Optical thickness determination on transparent granulates

Intermediate layers with thicknesses in the nanometer range (0.5 - 100 nm) play a crucial role in several technological processes as e. g. microelectronic multilayer systems and magnetic storage. In these cases the determination of the layer thickness, which controls the physical properties of the coatings to a high degree, is, although not trivial, well established. On the other hand interlayers are applied to an increasing extent to modify the surface properties of fibrous, granular and other irregularly shaped bodies. Here the determination of the thickness of the modifying layer is much more complicated and reliable methods of thickness measurement are rare. Prominent examples in this respect are copper diamond composites with their extremely high thermal conductivities of more than 600 W/mK in combination with a Coefficient of Thermal Expansion (CTE) which can be matched to the CTE of commonly used electronic materials (4 - 8 ppm/K). The weak point of this material is the Thermal Contact Resistance (TCR) at the interface between the Copper matrix and the diamond particles dispersed therein. A method to influence the TCR is the introduction of an intermediate layer which improves the thermal transfer between the constituents. The thickness of this layer is of crucial importance. It has to be thick enough to significantly alter the physical properties of the interface and thin enough not to act as a contamination of the composite. These requirements lead to interlayer thicknesses in the range of 100 nm and below which means that also metallic layers become optically transparent. It is the aim of this project to exploit the fact that both, the granulate (diamond particles) and the surface modification layer are transparent to electromagnetic radiation in the vilsible range. The aim of the project is to develop a fast, cost effective method to determine film thicknesses on a large number of diamond particles with high spatial resolution. The optical density of the coated diamond particles is dependent on the thickness of the surface modification layer and can be quantifued with high spatial resolution (as low as 8 m) with a high quality optical scanner. Starting from existing pattern recognition techniques for the characterization of granular and fibrous materials algorithms shall be developed which not only yield information on the coating thickness on single diamond grains but also on statistically significant ensembles of grains which allow for the determination of e. g. coating uniformity. Based on these results also the thickness of transparent coatings on spherical and fibrous transparent particles will be investigated for statistically significant particle ensembles.

Intermediate layers with thicknesses in the nanometer range (0.5 - 100 nm) play a crucial role in several technological processes as e. g. microelectronic multilayer systems and magnetic storage. In these cases the determination of the layer thickness, which controls the physical properties of the coatings to a high degree, is, although not trivial, well established. On the other hand interlayers are applied to an increasing extent to modify the surface properties of fibrous, granular and other irregularly shaped bodies. Here the determination of the thickness of the modifying layer is much more complicated and reliable methods of thickness measurement are rare. If the coated granulate is used in a composite material, the interlayer has to be thick enough to significantly alter the physical properties of the interface and thin enough not to act as a contamination of the composite. This often means thicknesses in the range of 100 nm and below which renders also metallic layers optically transparent.It was the aim of this project to exploit the fact that several granulates (e. g. diamond particles or hollow glass microspheres), as well as the coatings thereon are transparent in the visible range. By using a high resolution optical scanner, the thickness of metallic layers on the above-mentioned particle classes could be quantified. Starting from existing pattern recognition algorithms it was possible to quantify the coating thickness on single particles. Also thickness distributions on large ensembles of particles could be assessed, which e. g. allows for the evaluation of different coating techniques in respect to thickness uniformity. To investigate also smaller particles the measurement technique was adapted for an optical microscope, so that metallic coatings on particles of a minimum diameter of approx. 5 m can be quantitatively determined. As a by-product of the quantitative assessment of the film thickness calibration methods were developed which allow for the quantification of the sensitivity of the CCD-sensors of all involved optical data acquisition devices (e. g. scanner or digital camera). For this task special wedge shaped Fabry-Perot interferometers are used. With these calibration data and with the knowledge of the dielectric function of the granulate material and the coating the thickness of absorbing (i. e. metallic) coatings can quantitatively be assessed. Unfortunately this is not possible for transparent dielectric coatings because thickness and optical transmission cannot be correlated unambiguously. The software package "DiamondView" developed within the project is now successfully used for the thickness determination on several types of granulates and an industrial application (e. g. for coated diamond particles) is possible.