Open Access
Add to BibSonomy
Abstract
A reaction chamber of atomic layer deposition (ALD) was developed for simultaneous coating on the inner and outer surfaces of a large-size and strongly curved glass bowl. The inner surface ALD process was in a showerhead reaction mode and the outer surface ALD process was in a cross-flow reaction mode. Blue reflection (BR) film of 400 nm wavelength and broadband antireflection (BBAR) film of 400-700 nm wavelength were coated on different glass bowls by ALD. The spectral uniformity of both coated bowls was studied. The measured spectra at multiple positions of the glass bowl with the BBAR coating show better spectral uniformity along the circumference than the depth. The spectral deviation is mainly caused by the non-uniformity of the film on the outer surface (<±3%), and the film on the inner surface has good uniformity along both the circumference and the depth (<±0.7%). The growth rate of the outer film was reduced by 10% on average compared to that of the inner film due to the different gas flow mode.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Corrections
18 May 2023: A typographical correction was made to the author affiliations.
1. Introduction
Various optical components are used for light beam shaping and distribution in optical systems and optoelectronic instruments. In order to improve energy transmission efficiency and eliminate ghost image induced by reflection on the surface of optics, or to modify the light beam distribution or spectral component, functional optical films need to be coated on the surfaces of optics [1]. Common optical coating methods, such as evaporation, sputtering, etc., as classified as physical vapor deposition (PVD), have a gradient deposition rate on curved surface due to the line-of-sight nature of PVD [2–4]. Normally, the film can be coated on only one side of the substrate, and it is difficult to perform uniform coating for strongly curved surface [5]. The non-uniformity of coating on curved surface makes the performance of the practical optical film significantly different from the theoretical design, sometimes even harmful to the optical system [6]. In optical system or equipment, the strongly curved glass surface helps to improve the beam shaping and collecting efficiency [7]. The hemispherical glass dome can expand the field of view while effectively isolating the optical instrument from the harmful external environment, such as water, toxic gas, etc. [8]. It ensures a stable and suitable working environment, prolongs life-span and improves performance. Improving the uniformity of coating on strongly curved surface has always been one of the research hotspots in optical coating field [3,5,9].
Atomic layer deposition (ALD) is a modified type of Chemical Vapor Deposition (CVD) with reaction only existing on surface [10,11]. During the deposition process, precursor pulses are exposed on surface of substrate to form saturated adsorption or to perform self-limiting reaction, separately. In the interval of pulses, inert gas purging separates different reactant pulses to ensure reaction limited on the surface. Due to saturated chemical adsorption of single molecular layer on the surface, ALD technology promises a pinhole-free, dense and conformal layer growth. In the 1990s, it was realized that ALD could be used to fabricate optical films with unique advantages, but at that time the slow growth characteristics restricted its application in the field of industrial optics [12]. Nevertheless, a lot of works have been done on the ALD processes and properties of optical materials by ALD, including oxides, nitrides, sulfides, fluorides and elements [13,14]. It has been found that the initial ALD growth rates on different under-layers or substrates are a little different due to different distribution of hydroxyls [15]. Some optical films, such as antireflective films and bandpass filters, were fabricated to prove the feasibility of ALD for optical coating [16–19]. With the development of ALD technology and the growing need of conformal optical coating on nanostructure and curved surface, more practical and complex multilayer optical coatings by ALD were studied and fabricated. They were deposited on large-scale substrates and large-curvature surfaces. Sub-nanometer thickness controlling precision helps to precisely fabricate multilayer BBAR coatings [19]. Customized ALD reaction chamber was used to coat dense protective layer on meter sized silver mirror for astronomical observation and space remote sensing [20]. The characteristic of saturated absorption and self-limiting reaction of ALD provides uniform optical coating on strong curved surface [9,21].
In this work, we demonstrate simultaneous coating on the inner and outer surfaces of hemispherical glass bowls. Two representative films were coated on different bowls, respectively. They were a blue reflection (BR) film with quarter-wave dielectric layers of alternate high and low index and a BBAR film with non-quarter-wave layers. For optical film design, complex stacks can be decomposed into the above two. The spectral uniformity of coated glass bowls was studied. The results in this work should be useful to the improvement of film design and ALD process for uniform optical coating on whole surface.
2. Experiment details
2.1 Substrates
The official substrates are borosilicate glass bowl with refractive index ∼1.47 at 550 nm, which is equivalent to SCHOTT glass Borofloat 33. The bowl is similar to hemispherical shell, with an outer diameter 120 mm, an inner diameter 112 mm, an inner depth 45 mm, and a total height 53 mm. The substrates used in the single-layer experiments are P-type silicon wafers (<100>, 10∼25 Ω·cm, 15×15 mm). Because there is light loss inside the glass bowl and transmitted beam after multiple reflections on the curved surfaces may stray from the optical path to detector, the measured spectral transmittance of glass bowl is not accurate enough. In order to correct the measured spectral transmittance, planar glass B270 was also used as reference in BBAR coating experiment.
2.2 ALD reaction chamber
The films were deposited by thermal ALD method (TFS500, Beneq Oy, Finland). The ALD equipment was originally equipped with a cross-flow reactor (Fig. 1(a)), which was not suitable to simultaneous coating on the inner and outer surfaces. We reconfigured the reactor to adjust the flow mode of the inert gas and precursor vapor to suit double-sided coating of glass bowl. The schematic diagram of the gas flow in the reconfigured reactor is shown in Fig. 1(b). The gas flow enters the Φ100×40 cylinder space upon the sample chamber from the only one inlet, homogenizes, diffuses into the Φ200×60 sample chamber through the grid, and then distributes evenly on the circular sample tray. In the purging time, excess reactants and by-products enter into the annular gas collection channel, and then exhaust through the outlet. Two inert gas flow (N2) with flux 300 sccm are used as carriers for A and B reactants. The pressure is maintained about 1.1 mbar (110 Pa) inside the reaction chamber, and about 8 mbar (800 Pa) outside the reaction chamber by filling 200 sccm N2 into the vacuum chamber.
Fig. 1. Schematic diagram of the gas flow in the reactor, (a): the original one, (b): this work, (c): top view of the new reactor, (d): bottom view of the new reactor.
2.3 Film materials
The film materials used in this work were TiO2 (symbol H), Al2O3 (symbol M) and SiO2 (symbol L). They were all deposited by thermal ALD. The precursors of the metal elements are TiCl4, trimethylaluminum (TMA), and bis-tert-butylaminosilane (BTBAS), respectively, with purity > 99.99%. The oxidants are H2O, H2O and O3. The ALD process parameters are listed in Table 1.
Table 1. Parameters of ALD process for TiO2, Al2O3 and SiO2 deposition.
2.4 Film design
Film design was performed in commercial software (FilmWizard, Scientific Computing International, USA). The first optical film is BR film consisted of three layers HMH with quarter-wave thickness for each layer at wavelength 400 nm. The other film, BBAR film is consisted of six layers HMHMHL with non-quarter-wave thickness for each layer. For the H layers of BBAR, 1.5 nm Al2O3 interlayer was inserted per ∼15 nm to improve optical performance of TiO2 in the deep blue range (380-420 nm). The adopted film is Sub | 9.8Hp 38.46Mp (15Hp 1.5Mp)^4 7.6Mp (14.5Hp 1.5Mp)^2 88.56Lp | Air, where, p means physical thickness, the numeric values before layer symbols represent the physical thicknesses with unit nm. The designed film structure is based on the refractive indices determined in the single-layer experiments. The designed transmission spectra of the above two optical films are shown in Fig. 2.
2.5 Film measurement
Ellipsometry (V-VASE, J. A. Woollam, USA) was used to determine the thickness and parameters of dispersion formula of single layer by ALD, with spectral range 1-5.1 eV (wavelength range 243-1240 nm) per 0.1 eV and three incident angles 65°, 70° and 75°. The growth per cycle (GPC) and refractive indices of the three materials were then calculated.
The spectrum measurement of the glass bowl was performed using a self-built measurement system. In the measurement process, a wavelength adjustable and collimated monochromatic beam was generated by a grating monochromator with a bromine-tungsten lamp, and then incident at a position of the glass bowl. The transmitted light was received by a silicon detector. By dividing the transmitted light intensity by the incident one, the transmittance spectrum was acquired. The transmittance spectra at different positions of glass bowl were used to determine the non-uniformity of optical coating on the inner and outer surfaces. The non-uniformity is defined as (λmax- λmin)/2λaverage, where λ is the wavelength of specified transmittance.
In addition, for the reference substrate B270, commercial spectrophotometer (Lambda 950, PerkinElmer, USA) was applied to accurately determine the spectral transmittance, which was used to correct the peak transmittance of optical coating on glass bowl surface.
3. Results and discussion
3.1 Single layer
As shown in Fig. 1(b), there are two gas flow modes for the ALD process of the glass bowl, showerhead mode for the inner surface ALD and cross-flow mode for the outer surface ALD. In the showerhead mode, the precursor distribution is homogeneous, and the redundant precursor and product gas stays a long time in the reactor. In the cross-flow mode, the gas flows through reactor quickly, and the residual gas can be removed quickly. Due to the difference of the precursor retention time and distribution density in the reactor, the growth uniformity could be poor for some non-ideal ALD process, especially when the reaction product is chemical active for the deposited film or the absorbed precursor desorbs or decomposes as time goes by.
Single-layer experiments of the above three materials were done in different reactors. Multiple silicon substrates were placed on the Φ300 aluminum sample tray at positions P0-P8, and then loaded into the reaction chamber, as shown in Fig. 3. For the cross-flow mode, the samples at P0-P8 were measured and analyzed. For the showerhead mode, the available samples include P0-P4. The adopted GPC was the average value of the maximum and minimum GPC of the available samples. Generally, it is close to the GPC at the P0 position. Thickness non-uniformity is defined as the ratio of the difference between the maximum thickness and the minimum thickness to the sum of the two. The optimized parameters of ALD process were the pulse and purge time. The reaction temperature and the inert gas fluxes were kept constant. The pulse time was adjusted to ensure the saturated adsorption and full reaction of reactants, and the purge time was kept long enough to assure that the excess reactants and by-products were fully purged out of the reaction chamber. After the process parameters were determined, we carried out the single-layer ALD processes with different cycles and did repetitive experiments with the same cycles.
Fig. 3. Schematic diagram of sample alignment in single-layer experiments, (a): cross-flow mode, (b): showerhead mode.
The results of the single layer experiments were listed in Table 2. Al2O3 and SiO2 have good uniformity, while the uniformity of TiO2 is relatively poor, especially in the cross-flow reactor. The reaction products of TiCl4 and H2O include HCl, which is chemical reactive with TiO2 and semi-product of ALD reaction, and it causes decreasing GPC of ALD TiO2 along the gas flow [22]. In the single-layer experiments, the GPC is 10%∼20% different in different reactors, but basically same in different runs with the same process conditions. The dispersion curves of the three materials are depicted in Fig. 4. As seen in Table 2, the change in the refractive index of the single layer caused by the gas flow mode and the layer thickness is about ±1%. The results tell that the major factor inducing spectral change of optical coatings is the film thickness. Since the glass bowl in the reactor affects the gas flow mode, and the substrate or the under-layer induces deviation of the initial growth rate of ALD layer, the accurate growth rate of each layer of optical film on glass bowl is hard to determine. In practical fabrication of optical film, reverse engineering was applied for calibration based on the spectral performance of the coating.
Table 2. Growth rate and thickness non-uniformity of ALD TiO2, Al2O3 and SiO2 single layer.
3.2 BR coating
Based on the designed film HMH with reference wavelength 400 nm, the BR film was deposited on double sides of a glass bowl simultaneously. The coated glass bowl reflects blue light and shows good color uniformity, as shown in Fig. 5.
Fig. 5. Photographs of glass bowl with double-sided BR film, (a): the inner surface, (b): the outer surface.
At different positions along the circumference and the depth of the glass bowl, the transmittance spectra were measured at the normal incident angle, as shown in Fig. 6 and Fig. 7. The results show better spectral uniformity along the circumference than the depth. The spectra show different valley values, because there is light loss such as light scattering and absorption inside the glass, and also, after multiple reflections between the curved surfaces of the glass bowl, some of the transmission light is not caught by the detector.
Fig. 6. Measured transmittance spectra (AOI=0°) of glass bowl with double-sided BR coatings at different positions along the circumference.
Fig. 7. Measured transmittance spectra (AOI=0°) of glass bowl with double-sided BR coatings at different positions along the depth.
In order to quantify the non-uniformity of the spectra, the transmittance spectra were normalized by setting the peak value as 100% and the valley value as 0%, and then the wavelength at transmittance 50% was used to characterize spectral position, as shown in Fig. 8. The spectral non-uniformity is about ±1.4% along the circumference and about ±3% along the depth. The results meet the estimate based on the non-uniformity of ALD single layer. Since the inner surface ALD is in showerhead mode, it is believed that the uniformity of inner coating is good, and the spectral deviation is mainly caused by the outer ALD film in cross-flow mode. In the case of BR coating, it is difficult to determine the non-uniformity of the inner and outer films separately by the transmission measurement. So, assuming that the inner coating is ideally uniform, reverse engineering calculation shows that the non-uniformity of TiO2 on the outer surface is ∼±15% (with uniform Al2O3) or ∼±12% (assuming Al2O3 non-uniformity ±2%), which is much the same of the results of single layer experiments in cross-flow reactor.
Fig. 8. Quantified deviation of spectra of glass bowl with double-sided coated BR film at different positions, (a): along the circumference, (b): along the depth.
3.3 BBAR coating
Due to different gas flow modes of ALD, the thickness of the inner and outer films on glass bowl is different. In this situation, the non-uniformity of the inner and outer coatings can be calculated out based on the change of antireflective spectral range of the double-sided coated glass bowl. So, BBAR coating is suitable to study the uniformity of double-sided coatings.
Due to the optical loss of the glass bowl, the measured transmittance of the glass bowl is lower than that of non-loss substrate. So, in the experiment of BBAR coating on glass bowl, B270 planar glass was used to deposit the same film on one side, and the measured transmittance spectrum was used to correct the transmittance of the film on glass bowl. Figure 9 depicts the transmittance of B270 with single-sided BBAR coating. The peak value of measured spectrum is very close to the ideal antireflection coating (none residual reflection at all wavelengths). The measured spectrum is a little different from the originally designed spectrum. Applying reverse engineering, it is found that the deposited first layer is thinner than design [19]. By changing the thickness of the first layer from 9.8 nm to 7.8 nm, and also the reference wavelength from 400 nm to 395 nm, the spectrum of modified design is close to the measured. As showed in Fig. 9, the average transmittance of 400∼700 nm is greater than 95%, which means that the residual reflectance of the BBAR film is less than 0.5%, and the residual reflectance around 425 nm and 640 nm is close to 0%.
Similar to the previous experiment of BR coating, we also measured the transmittance spectra of the glass bowl at different positions along the circumference and the depth at normal incident angle, as shown in Fig. 10 and Fig. 11.
Fig. 10. Measured transmittance spectra (AOI=0°) of glass bowl with double-sided BBAR coatings at different positions along the circumference. The measured transmittance of uncoated glass bowl and standard deviation were derived from spectra at multiple positions along the circumference.
Fig. 11. Measured transmittance spectra (AOI=0°) of glass bowl with double-sided BBAR coatings at different positions along the depth. The measured transmittance of uncoated glass bowl and standard deviation were derived from spectra at multiple positions along the circumference.
As shown in Fig. 9, the peak transmittance of the film is close to ideal one (with zero residual reflectance). In order to quantify the deviation of the spectra at different positions, we normalized all the measured spectra of coated glass bowl by setting the peak transmittance as 100%. Since thicker film induces red shift of spectrum and the inner coating is thicker than the outer coating, it is reasonable to determine that the shorter wavelength at 95% transmittance is approximately proportional to the thickness of the inner coating and the longer wavelength at 98% transmittance is approximately proportional to the thickness of the outer coating. Then, these two wavelengths were used to characterize spectral position of the inner and outer coatings. The spectral deviation of the BBAR coating on both surfaces of the bowl is shown in Fig. 12. The non-uniformity of the inner and outer coating is ±0.55% and ±1.01% along the circumference, and ±0.39% and ±2.4% along the depth. The little spectral deviation of inner coating verifies good uniformity in showerhead mode. The non-uniformity of the outer coating is due to ALD TiO2 layer in cross-flow mode. The non-uniformity of TiO2 is ±5.5% estimated by reverse engineering assuming that Al2O3 and SiO2 layers are uniform. Compared to that of BR coating, TiO2 with Al2O3 interlayer is more uniform.
Fig. 12. Quantified spectral deviation of the BBAR coating on both surfaces of the bowl at different positions, (a): along the circumference, (b): along the depth.
The transmittance spectra along the circumference of the glass bowl mouth were averaged and normalized to remove the light loss caused by the glass bowl. The averaged and normalized spectrum was compared with spectra of two design schemes, as shown in Fig. 13. In both design schemes, the thickness of the first layer was corrected. In the first design scheme, the same reference wavelength 400 nm was adopted for double-sided films. The calculated spectrum is quite different from the measured spectrum, and shows a wider antireflection range. In the other design scheme, the reference wavelength of inner film is 406 nm, and the outer one is 375 nm. As shown in Fig. 13, the calculated spectrum after modifying the reference wavelength is close to the measured one. It means that around the mouth of the bowl, the outer film is 7.6% thinner than the inner film. As mentioned above, the thickness of the outer film gradually decreases along the depth. So, the outer film is about 10.0% thinner than the inner film on average. This result is consistent with single-layer experiment results that the growth rate in cross-flow reaction mode is about 10% lower than that in showerhead reaction mode.
Fig. 13. Normalized measured spectrum of glass bowl with double-sided BBAR coatings, compared with two types of film design with reference wavelength modified or not.
After calculation and analysis, the total non-uniformity of the coating on glass bowl was listed in Table 3. These results are obtained based on the ALD coating on large-size glass bowl, which is comparable in size with the reaction chamber. The major thickness deviation is caused by the change of ALD gas flow mode. For a small-sized substrate, the gas flow mode should be considered same, and the total non-uniformity of double-sided coating on the substrate should be much better than that in this work. In addition, if many small-sized substrates are placed in the reactor, the difference between individuals should be similar to the results of this work, because the gas flow mode may be different at different positions.
Table 3. The non-uniformity of BBAR film on inner and outer surfaces of glass bowl.
4. Conclusion
Based on ALD technology, the simultaneous coatings on both sides of large-size and strongly curved glass bowl were performed in a self-built reaction chamber. Two types of optical film coatings were demonstrated. The BR coating on the glass bowl shows uniform color at different positions. The BBAR film of 400∼700 nm wavelength has an averaged residual reflectance less than 0.5%. The results of spectral measurement on multiple positions show that the non-uniformity of the inner BBAR film deposited in showerhead ALD mode is less than ±0.7%, and the non-uniformity of the outer BBAR film deposited in cross-flow ALD mode is less than ±3%. The non-uniform growth of TiO2 is the main reason for the non-uniformity of the optical film. Al2O3 interlayer helps to improve the uniformity of TiO2 layer and the BBAR film. The gas flow mode of the precursor evidently affects the growth rate of ALD, resulting in 10% different thickness of the film on the inner and outer surfaces of the glass bowl. These results are useful for the improvement of optical film design and ALD process. In the aspect of film design, more attention should be paid to the tolerance of reference wavelength to deal with the thickness difference caused by gas flow mode. In the aspect of ALD process, the design of the reaction chamber is the key to improve the uniformity of ALD film, and it is important to ensure the same gas flow mode for all surfaces.
Funding
National Natural Science Foundation of China (61805267, 61775042, 61605089, U1737111, 61975223); Science and Technology Commission of Shanghai Municipality (18ZR1445400, 18ZR1446000); Youth Innovation Promotion Association of the Chinese Academy of Sciences (2019241); Hundred Talents Program of the Chinese Academy of Sciences (20181214).
Disclosures
The authors declare no conflicts of interest.
References
1. H. A. Macleod, Thin-Film Optical Filters, 3rd ed. (Institute of Physics, Bristol, UK, 2001).
2. H. K. Pulker, “Optical coatings deposited by ion and plasma PVD processes,” Surf. Coat. Technol. 112(1-3), 250–256 (1999). [CrossRef]
3. C. D. Liu, M. D. Kong, C. Guo, W. D. Gao, and B. C. Li, “Theoretical design of shadowing masks for uniform coatings on spherical substrates in planetary rotation systems,” Opt. Express 20(21), 23790–23797 (2012). [CrossRef]
4. M. Gross, S. Dligatch, and A. Chtanov, “Optimization of coating uniformity in an ion beam sputtering system using a modified planetary rotation method,” Appl. Opt. 50(9), C316–C320 (2011). [CrossRef]
5. Z. X. Wang, P. R. West, X. G. Meng, N. Kinsey, V. M. Shalaev, and A. Boltasseva, “Angled physical vapor deposition techniques for non-conformal thin films and three-dimensional structures,” MRS Commun. 6(1), 17–22 (2016). [CrossRef]
6. P. Kelkar, B. Tirri, R. Wilklow, and D. Peterson, “Deposition and characterization of challenging DUV coatings,” Advances in Thin-Film Coatings for Optical Applications V7067, (2008).
7. M. Baba, T. Sasaki, M. Yoshita, and H. Akiyama, “Aberrations and allowances for errors in a hemisphere solid immersion lens for submicron-resolution photoluminescence microscopy,” J. Appl. Phys. 85(9), 6923–6925 (1999). [CrossRef]
8. F. Menna, E. Nocerino, F. Fassi, and F. Remondino, “Geometric and Optic Characterization of a Hemispherical Dome Port for Underwater Photogrammetry,” Sensors 16(1), 48 (2016). [CrossRef]
9. K. Pfeiffer, U. Schulz, A. Tunnermann, and A. Szeghalmi, “Antireflection Coatings for Strongly Curved Glass Lenses by Atomic Layer Deposition,” Coatings 7(8), 118 (2017). [CrossRef]
10. S. M. George, “Atomic Layer Deposition: An Overview,” Chem. Rev. 110(1), 111–131 (2010). [CrossRef]
11. R. L. Puurunen, “Surface chemistry of atomic layer deposition: A case study for the trimethylaluminum/water process,” J. Appl. Phys. 97(12), 121301 (2005). [CrossRef]
12. D. Riihela, M. Ritala, R. Matero, and M. Leskela, “Introducing atomic layer epitaxy for the deposition of optical thin films,” Thin Solid Films 289(1-2), 250–255 (1996). [CrossRef]
13. V. Miikkulainen, M. Leskela, M. Ritala, and R. L. Puurunen, “Crystallinity of inorganic films grown by atomic layer deposition: Overview and general trends,” J. Appl. Phys. 113(2), 021301 (2013). [CrossRef]
14. S. Shestaeva, A. Bingel, P. Munzert, L. Ghazaryan, C. Patzig, A. Tunnermann, and A. Szeghalmi, “Mechanical, structural, and optical properties of PEALD metallic oxides for optical applications,” Appl. Opt. 56(4), C47–C59 (2017). [CrossRef]
15. M. Ylilammi, “Monolayer thickness in atomic layer deposition,” Thin Solid Films 279(1-2), 124–130 (1996). [CrossRef]
16. Z. C. Liu, S. L. Chen, P. Ma, Y. W. Wei, Y. Zheng, F. Pan, H. Liu, and G. Y. Tang, “Characterization of 1064 nm nanosecond laser-induced damage on antireflection coatings grown by atomic layer deposition,” Opt. Express 20(2), 854–863 (2012). [CrossRef]
17. A. Szeghalmi, M. Helgert, R. Brunner, F. Heyroth, U. Gosele, and M. Knez, “Atomic layer deposition of Al2O3 and TiO2 multilayers for applications as bandpass filters and antireflection coatings,” Appl. Opt. 48(9), 1727–1732 (2009). [CrossRef]
18. K. Pfeiffer, S. Shestaeva, A. Bingel, P. Munzert, L. Ghazaryan, C. van Helvoirt, W. M. M. Kessels, U. T. Sanli, C. Grevent, G. Schutz, M. Putkonen, I. Buchanan, L. Jensen, D. Ristau, A. Tunnermann, and A. Szeghalmi, “Comparative study of ALD SiO2 thin films for optical applications,” Opt. Mater. Express 6(2), 660–670 (2016). [CrossRef]
19. Y. H. Li, W. D. Shen, Y. G. Zhang, X. Hao, H. H. Fan, and X. Liu, “Precise broad-band anti-refection coating fabricated by atomic layer deposition,” Opt. Commun. 292, 31–35 (2013). [CrossRef]
20. D. M. Fryauf, A. C. Phillips, M. J. Bolte, A. Feldman, G. S. Tompa, and N. P. Kobayashi, “Scaling Atomic Layer Deposition to Astronomical Optic Sizes: Low-Temperature Aluminum Oxide in a Meter-Sized Chamber,” ACS Appl. Mater. Interfaces 10(48), 41678–41689 (2018). [CrossRef]
21. K. Pfeiffer, L. Ghazaryan, U. Schulz, and A. Szeghalmi, “Wide-Angle Broadband Antireflection Coatings Prepared by Atomic Layer Deposition,” ACS Appl. Mater. Interfaces 11(24), 21887–21894 (2019). [CrossRef]
22. M. Ritala, M. Leskela, E. Nykanen, P. Soininen, and L. Niinisto, “Growth of Titanium-Dioxide Thin-Films by Atomic Layer Epitaxy,” Thin Solid Films 225(1-2), 288–295 (1993). [CrossRef]
