Research

Modelling

Within the Centre there is considerable activity in the area of plasma process measurement and modelling. Plasmas have a wide variety of applications and are used in numerous industries including, semiconductor and microsystem manufacture, surface engineering of mechanical components and medical devices, environmental, lighting and energy.

Researchers in the NCPST are also investigating the success and quality of orthopaedic Plasma coatings by synthesising HA powder and examining its characteristics such as; phase composition, crystallinity, particle size and powder morphology. These powders are converted into HA Plasma coatings (applied onto femoral implants) via Plasma spraying. This complex process is also been researched within the NCPST, as prior to this practice has preceded understanding. HA coated implants that give excellent short term results can be produced using this method. The long term results are still disappointing, with implant longevity being at best only about 15 years. In order to produce higher quality HA coatings a clearer understanding of the structure-property-process relationship is necessary. Dr. Joseph Stokes and Dr. Lisa Looney are very active in this area.

Plasma processes are highly complex, non-linear, and difficult to control and characterise, requiring complex models and advanced sensors and diagnostic techniques for the most demanding applications. Within the NCPST we are primarily interested in the plasmas used in the fabrication of miniaturised electronic and photonic devices, surface coatings and plasmas used for nuclear fusion experiments.

In order to be able to understand, control and optimise these plasma processes we need first to be able to measure with a high degree of accuracy, resolution and repeatability the internal plasma parameters. These parameters can be observed optically or electrically using a variety of techniques that depend on the specific application. To understand how these measured parameters relate to the characteristics of the plasma itself and how it interacts with, for example, a substrate surface in contact with it, models of the plasma process or sub-process need to be used.

Researchers within the NCPST develop instruments and sensors for measuring internal plasma parameters, and models to help understand how the plasma behaves and how to control it.

List of Projects within programme

Current

• Production of HA for Plasma Spraying
• Plasma Deposition of Bio HA Powders
• HVOF Deposition of Functionally Graded Deposits
• Modelling HA Production
• Intel-Collaboration, Harmful Process Reflective Reduction, Intel Funded
• Plasma Surface Modification of Polymer for Brro Fabrication,
• Nano-plasma - Process Control for surface nano-roughness optimisation
• DVD Process Condition Monitor,
• PlasMAC, MiniMAC
• PlasMAC, Modelling
• Dual and High Frequency Discharge Modelling
• Langmuir Probes for Flowing Plasmas
• Probing Colliding Plasmas with time resolved imaging and spectroscopy

Proposed

• Modelling of Bio Materials using Plasma Techniques to Carry Drug Delivery Systems
• Modelling Plasma/HVOF Process using Computational Fluid Dynamics
• Neutral species temp measurement in low temp plasmas
• New numerical algorithms for system of differential equation modelling plasma sheaths

Technical Outline

To date we have been active in modelling, measuring and controlling typical low-pressure plasmas used in integrated circuit fabrication for surface etch processes and thin film deposition processes. We are striving to understand plasma tool and gas phase plasma states, plasma-surface physics and the development of surface features, and the key parameters affecting the plasma processed substrate.

Modelling and simulation is an activity of generic significance to our research programme. This activity is lead by Prof. Miles Turner whose generic approach is to use experimental data to calibrate kinetic simulations that form the basis of analytic models. We have recently purchased a parallel cluster computer that allows us to routinely carry out 2D Particle in Cell (PIC) simulations. Recently we have initiated a significant program to investigate the feasibility of real-time closed loop control of plasma processes. Under the guidance of Prof. John Ringwood at NUI Maynooth . an expert in system control engineering, we have begun to develop simple global models and design subsystem controllers for closed loop control. The above mentioned 2D PIC simulations are used to test the performance of proposed control algorithms. Real-time closed loop control for plasma assisted semiconductor manufacturing has been the subject of academic research for over a decade. However, due to process complexity and the lack of suitable real-time metrology, progress has been elusive and genuine real-time, multi-input, multi-output control of a plasma has yet to be achieved in an industrial setting.

The emphasis of our work has been on radio frequency plasmas and we have worked with a number of plasma diagnostic techniques, including Hairpin Probe, Frequency Domain Spectroscopy, Langmuir Probe, B.dot Probe, Plasma Impedance, optical emission spectrometry, cavity ring-down spectroscopy and mass spectrometry. These techniques measure tool-scale internal plasma parameters and with varying sensitivity, repeatability and practicality, and can be used as inputs to reactor models and control algorithms.

In todays semiconductor industry, variation in critical plasma process performance from one tool to another can lead to a reduction in die yield. Together with Intel we are investigating techniques for fault location and post-maintenance qualification of chamber hardware. The primary technique we are using is .frequency domain reflectometry.. Frequency domain reflectometry is well established as a technique in the power industry for distance-to-fault location and one-port cable loss measurements. Within the NCPST we have pioneered the development of this technique to complex frequency band limited systems (plasma-tools and DC plasma).

Laser processing of materials is used in manufacturing environments to produce sharper or more wear or corrosion resistant surface coatings and also for laser micromachining. Professor Saleem Hashmi, Dr. Dermot Brabazon and Dr. Joe Stokes, DCU work in these areas.

Laser-induced breakdown in materials (gases, solids, liquids) can be defined as the generation of a practically totally ionised gas (plasma) that can be observed as a glow or flash in the focal region. There are two main mechanisms for electron generation and growth. The first mechanism involves absorption of laser radiation by electrons when they collide with neutrals. If the electrons have sufficient energy, they can impact ionise the gas or solid. This can lead to cascade breakdown and the electron concentration will increase exponentially with time. There are two necessary conditions, firstly that there exists an initial electron in the focal volume, and secondly that the electrons acquire energy greater than the ionisation energy of the gas (or band gap of the solid). The second mechanism is called multiphoton ionisation (MPI), involves the simultaneous absorption by an atom or molecule of a sufficient number of photons to cause its ionisation. Both cascade and multiphoton ionisation require high laser irradiances, usually in excess of 108 W/cm2. However, breakdown of solids has been observed at irradiances as low as 106 W/cm2. If the solid is transparent to the laser radiation, apparently lower breakdown irradiances may be due to a nonlinear phenomenon called self-focusing. If enough laser energy is deposited into the material by these nonlinear excitation and absorption mechanisms, permanent damage is induced. The damage takes a void form when a laser beam is a tightly focused beam inside a transparent material.

Current work includes investigations of the process of fabricating micro-channels on the surface of silica glass by focusing a CO2 pulsed laser beam on the samples and translating them relative to the stationary 190 .m focused laser beam and a Nd:YVO4 laser focused to 10 .m. The width, depth and surface roughness of the channels were the response parameters investigated. After fabrication, the samples were analysed, and significant mathematical models for the process were developed, based on response surface methodology (RSM). The models describe the influence of the process parameters on the responses and predict responses within the limits of the variables being studied. Thermal models also allow correlation of breakdown material temperatures to processing parameters. The effects of the process variables are presented graphically using contour plots and 3D graphs.

An example of surface coating work that is performed in this group includes thermal spraying, DLC, and magnetron sputtering. Magnetron sputtering is a physical vapour deposition process (PVD). The magnetron device is used to deposit thin films on substrates. These thin films impart improved properties to the product such as improved wear resistance, sharpness, hardness, and strength. The magnetron operates by using a magnetic system on the back of a cathode which deflects electrons. Accelerated ions transfer there momentum to particles of the coating material which are then deposited on the substrate. This is achieved by applying a high voltage to a target plate (e.g. Titanium) within a vacuum space filled with an inert gas (e.g. Argon). This technique is commonly used to coat drills and cutting heads, but is also useful for manufacturing semiconductor IC.s and even food packaging. In this project, the system will be completely automated. This includes automating the vacuum and electrical parts of this system. The effectiveness of this automation will be tested with a screening test sequence which will include previously untested processing parameters.

Scientific Outputs

• Closed-loop plasma process control
• Diagnostic techniques sensitive to surface condition
• Diagnostic techniques sensitive to tool condition
• Fundamental understanding of dual frequency and high frequency discharges
• Plasma chemical dynamics
• Structure of electronegative discharges
• Fundamentals of plasma heating and EEDF formation
• Plasma dynamics near electrostatic probes

Educational Outputs

This activity feeds into a number of undergraduate, postgraduate modules and undergraduate projects. An example of this is the use of thermal modelling and experimental measurements of laser plasma production during manufacturing in Advanced Manufacturing modules at undergraduate final year level (MM461) and taught masters level (MM555).

Commercial Outputs

In the recent past, two technology companies have spun out of the activity in this area, Plasma Ireland (recently acquired by Dow Corning) and Scientific Systems (now rebranded as Straatum). More recently Lexas IT, a successful services company within the semiconductor manufacturing industry, has spun in its R&D activity into the university Invent Centre, in order to be in close proximity to researchers within the NCPST. This .spin-in. . Lexas Research, recently won the Mallin Invent award for their advanced plasma process endpoint system. Two other start-up companies are in discussions with the University with a view to establishing formal relationships and IP agreements - Impedans Ltd and Phive Ltd.

Within the past 3 years, researchers in this programme within the centre have had close working relationships with a number of blue chip companies, including Intel, Lam Research, Hewlett Packard and Oxford Instruments.