Plasma deposition of nanocomposite thin films : process concept and realisation

Abstract

Recent developments in materials technology, fuelled by the growing hype surrounding nanotechnology, have given rise to a new breed of materials known as nanocomposites. Nanocomposite materials (a subgroup of hybrid materials) are formed from standard polymers impregnated with nanometre sized ceramic or inorganic particles. Bulk polymeric properties, such as toughness and low weight are retained, while the incorporation of ceramic/inorganic nano-particles adds additional functionality such as improved scratch resistance and reduced permeability and makes hybrid materials interesting for a wide variety of applications. In this thesis, the possibility to incorporate controlled porosity into nanocomposite thin films as low-k material for dielectric interlayers in semiconductor devices is successfully investigated. Wet chemical process technologies are currently the only methods used to synthesize nanocomposite materials on an industrial scale. However, high curing temperatures, multiple process steps and use of environmentally harmful solvents make wet chemical techniques unattractive for production on the scale necessary to fulfil industrial requirements. In contrast, gas phase process technologies are typically one-step solvent free processes and therefore offer an attractive alternative to wet chemical processes. In addition to these benefits, the extensive use of vacuum and gas phase processing equipment in the semiconductor processing industry make gas phase synthesis of nanocomposite films for semiconductor applications particularly attractive. The primary aim of this work has been the conception and realisation of a process for synthesizing hybrid materials using solely gas phase synthetic methods. The emphasis is on plasma-based synthesis of nanocomposites since for these processes the substrate temperature can be kept low. The main theme through the thesis has been the continued design, testing and adaptation of a reactor system conceived from reviewing literature regarding a range of materials and deposition processes. The development of the reactor concept progresses with increased understanding of the plasma physics and chemistry as obtained from applying in situ plasma diagnostics to monitor the plasma and particle formation, and from techniques related to film formation. Alongside the reactor development, the research has also investigated reactive ion etching and adhesion of hybrid coatings, chemical kinetics and the physics of dusty plasmas. The thesis begins with an extensive review of relevant literature regarding a variety of hybrid film deposition techniques and material descriptions. Results of this review prompted the design of the initial reactor concept consisting of a dual inductively coupled/capacitively coupled (ICP/CCP) plasma reactor, i.e. the initial approach chosen was to separate the production of nanoparticles spatially from the film formation process on the substrate. Due to its ability to form both hard silica and flexible silicone based materials, silicon based process chemistry was chosen as the main chemistry with which to deposit the hybrid films. Several silicon Summary 130 precursors were evaluated for producing silica and silicone materials using the ICP and CCP plasmas. Preliminary findings showed the precursor combinations O2/Tetraethoxysilane (TEOS) and 1,2-bis(trimethyl)siloxyethane (TMSE) were optimal for synthesizing the inorganic and organic film fractions respectively. Both inorganic and polymeric thin films were successfully synthesized and compared well to wet chemically synthesized films. With the early success in synthesizing hybrid layers, the more challenging task of synthesizing nanocomposite materials became the objective of the research. Unfortunately all initial attempts at synthesizing nanoparticles with the ICP were unsuccessful. An investigation into dusty plasma physics and the plasma chemistry of the O2/TEOS precursor system identified negative ion formation and trapping as being the key steps in the formation of particles in low pressure discharges. The low residence time in the ICP tube caused by the absence of confining electric fields was the primary reason why particles were not formed in this plasma. On the other hand, in the CCP, due to the presence of confining electric fields, successful generation of dust particles was demonstrated. Characterisation of these particles, both in situ using Fourier transform infrared absorption spectroscopy (FTIR) and of particles collected after the experiments using a range of physical and chemical characterization techniques, indicated the particles to be porous and silica-like. It was demonstrated that these particles could be grown, manipulated and kept for a long time above the electrode of a CCP. Combining the finding of the dusty plasma and silicon chemistry studies resulted in the design of a second dual plasma reactor consisting of two capacitive plasmas. In addition, the nanoparticle production and the film deposition were temporally separated by the use of pulsed plasma operation. This reactor was then used in combination with the O2/TEOS and TMSE precursor systems to deposit nanocomposite films with controlled porosity for low dielectric constant films in semiconductor devices. This approach proved to be successful and low-k films exhibiting dielectric constants as low as 1.82 ± 0.02 were demonstrated. On this finding a patent application has been based.