Coating Materials
Martin Friz1 and Friedrich Waibel2
1Merck KGaA, D-64579 Gernsheim, Germany
2Umicore Materials AG, FL-9496 Balzers, Liechtenstein
1 Coating Materials for Different Deposition Techniques
A large number of different deposition techniques are used for the production of thin films for optical applications, as outlined in chapter 'Thin Film Deposition Techniques' by H. K. Pulker. The two most important categories are physical va-pour deposition (PVD), namely thermal vaporisation and sputtering, and chemical vapour deposition (CVD). It is obvious, that for each deposition technique suitable coating materials are required.
The PVD processes normally use inorganic elements or compounds and gases, whereas the CVD processes, dip coating and spinning, use liquid inorganic and organic compounds and gases.
Liquid compounds and gases are normally purchased directly from the pro-ducer, because they need no special preparation.
Solid materials have to be compact and in the appropriate form or shape, free of gas inclusions or even be prepared according to a special recipe. Targets must also fulfil structure requirements (grain size, texture, precipitation). These operations are the task of companies specialised in the production of coating materials and targets.
This chapter focuses on solid coating and sputtering materials. The require-ments on these materials are discussed, their properties listed and their production described.
Earlier summary articles or chapters in monographs on coating materials have been written by H. A. Macleod (Macleod 1986) , H.K. Pulker (Pulker 1979, 1999) and E. Ritter (Ritter 1975).
106 Martin Friz and Friedrich Waibel
2 Requirements for Coating Materials
2.1 Evaporation Materials
Apart from the elements only a relatively small number of inorganic compounds
can be evaporated to produce a coating with the same composition as the starting material. Suitable c
ompounds are found mostly among the chalcogenides and hal-ides. In some cases it is necessary to change the evaporation process (e.g. reactive evaporation, evaporation from several sources or flash evaporation) in order to en-sure that the resultant coating has the desired stoichiometry.
Evaporation materials have to fulfil a series of requirements in order to meet the demands of the coating process and to achieve the required film properties.
2.1.1 Chemical Purity
The chemical purity of vapour deposition materials influences not only the coating properties but also the way the material behaves during evaporation. As will be shown, a purity of least 99.99 %, normal in metallurgy and often specified by op-tical coating companies, is indeed a step in the right direction. However, it does not indicate whether the material will be suitable, nor how the material compares to materials of a different origin.
Even minute concentrations of transition elements can have a marked effect on the transmission properties of dielectric layers. Unfortunately, there are only few publications that address this topic. Nowhere is there a description, for instance, of the relationships between levels of contaminants and optical absorption in thin films. The most detailed articles containing quantitative accounts of impurity-s
pecific losses can be found in the literature on optical waveguides made of silicon dioxide (SiO 2) (Newns et al. 1973; Schultz 1974). Particularly troublesome in the visible range are minute quantities of cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), nickel (Ni) and vanadium (V). Cobalt, chromium and vanadium exhibit the greatest molar absorption or extinction coefficient k in the order of 100 li-tres ⋅mol -1⋅cm -1. Using Beer’s law of absorption
A = 1-e -k ⋅c ⋅d
enables an absorptance A of 10-3 % to be estimated for a layer thickness d = 100 nm and an impurity level c = 100 ppm (equivalent to about 0.003 mol/litre). Compared to this value, other causes of optical losses in the layer, e.g. scattering or absorption caused by variations in stoichiometry, often have a much greater effect. This means the aforementioned purity of 99.99 % is almost invariably adequate. For many important applications (e.g. AR coatings for spectacles and photographic lenses, filters, cold-light mirrors) levels of impurities even totalling 1000 ppm may be tolerated.
Prerequisite is, of course, that the strongly absorbing impurities mentioned are actually detected during chemical analysis of the evaporation material. It is there-fore necessary to define limits for the interfering impurities – usually the elements listed above. The methods of analysis most widely used nowadays for detecting
Coating Materials107 impurities are flame atomic absorption spectrometry (FAAS) and optical emission spectrometry with plasma excitation (ICP-OES). For substances with light cations, e.g. for magnesium fluoride (MgF2), aluminium oxide (Al2O3), silicon monoxide (SiO) and silicon dioxide (SiO2), x-ray fluorescence may also be used with excel-lent limits of detection for heavier elements.
Chemical analysis must also be able to detect foreign substances, which, for in-stance, might have a detrimental effect on deposition behaviour or the structure of the coating. A well known effect is that of a high oxide content in fluorides, most noticeable perhaps in MgF2, as it can result in splashing during evaporation. Many oxides have lower vapour pressure than the corresponding fluorides, and form an oxide layer over the melt during evaporation. This causes the melt to overheat and the oxide layer to tear, the actual cause of splashing. Determination of non-evaporable residue is one possible way of analysing the oxide content. Another option is inert gas extraction.
Evaporation behaviour and structure of the layer may be influenced by high re-sidual gas content in the evaporation material. In such cases it is best to determine the residual gas content by heating the substance in a high-vacuum recipient while monitoring pressure versus time. A mass spectrometer should be used to test the liberated gases, to be able to find the precise cause.
2.1.2 Physical Properties
Testing of dimensions, particle sizes and weights is a vital part of quality control. For tablets it is quite easy to deduce their density, which in the case of substances with poor melting properties, e.g. zirconium oxide (ZrO2), is a critical measure of the stability of the tablet. Extreme differences in temperature within a tablet dur-ing electron beam evaporation lead to significant heat stress. In such cases cracks in the tablet can not be completely avoided, and disintegration can only be pre-vented by keeping the porosity of the tablet sufficiently high; yet the porosity must not be so high as to make it impossible to maintain a low adsorbed gas content.
Some substances, e.g. some titanium suboxides, are relatively brittle and pro-duce large amounts of abraded particles during storage and transport. The problem can be lessened by optimising the packaging with padding or vacuum wrapping. An abrasion test in a running wheel is needed to allow various production batches to be assessed and compared.
Various vapour deposition materials with favourable properties are prepared from substance mixtures, whereby new compounds are formed by solid state reac-tions during the production process. The melting and evaporation properties of these materials depend to a very large extent on the completeness of the chemical reaction. X-ray diffraction (powder diffraction) is used to check the phase purity of the end product. Also, differences in the stoichiometry of a compound may negatively a
ffect evaporation characteristics and coating properties. A good exam-ple of this are the various titanium sub-oxides, where determination of stoichiome-try may well be an appropriate part of quality control.
108 Martin Friz and Friedrich Waibel
2.1.3 Process Suitability
Decades of experience in the manufacture and use of evaporation materials have shown that factors affecting process suitability are so complex that they do not al-low quality to be defined simply by physical and chemical properties. Therefore, substance testing under actual conditions of use continues to be a major feature of quality control. The assessment criteria differ from substance to substance. Com-mon criteria are, however, the way the material behaves during the heating or melting phase, the pressure profile for the system and the heating power needed for a given rate of evaporation. In the case of oxides, in particular during reactive evaporation, it is also necessary to record the evaporation data (pressure and rate fluctuations) over a prolonged period. In the case of mixed oxides, the reflectance of the coating is measured as it builds up, and the refractive index determined as an additional control for correct stoichiometry. In certain cases, such as when haf-nium
oxide (HfO2) is used in UV applications, spectrophotometric measurements are taken to confirm the suitability of the material for this spectral range.
2.2 Sputtering Targets
Sputtering of metallic layers is straightforward. For the sputtering of dielectric layers there are two different approaches. The first is to use ceramic target materi-als of the desired layer composition. The second is a reactive sputtering process, where a metallic target is sputtered in a reactive atmosphere of oxygen or nitrogen. Most sputtering processes for interference coatings use metallic target materials. In reactive sputtering processes there is not only a reaction on the substrate but also on the surface of the metallic sputter target. A stable sputtering process needs optimised management of reactive gas and cathode power. Most processing prob-lems (such as arcing) result from dielectric layers grown on the target surface. Ceramic targets are seldom used because of two severe disadvantages. One disad-vantage is that most of the ceramic materials are non-conductive and cannot be operated using Direct Current (DC) power. To install RF (Radio Frequency) power is expensive, it induces a low sputter rate and is complicated. The second disadvantage is the high production cost of ceramic target materials.
Ceramic targets are only used where an extremely precise stoichiometry of the dielectric layer is required . The best example is the production of ITO-layers with In2O3/SnO2 targets. The use of conductive TiO2 targets is also increasing (Weigert 2001).
Materials for sputtering targets have to fulfil specific requirements: The most important characteristics include:
•purity
•chemical homogeneity in case of alloys (stoichiometry)
•homogeneity of the structure (grain size, texture, precipitates, crystallinity) •target density
•no oxide or other impurity inclusions (metals and alloys)
•no pores and voids
Coating Materials109•good electrical conductivity (ceramics and semiconductors)
•no surface contamination
•dimension, target surface roughness
•mechanical stability of the compound target
The overall purity of sputtering targets for interference coating applications is 99.99%. It allows a sum of metallic impurities of 100 ppm. This is not very high compared with microelectronic applications (99.9999%, sum of impurities of 1 ppm in maximum).
Several of the most common analytical techniques to control impurity values and contents of alloying elements are XRF, ICP, GDS, GDMS, and CGHE. (Ven-zago and Weigert 1994; Wilhartitz et al.1990).
The optimised performance (purity, structure, homogeneity, etc.) for each tar-get, from target-to-target, and from lot-to-lot is very important for good sputter re-sults. The critical elements that influence the absorption levels are Co, Cr, Cu, Fe, Mn, Ni, V, etc.
Studies show a direct correlation between grain size and target performance. A small and homogeneous grain size induces a constant sputter rate over the entire target surface and produces consistent film thickness uniformity. It also affects the control of second phase precipitates (Marx and Murphy 1990). Precipitates larger than 10 µm cause particle generation during the sputtering process.
These quality factors are strongly influenced by the way the material is pro-duced (see Chap. 4).reactive substance
Nearly all modern sputtering processes are based on magnetically enhanced cathode processes. For the planar magnetron the sputtering target is just a sheet of metal or ceramic. Very often it is produced as a compound target bonded on a cooling or a special backing plate, to produce a mechanically stable system (weak or brittle materials) or to save costs of expensive target materials. Rotating cath-odes are increasingly used for large area coatings (tube as target geometry).
The criteria for the choice of technology include target lifetime, target utilisa-tion, material recycling and uptime of the sputtering process or stability of the process during target lifetime.
3 Survey on Materials, Properties and Applications Properties of thin films including optical, mechanical, electrical, thermal, and chemical properties are influenced by deposition parameters. Therefore refractive indices and transmittance range given in this chapter are only meant as guidance values. Detailed information on the influence of deposition parameters on film properties is given in chapters ‘Some Fundamentals of Optical Thin Film Growth’ by N. Kaiser and ‘Film Deposition Methods’ by H.K. Pulker.
版权声明:本站内容均来自互联网,仅供演示用,请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198,我们将在24小时内删除。
发表评论