PLASmas for thick highly BOron Doped DIamond : Analysis and Modelling – PLASBorDDiAM

A switching device is a key component for high power electronics commutation. A high breakdowng voltage associated with a low conduction resistance, and a fast switching capability with a low switching loss are needed for high power commutation device. Wide band gap semiconductors are the best materials for these devices, and among them, diamond is the most promising candidate. Although n-type doping is still difficult to obtain due to the lack of an efficient donor, p-type doping can now be reliably achieved using boron as an acceptor impurity, thus potentially opening the way to the fabrication of unipolar power devices, entirely made in diamond. To reach this goal, the question is how can we elaborate a single diamond crystal presenting high purity-high quality bulk diamond, vertically organised in multi-layers with variable electronic properties (i, p+, p-, for unipolar devices) ? This constitutes one of the most promising ways to fabricate the next generation of power devices to manage electrical networks and high-power systems.
The response of this question comes from the ability in mastering the H2-CH4-B2H6 plasma responsible for boron doped diamond growth. For fabricating an unipolar vertical device all in diamond, one needs to elaborate typically 100 µm thick highly doped diamond (1020 atom. cm-3) over around 1 to 10 µm thick lightly doped diamond. For obtaining such thicknesses associated with a high level of doping and a high microstuctural diamond single crystal, one needs using very energetic plasma. In such conditions, mastering the doping is still yet a challenge although moderated boron doped thin diamond films are daily made around the world. One of the bottlenecks is the formation of soot just at the plasma border, that absorb progressively an important part of the plasma energy and radiate, leading to an overheating of the quartz windows, and eventually to its break. Another one is to be able to determine the boron containing species responsible for boron incorporation in diamond.
The goal of the project is understanding high power density (50-200 Wcm-3) H2-CH4 plasma containing 0 to 10000 ppm of B2H6. The questions we want to answer are : (1) what is the chemical kinetics scheme for diborane dissociation in this kind of plasma and what is the key species for incorporation of boron into diamond ? (2) what are the aero-thermo-chemical effects onto the species fields, inside the plasma as well as its periphery? (3) what are the soot formation mechanisms and their governing parameters, including hydrodynamics ? (4) what conditions may we use to overcross actual limitations, keeping constant both diamond microstructure and purity ?
Modelling and laser based spectroscopic means will be developed in order to reach answers to these questions. The development of complete 2D/3D model able to described the microwave energy deposition into the plasma, aero-thermo-chemical phenomena, soot formation, as well as surface processes would constitute an ideal. Due to the complexity of the problem, we propose here to build tools that will bring us a better understanding of the processes, and that could be integrated all together later on in an « ideal » numerical code. The project will be dedicated (1) in developing an aero-thermo-chemical plasma code, that includes a chemical kinetics model and the main species – surface interaction basic data, that both will be established during the ANR project, and (2) to soot formation modelling. Ultra-violet (UV) to infrared (IR) laser (LIF and absorption) spectroscopic measurements will carried out in H2-CH4-B2H6 plasma.
The project associates four very complementary laboratories : LSPM (former LIMHP), GREMI, LIMSI and EM2C.