|
It is now possible to deposit thin coatings of diamond onto a substrate of choice using a gas phase chemical reaction called 'Chemical Vapour Deposition' (CVD). In CVD, a carbon containing molecule, such as methane gas, is passed into a vacuum chamber along with a mixture of other gases, such as hydrogen and argon, where they are supplied with energy from a microwave generator. This energy heats the gases and causes them to break apart and diffuse to a surface to form a diamond coating. The longer the reaction continues, the thicker the diamond coating becomes. Such diamond coatings are beginning to find applications in cutting tools, wear-resistant coatings, medical implants, optics and heat spreaders. But what is missing is a deeper understanding of the chemistry behind the CVD growth process. Without this, the film properties cannot be optimised and process development becomes a matter of trial-and-error. The Bristol diamond CVD group has a great deal of experience in understanding the chemistry of the complex plasmas used to deposit diamond films. In 1994 we designed and built the first generation of an apparatus capable of measuring the concentrations of the important stable and radical species within these plasmas during diamond deposition - this is a molecular beam mass spectrometer, which can suck gas from the plasma through a sampling probe and pass it into a detector which identifies each gaseous component. We have used this apparatus very successfully to understand the processes occurring in various diamond-growing gas mixtures, and have published 10 papers based on its findings. However, this unique apparatus has a number of limitations, which became most obvious when we recently began trying to probe commercially more important microwave plasma environments. We now wish to build on these lessons and develop a new, more versatile, and much more sensitive, molecular beam mass spectrometer (MBMS) experiment. What new science will an improved 2nd generation reactor-MBMS system enable? For the first time, we will be sampling process gas at the surface of the growing diamond film in a MW reactor / which has been designed to approach industry standards. We anticipate a >1000x sensitivity gain compared to our previous apparatus, as a result of careful consideration of orifice sizes, propagation distances in the various pressure regions, much improved pumping speeds, and the incorporation of a properly-designed chopper. Just in the context of the basic CH4/H2 (and CH4/Ar/H2) gas mixtures, this should afford us the opportunity to make the first careful studies of the relative (and absolute) abundances of reactive species like C, CHx, C2, C2Hx, C3, C3Hx / in parallel, and as a function of process conditions. The plasma chemical model and thermochemical reaction mechanism that we have developed thus far, and which has been used in all of our most recent comparisons with experiment, still has a number of recognised limitations / which can be tested, validated or improved, as necessary, by experiments of this kind. We can also use the new system to probe gas mixtures that have so far eluded diagnosis, due to the pressure being too high or lack of sensitivity for low concentration species. For example we wish to study the role of B and N in the gas phase, for use in doping diamond to make electronic devices. We also want to study the role of Ar and CH4 in depositing so-called ultrananocrystalline diamond films, which are currently being suggested as candidates for biochemical sensors and other detectors.
|