We experimentally demonstrate that part of the absorptance of coated multicomponent glass substrates results from a photoinduced absorptance of the substrate during the evaporation process. This in- crease in the glass substrate absorptance is due to a solarization pro- cess by UV radiation, which is produced in the coating plant when elec- trons emitted by the electron gun interact with the material to be evaporated. Thus, we can experimentally show the dependence of the multicomponent glass substrate absorptance on the evaporation time. An original method of front and back illumination of samples is presented that enables the separation of each part of the in-depth absorptance using photothermal measurements: bulk and interface absorption of the thin film and substrate photoinduced absorption. This new method over- comes the difficulties presented by other measurement techniques and caused by the dependence of substrate and interface absorptances on the evaporation time. Absorption losses in different single-layer films are balanced.

Absorptance measurements of optical coatings: a round robin

An international round robin study was conducted on the absorption measurement of laser-quality coatings. Sets of optically coated samples were made by a reactive DC magnetron sputtering and an ion beam sputtering deposition process. The sample set included a high reflector at 514 nm and a high reflector for the near infrared (1030 to 1318 nm), single layers of silicon dioxide, tantalum pentoxide, and hafnium dioxide. For calibration purposes, a sample metalized with hafnium and an uncoated, superpolished fused silica substrate were also included. The set was sent to laboratory groups for absorptance measurement of these coatings. Whenever possible, each group was to measure a common, central area and another area specifically assigned to the respective group. Specific test protocols were also suggested in regards to the laser exposure time, power density, and surface preparation.

Effects of deposition conditions on thin film bulk and interface absorption

The Photothermal Deflection Technique is used for mapping absorption in optical coatings. We have observed a strong influence of the nature of the substrate material on measured absorptance. Different single layer films deposited in the same conditions, at the same time, on different substrates located at the same distance from the rotating axis, have very different absorptances: the films deposited on fused silica or calcium fluoride substrates absorb weakly while those deposited on glass substrates (BK7, C2036*, D2050*) can absorb five to fifty times more. This effect is observed with different coating materials: TiO2, Ta2O5, SiO2 prepared by different techniques: I.A.D., I.P. and E.B.. We interpret these phenomena in terms of film contamination by the substrate: metallic ions present in glass substrates can be responsible for enhanced absorptance. A simple model taking into account thin absorbing interface layers is developed. We determine both interface and bulk absorptance of the studied thin films.

Interpretation of measurements of both losses on guided propagation and absorption from a model of absorbing transition layers

It has been shown that a good interpretation of the absorption measurements performed from the photothermal deflection (PD) technique is given from a model including interface absorption. This model is applied to analyze the measurements of the losses on guided propagation, and a comparison with results obtained from PD technique is mae for various single-layer films. W show that the same parameters can explain satisfactorily both measurements. This study confirms the existence of absorbing transition layers close to the glass substrate and at the interfaces of the film.