Thermal Wave and Raman Characterization of Diamond Films

Pryor, R.
Kuo, P. K.
Wei, Lanhua
Thomas, R.
Major Professor
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue

Diamond films possess many of the attractive properties of bulk diamond such as hardness, thermal conductivity and wide band transparency. This fact plus the recent progress in making these films inexpensively[1,2] has attracted much renewed interest in using them in many different applications which include coatings to machine tools and optical components, heat sinks for high power semiconductor devices. The challenge is then placed on material characterization techniques intended to measure their electrical, optical, thermal, and elastic properties. The challenge on the measurement of thermal properties is especially acute because none of the conventional techniques are appropriate for thin films. The films are usually very thin (of the order of a few to tens of microns) and in intimate thermal contact with the substrate. Even though the diamond films are supposed to have superb thermal conductivity, their contribution to the thermal conductivity of the combined film/substrate composite may still be too small to be detectable by conventional methods. Lifting the film from its substrate and measuring its thermal properties in isolation is not sufficient, because not only is this a destructive procedure, but also it misses the main point. For many applications, it is the in situ thermal properties, together with the coupling to the substrate, which constitute the main focus of interest. The thermal wave mirage method of measuring the thermal diffusivities of solid materials[3,4] meets this challenge very well. In this method a thermal wave is launched at the surface of the sample by a periodic, focused laser beam. The amplitude and phase of the gradient of the temperature field in the surrounding area are then measured with the mirage technique. The thermal properties of the sample/substrate are then deduced by comparing measured values with theoretical model predictions.