The antireflection coatings which are used on lenses and other optical interfaces can have a profound effect on the image quality formed by an optical system. This paper evaluates the effects on optical performance of s- and p-phase shifts due to coatings. Coating design and lens design are usually treated as separate tasks, since the coating-induced wavefront aberrations are considered insignificant. As demands on optical systems increase (NA > 0.9, wavelength < 400nm, strehl ratio > 95%), it is shown, that coating induced-wavefront aberrations can be the dominating aberration within an optical system. The wavefront aberrations induced by different antireflection coatings on an optical imaging system are numerically evaluated as an example. It is concluded that coating design and optical design have to be optimized in one task to achieve the best optical performance.

A combination of ellipsometric and photometric measurements provides a convenient and accurate method for the determination of the optical properties of thick polymer films. Of course, a reasonably good surface and layer quality is necessary. Multiple reflections in the thick film and thus incoherent superposition causes partial depolarization of the reflected or transmitted light. Therefore, the Mueller matrix elements of these layers are measured and compared to calculated ones. A relationship between Jones and Mueller matrices originally given for random media is used to derive the formulas for the Mueller matrix elements of these anisotropic thick films. This relationship can also be applied to other incoherent effects in spectroscopic ellipsometry, such as depolarization upon reflection by a sample with a varying film thickness or by a layer with fluctuating optical constants.

A relationship between Jones and Mueller matrices originally derived for random media is applied to incoherence effects in photometric ellipsometry. Such effects are for example depolarization after reflection from a sample with varying film thickness or from a layer which is thicker than the coherence length of the incident light. The main task is to calculate the expectation value of a statistical ensemble. For the important case of thick layers, this expectation value is derived in a symbolic form. Results calculated with this method for transmission ellipsometry, ellipsometry at the back surface of the substrate and the determination of the optical constants at the substrate—layer interface are compared with measurements.

Reflections from the back surface of a transparent substrate influence the evaluation of optical constants of thin films from ellipsometric measurements. If the thickness of the substrate is large compared with the coherence length of the light, the relative phase between the p and s mode, which commonly is measured by ellipsometry, cannot be defined properly. We show how the reflections from the back surface of the substrate are taken into account in ellipsometric measurements by calculating the intensities of reflections for arbitrary angles of polarization. Applications of the new method, such as transmittance ellipsometry, ellipsometry at the back surface of the substrate, and the determination of the optical constants at the substrate-layer interface, are compared with measurements.

The optical constants of thin sputtered molybdenum layers, embedded in a ceramic-metal composite produced by a batch sputtering deposition system were analyzed. This was accomplished by assuming a multilayer system for a tin oxide-molybdenum cermet and calculating the optical constants from angular and polarization dependent reflection and transmission spectra. These optical constants differ strongly from those for sputtered bulk material obtained ellipsometrically. A good agreement between measured effective refractive indices for cermets and effective medium calculation was found, if these optical constants for molybdenum were used in the effective medium calculations. Differences to the optical constants of the cermet determined ellipsometrically were explained by the birefringence of the cermet. The size and the shape of the embedded particles were investigated with an atomic force microscope.

An apparatus was built to measure the spectral hemispherical reflectance and transmittance at variable angles of incidence. The apparatus consists of a Fourier-transform spectrometer, polarizers, and two integrating spheres. With one of the spheres, transmittance measurements can be performed, with the other, reflectance or absorptance measurements depending on the transmittance of the samples. The measurement range for transmittance measurements is from 400 nm to 2500 nm and for reflectance or absorptance measurements from 400 nm to 1900 nm. The design of the spheres and the optical set-up is described. The measurement accuracy was determined by measuring well-defined samples. Examples of measurements of different solar selective absorber coatings and anti-reflection layers are shown.

The optical constants of thin molybdenum layers and molybdenum/tin oxide cermets, produced by a batch sputtering deposition system, were determined by spectroscopic ellipsometry. The geometrical structure of the layers and the embedded particles was analysed with an atomic force microscope. The results were used for the investigation of a suitable effective medium theory to describe the optical constants of cermets with relatively high filling factors.