With the development and construction of high peak power lasers it has become more and more important to improve the laser-induced damage threshold (LIDT) of optical coatings. In this paper, ZrO2 coatings were deposited by sol-gel dip-coating method and further treated by conventional furnace annealing (CFA) and rapid thermal annealing (RTA) at different temperatures. By measuring the Raman spectra, optical constants and LIDT, the influence of annealing on the crystal structures, refractive indices, laser-induced damage characters of ZrO2 coatings were analyzed. The results show that RTA is effective in tuning the crystal structures of ZrO2 coatings. Lattice mismatch between monoclinic and tetragonal phases happened on CFA treated film reduces its refractive index, hence the film annealed by RTA at 800 °C realizes a higher refractive index. Compared with CFA annealed films, RTA annealed films were no more susceptible to laser damage due to their crystal structure difference caused lager band gap.
©2012 Optical Society of America
The power handling capability of optical materials and coatings was often encountered as a major bottleneck, which limits scaling up high power laser systems for future laser fusion technology . Due to the high LIDT, wide spectral region of transparency, as well as high chemical and thermal stabilities, sol-gel ZrO2 is widely used as high refractive index material in multilayer optical filters, such as highly reflective coatings  and antireflective coatings  in high power laser system. A number of work including polyvinyl pyrrolidone modification  and hydrothermal synthesis  have been done to improve the laser resistance of these coatings. However, these modified ZrO2 film exhibits an organic-inorganic composition that resulted in a decrease of refractive index. Thus, to obtain ZrO2 films for common optical application, the as-deposited film usually experiences a thermal annealing procedure to realize high refractive index. However, Liang et al  found that the LIDT of high temperature annealed samples decreased significantly. Therefore, it is crucial to reduce the annealing-induced attenuation in LIDT of sol-gel ZrO2 films.
It was known that the annealing kinetics was the most important parameter to control in order to obtain decent optical performances and high LIDT coatings . In this work, we applied RTA technique on the thermal treatment of the as-deposited sol-gel ZrO2 films to control the crystal structures. The optical properties, crystal structures and laser resistance capability of the coatings treated by different thermal annealing techniques are then investigated and discussed.
The ZrO2 sol was prepared using zirconium butoxide (Zr(OC4H9)4, TBOZ) as precursor. TBOZ was mixed with anhydrous ethanol (EtOH), distilled water, acetylacetone (CH3COCH2COCH3, AcAc) and acetic acid (CH3COOH, HAc) for 2 h. The molar ratio was TPOZ: EtOH: H2O: HAc: AcAc = 1:15:3:1.6:0.2. The mixture was finally aged in stable environment (with humidity lower than 30% and temperature of 20 ~25 °C) for 72 h.
The synthesized sol was deposited on the well-cleaned fused silica and silicon wafer substrates by dip-coating method. All the films in the paper are single dip-coated on the substrates. The as-deposited films were firstly pretreated at 100 °C for 1h, and then heat-treated by CFA or RTA at different temperatures ranging from 200 ~800 °C. In CFA, the films were annealed in muffle furnace for 2 h. For RTA, the samples were heated in a rapid thermal processor (AG Associates Heatpulse 610) in air atmosphere. In each of the annealing temperature, the coating was firstly heated to designated temperature with a ramp-up rate of 15 °C/s and then stayed for 60 s. Finally the sample was cooled down through nitrogen gas and watercooling system at a ramp-down rate of 5 °C/s.
Raman spectra were recorded in the backscattering geometry using a Jobin-Yvon micro-Raman apparatus (HR-800) equipped with a 30 mW He-Cd laser (Kimmon Koha, IK3301R-G) emitting at 325 nm and a microscope (Olympus BX41). An edge filter was used in the Raman setup to block the Rayleigh scattering light and stray laser bands. The laser beam irradiating the sample was attenuated to below 0.1 mW in order to avoid laser-induced heating. Before each measurement, a silicon wafer was used for the wavenumber calibration of the spectrometer. The transmittance and reflectance spectra of the films were measured in the 400 ~1000 nm region using a Jasco-V570 UV-VIS-NIR double beam spectrometer. The LIDT of the samples were measured in the “1-on-1” regime according to ISO standard 11254-1.2 . A Q-switched Nd: YAG single mode laser with 1064 nm wavelength and 10 ns pulse duration was used as the pump source. Damage was determined if any visible change was observed by an in situ microscope (magnification from 25 × to 1000 × ). The LIDT was defined as the maximum incident pulse’s energy density for which the possibility of damage is 0%. The experimental setup and the detailed laser parameters of the laser damage threshold measurement can be seen in Ref .
3. Results and discussion
Raman spectroscopy, as a fast and nondestructive characterization tool, is extensively applied to the identification of crystal structures. Figure 1(a) shows the Raman spectra of ZrO2 films annealed by CFA at different temperatures. The first spectrum curve corresponds to the 300 °C annealed zirconia film, which is essentially amorphous. When the zirconia film is heated to 400 °C, some broad lines at 271, 466, 643 cm−1 develop which correspond to the formation of tetragonal zirconia  (t-ZrO2). For temperature up to 500 °C, these Raman peaks get sharper and shift towards lower frequencies. Meanwhile, the Raman modes at 179, 190, 222, 333, 347, 381, 476, 556, 616 cm−1 that come from the monoclinic zirconia  (m-ZrO2) are also observed, which means that the films were crystallized into a mixed phases of tetragonal and monoclinic. With increasing annealing temperature, the Raman peaks of the monoclinic phase further get stronger and sharper while that of tetragonal phase get weaker, indicating the transformation of tetragonal to monoclinic. However, the ZrO2 film still co-exists with two phases until 800 °C. Due to the high penetration depth of the laser line, the very strong peaks at 520.7, 320 cm−1 which come from the LO phonon and overtone of 2TA(X)  of silicon wafer are also visible in the Raman spectra.
The Raman spectra of ZrO2 thin films obtained by RTA are shown in Fig. 1(b). It can be recognized from the Raman spectra that crystallization does not start until the temperature reaches to 500 °C. The appeared Raman modes at 274, 461, 647 cm−1 are related to t-ZrO2. The peak sharpening after increasing the annealing temperatures reveals the better crystallization of tetragonal zirconia. At 800 °C, still no monoclinic zirconia phase is detected and only tetragonal zirconia is observed. Based on these results, we successfully tuned the crystal structure of ZrO2 films through RTA.
The optical constants of ZrO2 films were determined through fitting simultaneously the theoretical transmittance and reflectance curves to experimental ones using Cauchy model over the 400 ~1000 nm region. To judge the quality of the fitting, the root mean squared error (RMSE) was defined by:11]. A representative fitting data of 300 °C RTA annealed film is shown in Fig. 2 . The experimental and calculated values look in very good agreement over the whole spectrum, which indicates the Cauchy formula can well describe the dispersion relation of the ZrO2 film in 400~1000 nm region. The thickness and optical constants of the film at 400~1000 nm are then determined. The calculated thickness and RMSE parameters of all the films are listed in Table 1 . All the RMSE values are less than 1, which ensure the reliability of the determined optical constants.
The determined refractive indices at He-Ne laser wavelength of 633 nm are chosen and plotted in Fig. 3 to study the variation during annealing. In CFA, we can see that the amorphous zirconia films have low refractive indices compared with those of the crystallized films. The refractive index of the 200 °C annealed film is 1.675, which is approximately same as that reported by Guo et al  and Gueto et al . As a result of decomposing and evaporating the organic groups with increasing annealing temperatures, the porous structure of the amorphous films is destroyed and hence the packing density became larger which then increases the refractive index distinctly. After annealing at 400 °C, the film is crystallized to tetragonal phase and shows a refractive index of 1.870. In addition, the refractive index first decreases and then increases with the watershed temperature of 500 °C (as shown in red circle part of Fig. 3). Such a behavior is different from that reported by Liu et al  and Liang et al . This anomalous evolution can be explained by the phase transformation of zirconia. After being annealed at 500 °C, the film is composed of t-ZrO2 and m-ZrO2 mixed phases from Raman analysis. The lattice mismatch between the new monoclinic phase and parent tetragonal phase increases the porosity of the film. This porosity increase is larger than the decrease induced by thermal densification, which in turn then reduces its refractive index. Increasing annealing temperature (600 ~800 °C) promotes the densification of the film and further transformation of t-ZrO2 to m-ZrO2, which then results in the increase of the refractive index continuously again. The thickness of the coatings (as shown in Table 1) decreased from 198.3 nm (annealed at 200 °C) to 91.0 nm (annealed at 800 °C). And it is found the decrease between 200 °C and 400 °C is larger than that between 400 °C and 800 °C, indicating the densification of amorphous ZrO2 is larger than that of crystallized ZrO2 films. This is consistent with the refractive index change during annealing. At 500 °C, the lattice mismatch induced porosity change increased the refractive index of the film, but did not expand the volume of the film. Therefore, the thickness still slightly decreased due to the thermal densification induced by temperature increase.
Also, it is noticeable from Fig. 3 that different annealing procedure caused a significant change in refractive index variation of the samples during annealing. From 200 to 400 °C by RTA, the residual organic segments have a low refractive index and also preserve the film with a high porosity, hence make the film a low refractive index. Thermally induced densification of the film only results in a slight increase of refractive index from 1.575 to 1.594. After annealed at 500 °C, the pores left by the removal of organic segments decrease the refractive index while crystallization increases it. Therefore, the refractive index of the film only increases to 1.606. Upon annealing at higher temperatures, the densification and increased crystallinity of the films give rise to a rapid increase of the refractive index, from 1.606 to 1.929 between 500 and 800 °C. Only tetragonal phase is present for the RTA treated films in the whole annealing process. Thus, the refractive index grows up consecutively. Correspondingly, the thickness of the films (Table 1) decreased continuously with the increase of the annealing temperature. The decrease of the thickness between 200 and 500 °C is slower than that between 500 and 800 °C, which is consistent with the variance of refractive index.
The above-mentioned decrease of the refractive index induced by the lattice mismatch can also be explained from the films composed of different crystal phases at the same temperature. Compared with CFA treated films, the refractive index of the film annealed at 800 °C by RTA is 1.929, which is larger than that of treated by CFA (n = 1.898). Considering that in RTA process, the film annealed at 800 °C is tetragonal phase, which would not have the lattice mismatch induced porosity change that happened in CFA treated films. Therefore, the refractive index in RTA is larger than that of CFA.
Figure 4 shows the LIDT values of the samples, which correspond to the highest energies at which no damage observed. We can see that the amorphous coating (200 °C annealed) exhibits the highest LIDT of 21 J/cm2. This can be mainly attributed to the large optical band gap and network structure that permits a sufficiently fast energy relaxation within the films. After annealing at 400 °C, the porous network structure of the coating is destroyed, and simultaneously a rapid densification and an increase in the tensile stress of films occurs. Therefore, after such a densification, the absorbed laser energy must be restricted within a much smaller region on the tested films that results in the sharp decrease of LIDT from 21 to 11.6 J/cm2. With a further increase in the annealing temperature to 800 °C, the crystal transformation induced monotonous decrease in band gap  and the gradually increased internal stress lead to a further decrease in LIDT of the films.
Here, one may doubt that why we don’t use low temperature annealed amorphous ZrO2 coating that has a higher LIDT. However, we have to interpret that according to the principle of thin film optics , the higher refractive index it has, the less layer numbers it needs for the multilayer optical filters, which will simplify the coating process and also reduce the potential laser induced damage possibility to the coating filters. To our surprise, it is found that after crystallization, the coatings treated by RTA always have a higher LIDT than CFA treated films. Especially at 800 °C, the coating annealed by RTA not only achieves a higher refractive index than CFA annealed coating, but also has a significantly improved LIDT value of 13.8 J/cm2. Meanwhile, the LIDT of coating treated with the same temperature by CFA is only 4.2 J/cm2. One apparent reason for this huge difference of LIDT is the crystal structures difference. Although the complexity of the mechanisms that cause laser damage to an optical coating makes it too difficult to establish a quantitative correlation, there is still a generally accepted viewpoint that materials with larger band gap usually exhibits weaker absorption and then a relatively high LIDT . In our work, 800 °C RTA annealed coating still presents as t-ZrO2 structure, while that of CFA annealed sample exhibits t-ZrO2 and m-ZrO2 mixed crystal structure. This crystal structure difference results a larger band gap of RTA annealed coatings thereby an improved LIDT. In addition, the LIDT value of 800 °C RTA annealed ZrO2 is also higher than that of the films deposited by e-beam evaporation (8.0 J/cm2) and by sol-gel process (11.0 J/cm2) .
In summary, we studied the crystal structures, refractive indices and LIDT of sol-gel ZrO2 coatings annealed by CFA and RTA. It was found that the evolution of crystal structures and refractive indices of ZrO2 coatings treated by RTA are completely different from that of CFA annealed coatings. The RTA treated ZrO2 coatings exhibit higher LIDT values compared with the CFA annealed samples. This improvement of the laser resistance of ZrO2 films shows great importance for the further development of high power performance.
We thank Prof. Jingxin Yang for the LIDT measurements. This work was financially supported by National Natural Science Foundation of China (No.11074189, U1230113).
References and links
1. X. Wang and J. Shen, “A review of contamination-resistant antireflective sol-gel coatings,” J. Sol-Gel Sci. Technol. 61(1), 206–212 (2012). [CrossRef]
2. Q. Zhang, X. Li, J. Shen, G. Wu, J. Wang, and L. Chen, “ZrO2 thin films and ZrO2/SiO2 optical reflection filters deposited by sol-gel method,” Mater. Lett. 45(6), 311–314 (2000). [CrossRef]
3. H. Li, H. Xiong, Y. Tang, and L. Hu, “ZrO2/SiO2 Two-Layer Antireflective Coatings by Sol-Gel Process,” Rare Metal Mat. Eng. 39, 145–148 (2010).
4. H. Li, H. Xiong, and Y. Tang, “Study on the laser-induced damage threshold of sol-gel ZrO2-PVP coatings,” Chin. Opt. Lett. 8(2), 241–243 (2010). [CrossRef]
6. L. Liang, Y. Xu, L. Zhang, Y. Sheng, D. Wu, and Y. Sun, “Annealing effect on the optical properties and laser-induced damage resistance of solgel-derived ZrO2 films,” J. Opt. Soc. Am. B 24(5), 1066–1074 (2007). [CrossRef]
7. ISO 11254–1.2, “Lasers and laser-related equipment: Determination of laser-induced damage threshold of optical surfaces-Part 1: 1-on-1 test” (2000).
8. J. Hu, J. Yang, W. Chen, and C. Zhou, “Experimental investigation of enhancing the subsurface damage threshold of Nd-doped phosphate glass,” Chin. Opt. Lett. 6(9), 681–684 (2008). [CrossRef]
9. A. Kuwabara, T. Tohei, T. Yamamoto, and I. Tanaka, “Ab initio lattice dynamics and phase transformations of ZrO2,” Phys. Rev. B 71(6), 064301 (2005). [CrossRef]
10. X. Wang, J. Shen, and Q. Pan, “Raman spectroscopy of sol-gel derived titanium oxide thin films,” J. Raman Spectrosc. 42(7), 1578–1582 (2011). [CrossRef]
11. M. G. Sandoval-Paz and R. Ramírez-Bon, “Analysis of the early growth mechanisms during the chemical deposition of CdS thin films by spectroscopic ellipsometry,” Thin Solid Films 517(24), 6747–6752 (2009). [CrossRef]
12. Y. J. Guo, X. T. Zu, X. D. Jiang, X. D. Yuan, S. Z. Xu, B. Y. Wang, and D. B. Tian, “Experimental research of laser-induced damage of the monolayer ZrO2 PVD and sol-gel films,” Opt. Laser Technol. 40(5), 677–681 (2008). [CrossRef]
13. L. F. Cueto, E. Sánchez, L. M. Torres-Martínez, and G. A. Hirata, “On the optical, structural, and morphological properties of ZrO2 and TiO2 dip-coated thin films supported on glass substrates,” Mater. Charact. 55(4-5), 263–271 (2005). [CrossRef]
14. W.-C. Liu, D. Wu, A.-D. Li, H.-Q. Ling, Y.-F. Tang, and N.-B. Ming, “Annealing and doping effects on structure and optical properties of sol-gel derived ZrO2 thin films,” Appl. Surf. Sci. 191(1-4), 181–187 (2002). [CrossRef]
15. H. Jiang, R. I. Gomez-Abal, P. Rinke, and M. Scheffler, “Electronic band structure of zirconia and hafnia polymorphs from the GW perspective,” Phys. Rev. B 81(8), 085119 (2010). [CrossRef]
16. H. A. Macleod, Thin-Film Optical Filters, 3rd ed. (Institute of Physics Publishing, Bristol and Philadelphia, 2001).
17. D. Ristau, M. Jupé, and K. Starke, “Laser damage thresholds of optical coatings,” Thin Solid Films 518(5), 1607–1613 (2009). [CrossRef]
18. Y. Jian-Ke, L. Hai-Yuan, F. Zheng-Xiu, T. Yong-Xing, J. Yun-Xia, Z. Yuan-An, H. Hong-Bo, and S. Jian-Da, “Comparison of TiO2 and ZrO2 films deposited by electron-beam evaporation and by sol-gel process,” Chin. Phys. Lett. 24(7), 1964–1966 (2007). [CrossRef]