Physical and chemical cleaning of the surface of power metal optics for the purpose of increase in beam firmness

Features of the choice of solvents for the physical and chemical cleaning of power optical elements in accordance with the parametric solubility theory were investigated. When cleaning the surface from the model contaminant with halogenated hydrocarbons, visually observed white film of alkali and alkaline earth metal salts that are not soluble by this class of solvents and iridescent bands from the interaction of hydrocarbons with the metal remain on the surface. All this greatly impairs the optical properties of the surface. It was shown that, when using solvents, it is necessary to inhibit the interaction of halogenated hydrocarbons with mirrors by stabilizing solvents or by selecting the regimes for carrying out the physicochemical purification process, or by thereof combination.


Introduction
In industry, in science and technology, powerful CO2 lasers emitting at a wavelength λ = 10.6 μm are used [1]. In the middle infrared region, CO and HF (DF) lasers are used [2]. Reflectors are usually cooled mirrors [3][4][5][6][7][8][9] made of copper, aluminium, beryllium, molybdenum, silver, etc. They have high thermal conductivity and a reflection coefficient in the operating range, which makes it possible to use mirrors without additional interference coatings that reduce the radiation strength. Cooling reduces the effect of thermomechanical distortions in the shape of mirrors, which require regular cleaning to restore performance.
Optical durability of power optics is the main requirement for laser systems operating at high radiation intensities. Optical stability in the continuous mode of radiation is due to a combination of absorptivity and thermomechanical properties of the material [10]. In the pulsed mode, plasma of lowthreshold optical disruption of the gas, initiated by evaporation of the ions of the material from the surface, can occur near the surface, which can sometimes be damaged with the appearance of irreversible changes [3][4][5].
In the framework of the theory of thermal optical breakdown of solid transparent dielectric inclusions, the reason for the appearance of luminous spots on the metal under the action of laser radiation, which are distinguished by their brightness on the general background of the irradiated zone,  [5,[11][12][13][14]. On these formations, the reflection of light is markedly reduced, and the temperature of their melting and evaporation considerably exceeds the corresponding values of the metal. The effect is enhanced by the interaction of the laser pulse with surface electromagnetic waves [15], since up to 100% of the radiation can be absorbed in this case.
In the process of using mirrors with a gradual increase in power, appears the effect of "laser cleaning". [15][16][17][18]. Due to the evaporation and burning out of the absorbing centres, the optical stability threshold increases. However, the selection of the regime is very complicated, since "laser cleaning" can be accompanied by oxidation of the metal.
The parameters of technological contaminations on the surface of metallic mirrors formed during optical processing and during operation and decreasing the optical stability of the mirror are revealed in this work. The efficiency of solvents for cleaning optics was also investigated.
The degree of impurities contamination of the surface of the irradiated optical element affects the occurrence of irreversible changes in the material, in particular, the absorption capacity, and, consequently, the thresholds of gas breakdown at the surface. It was noted in [11][12][13][14] that under the action of the first pulse on the unpurified surface, the threshold intensity is always several times lower than under the action of "subsequent" pulses. In [12], when the metal surface was irradiated consecutively with a series of pulses from a CO 2 laser, the energy expenditures for the formation of the plasma rose with increasing in the number of pulses to a certain value, at which they remained constant, i.e. the surface was cleaned in the presence of an air-breakdown plasma by removing the defects which absorb radiation. It was shown in [13] that the initial absorbing power for copper and aluminum targets, was 10-12 % and 6 %, respectively, but after laser cleaning it was reduced to 2-3 %, which is typical for pure unpolished samples.

Materials and methods
In reality, metals are covered with a sorbet layer of dielectric, mostly organic, contaminants. These may be areas of the surface covered with a continuous thin film or single ("island") film formations.
Since purification is a reduction of surface contamination to a certain level (surface modification), it is interesting to investigate the plasma formation threshold caused by this contamination as a function of the contamination radius (provided that it is small in comparison with the depth of the optical absorption).
Analysis the increase of the temperature of dielectric impurities on the basis of the theory [19] makes it possible to determine the initial heating of the absorbing contaminants by formula where ω is the circular frequency of light; ″ = ″ 4 is the imaginary part of the susceptibility; t c is the heat capacity; E0 is the field intensity; t is the time of exposure. The depth of optical absorption is the reciprocal of the absorption coefficient where λ is the wavelength in vacuum; n is the refractive index. For short radiation pulses of duration tp, it is necessary to use the equation for ΔT, in which t = tp. In many cases, t is determined by the time tc of heat transfer from the inclusions. For stationary heating where D τ is the thermal diffusion coefficient. The value D τ for many solids at room temperature is of the order of ̴ 10 -2 cm 2 c -1 , as a result, tс ̴ 1 μs for the particles with a radius of ̴ 1 μm. If the particle radius is ≤ 0.1 μm, and t ≥ 10 -8 s, the increase of the inclusion temperature can be recorded It follows from equation (4) that the field strength required to achieve a critical temperature for inclusions of ̴ 0.01 μm should be ̴ 2.1·10 6 Vcm -1 (for λ = 10.6 μm), which is comparable with the plasma formation of a pure material (for example, of glass). Consequently, inclusions with a radius of ̴ 0.01 μm from the point of view of lowering the threshold are less dangerous compared to larger ones. Inclusions ≥ 1 μm can be easily detected and removed. The most dangerous from the viewpoint of reducing the threshold of plasma formation are inclusions with a radius of ̴ 0.1-1.0 μm located in the surface layer and having an optical absorption depth of 10 -3 ÷ 10 -4 cm. The critical radius (rcr) of the spherical inclusion, less than which its absorption cross section begins to decrease, is at the wavelength λ = 10.6 μm for dielectrics. The critical radius (rcr) of the spherical inclusion, less than which its absorption cross section begins to decrease, is at the wavelength λ = 10.6 μm for dielectrics where μ is the radiation absorption coefficient. Equation (5) defines a minimum limit, indicating below what size on the treated surface after physical and chemical cleaning inclusions may be present. Thus, for λ = 0.63 μm, the maximum impurity size below which it is impractical to clean the surface is ~ 0.1 μm. The presented evaluations have largely a qualitative nature, since under pulse actions the index ( ) can increase to 10 6 times [3].
Structural defects (grain boundaries, dislocations, vacancies, etc.) contribute to a decrease in the plasma formation threshold by more than 2 orders of magnitude [14].
Surface inclusions, having a much larger (compared to bulk) absorption, along with a decrease in the plasma gas formation threshold near the metal, are also responsible for the destruction thresholds caused by their thermal explosion [11].
The choice of solvents for the physical and chemical cleaning of power cells in accordance with the parametric solubility theory was carried out as follows. An analysis of the process contaminants present on the surface of the metal showed that their bulk (up to 85 % by weight) mainly includes rosin pitch (with a rosin content of up to 70 % by weight), used to manufacture a polisher for processing optics. The pitch is a polymerization type compound whose solubility parameter (δ) is not determined from the heat of evaporation or other quantities. Therefore, it is determined experimentally by the nature of solubility in solvents. For rosin, δ = 18 J 1/2 cm -3/2 [20].

Result and discussion
At the first stage, a kinetic analysis of the physical and chemical process of dissolving the rosin pitch was carried out in solvents with an indicator (δ) close to rosin. The choice of solvents was carried out from various classes of selectivity according to Snyder [21,22]. Then we carried out a kinetic analysis of the physical and chemical process of dissolving the rosin pitch in these solvents, for which 1 g of resin was placed in a container and dissolved in solvents. The amount of resin dissolved after each cycle of contact was determined on an EF-3MA electronic fluorometer (table 1).
The data in Table 1 show that acetone and diethyl ketone have the maximum solubility with respect to rosin pitch (after 4-5 cycles of dissolution, the rosin is not detected, completely passing into solution). Somewhat worse, its dissolution occurs in diethyl carbonate, xylol, methylene dichloride, 2-butanol. The bulk of the rosin pitch dissolves after 6-7 cycles (only insignificant amount of visually observed suspension remains undissolved).
As a model contamination, we used spindle and diffusion (VM-5) oils and grease from human hands. These impurities (total amount of 100 mg, taken in equal proportions) were applied to copper mirrors (Mob grade copper), previously purified by acetone. Data on the amount of pollutant that passed into the solvent after each cycle of their contact were determined on the EF-3MA fluorometer (table 3).