3.1 Design of UV Broadband Double-Layer AR Coating
Computer aided design of optical thin films is a simple way to improve optical performance of multilayer coatings. At present, the commonly used film design software includes TFcalc, Essential Macleod, and Filmstar. In this work, the UV double-layer broadband AR coating was designed with the aid of the software Filmstar.
Based on former exploring experiments, the possible refractive index (n) of the top layer was manually input into Filmstar software. The central wavelength of the design was designed at 330 nm. The refractive index of the bottom layer and the thicknesses of the two layers were automatically adjusted for the architecture of the λ/4-λ/2 double-layer AR coating with the targeted transmittance of 100% at 250–400 nm. It should be noted that refractive index dispersions for the two layers were temporarily ignored so as to quickly obtain the preliminary design best suited for the design targets. As shown in Table 1, with the refractive index of the top layer increasing from 1.22 to 1.33, the optimized refractive index of the bottom layer also shows an upward trend. It can be seen that the top and bottom layers with refractive index of 1.28 and 1.64 and thickness of 59.54 nm and 91.26 nm are the best choices to achieve the average transmittance of 99.23% for the target wavelength region. The optimized design of this double-layer coating can offer a clear guidance for the preparation of sols and thin films.
Table 1
Refractive index, thickness, and transmittance of double-layer AR coatings based on λ/4-λ/2 double-layer structure with respect to the central wavelength 330 nm
Top layer | Middle layer | Tmax (%) | Tavg. 250−400 nm (%) |
n | d (nm) | n | d (nm) |
1.22 | 62.36 | 1.57 | 99.36 | 99.91 | 99.73 |
1.25 | 61.08 | 1.62 | 95.13 | 99.93 | 99.78 |
1.28 | 59.61 | 1.64 | 91.84 | 99.96 | 99.83 |
1.31 | 56.63 | 1.67 | 90.34 | 99.94 | 99.65 |
1.34 | 52.32 | 1.73 | 83.67 | 99.92 | 99.59 |
3.2 Optical Properties of Monolayer films
According to the strict theoretical formula, a fitting program was constructed in Matlab to fit transmittance data of thin film [15]. The Tauc-Lorentz (TL) model [18], suitable for amorphous or crystalline semiconductor materials, was used to fit the optical properties of ZrO2-SiO2 hybrid gel films. In this program, the combination of the genetic algorithm and the Isqcurvefit algorithm was applied to adjust the fitting parameters such as film thickness and optical constants until the theoretical spectrum closest to the measured data was obtained.
The transmittance curves of ZrO2-SiO2 hybrid films are shown in Fig. 1a, and the experimental data (symbols) and fitted data (solid lines) show a good consistency. The corresponding refractive index is shown in Fig. 1b. The refractive indices of Z20S80, Z23S77, and Z25S75 at the central wavelength of 330 nm are 1.616, 1.638, and 1.654, respectively. It can be seen that film refractive index exhibits a more evident dispersion in the range of 250–400 nm than in the visible range. However, the change of refractive index in the working region is still fairly moderate compared with the results of TiO2-SiO2 films [9]. The curves of the extinction coefficients k of the three groups of films are very close to each other, and the results of film Z23S77 are typically presented in Fig. 1b. As expected, the extinction coefficient k of film Z23S77 remains low at 250–1200 nm and notably increases only in the shorter wavelength range, which is beneficial for realizing high performance of the UV AR coating.
Figure 2a shows the transmittance spectrum of the F127-SiO2 film. The Cauchy equations [19] were used to fit the entire spectrum of the F127-SiO2 film, and the achieved optical constants are given in Fig. 2b. The refractive index of the film F127-SiO2 at 330 nm is 1.282, indicating that the film can meet design requirement of the top layer. Moreover, the degree of index dispersion of the F127-SiO2 film is slight lower than that of the ZrO2-SiO2 hybrid films, and the extinction coefficient k with the order of 10− 5 is very low, which is conducive to the preparation of UV AR coating.
3.3 Assembly of Double-Layer AR Coating
Accurate analysis and control of the optical parameters of each layer is the key to the preparation of the double-layer AR coating. In this study, the F127-SiO2 film was used as the top layer. Based on the optical constants of the prepared ZrO2-SiO2 films, the actual performance of the corresponding double-layer coatings were investigated with the software Filmstar. The theory design shows that the film Z23S77 should be chosen for refractive index matching to assemble the UV AR coating. The optimized structure of the λ/4-λ/2 AR coating with the top and bottom layer thicknesses of 59.54 nm and 91.26 nm is shown in Fig. 3. According to the theory design, the thickness of each layer was nicely controlled by adjusting the withdrawal speed of dip coating, and the films Z23S77 and F127-SiO2 were deposited successively on the quartz substrate to form the double-layer AR coating.
The transmittance spectrum of the double-layer AR is shown in Fig. 4. It can be seen that the prepared double-layer coating exhibits high transmittance from ultraviolet to visible band with a maximum transmittance of 99.79% at 336 nm, and the average transmittance is 99.34% in the wavelength range of 250–400 nm. On the whole, the prepared double-layer AR coating significantly enhances the transmittance at the target wavelengths compared with about 93.0% transmittance of bare glass. The good agreement between the experimental and design spectral curves mainly benefits from the accurate determination of film optical parameters and the precise control of film thickness.
The cross-sectional SEM image of the double-layer coating is illustrated in Fig. 5. It can be identified that the thickness of each layer is very close to the designed value. Figure 6 shows the AFM images and root-mean-square roughness (Rq) values of the bottom layer, top layer, and double-layer coating, respectively. The bottom and top layers have rather smooth surfaces with low Rq values of 0.98 nm and 0.74 nm, respectively. Moreover, the Rq value of the surface of the double-layer AR coating is 1.14 nm, slightly larger than that of the top layer. Thus, it is reasonable to presume that the accumulative effect between the two layers is very limited, and the surface and interface scattering has almost little impact upon the performance of the AR coating.
Fig .6 AFM images of (a) single bottom layer, (b) single top layer, (c) double-layer coating
3.4 Contamination-resistant test
In high power laser systems, many non-volatile organic compounds can be converted into steam under the huge negative pressure of high vacuum. The adsorption of pollutants is bound to increase film refractive index, resulting in the decline of optical performance of AR coatings. In this experiment, the surface treatment and modification steps of the AR coating with PFDS were introduced. The transmittance curve of the double-layer AR coating before and after the pollution test is shown in Fig. 7. After 15 d of exposure to PDMS vapor in a vacuum of 10− 3 Pa, the average transmittance of the modified AR coating in the 250–400 nm region decreases only by 0.6%, as shown in Fig. 7a. In contrast, the average transmittance of the coating omitting the modification has a noticeable decrease of 2.3%, as shown in Fig. 7b. Therefore, the stability of the double-layer AR coating in this study can meet the demand of practical stability in potential applications.
3.6 Measurement of LIDT of the coatings
According to ISO 11254-2:2000, the LIDT of film was tested by Nd: YAG pulsed laser (355 nm, 6.4 ns) in "R-on-1" mode. Based on the standing wave theory, Chen et al. [20] used a theoretical model of high energy absorption at the surface and interface to describe laser damage to optical multilayers. The theoretical results show that the energy absorption mainly depends on the distribution of the standing wave and extinction coefficient of the material. The damage process of film is manifested as the formation of plasma flash at the laser damage point on film surface. At 355 nm, the damage mechanism is dominated by intrinsic absorption, and the damage morphology presents a uniform fusion zone [21]. The morphologies of the samples were observed using an optical microscope to understand the type of damage occurring on the surface of the coating. Figure 8a shows the cracked surface of the quartz substrate by laser irradiation. The damage morphology of the double-layer coating under 355 nm intense laser is presented in Fig. 8b. It can be seen that the center of the film is in a molten state, and the damage is mainly caused by the bulk damage from the substrate. Other film samples have similar damage morphology under the triple frequency intense laser.
The LIDT values of the double-layer coating as well as the results of the monolayer films are depicted in Fig. 9. The LIDT value of the quartz substrate is 2.87 J·cm− 2, which is slightly higher than those of the films. The LIDT value of the double-layer AR coating is 2.24 J·cm− 2, slightly lower than that of the bottom layer of 2.58 J·cm− 2 and the top layer of 2.32 J·cm− 2, which may be due to the different thermal energy absorption of different films [22]. It should be noted that the LIDT of the double-layer coating matches the reported values of 1.5–10 J•cm− 2 (355 nm, 6–10 ns) for oxide optical materials [23–25], and the proposed coating can meet the basic application requirements of ultraviolet lasers.