Originally, optical coatings were only deposited by evaporating compounds using resistive heating or electron beam evaporation sources. The resulting coatings are always non-stoichiometric. To improve the stoichiometry and thus the refractive index of the coating, the reactivity was then followed by evaporation techniques. Next, plasma-enhanced evaporation techniques were developed for better stoichiometric control. A technique called the activation reaction evaporation process discussed earlier has been used to deposit a variety of optical films. High-threshold optical films of TiO2, ZrO2 and HfO2 have been synthesized.
While color is the essential characteristic of any decorative coating, hardness is of secondary importance. When it comes to matching these two aspects, it is often necessary to accept compromises. Significant deviations in color from golden yellow to brown, purple, reddish and gray are described in the literature. TiC-TiN-TiO ternary system. This was observed on bulk samples obtained by high temperature sintering, thus very close to a 50:50 patterning. According to this data, all TiN-TiO with binary system is yellow; by mixing some TiC, the color changes from brown and blue to metallic gray. PVD gave somewhat different results because of the low substrate temperature and the ease of obtaining non-equilibrium compositions with large deviations from stoichiometry. Compositions where the sum of nitrogen and oxygen is less than 50% appear yellowish or even metallic white, while small excesses of these two components produce dark shades and increase light absorption.
X-ray spectroscopy indicated that no true solid solution occurred during the low-temperature synthesis. This assumes that large excesses of nitrogen or oxygen are not incorporated into the basic cubic structure, but are instead deposited on defect sites that highly perturb the lattice and grain boundaries, thereby increasing light absorption.
The reflectance spectra of ZrN and TiN based coatings obtained by reactive magnetron sputtering are shown in Fig. 32.13.58, and the stoichiometrically close ZrN coatings have a golden yellow color. TiN-based coatings with increased amounts of oxygen and nitrogen have been investigated. Two effects were observed: an excess of nitrogen reduces the reflectance on the long wavelength side; the paint turns a deep yellow (old gold) and then bronze. The reflectance minimum, located in the yellow sample, shifts near-UV light to longer wavelengths. Added oxygen combined with excess nitrogen acts in the same direction, reducing the red side reflectance to very low values. Due to the displacement of the minima, the blue side reflectance increases, neutralizing the red and yellow components and giving black and even blue tints. Carbon is added primarily to give deeper blacks by further reducing the total carbon reflectance. Recently, some other binary and ternary systems have been explored, when titanium is replaced by aluminum and vanadium.
The activated reactive ion plating process is used to produce well adhered, dense nitride, carbide and oxide films at relatively low substrate temperatures. They are used commercially to make film optics applications.
Using the ion-plating oxidation process, simple photometric intensity measurements of absorption in the high transmittance range cannot be used. Measurements of the refractive indices of thin films give values close to those of the bulk material. In all cases, the refractive index is much higher than that of evaporated films. The various values are listed in Table 32.6 for comparison.
Optical properties remain unchanged even in repeated heat treatment cycles. Heating ZrO2–SiO2 and Ta2O5–SiO2 multilayer glasses at 400°C for several hours and then immersing them in water for 3 days did not produce changes in optical and mechanical properties.
Matching optical and mechanical properties requires some compromises to be made. This is especially true when the golden-yellow hue turns darker, due to the hyperstoichiometric composition. In the case of yellow, start from very high values (over 2500 VH) and for dark coatings the hardness can be as low as 1000 VH. This is still acceptable from the rubbing of a protective cloth or the wear and tear caused by traditional cleaning products.
Obviously, the adhesion needs to be "good", i.e. there should be no peeling, even at deep scratches. In plasma-assisted PVD methods, adhesion is usually very good. Adhesion does not depend on the coating alone nor on the substrate. When deposited on soft metals, the hardest coatings will perform poorly due to very limited resistance to impact and scratches. At the same time, problems may arise with corrosion protection; the coating may not be continuous and pinhole-free as deposited, or may become discontinuous after mechanical damage.
Good watch cases are made of a carbide substrate (case) with a hardness of 1500 VH or higher, but dark gray, gold-plated yellow TiN or ZrN are even harder than this (2500 to 3000 VH). Unfortunately, this solution is very expensive.
Attempts have been made to improve the situation by first using low-pressure plasma nitriding in an enhanced glow discharge to obtain a hardness gradient within the first micron below the surface of the alloyed steel substrate, followed by a further increase in hardness and corrosion resistance by depositing a TiN or ZrN coating sex.
Depositing hard coatings directly on brass and similar soft alloys will result in a dull and poor quality coating that protects against scratches and impacts. Hard chrome undercoats up to several microns in thickness, deposited by well known electrochemical methods, give acceptable solutions.
Activation of reactive ion plating processes is becoming increasingly important for optical applications. Bombardment during ion deposition induces a cascade of atomic collisions in the growing film. The receding and displaced atoms cause a continuous atomic mixing that also enhances surface migration. This results in filled voids and smooth or graded grain boundaries. Another consequence of ion bombardment is a high concentration of point defects. These are frozen in the structure under growth under conditions comparable to rapid quenching. High compressive internal stresses are a direct consequence of these defects.
Density, high bond strength and compressive stress characterize the great success of these thin, brittle films in wear protection applications . Adhesive strength is required to compensate for the buckling force induced by the high compressive intrinsic growth stress, and this stress is required as a mechanical prestress to be relaxed by the mechanical load imposed on the film in actual use. The consequence of this prestressing is that such loading will not damage the film as long as the deformation of the substrate remains in the elastic range.
When the thermal expansion coefficient of the film is less than the temperature of the substrate and when the film is at or above the maximum expected operating temperature. Thermal stresses are also relaxed in service when the operating temperature is higher than room temperature, which is usually the case. The optical thin films produced in the process without ion bombardment showed mobility with low condensed atoms and molecules due to their low thermal energy between 0.1 and 0.2 eV. Since no current can flow in or on an insulating substrate, ion bombardment of such substrates or thin films requires that exactly equal numbers of positive and negative charges hit the upper surface of each point to achieve charge balance.
Plasma-assisted ion plating utilizes a negative self-bias potential and not only occurs on the surface of a thin insulating substrate placed on an RF electrode. In DC plasmas, a significant difference between the average velocities of electrons and ions in the plasma is that the self-bias potential is more than 10 V relative to the plasma. This bias accelerates the positive ions towards the substrate surface. Their energy is usually not high enough to cause sputtering, but they are at least 50 times more thermally energetic than vapor atoms and molecules, and higher than the crystalline binding energy.
Dissociation is a problem for composite films. Even upon evaporation, which is mild during PVD, the compounds dissociate to some extent. Due to their low adhesion coefficients, gaseous components can be pumped away, resulting in films of substoichiometric composition of the sediment . In reactive deposition, gaseous components are continuously replaced. Because of the high reactivity of oxygen, only oxide films have been successfully produced industrially by reactive evaporation. The oxide or suboxide is used as the evaporation material and the oxygen partial pressure in the chamber is stabilized at about 10-2 Pa using a controlled inlet valve. The actual oxidation occurs to a great extent on the surface of the substrate. Oxygen chemisorption rate is a key factor in well completion response.
However, optical coatings produced by reactive evaporation alone are usually still slightly substoichiometric and thus slightly absorptive. These films have a rough surface and a columnar or spongy microstructure with large void volume and large internal surface area. Due to the low density , the refractive index of these films is much lower than that of bulk oxides. These films absorb water vapor and other gases from the atmosphere, changing the refractive index and other physical properties. They have poor adhesion to the substrate, low wear resistance and low hardness. By heating the substrate to approximately 300 °C before the reaction evaporates. In fact, substrate heating is a standard procedure in this process.
Years of experience have shown that bombardment of growing thin films with primarily atomic species of film-forming has many advantages. Therefore, ion plating and those activation reactive evaporation processes using anodically or cathodically active coating materials as evaporation sources lead to processes that produce extraordinary results. Activated reactive ion plating includes not only bias activated reactive evaporation processes, but also three important industrial PVD processes for depositing wear resistant coatings. They are based on arc discharges, i.e. discharges of gases in which the majority of electrons are produced by electron emission from the hot spot of the cathode. After ignition, this hot spot can be maintained by heating by ion bombardment (self-sustaining arc), or it can be initiated, maintained and localized by an independent energy source (such as a heating filament) (thermionic arc). The working conditions (pressure, temperature) during deposition on various metal objects tend to give rise to typical columnar structures. One consequence of this texture is microscopic roughness, which is roughly proportional to the coating thickness. In the case of decorative coatings, which are usually thinner than those deposited on cutting tools, this phenomenon is not noticeable but occurs when touching a fresh surface with a finger.
Fingerprints cannot be wiped off simply by rubbing, due to traces of grease deposits in the microscopic depressions between the single columnar crystals. This roughness, in the worst case up to 10% of the total coating thickness, can significantly reduce texture in isometric conditions and is favored. It is well known that during the activation reaction, strong low-energy ion bombardment ion plating creates favorable conditions for this texture, resulting in a brighter surface.
To move the golden stoichiometric nitrides, additional nitrogen, oxygen, and even carbon need to be added to the coating. When the nitrides are deposited alone, the final nitrogen content is approximately proportional to the fraction. the gas pressure present. To get a darker colored oxynitride, the easy way is to start with a yellow nitriding and add some oxygen. Titanium and zirconium have a significantly higher affinity for this gas than nitrogen, and the first gas is more collected by the metal vapor than the second gas. One consequence is that relatively small amounts of oxygen are required.
Clearly, carbon cannot be introduced in elemental form. One possibility is to use gaseous hydrocarbons. Partial decomposition of hydrogen from these molecules leaves free radicals behind. The probability of a reaction between these radicals to generate new organic molecules is higher than the formation of metal carbides. A series of organic synthesis takes place in the plasma, heavily polluting the chamber and condemning the next experiment to failure in color reproduction. Part of the solution to this problem lies in the use of carbon oxides, and if the extraction unit is compatible with these gases, carbon oxides of nitrogen and even halocarbons are unavoidable.
Obtaining precise quantitative prescriptions for each color is obviously impossible. Parameters require experimentation to find out what works for each coating system by trial and error. Further development trends point to new binary or ternary compositions offering new colors and possibly even better mechanical and chemical properties.
