Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Annealing effects on microstructure and laser-induced damage threshold of quasi-rugate filters

Open Access Open Access

Abstract

Ta2O5/SiO2 quasi-rugate filters with high damage thresholds were deposited by ion-beam sputtering and then annealed at temperature of 200-800°C. The relations between microstructure, optical properties, chemical composition, weak absorption, and laser-induced damage threshold (LIDT) were studied. It was found that the transmittance spectra shifted to short wavelength as the annealing temperature increased. Three evolution courses of the films in the annealing process were analyzed by Atomic Force microscopy (AFM), Zygo interferometer measurement and Focused Ion Beam microscope (FIB). The decreased weak absorption during annealing process was found with significant effect on the LIDT. As the annealing temperature increased to 600°C, the weak absorption of films decreased from 39.99 to 7.2 ppm and the 50%-LIDTs increased from 59.32 to 158.87J/cm2. Distinct damage micrographs of the films annealed at different temperature were obtained. A combination of substoichiometric defect and structural defect dominant description was used to illustrate the aggravation of laser-induced damage.

© 2016 Optical Society of America

1. Introduction

Rugate mirror is a specific high reflector in rugate notch filter design. It exhibits the property of reflecting a narrow spectral region and transmitting all others. The continuous changing of refractive index is very difficult to be realized [1]. For convenience of technical realization, quasi-rugate filter through a stepped index profile is a suitable approach [2,3].

Ta2O5 is one of the most important high index materials for optical coatings with high damage thresholds. However, Ta2O5 film is liable to form substoichiometry in the deposition process, thereby annealing is regarded as an important method for achieving better stoichiometry [4,5]. Moreover thermal annealing is often required to further improve the mechanical, electrical or optical properties by structural relaxation. The relaxing thermal diffusion mechanism is more efficient as the annealing temperature increases, but this process is limited by the risk of degradation of the amorphous material properties because of crystallization [6–8].

To our knowledge, there was no specific study on laser-induced damage behavior of co-sputtered Ta2O5 and SiO2 rugate filters. In this paper, focus was placed on obtaining Ta2O5/SiO2 rugate filters with high damage thresholds. The effect of high temperature annealing on the laser-induced damage threshold was reported. A combination of substoichiometric defect and structural defect dominant description was used to illustrate the aggravation of laser-induced damage.

2. Experimental procedure

The rugate design was acquired and optimized by a thin film design software developed by Laser Zentrum Hannover e.V. Ta2O5/SiO2 rugate filters were prepared by ion-beam sputtering using two metallic targets of pure silicon and pure tantalum. The mixture ratio was controlled by the relative position of targets in the ion source beam spot, as shown in Fig. 1(a). Furthermore, the Lorentz-Lorenz model and a cosn-distribution of the sputtered particle flux were used to simulate the refractive indexes as a function of different target positions, as shown in Figs. 1(b) and 1(c). The Lorentz-Lorenz model was described in Eq. (1):

 figure: Fig. 1

Fig. 1 (a) Experimental setup for the deposition of Ta2O5/SiO2 rugate filters. (b) The refractive indexes as a function of different target positions. (c) Relationship between refractive index and volume ratio of Ta2O5.

Download Full Size | PDF

n21n2+2=fana21nb2+2+fbnb21nb2+2

In the expression above, na、nb were the refractive indexes of the two pure materials. The Lorentz-Lorenz model was suit for materials consisting of induced atomic point dipoles in vacuum.

Annealing processes of the films were performed in air at 200, 400, 600 and 800°C for 4 h, respectively. Transmittance spectra of the samples were measured using a Lambda 950 spectrophotometer. The damage performances of samples were investigated via S-on-1 tests. The beam (5.5 ns FWHM) generated from an Nd: YAG laser was lens-focused to form a spot of 0.67 mm 1/e diameter on the sample plane. In our experiment, the incidence angle at sample surface was slightly off (3~5 degrees) normal so as to avoid additional effect from the surface reflection. For S-on-1 tests, over 200 sites were chosen with an interval of 2 mm on each sample. Each site was irradiated with 200 pluses at certain energy and its surface status was recorded by online detection unit. Every site was examined by Nomarski microscope after irradiation to check whether damage happened or not. Weak absorption was measured by Laser Calorimetry according to ISO 11551 [9]. A set of NTC thermistor was employed to perform an absolute temperature measurement. A reflecting mirror triggered by computer was used as a shutter to block the beam. With this configuration it was possible to detect the increase of temperature well below 1 mK.

3. Results and discussion

The refractive index and electric field distribution of the Ta2O5/SiO2 rugate filters were shown in Fig. 2. It was found that the surface electric field intensity was reduced to 0 by the optimization of refractive indexes and thickness values.

 figure: Fig. 2

Fig. 2 Refractive index and electric field distribution of the Ta2O5/SiO2 rugate filters.

Download Full Size | PDF

The transmittance spectra of the as-deposited and annealed films at different temperatures were illustrated in Fig. 3. The spectral shifts of the as-deposited and annealed films were shown in Fig. 3(b). It was found that the spectra shifted to short wavelength as the annealing temperature increased. Previous work revealed that the “blue shift” was due to the increase of refractive indexes (nTa2O5) [10]. The refractive index increased with the increase of annealing temperature, which could be attributed to the decrease of structural defects affected by the thermal modification during annealing. The relation between the refractive index nf and the packing density P was illustrated in Eq. (2) [11]:

 figure: Fig. 3

Fig. 3 (a) Transmittance spectra of the as-deposited and annealed films at different temperatures. (b) Blue shifts of the as-deposited and annealed films. (c) Reflectances of the as-deposited and annealed films at 1064nm.

Download Full Size | PDF

nf=Pns+(1P)nν

Where ns and nv were the refractive indexes of the solid part of the film and the vacancies, respectively.

The reflectances of quasi-Rugate filters achieved above 99.8% at 1064nm, as shown in Fig. 3(c). Meanwhile, the quasi-Rugate filter had a weaker second harmonic peak and narrower stopband.

Laser Calorimetry was a sensitive method for detecting the thermal absorption of films. It was worth noting that the weak absorption was affected by the combination of defect absorption and microstructure of films. The weak absorption data measured by Laser Calorimetry was shown in Fig. 4.

 figure: Fig. 4

Fig. 4 Weak absorptions of the as-deposited and annealed films at different temperatures: (A), (B) and (C) represent the evolution of the films in the annealing process individually.

Download Full Size | PDF

Ion beam sputtering (IBS) deposition had proven itself to be an excellent method for depositing low scattering coatings, because it involved an atom-by-atom (or molecule-by-molecule) transport process of high-energy materials in a relatively low oxygen pressure and low temperature environment. However, IBS was a slow deposition process and might produce oxide films with stoichiometric problems, especially when the targets were metal materials (such as Ta, Si). In the course of annealing below 400°C, oxygen diffused and reacted with the non-oxidized Ta inside the film. The Gibbs free energy of formation (ΔGo) for this reaction was described in Eq. (3) [12]:

(45)Ta+O2(25)Ta2O5ΔGo=196+0.04(kJ/mol)

As ΔGo<0, it was favorable for the reaction to occur based on thermodynamical arguments. Therefore, the annealing treatment was responsible for the repair of both substoichiometric defects and microstructure defects. When the annealing temperature increased to 400°C, the O/Ta ratio reached the theoretical value of 2.50 and the substoichiometric defects disappeared [13]. Meanwhile, the AFM (3D) patterns of quasi-rugate films unannealed and annealed at 400°C were shown in Fig. 5(a), respectively. It was found from the microstructure that the annealing treatment could be conducive to obtain good surface topography with good grain growth and smooth surface morphology. The film annealed at 600°C achieved the minimum absorption of 7.2 ppm, as shown in Fig. 4. The weak absorption of the film annealed at 600°C decreased a litter bit, compared with that of the film annealed at 400°C. These phenomena could be due to film-stress-variation-induced surface microstructural evolution. The curvature radii of quasi-rugate films were obtained by Zygo interferometer measurements, as shown in Fig. 5(b). From the curvature of the substrate R, the residual stresses in the deposited thin films were evaluated by the formula:

σ=Es6(1υs)ts2tf(1R21R1)
where Es/(1υs)=Es was the substrate biaxial modulus, Es and νs were Young,s modulus and Poisson ratio, ts was the substrate thickness, tf was the film thickness. R1 and R2 were the radii of curvature of substrates measured by means of an optical interferometer before and after deposition of films, respectively. According to mathematical formula (4), the residual stress of the film changed from the compressive stress (σ400 = −75.6 MPa) to the tensile stress (σ600 = 119.8 MPa) by increasing the annealing temperature.

 figure: Fig. 5

Fig. 5 (A) AFM (3D) images of the quasi-rugate films: (a1) unannealed film and (a2) annealed film at 400°C. (B) The curvature radii of quasi-rugate films obtained by Zygo interferometer measurements: (b1) annealed film at 400 °C and (b2) annealed film at 600°C. (C) Profiles of the quasi-rugate films imaged by FIB/SEM: (c1) annealed film at 600°C and (c2) annealed film at 800°C.

Download Full Size | PDF

The weak absorption of the film annealed at 800°C was 49.39 ppm, which was the largest compared with other annealed films. Previous work revealed that transition from amorphous material (a-Ta2O5) to low temperature (LT) crystal occurred at approximately 650°C [14]. So the increase of absorption was attributed to the scattering loss of grain boundaries and the reduced thermal diffusion resulted from the appearance of grain boundaries in the course of crystal growth. Focused Ion Beam (FIB) microscope was used to characterize the microstructures of the rugate filters. The cross-section electron micrographs of rugate filters were shown in Fig. 5(c). The quartz substrate was at the right side. The crystal growth was found in the upper stack, as shown in Fig. 5(c2).

Laser damage thresholds of Ta2O5/SiO2 rugate filters under multi-pulse irradiation had been tested by S-on-1. The results indicated that the 50%-LIDTs of all samples decreased with increase of pulse number following an exponential decay rule, as shown in Fig. 6. This phenomenon had also been observed in various optical materials in previous studies and there was an empirical formula used to fit the experiment data [15]:

 figure: Fig. 6

Fig. 6 S-on-1 performances of the as-deposited and annealed films at different temperatures.

Download Full Size | PDF

F(N)=Aexp(N/B)+Φ

F(N) was the 50%-LIDT while N was the pulse number and Φ∞, A and B were the fitting parameters. In some cases this formula was used for roughly estimating the lifetime of thin films under multi-pulse irradiation. After a number of pulses, the 50%-LIDTs of all samples were stable. The annealed samples achieved higher 50%-LIDTs than the as-deposited sample as shown in Fig. 6. The 50%-LIDTs increased dramatically when the films were annealed at 400°C and 600°C, which stabilized at 150.37J/cm2 and 158.87J/cm2. The 50%-LIDTs increased about 265% than that of the as-deposited film. However, when the annealing temperature increased further, the LIDTs started to decrease. The LIDTs of films annealed at 800°C were 49.51J/cm2.

The damage morphologies were obtained by Leica-DMRXE Microphoto. The damage morphologies of as-deposited and annealed films at different temperatures were shown in Fig. 7. All the damaged sites of the samples were centered on defect points, which implied that the defect-induced damage mechanism was not altered after annealing. But fewer absorption points which were induced by defects in the films such as substoichiometry and imperfect structure were available in the damage areas of annealed films compared to as-deposited film. This could be proved by the data of weak absorption. When the annealing temperature increased to 600°C, the absorption of film decreased from 39.99 to 7.2 ppm and the LIDTs increased from 59.32 to 158.87J/cm2.

 figure: Fig. 7

Fig. 7 Typical damage morphologies of the as-deposited and annealed films at different temperatures.

Download Full Size | PDF

A combination of substoichiometric defect and structural defect dominant description was used to illustrate the aggravation of laser-induced damage. As the activation energy of substoichiometric defect was only 0.6 eV which was much smaller than the band gap of Ta2O5 (4.3 eV), this defect easily turned into the damage center under laser radiation [16]. The O/Ta ratio of the film annealed below 400°C was substoichiometry. Therefore, the damage initiation of the film was dominated by substoichiometric defects. When the films were annealed in low temperature, the structural defects could not be eliminated. Due to thermal heating induced by laser irradiation, the structural defects formed the “thermal resistance ring” because of lower thermal conductivity compared to that of perfect structure. The formation of thermal resistance ring could accelerate the damage of the films. During annealing at high temperature, the structure of film underwent a modification driven by the thermal energy. It was accompanied with the decrease of the number of voids and the formation of new stable bonds, so the density of structural defects decreased. “Thermal conductance net” which was formed by approximately perfect structure accelerated the dissipation of thermal heating and moderated the damage of the films, as shown in Fig. 7.

During annealing at 800°C, the structure of Ta2O5 transformed from amorphous to crystalline phase, accompanied with the appearance of grain boundary in the crystal growth course. So the thermal dissipation of films was decreased. In this case, the decrease of LIDTs might be due to more grain boundaries induced by higher annealing temperature, as shown in Fig. 8.

 figure: Fig. 8

Fig. 8 Typical damage morphology of annealed film at 800°C.

Download Full Size | PDF

4. Conclusion

In this study, Ta2O5/SiO2 rugate filters with high damage thresholds were prepared by ion-beam sputtering. Results showed that the annealing temperature had significant effects on the optical properties and LIDTs of the Ta2O5/SiO2 rugate filters. The transmittance spectra shifted to short wavelength as the annealing temperature increased. The three evolution courses (Substoichiometric to stoichiometric transformation, Compressive to tensile transformation and Amorphous to Crystalline transformation) were analyzed in the annealing process. The decreased weak absorption during annealing was found having significant effect on the LIDT. When the annealing temperature increased to 600°C, the absorptions of films decreased from 39.99 to 7.2 ppm and the LIDTs increased from 59.32 to 158.87J/cm2. The aggravation of laser-induced damage was due to the combination of substoichiometric defects and structural defects.

References and links

1. W. H. Southwell, “Spectral response calculations of rugate filters using coupled-wave theory,” J. Opt. Soc. Am. A 5(9), 1558–1564 (1988). [CrossRef]  

2. P. Baumeister, “Simulation of a rugate filter via a stepped-index dielectric multilayer,” Appl. Opt. 25(16), 2644–2645 (1986). [CrossRef]   [PubMed]  

3. B. Wu, U. Bartch, M. Jupé, L. Jensen, M. Lappschies, K. Starke, and D. Ristau, “Morphology investigations of laser induced damage,” Proc. SPIE 6403, 640319 (2007). [CrossRef]  

4. J.-P. Masse, H. Szymanowski, O. Zabeida, A. Amassian, J. E. Klemberg-Sapieha, and L. Martinu, “Stability and effect of annealing on the optical properties of plasma-deposited Ta2O5 and Nb2O5 films,” Thin Solid Films 515(4), 1674–1682 (2006). [CrossRef]  

5. X. L. He, J. H. Wu, X. M. Li, X. D. Gao, X. Y. Gan, and L. L. Zhao, “Effects of the post-annealing ambience on the microstructure and optical properties of tantalum oxide films prepared by pulsed laser deposition,” J. Alloys Compd. 478(1–2), 453–457 (2009). [CrossRef]  

6. Y. N. Zhao, Y. J. Wang, H. Gong, J. D. Shao, and Zh. X. Fan, “Annealing effects on structure and laser-induced damage threshold of Ta2O5/SiO2 dielectric mirrors,” Appl. Surf. Sci. 210(3–4), 353–358 (2003). [CrossRef]  

7. Y. S. Kim, S. H. K. Park, J. Y. Sun, and S. K. Jung, “Effect of rapid thermal annealing on the structure and the electrical properties of atomic-layer-deposited Ta2O5 films,” J. Korean Phys. Soc. 37(6), 975–979 (2000). [CrossRef]  

8. H. Grüger, Ch. Kunath, E. Kurth, S. Sorge, W. Pufe, and T. Pechstein, “High quality r.f. sputtered metal oxides(Ta2O5, HfO2) and their properties after annealing,” Thin Solid Films 447, 509–515 (2004). [CrossRef]  

9. ISO 11551: “Optics and optical instruments-Lasers and laser-related equipment-Test method for absorptance of optical laser components” (2003).

10. Ch. Xu, H. C. Dong, L. Yuan, H. B. He, J. D. Shao, and Z. X. Fan, “Investigation of annealing effects on the laser-induced damage threshold of amorphous Ta2O5 films,” Opt. Laser Technol. 41(3), 258–263 (2009). [CrossRef]  

11. H. A. Macleod, “Structure-related optical properties of thin films,” J. Vac. Sci. Technol. A 4(3), 418–422 (1986). [CrossRef]  

12. H. Shinriki, T. Kisu, S. Kimura, Y. Nishioka, Y. Kawamoto, and K. Mukai, “Promising storage capacitor structures with thin Ta2O5 film for low-power high-density DRAM’s,” IEEE Trans. Electron Dev. 37(9), 1939–1947 (1990). [CrossRef]  

13. Ch. Xu, Y. H. Qiang, Y. B. Zhu, T. T. Zhai, L. T. Guo, Y. L. Zhao, J. D. Shao, and Z. X. Fan, “Laser-induced damage threshold at different wavelengths of Ta2O5 films annealed over a wide temperature range,” Vacuum 84(11), 1310–1314 (2010). [CrossRef]  

14. C. Joseph, P. Bourson, and M. D. Fontana, “Amorphous to crystalline transformation in Ta2O5 studied by Raman spectroscopy,” J. Raman Spectrosc. 43(5), 1146–1150 (2012). [CrossRef]  

15. Z. Liu, S. Chen, P. Ma, Y. Wei, Y. Zheng, F. Pan, H. Liu, and G. Tang, “Characterization of 1064nm nanosecond laser-induced damage on antireflection coatings grown by atomic layer deposition,” Opt. Express 20(2), 854–863 (2012). [CrossRef]   [PubMed]  

16. Ch. Xu, Q. L. Xiao, J. Y. Ma, Y. X. Jin, J. D. Shao, and Z. X. Fan, “High temperature annealing effect on structure, optical property and laser-induced damage threshold of Ta2O5 films,” Appl. Surf. Sci. 254(20), 6554–6559 (2008). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (8)

Fig. 1
Fig. 1 (a) Experimental setup for the deposition of Ta2O5/SiO2 rugate filters. (b) The refractive indexes as a function of different target positions. (c) Relationship between refractive index and volume ratio of Ta2O5.
Fig. 2
Fig. 2 Refractive index and electric field distribution of the Ta2O5/SiO2 rugate filters.
Fig. 3
Fig. 3 (a) Transmittance spectra of the as-deposited and annealed films at different temperatures. (b) Blue shifts of the as-deposited and annealed films. (c) Reflectances of the as-deposited and annealed films at 1064nm.
Fig. 4
Fig. 4 Weak absorptions of the as-deposited and annealed films at different temperatures: (A), (B) and (C) represent the evolution of the films in the annealing process individually.
Fig. 5
Fig. 5 (A) AFM (3D) images of the quasi-rugate films: (a1) unannealed film and (a2) annealed film at 400°C. (B) The curvature radii of quasi-rugate films obtained by Zygo interferometer measurements: (b1) annealed film at 400 °C and (b2) annealed film at 600°C. (C) Profiles of the quasi-rugate films imaged by FIB/SEM: (c1) annealed film at 600°C and (c2) annealed film at 800°C.
Fig. 6
Fig. 6 S-on-1 performances of the as-deposited and annealed films at different temperatures.
Fig. 7
Fig. 7 Typical damage morphologies of the as-deposited and annealed films at different temperatures.
Fig. 8
Fig. 8 Typical damage morphology of annealed film at 800°C.

Equations (5)

Equations on this page are rendered with MathJax. Learn more.

n 2 1 n 2 +2 = f a n a 2 1 n b 2 +2 + f b n b 2 1 n b 2 +2
n f =P n s +( 1P ) n ν
( 4 5 )Ta+ O 2 ( 2 5 )T a 2 O 5 Δ G o =196+0.04( kJ / mol )
σ= E s 6( 1 υ s ) t s 2 t f ( 1 R 2 1 R 1 )
F( N )=Aexp( N /B )+ Φ
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.