Ana Lacoste

Multi-dipolar plasmas for uniform processing: physics, design and performance

The scaling up of conventional distributed electron cyclotron resonance plasmas presents limitations in terms of plasma density, limited to the critical density, and of uniformity, due to the difficulty of achieving constant amplitude standing wave patterns along linear microwave applicators in the metre range. The alternative solution presented in this study is the extension of the concept of distribution from one- to two-dimensional networks of elementary plasma sources sustained at electron cyclotron resonance (ECR). With the so-called multi-dipolar plasmas, large size and uniform low-pressure plasmas are produced from a two-dimensional network of elementary, independent plasma sources sustained at ECR. Each elementary plasma source consists of a permanent magnet on which microwaves are applied via an independent coaxial line. The plasma is produced by the electrons accelerated at ECR and trapped in the dipolar magnetic field of the magnet acting as a tri-dimensional magnetron structure. Large-size uniform plasmas can be obtained by assembling as many such elementary plasma sources as necessary, without any physical or technical limitations. Examples of two-dimensional networks are described and the performances in terms of density and uniformity of such plasma sources are presented. The interesting characteristics and advantages of multi-dipolar plasmas over distributed ECR plasmas are listed and the perspectives for plasma processing emphasized.

PBII processing of dielectric layers: physical aspects limitations and experimental results

Processing of dielectric layers using a plasma-based ion implantation (PBII) technique has general implications in terms of plasma specifications and pulse characteristics. In particular, the different aspects of the processing of dielectric layers are discussed as functions of plasma density, pulse duration, and layer characteristics (thickness and permittivity). Clearly, severe limitations (true implantation energy, arcing) may appear for high-density plasmas as well as for long pulse durations when processing dielectric layers with thicknesses in the millimeter range. Typical examples of ion implantation in dielectric materials are presented, e.g. oxygen ion implantation in polymer sheets (for hydrophilic or adhesion treatments) and nitrogen implantation of polysilicon films on glass. The experimental results demonstrate the feasibility of processing dielectric layers with the PBII technique, but with severe limitations resulting from the process itself.

Determination of the EEDF by Langmuir probe diagnostics in a plasma excited at ECR above a multipolar magnetic field

In order to better understand the mechanisms of plasma production above multipolar magnetic fields via electron cyclotron resonance, the electron energy distribution function (EEDF) of an argon plasma in the magnetic field of a planar magnetron-like structure is determined by using optical emission spectroscopy and a cylindrical Langmuir probe. After a brief recall of the validity conditions for probe measurements in a magnetic field, probe characteristics generally allow the determination of the whole EEDF while emission spectroscopy can only provide the integral of the distribution function above the threshold energy of the selected optical transitions. The probe results show that the EEDF in fact appears as the sum of two Maxwellian electron populations. The first one is the population of fast electrons, accelerated at electron cyclotron resonance and which produces the plasma, and the second one corresponds to the cold, plasma electrons produced by the fast electrons. The variations in the parameters which characterize these two electron populations, i.e. density and electron temperature, as a function of the position in the multipolar magnetic field clearly demonstrate that the fast electrons remain trapped in the magnetic field close to the multipolar structure while the population of the slow, cold plasma electrons diffuses away from the magnets with a nearly constant electron temperature. The variations as a function of external parameters, gas pressure, microwave power, microwave frequency or magnetic field configuration are also discussed. In all cases, the maximum of optical emission corresponds to the region in the magnetic field where the fast electrons, accelerated at electron cyclotron resonance, are trapped and oscillate within two field lines between two mirror points in front of two adjacent poles of opposite polarity. Finally, simulation of the plasma production, as deduced from the experimental values, allows the determination of the ionization frequency of fast electrons. The results are perfectly consistent with literature data on magnetron plasmas.

High density distributed microwave plasma sources in a matrix configuration: concept, design and performance

Plasma scaling up can be achieved by distributing elementary microwave plasma sources over two or tri-dimensional networks. This concept is applied to a planar reactor comprising 4 × 3 microwave plasma sources distributed according to a square lattice matrix configuration. In each elementary plasma source, the plasma is produced at the end of a coaxial applicator implemented perpendicularly to the planar source. An argon plasma can be sustained in the medium pressure range from 7.5 to 750 Pa. The sheet of plasma thus obtained becomes uniform at a distance from the source plane of 15 mm, i.e. less than half the 40 mm lattice mesh. Using a cylindrical Langmuir probe, plasma density and electron temperature have been investigated as functions of pressure and microwave power. Results show that the plasma can reach densities between 1012 and 1013 cm−3 with a uniformity better than ±3.5%.

New trends in DECR plasma technology: applications to novel duplex treatments and process combinations with extreme plasma specifications

After a brief review of the plasma production and diffusion mechanisms above multipolar magnetic field structures, the possible means of sustaining magnetron discharges are listed. In particular, various designs of distributed electron cyclotron resonance (DECR) plasma reactors are described. At the industrial level, besides gas and pumping distribution, the control of plasma uniformity (up to square meters) is a necessary condition to obtain the desired uniformity of reactive species. Another process parameter, controlled via independent substrate biasing, is the energy distribution function of ions onto surfaces. Plasma-based ion implantation (PBII) in the 100-keV range requires large volumes (cubic meters) of very low-pressure plasma (below 10−4 torr), whereas ion bombardment in the very low energy range (a few eV) can only be obtained with quiescent, magnetic field-free plasmas exhibiting a very low electron temperature. In contrast, increasing the electron temperature of the diffusion plasma can enhance ionization of metallic vapors present in the central volume of a DECR reactor. The control of plasma parameters in magnetron-like plasmas opens new possibilities for complex treatments associating successive processes with extreme plasma specifications. As examples, thermochemical processing at low ion bombardment energy can be performed in DECR plasmas. Results on plasma nitriding of stainless steel for industrial applications are reported. More generally, plasma-assisted deposition (PAD) or PBII and deposition (PBIID), using chemical (CVD) or physical vapor deposition (PVD), can be operated with magnetron-like discharges: PACVD and PBIICVD in DECR plasmas, PAPVD and PBIIPVD in hybrid DECR magnetron reactors.

Plasma-based ion implantation: a valuable technology for the elaboration of innovative materials and nanostructured thin films

Plasma-based ion implantation (PBII), invented in 1987, can now be considered as a mature technology for thin film modification. After a brief recapitulation of the principle and physics of PBII, its advantages and disadvantages, as compared to conventional ion beam implantation, are listed and discussed. The elaboration of thin films and the modification of their functional properties by PBII have already been achieved in many fields, such as microelectronics (plasma doping/PLAD), biomaterials (surgical implants, bio- and blood-compatible materials), plastics (grafting, surface adhesion) and metallurgy (hard coatings, tribology), to name a few. The major advantages of PBII processing lie, on the one hand, in its flexibility in terms of ion implantation energy (from 0 to 100 keV) and operating conditions (plasma density, collisional or non-collisional ion sheath), and, on the other hand, in the easy transferrability of processes from the laboratory to industry. The possibility of modifying the composition and physical nature of the films, or of drastically changing their physical properties over several orders of magnitude makes this technology very attractive for the elaboration of innovative materials, including metastable materials, and the realization of micro- or nanostructures. A review of the state of the art in these domains is presented and illustrated through a few selected examples. The perspectives opened up by PBII processing, as well as its limitations, are discussed.

New trends in PBII technology: industrial perspectives and limitations

Abstract The two general specifications required for plasma-based ion implantation are low pressure large size plasmas and high voltage high current pulse generators. Due to the wide ion sheath expansion (up to a few tens of cm), large volumes of plasma are mandatory around the substrate. Multipolar discharges, which produce a peripheral ionization facing the substrate and can be easily scaled up, are well suited to PBII processing and begin to be widely used. However, hot filaments to sustain plasmas of reactive gases in multipolar magnetic field structures must be ruled out in favor of distributed electron cyclotron resonance (DECR) plasma sources. In order to produce the high voltage high current pulses necessary for PBII processing, generators using pulse transformers, where the voltage at the primary is provided by transistor switches and where the energy is stored at a low voltage level, appear particularly well-adapted to fulfill most of the PBII requirements in terms of reliability, compactness, cost and safety. At the industrial level, a very great advantage of PBII over ion beam implantation lies in achieving sequential processing in the same reactor, such as cleaning, etching and deposition prior to, during, or after the implantation process. As examples, thermochemical processing can be performed via PBII with or without external independent heating. More generally, the combination of DECR plasmas and magnetron discharges in the same reactor opens new possibilities for complex treatments such as PBII/CVD (chemical vapor deposition) in DECR plasmas or PBII/PVD (physical vapor deposition) in hybrid DECR-magnetron reactors. However, the transfer of processes from the laboratory to industry is mainly limited to very specific and low energy applications. In fact, mass production using high voltage PBII processing requires production tools still under development. Due to huge secondary electron emission and sheath thickness above 100 kV pulse voltages, large volume reactors (a few cubic meters) on one hand, high power pulse supplies (100 kV–1000 A/100 MW) on the other hand, are mandatory for the rise of PBII at the industrial scale.

Characterization of high density matrix microwave argon plasmas by laser absorption and electric probe diagnostics

Microwave plasma sources distributed on a planar matrix configuration can produce uniform bi-dimensional plasmas free from magnetic field in the 100 Pa pressure range. In argon, such uniform sheets of plasma have been obtained with ion densities in the range of 1012 to 1013 cm−3 with microwave power ranging from 0.4 to 2 kW. The electrical characterization of the plasma has been investigated using a cylindrical Langmuir probe. A first feature concerns the plasma potential that exhibits quite high values due to the large increase in the electron temperature towards the source plane where the microwave electric field is applied. Secondly, the decrease in the electron temperature observed when increasing the microwave power can be justified by the apparition of multi-step ionization mechanisms via metastable and higher excited states of Ar atom. The concentration and temperature of Ar(3P2) metastables have been measured by laser diode absorption spectroscopy. The results indicate that the presence of Ar(3P2) metastables is significant at 2 cm from the source plane with concentration and temperature values varying from 1010 to 1011 cm−3 and from 500 K to 1300 K, respectively, as functions of argon pressure and microwave power. An analytical model using a few simplifying assumptions provides a plasma picture in good agreement with the experimental results.

Compact Waveguide-Based Power Divider Feeding Independently Any Number of Coaxial Lines

The device described in this paper has been designed to enable the feeding of many individual plasma sources from a single microwave generator, providing a noninterfering and constant supply of power to each coaxial line driving these plasma sources. The power coming from the generator flows through a waveguide under standing-wave conditions provided by the presence of a conducting plane located at the waveguide end opposite that linked to the generator. Power is extracted from the waveguide, at the maximum of intensity of the E-field standing wave, by a waveguide-to-coaxial-line transition designated as a probe. One or two probes can be set at each such maximum of field intensity (and this on both sides of the waveguide wide wall), yielding a compact power divider. Each coaxial line feeds a microwave field applicator, sustaining plasma, through a matching circuit comprising a tuning means and a ferrite isolator (circulator with a matched load), the latter ensuring that whatever happens to the plasma source, the other feeding lines are not affected. The conditions required for a perfect match of the microwave generator to the power divider are elaborated and examples of actual designs are presented

Characterization of amorphous hydrogenated carbon films deposited from CO–C2H2 mixtures in a distributed ECR plasma reactor

Abstract Amorphous hydrogenated carbon (a-C:H) films have been deposited on silicon substrates at the floating potential from CO–C 2 H 2 mixtures using a distributed electron cyclotron resonance (DECR) plasma reactor. The deposition rate of films and chemical composition determined by X-ray photoelectron spectroscopy (XPS) were investigated as functions of the CO concentration in the gas phase. The concentration of CC, CO and CO bonding configurations in the films was deduced from XPS measurements. The hybridization state of carbon atoms was determined by Raman spectroscopy. Analyses by infrared absorption spectroscopy were performed to investigate the O and H bonding configurations in the films produced from CO–C 2 H 2 mixtures containing 25–75% CO. The effect of the CO concentration in the gas discharge on the characteristics of a-C:H films was analyzed and discussed in this paper.