Plasmas micro-onde distribués à conditions opératoires étendues – PLASMODIE

B-1 Context and objectives A constant challenge in plasma technology lies in the difficulty to scale up plasma sources in view of uniform plasma processing, e.g. for surface treatments in the meter range scale. A necessary condition to meet such a requirement is to develop uniform plasma sources. Conceptually, producing uniform plasmas with a single plasma source can only be achieved by applying uniform electric fields to a gas submitted to a uniform static magnetic field. Practically, this can only be achieved with DC discharges free from magnetic field (B0 = 0). With HF electric fields (with or without magnetic fields), this objective can be considered as impossible mission. Only clever palliatives may be used in order to obtain uniform plasma sources (e.g. standing waves with constant amplitude pattern). Therefore, a recently adopted method to circumvent this difficulty is to obtain the scaling up of plasma sources by assembling as many elementary and independent plasma sources as necessary on bi- or tri-dimensional networks. This concept has been successfully applied with the so-called multi-dipolar plasmas [1,2], which operate in the very low-pressure range (10-2 to 1 pascal), and matrix plasmas [3,4], which operate at higher pressures (10 to 103 pascal). In a multi-dipolar plasma reactor, the plasma produced by the elementary plasma sources is sustained by microwaves in presence of a magnetic field at electron cyclotron resonance (ECR). In this case, the coupling mode is a resonant mode that requires low collision frequencies to be efficient. On the contrary, in matrix plasmas, which are free from magnetic field, the coupling of microwaves with electrons is obtained by collisional absorption. In both cases, uniform plasmas are obtained at a distance from the elementary plasma sources of the order of the lattice mesh of the network. Therefore, large area planar or cylindrical plasma sources can be obtained without any physical limitation. Another quest in plasma technology is the increase in the flexibility for plasma processing through the extension of operating conditions, particularly in terms of the pressure range and frequency range available. An improved flexibility and a total control of plasma parameters open new possibilities to develop sequential processing by combination of elementary operations, e.g. surface cleaning, etching, PACVD and/or PAPVD deposition, plasma-based ion implantation (PBII), PBII and deposition (PBII & D). Among plasma technologies able to cover a wide range of pressure, from a few 10-1 pascal up to above atmospheric pressure, surface wave discharges (SWD) and inductive coupled plasmas (ICP) both exhibit a very high flexibility. In the case of ICP plasmas, the flexibility of plasma density is obtained via different coupling modes, i.e. capacitive coupling (E - mode) in the low density regime, and inductive coupling (H - mode and W - mode) in the high density regimes. Of course, in the very low pressure range (around 10-1 pascal), the help of a magnetic field is mandatory in order to sustain the plasma through its confinement and resonant coupling modes (ECR for SWD, helicon / W - mode for ICP). Besides their flexibility in term of pressure range available, SWD also exhibits a very high flexibility in term of frequency (and, as a consequence, of plasma density) since SWD can be sustained (nevertheless using different electric field applicators) with frequencies from 13.56 MHz up to 10 GHz or more. Although SWD and ICP can cover a wide range of pressures, they present a few limitations, such as the use of frequencies in the 10 MHz range for ICP, the impossibility to produce low density plasmas below the critical density (defined by wpe = w0, where wpe is the angular electron plasma frequency and w0 the microwave angular frequency) with SWD discharges, the lack of flexibility of the plasma source shape with SWD and ICP, or the use of dielectric walls. However, the most important disadvan.