Open Access
Add to BibSonomyAbstract
Broadband low dispersion (BBLD) mirrors are an essential component in femto-second (fs) pulse laser systems. We designed and produced Ta2O5-HfO2/SiO2 composite quarter-wave and non-quarter-wave HfO2/SiO2 BBLD mirrors for the 30fs petawatt laser system. The laser damage properties of the BBLD mirrors were investigated in an uncompressed sub-nanosecond laser pulse. It showed that the Ta2O5-HfO2/SiO2 composite BBLD mirror possessed higher LIDT due to the low electric-field intensity (EFI) in the case of the coating without artificial nodules. Nevertheless, the LIDT of the composite mirror was significantly lower than the non-quarter-wave HfO2/SiO2 mirror when the nodules exist. The EFI simulation and damage morphology of the nodules analysis demonstrated that the nodule leading to the light intensification in the middle of the boundary between the nodular and the surrounding coating, thus the outermost HfO2/SiO2 layers cannot protect the Ta2O5/SiO2 layers, and resulting to the significantly low LIDT. This study shed some light on the development of high-laser-damage BBLD mirrors for pulse compression laser systems.
© 2017 Optical Society of America
1. Introduction
There are over 50 petawatt (PW) class lasers currently operational, under construction worldwide, such as Vulcan laser facility, European light infrastructure, Qiangguang in China and so on [1]. These ultrashort laser systems are used for a number of research projects ranging from inertial confinement fusion, particle acceleration and studying materials at high temperature and pressure. Optical coatings that afford broadband spectral high-reflection (HR), high laser induced damage threshold (LIDT), and well-controlled group delay dispersion (GDD), play an essential enabling role for PW lasers with pulses in the femtosecond (fs) regime [2].
Since the coatings for fs laser systems require more-stringent performance criteria than those for longer pulses, researchers have made much effort to study and improve the property of the broad bandwidth low dispersion (BBLD) mirrors. H.Takada proposed a broadband high-energy mirror for chirped-pulsed amplification of 10fs pulses. This mirror consists of broadband TiO2/SiO2 coatings and optimized high-damage-threshold ZrO2/SiO2 coatings, so that high damage-threshold and relatively low dispersion are traded off against each other [3]. A significant contribution to this line of research is the work of J.Oliver [4]. They have developed multilayer BBLD mirrors with all-dielectric coating layers for S polarization and a mixture of dielectric and metal coating layers for P-polarization that are suitable for deposition on meter-size optics. And the laser damage property of the multilayer BBLD mirrors with different coating materials is investigated in femto-second laser pulse. J. Bellum reported the TiO2/SiO2 BBLD mirror with special coating design to reduce the electric-field intensity (EFI) in the high-index coating materials, it handles both S-polarization and P-polarization situations, and the laser damage test in sub- nanosecond and fs pulse showed with an improved LIDT [5,6]. Moreover, a laser damage competition for BBLD mirrors was held recently, and it showed that the development of BBLD was still a challenge, and the LIDT of different materials with different coating process was discussed [7].
The above studies have demonstrated that the coating material and EFI play a significant role in the LIDT of the broadband low-dispersion mirrors. However, two important issues were not addressed in previous researches. First, in order to achieve the spectral, dispersion bandwidth and low electric field distribution, the previous researches mainly consider the quarter-wave HR coating with the high-index coating materials of Nb2O5, TiO2 and Ta2O5. However, the HfO2/SiO2 coating which is verified to possess the highest LIDT [7], is rarely used due to the narrow bandwidth. So it is natural to think that whether it is possible to design BBLD mirror with HfO2/SiO2 coating to extend the bandwidth and improve the LIDT. Moreover, as it is generally accepted, the nodule defect is the main limiting defect for the multilayer reflective coatings working at the near-infrared region in both ns and fs pulse lasers [8–10], but the effect of nodular defect in the damage of the BBLD coating is not investigated.
In this paper, we designed and produced the broadband low-dispersion mirrors for a 30fs optical parametric chirped-pulse-amplification (OPCPA) laser system using the HfO2 coatings. The laser damage property of the BBLD mirrors was investigated in the sub-nanosecond pulse, and finally the laser damage behavior of nodular defects with artificial seeds in different coating was demonstrated.
2. Experiments
2.1 Design of broadband low dispersive mirrors
We aimed to develop the high laser-damage-threshold BBLD mirrors for a 30fs OPCPA PW laser system. The mirrors should meet a minimum reflection of 99.5% at 45 degrees incidence angle at p-polarization with a group dispersion delay of < ± 50fs2 over a spectral range of 730-830nm.
The bandwidth of the reflector is typically determined by the ratio of the high- and low-index materials. SiO2 coating is the primary choice for the low-index materials, according to the theoretical relationship of the bandwidth and the refractive index difference, Nb2O5, TiO2 and Ta2O5 coatings can be used as high-index materials to achieve the spectral and GDD requirement with a quarter-wave (QW) structure. The HfO2/SiO2 QW mirror does not have enough bandwidth, however, due to high LIDT of the HfO2 material, in this study we will design two low-dispersive coatings consisting of the HfO2 coatings.
The first design was based on a 44 layers Ta2O5/SiO2 QW reflector with a half-wave outermost SiO2 layer, at the center wavelength of 780nm. Then we replaced the 4 outermost layer pairs with HfO2/SiO2 layers, and the bandwidth of the composite QW mirror was 723-837nm. The high-intensity electric fields quench substantially in the outer HfO2 layers before reaching those inner Ta2O5 layers, as shown in Fig. 1(a). The maximum EFI is 90% and 12% in HfO2 and Ta2O5 layers respectively, and this EFI distribution allows the increase of the LIDT of the coatings [11]. Here we take a 36 layers HfO2/SiO2 QW mirror for comparison, it possesses the bandwidth of 735-825nm, and the maximum EFI is also about 90% in HfO2 layers. It seems that the composite BBLD mirrors will has almost the same LIDT compared to the HfO2/SiO2 QW mirror.
Then we try to design non-quarterwave HfO2/SiO2 mirror to meet the requirement of the spectral and GDD. The bandwidth can be extended by the using multiple reflectors or a geometric stack, such a structure will lead to the strong intensity peaks that in turn make the coating more susceptible to laser damage at the lower fluences [12]. For looking to design options based not only on meeting the spectral and GDD requirement but also on meeting the requirement that the optical EFI behavior within the high-index materials show moderate intensity peaks. We included the maximum EFI in the high-index materials in the merit function of refinement, and the optimization of the design was obtained by Optilayer software [13]. Finally we get the 56 layers modified HfO2/SiO2 design with thickness of 7.6um, and the bandwidth is 724-836nm. The coating structure and the EFI are shown in Fig. 1(b). There is some high EFI in the top layer pairs due to the strong interference effect. Nevertheless, the maximum EFI is about 120% in HfO2 layers, not quite high compared to the value in the QW mirror. It indicates that such a non-quarter-wave coating will possess a moderate LIDT.
2.2 Preparation of broadband low dispersion mirrors and artificial defects
The two BBLD mirrors were deposited by e-beam evaporation (Optorun e-beam deposition plant) on BK7 substrates with a diameter of 25 mm. The HfO2/SiO2 QW mirror was also manufactured and studied in order to compare the LIDT properties. Ta, Hf and SiO2 were used as the evaporation materials. The substrate temperature was kept at 200°C during deposition, and the deposition rates were 0.3 nm, 0.2nm and 0.6nm for the Ta2O5, HfO2 and SiO2 layers respectively. During deposition of Ta2O5/SiO2 coating, the layers were densified with an RF-type ion source. Quartz crystal monitoring was mainly used as feedback for stabilizing evaporation rates and also for registering layer thicknesses. The indirect monochromatic optical monitor with modified monitoring strategy was ultilized to control thickness [14]. In order to analysis of the LIDT behavior of nodular defects, the three HR coatings on substrates with mono-disperse silica microspheres of three different sizes, 0.5, 0.9, 1.45 μm was also manufactured and studied. A detailed description of the manufacturing and characterization of the samples used in the present study can be found in [15].
The spectral performances of the prepared low dispersion HR coatings were measured using a Cary-5000 spectrophotometer. GDD measurements were performed by CHROMATIS from KMLabs over a wavelength range of 600-1000nm. Figure 2 shows the measured and the theoretical GDD curves of the composite and modified BBLD mirrors, the spectral bandwidth is a little narrower compared to the theoretical, but it still shows an excellent agreement and meets the requirement. It means that the coating structure and EFI distribution in the prepared BBLD mirrors are also the same with the theoretical one, which lays a good foundation for the analysis of subsequent laser damage studies.
Fig. 2 Comparison of measured GDD data (red crosses) and theoretical GDD (solid black curve) of the low-dispersive mirrors.
2.3 Laser damage test and characterization techniques
Damage testing was performed with uncompressed pulses from a Ti:Sapphire regenerative amplifier, which has the laser wavelength of 780nm with a centered spectral bandwidth of 40nm. The laser pulse has a TEM00 mode, a beam diameter of 100 microns at 1/e2, pulse duration of 0.2ns, and a repetition rate of 1000Hz. The raster scan method [16] was used to determine the LIDTs of optical coatings. The samples were raster scanned over a 3mm by 3mm area starting at a fluence of 1J/cm2 and increasing in 2J/cm2 increments. A new area was scanned at each higher fluence to minimize the potential for laser conditioning. With a scanning speed of 1mm/s and repetition rate of 1000Hz, each site was exposed to roughly 25 shots at 90% peak fluence indicating the opportunity to grow the laser damage. An in situ camera was used for damage detection, and the accuracy of the fluence was 10 percent. When testing the coatings without nodular defects, the raster scan was stopped as catastrophic damage occurred. When the coatings with nodular defects were tested, the laser fluence when more than 10 nodules were ejected was recorded as the ejection threshold, and the raster scan was stopped when damage grew, recorded as damage growth threshold. The damage morphologies were observed and characterized by a Nomarski microscope, a scanning electron microscope (SEM) and a focus ion beam (FIB) equipment.
3. Results and discussion
3.1 Laser induced damage without nodular defects
LIDTs of three HR coatings without artificial nodular defects are given in Table 1. The LIDT of the QW HfO2/SiO2 QW mirror and the Ta2O5-HfO2/SiO2 composite mirror are 19 ± 2 J cm−2, and the LIDT of the modified HfO2/SiO2 mirror is 15 ± 2 J cm−2. It is in accordance with our prediction that the modified HfO2/SiO2 mirror has a higher EFI in the HfO2 layers. The represent damage morphology was observed to be delamination triggered by nano-sized absorbers, and it easily grew to catastrophic damage due to the multi-pulses irradiation. Since the laser test area was selected with no obvious defects, it is reasonable to assume that the strong EFI at the HfO2 layers to trigger the damage from the nano-absorbers [17]. This result shows that the composite mirror is a good compromise of wide bandwidth and high LIDT, and the LIDT of the modified HfO2/SiO2 mirror is also acceptable.
Table 1. LIDTs of three HR optical coatings without nodular defects
3.2 The influence of the nodular defects to the LIDT of the low-dispersive mirrors
Then we test the three HR coatings with artificial silica seeds. Figure 3 shows the statistical ejection fluences of artificial nodules of the coatings. One can see that the ejection fluences of nodules are all much lower compared to the LIDT of coatings in the clean substrate, and monotonically decrease with increasing diameter of silica microspheres for all the three coatings. It is similar to the behavior of the results in the previous study [18]. And it is interesting to find that modified HfO2/SiO2 mirror has the highest ejection fluence, which is more than two times higher than the Ta2O5-HfO2/SiO2 composite mirror, which is totally different to the LIDT results without artificial nodules. This result illustrates that the LIDT of composite mirror decreased significantly due to the existence of the nodule defects.
It is verified that the nodule defect will lead to significant EFI enhancement, and the EFI enhancement has a direct link to the damage behavior [9]. In order to understand the reason why the LIDT of the mixed BBLD mirror behaves so different with and without the nodules, the EFI distribution due to the nodule was simulated and compared with the cross-section damage morphologies of ejected nodules. We first use the FIB instrument to examine the cross-section morphologies of nodules to confirm the structure of the nodules. Figure 4(a1-a3) show the FIB figures of the nodules in the three HR coatings. The results showed that nodules in HfO2/SiO2 mirrors prepared by EBE process exhibited a D = sqrt(4dt) aspect-ratio. For the composite mirror, a smaller aspect-ratio of D = sqrt(2.5dt) was found because the Ta2O5/SiO2 stack was deposited by ion-assisted technique. Moreover, the nodular boundary continuity between the nodules and surrounding films was significantly improved in the composite mirror and modified HfO2/SiO2 mirror. Such a more continuous boundary will lead to stronger mechanical and environment stability. The results are correspondence to our previous study that the structure of the nodules depends on the deposition technology [9].
Fig. 4 Comparisons between simulated |E|2 distributions and damage morphologies of nodules in HfO2/SiO2 QW mirror, modified HfO2/SiO2 mirror and the composite mirror: (a1-a3) the cross-section morphologies of the nodules; (b1-b3) Typical damage morphologies of the nodules; (c1-c3) FDTD-simulated p-polarized |E|2 distributions where the white lines represent film stacks and the color scale is different for different nodules.
The |E|2 distributions were simulated using a 3D FDTD electromagnetic code described in [9], at the incident angle of 45 degree and the center wavelength of 780nm. Figure 4(c1-c3) gives the EFI distributions of nodules initiating from 0.9μm silica seeds, the p-polarized light is incident from the right side. There are two positions of EFI intensification were found in the nodule structures, one distributed in the top layers of the nodules due to the mechanism of the light interference effect, and the other one distributed in the middle of the left boundary between the nodular and the surrounding coating due to the light focusing effect. The mechanism of the EFI enhancement of nodules at the non-normal incidence will be discussed in our following paper.
Firstly, we compare the difference of the EFI distribution of the modified HfO2/SiO2 coating and HfO2/SiO2 QW coating. The maximum EFI of the two coatings are comparable, but it locates near the boundary for the HfO2/SiO2 QW coating, and it is in the top layers of the nodule structure for the modified HfO2/SiO2 coating. This different distribution can be explained that the HfO2/SiO2 QW coating has a narrow reflection bandwidth, it means more light penetrates to the nodule structure, leading to a more strong light focusing intensification. The modified HfO2/SiO2 coating has more strong standing EFI as shown in Fig. 1(b2), and will result to much more strong interference effect. Moreover, since the boundary continuity for the HfO2/SiO2 QW coating is very poor, it will have a weaker mechanical property and there will also be more absorption defect around the weak boundary. Therefore the ejection fluences were much lower compared to the modified HfO2/SiO2 coating. The damage morphologies of the ejected nodules exactly show agreement to the simulation. The coating materials were just peeled off near the maximum EFI region for the modified HfO2/SiO2 coating. And almost all the nodule structure is ejected for the QW HfO2/SiO2 coating, and the small sunken in the left boundary proved the EFI enhancement near this location.
Then we take a look at the EFI of the composite mirror, one can see that it has the highest peak EFI, and locating in the middle of the boundary due to focusing effect, this can be explained by the geometry and higher refractive index of the nodule structure according to our previous study [18]. Since this composite mirror only has 4 pairs of HfO2/SiO2 coating at the outermost, so the peak EFI is actually in the Ta2O5/SiO2 coating, it indicates that the outermost HfO2 layers do not protect the Ta2O5 layers when the nodules exist. Due to the low LIDT of Ta2O5 layers, the ejection fluences decrease significantly. The cross-section damage morphologies of ejected nodules is shown in the Fig. 4(b3), it is obviously that the ejection was initiated from the Ta2O5 layers, and the coating materials were just sprayed out near the melting region instead of the boundaries due to the good boundary continuity.
We will also discuss about the damage growth threshold that is very important in practical applications. Figure 5 shows the statistical the damage growth threshold of the coatings. It shows the same trends as the result of ejection fluences, except that the LIDT for the HfO2/SiO2 QW mirror is higher compared to the modified HfO2/SiO2 mirrors. As shown above, the nodules in HfO2/SiO2 QW mirror have low ejection fluences, however, after the ejection, the nodule of the HfO2/SiO2 QW mirror was almost all ejected, and the EFI enhancement would be quite low, it possessed high resistance to the following pulse similar as laser conditioning effect [19]. While for the modified HfO2/SiO2 mirror, the protective half-wave SiO2 layer was peeled and the main part of the nodule structure still remain, and it will possess EFI enhancement and be easier to grow with the following irradiation. Moreover, the composite mirror still had the lowest LIDT, that’s because after the top layers were ejected, the Ta2O5/SiO2 coating was directly irradiated by the following laser pulse, so it was easier to grow to the catastrophic damage.
This result illustrates that the LIDT of composite mirror decreased significantly due to the existence of the nodule defects. Is it possible to replace more Ta2O5/SiO2 coating pairs with HfO2/SiO2 coatings to improve the LIDT? One can estimate from the EFI figure that more than 10 pairs should be replaced, however, in this case the spectral and GDD bandwidth will not meet the requirement. Therefore we can predict that such a composite mirror will has a lower LIDT compared to the non-QW HfO2/SiO2 mirrors in applications, because the nodule defects are unavoidable in real-world optical components. However, it will be interesting to study the laser damage property of the composite mirrors using the planarization technology to minimize the impact of coating defects [20].
4. Summary
The laser damage characteristics of broadband low dispersion mirrors in sub-nanosecond pulse laser was been investigated. It was found that Ta2O5-HfO2/SiO2 composite mirror showed a good tradeoff of wide bandwidth and high LIDT in the case without nodule defect. However, the LIDT of the composite mirror decreased significantly due to the light focus in the Ta2O5 layers quite below the interface when the nodular exist. Moreover, it was found that the modified HfO2/SiO2 mirror exhibit almost the same LIDT compared to the HfO2/SiO2 QW mirror. This work illustrates that the non-quarter wave HfO2/SiO2 BBLD mirror is more vulnerable for application in the high power laser systems, and it verifies that it is meaningful to extend the bandwidth of dispersive mirrors with high LIDT materials. Since the uncompressed sub-ns laser contains a broadband spectrum, we will investigate the influence of the spectrum bandwidth to laser damage property of the dispersive mirrors in the future.
Funding
National Natural Science Foundation of China (NSFC) (61621001, 61522506, U1630124, 91536111, U1430130).
References and links
1. C. Danson, D. Hillier, N. Hopps, and D. Neely, “Petawatt class lasers worldwide,” High Power Laser Science and Engineering 3, e3 (2015). [CrossRef]
2. J. C. Bellum, P. Rambo, J. Schwarz, I. Smith, M. Kimmel, D. Kletecka, and B. Atherton, “Production of optical coatings resistant to damage by petawatt class laser pulses,” in Lasers-Applications in Science and Industry (InTech Open Access Publisher, 2011), pp. 23–52.
3. H. Takada, M. Kakehata, and K. Torizuka, “Broadband high-energy mirror for ultrashort pulse amplification system,” Appl. Phys. B 70(S1), S189–S192 (2000). [CrossRef]
4. J. B. Oliver, J. Bromage, C. Smith, D. Sadowski, C. Dorrer, and A. L. Rigatti, “Plasma-Ion-Assisted Coatings for 15 Femtosecond Laser Systems,” Appl. Opt. 53(4), A221–A228 (2014). [CrossRef] [PubMed]
5. J. C. Bellum, E. S. Field, T. B. Winstone, and D. E. Kletecka, “Low group delay dispersion optical coating for broad bandwidth high reflection at 45 incidence, P polarization of femtosecond pulses with 900nm center wavelength,” Coatings 6(1), 11 (2016). [CrossRef]
6. E. S. Field, J. C. Bellum, and D. E. Kletecka, “Laser damage comparisons of broad-bandwidth, high-reflection optical coatings containing TiO2, Nb2O5, or Ta2O5 high-index layers,” Opt. Eng. 56(1), 011018 (2016). [CrossRef]
7. C. J. Stolz, R. A. Negres, K. Kafka, E. Chowdhury, M. Kirchner, K. Shea, and M. Daly, “150-ps broadband low dispersion mirror thin film damage competition,” Proc. SPIE 9632, 96320C (2015). [CrossRef]
8. C. J. Stolz, M. D. Feit, and T. V. Pistor, “Laser intensification by spherical inclusions embedded within multilayer coatings,” Appl. Opt. 45(7), 1594–1601 (2006). [CrossRef] [PubMed]
9. X. B. Cheng, J. L. Zhang, D. Tao, Z. Y. Wei, H. Q. Li, and Z. S. Wang, “The function of electric-field in thermomechanical damage of nodular defects in dielectric multilayer coatings irradiated by nanosecond laser pulse,” Light Sci. Appl. 2(6), e80 (2013). [CrossRef]
10. L. Gallais, X. Cheng, and Z. Wang, “Influence of nodular defects on the laser damage resistance of optical coatings in the femtosecond regime,” Opt. Lett. 39(6), 1545–1548 (2014). [CrossRef] [PubMed]
11. J. Bellum, E. Field, D. Kletecka, and F. Long, “Reactive ion-assisted deposition of e-beam evaporated titanium for high refractive index TiO2 layers and laser damage resistant, broad bandwidth, high-reflection coatings,” Appl. Opt. 53(4), A205–A211 (2014). [CrossRef] [PubMed]
12. V. Pervak, M. K. Trubetskov, and A. V. Tikhonravov, “Design consideration for high damage threshold UV-Vis-IR mirrors,” Proc. SPIE 7504, 75040A (2010).
13. A. V. Tikhonravov and M. K. Trubetskov, OptiLayer Thin Film Software, http://www.optilayer.com.
14. J. Zhang, A. V. Tikhonravov, Y. Liu, M. K. Trubetskov, A. Gorokh, and Z. Wang, “Design, production and reverse engineering of ultra-steep hot mirrors,” Opt. Express 22(11), 13448–13453 (2014). [CrossRef] [PubMed]
15. X. Cheng, Z. Shen, H. Jiao, J. Zhang, B. Ma, T. Ding, J. Lu, X. Wang, and Z. Wang, “Laser damage study of nodules in electron-beam-evaporated HfO2/SiO2 high reflectors,” Appl. Opt. 50(9), C357–C363 (2011). [CrossRef] [PubMed]
16. K. Kafka, E. Chowdhury, R. Negres, C. Stolz, J. Bude, A. Bayramian, C. Marshall, T. Spinka, and C. Haefner, “Test station development for laser-induced optical damage performance of broadband multilayer dielectric coatings,” Proc. SPIE 9632, 96321C (2015). [CrossRef]
17. Z. Wang, G. Bao, H. Jiao, B. Ma, J. Zhang, T. Ding, and X. Cheng, “Interfacial damage in a Ta2O5/SiO2 double cavity filter irradiated by 1064 nm nanosecond laser pulses,” Opt. Express 21(25), 30623–30632 (2013). [CrossRef] [PubMed]
18. X. Cheng, A. Tuniyazi, Z. Wei, J. Zhang, T. Ding, H. Jiao, B. Ma, H. Li, T. Li, and Z. Wang, “Physical insight toward electric field enhancement at nodular defects in optical coatings,” Opt. Express 23(7), 8609–8619 (2015). [CrossRef] [PubMed]
19. S. R. Qiu, J. E. Wolfe, A. M. Monterrosa, M. D. Feit, T. V. Pistor, and C. J. Stolz, “Searching for optimal mitigation geometries for laser-resistant multilayer high-reflector coatings,” Appl. Opt. 50(9), C373–C381 (2011). [CrossRef] [PubMed]
20. C. J. Stolz, J. E. Mirkarimi, J. A. Folta, J. Adams, M. G. Menor, N. E. Teslich, R. Soufli, C. S. Menoni, and D. Patel, “Substrate and coating defect planarization strategies for high-laser fluence multilayer mirrors,” Thin Solid Films 592, 216–220 (2015). [CrossRef]
