Cleaning and Surface Modification Processes
Associate Professor Deb Kane, David Halfpenny, Alanna Fernandes
Department of Physics, Macquarie University.
The use of ultraviolet lasers for laser surface cleaning and modification is mushrooming internationally as the potential for generating clean surfaces without the use of environmentally hazardous solvents and the ability to create improved material characteristics and new structures, on scales from the bulk to sub-micron, is more fully appreciated in research and development of new technologies.
Currently, at Macquarie University we are pursuing three project areas.
1 UV Laser Cleaning
The motivation for the development of the new technique of laser cleaning (to be added to the large number of existing and standardised cleaning technologies ) has come from the widespread need to remove ever smaller (sub-micron) particles from substrates, to avoid the use of solvent techniques which may generate environmentally hazardous waste products (CFCs) or which may leave unwanted residues on small area substrates, to clean small particulates from multiangled surfaces in integrated optic devices and to reduce the water consumption in large scale semiconductor device manufacture.
Laser cleaning studies at Macquarie University commenced in 1995 as a solution to a technical problem of needing to clean the cleaved ends of optical fibres which have a diameter of just 100 micrometres. The technique developed by us, to date, is applicable to small area surfaces (with diameters of 70-150 mm) using high-pulse-rate, frequency-doubled, copper vapour lasers (UV-CVL) and modest powered excimer lasers. Larger areas can be studied by rastering the beam or the substrate. Excellent UV laser cleaning results have been obtained with the technique that has been developed. One hundred percent cleaning efficiency for particulate contaminants as small as 0.3 mm has been achieved using a single laser pulse in a dry laser cleaning technique. These results have challenged current thinking on laser cleaning which suggests that efficient laser cleaning occurs using multiple pulses and is best when solvent assisted. Additionally, we have pioneered new methodologies for quantifying the laser cleaning process.
(a) 
(b)
(c) 
(d)
Figure: Optical micrographs of contaminated surfaces before and after irradiation. (a) & (b) depict separate glass surfaces contaminated with, nominally, 3m m and 0.1 m m diameter particles respectively. (c) & (d) are these same surfaces after a single irradiating pulse with fluences of 180 & 160 mJ/cm2 respectively.
Our early motivation for the laser cleaning research as a technique for cleaning micro-optics remains one of our major aims. Additionally, there are enormous opportunities for developing laser cleaning for large area surface substrates. A large variety of contaminants (organic films, microparticles as small as 0.1 microns, metal flakes, oxides, dust, inks, etc) need to be cleaned from a wide range of substrates (semiconductor wafers and IC circuits, magnetic heads and other magnetic media surfaces, metals, glass, etc).
The research activity in this area also aims for a detailed understanding of mechanisms of laser cleaning, an issue which is lagging the technological developments at this time. Mechanisms which have been postulated include laser photodecomposition (single photon and multi-photon), laser ablation, surface vibration and photothermal effects, including explosive expansion of contaminants and mismatch of expansion coefficients of substrate and contaminant. The question of whether any of the mechanisms is dominant or whether several or all contribute significantly is an important one to be answered. There will be very useful and interesting developments in linking the thermal and mechanical physics theory directly with measurable quantities in the transient regimes relevant to the cleaning process. This knowledge will lead to further optimisation of the cleaning techniques by selection of the most appropriate fluence, power, irradiance and beam size for the UV laser source used, so that 100% cleaning is achieved without laser induced surface damage or microroughening.
2 Surface Modification Studies
In our research on UV laser cleaning of glass substrates the highly significant discovery has been made that UV laser irradiation also renders the glass surfaces resistant to the readhesion of particulate contaminants. This observation has been explained using results of surface science studies, carried out in collaboration with Professor Rob Lamb at UNSW, that show the silica surface is irreversibly dehydroxylated by the UV irradiation. The surface is thus changed from being a hydrated, hydrophilic surface to being a dehydroxylated, hydrophobic surface. We are the first in the world (to our knowledge) to observe and report this phenomenon. Thus, the UV laser irradiation technique has the potential to produce permanently clean glass of all sizes from large window panes to micro-optics. For example, a UV laser cleaned fused silica sample prepared more than two years ago still shows that 0.3 mm particles can be removed from it by tapping the glass lightly on a side edge. Such small particles adhere strongly to glass samples cleaned by a standard isopropanol/ethanol/ methanol solvent multiple wipe technique, used as a standard in optics laboratories around the world for cleaning high quality laser type optics.
Figure: Spatial map of silica surface showing SiOH concentration in irradiated (dark) areas, and un-irradiated (bright) areas.
Surface science studies give an ultra-sensitive measure of changes to the surface chemistry. We will continue to use these techniques, as we have in our studies to date, to establish any coarse change in elemental content of the treated surface compared to the untreated surface (Xray photoelectron spectroscopy (XPS)) and to do measurements of the relative abundances at the surface of key functional groups (Secondary Ion Mass Spectroscopy (SIMS)). Silanol and Siloxane are the species that have been relevant in our studies to date.
The future work will concentrate on identifying the parameters for UV irradiation that produce the maximum percentage of dehydroxylation for different types of glass surface, and determining the permanency of this effect for different glasses. The treatment will be applied to large areas and its potential application to window glass, as well as high quality optical glass, will be determined from the results of these studies.
3 Particle-Assisted UV Laser Ablation of Glass and Low Absorption Materials
In our studies, high magnification, high resolution (0.1 mm) images of the substrate before and after UV laser irradiation are collected. This has enabled us to start observations of the laser induced surface optical damage which is associated with the laser cleaning process. Our results to date show that the optical damage is in fact very much enhanced ablation of the glass substrate.
|
(a) |
(b) |
(c)
Figure: AFM images of damage to the surface caused by irradiation of a 0.1 m m diameter particle coated microscope slide with a fluence of 0.2 J/cm2. (a) shows a 2D view of a damage site, (b) shows a depth profile of the same area, while (c) shows an inverted AFM 3D image of this site.
While typical surface damage thresholds for excellent quality, clean fused silica for KrF laser irradiation (248 nm, 20ns pulse length, 1.5 mm beam diameter) are 7-10 J cm-2, this can be reduced to 0.1 - 2 Jcm-2 by the presence of the contaminant particles. AFM studies of the damage sites show that a pit of depths up to 1.5 mm has been ablated from the particle coated surface by a single laser pulse. Thus, these results show a dramatic enhancement of ablation efficiency that can be achieved using particle-assisted-ablation. This aspect of the research is being further developed as a potential technique for efficient laser machining of glass and other low absorption materials. Damage can be avoided in laser cleaning using more modest fluences and multiple laser pulses.
Associate Professor Deb Kane 18 Oct 1999