Open Access
Add to BibSonomy
Abstract
In synchrotron facilities, optics with multilayer coatings are used for beam monochromatization, focusing, and collimation. These coatings might be damaged by high heat load, poor film adhesion, high internal stress, or poor vacuum. Optical substrates always need high quality, which is expensive and has a long processing cycle. Therefore, it is desired to make the substrate reusable and the refurbished coating as good as a brand-new one. In this study, a W/B4C multilayer coating with a 2 nm Cr buffer layer was prepared on a Si substrate. The coating was successfully stripped from the Si substrate by dissolving the Cr buffer layer using an etchant. The roughness and morphology after the different etching times were investigated by measuring the GIXRR and 3D surface profiler. It is shown that the time required to etch the W/B4C multilayer coating with a Cr buffer layer, is quite different compared with etching a single Cr film. A layer of silicon dioxide was introduced during the fitting. After the new etching process, the roughness of the sample is as good as the one on a brand-new substrate. The W/B4C multilayer coatings with a Cr buffer layer were recoated on the etched samples, and the interface roughness was not damaged by the etching process.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Multilayer mirrors have been widely used as artificial one-dimensional crystals in synchrotron radiation [1,2], astronomical observation [3], and extreme ultraviolet lithography [4–6]. The length of the multilayer mirror could be shortened by increasing the glancing incident angle. The shorter mirror makes it easier to ensure smaller slope errors and figure errors. The cost and processing difficulty of the mirror will be reduced. For shorter mirrors, less deformation due to weight, temperature change, and vibration results in better mechanical stability. Shorter mirrors take up less space and can improve the integration of equipment such as astronomical telescopes [7,8]. To obtain the maximum possible glancing incident angles, the period thicknesses of the multilayers should be designed as short as possible using Bragg’s law (2dsinθ=mλ, where d represents the thickness of the bilayer, θ represents the glancing incident angle, m represents the Bragg order, and λ represents the wavelength) [9]. W/B4C is a promising candidate to form short-period multilayers [4,10,11]. This is because no reaction was observed at the W/B4C interfaces [12], which are smoother than those at the W/C [13,14], Mo/Si [15], Ru/B4C, and Mo/B4C systems [16]. The smooth interface reduces the interface roughness and increases the reflectivity of W/B4C multilayers [16,17]. Silicon is generally used as a substrate to deposit W/B4C multilayers, considering the elastic deformation, stability of the internal structure, resistance to heat load deformation, and other factors in the field of synchrotron radiation. To obtain high transmission efficiency and extremely small beam spots, the Si substrate is typically polished to a roughness of only a few angstroms before it can be used for coating [18].
Although multilayer optics in synchrotron radiation are placed in an ultra-high vacuum environment, the surface of these long-serving optics will still be roughened by carbon contamination [19], which causes the beamlines to diverge and reflectivity to decrease [20]. In addition to contamination, W/B4C multilayers on Si as consumable optics also experience aging and spalling owing to the extreme radiation environment [4], poor adhesion between the multilayers and substrate, and high internal stress in the multilayers [21]. To restore the flux and focusing capacity, the optics must be updated after a period of service. Because high-precision and large-size Si substrates are extremely expensive and have long production cycles, making Si substrates reusable is critical. DC plasma, plasma arc, laser, capacitively coupled RF plasma discharge [20], and low-pressure RF plasma [22] have been studied to chemically and selectively clean the pollution on the mirror surface. These methods can remove the pollution of specific elements successfully on the surface of coated optics but cannot guarantee the uniformity of the coated surface and may even destroy the surface. In addition to removing only the contaminating layer, the method of removing all coatings on the surface of the silicon substrate by dissolving the pre-deposited five-nanometer chromium buffer layer [21] between the substrate and monolayer film has also been studied. Macroscopic results showed that the monolayer could be peeled off with the dissolution of the Cr buffer layer in an hour. The substrate remained intact with usable surface roughness and did not require re-polishing. However, the etching solution would oxidize the surface of the silicon substrate, and it is necessary to study the effects of the etching process on the oxidation degree of the substrate. Whether the oxidized substrate could be recoated well needs to be studied experimentally.
In this study, a two-nanometer chromium buffer layer was fabricated between W/B4C multilayers and a Si substrate by magnetron sputtering. The W/B4C multilayers were stripped by dissolving the Cr buffer layer. The purpose of using a thinner buffer layer in this study was to reduce the roughness of the buffer layer and reduce the effect on the roughness of the following W/B4C multilayers [23]. To study whether the time required for the stripping of different coatings is the same, the Cr single-layer samples and the multilayer samples were all etched at different times. The roughness and morphology of the sample surface after etching at different durations were investigated. After etching, the samples were recoated, and the interface roughness of the recoated W/B4C multilayers was studied.
2. Experimental
2.1 Magnetron sputtering deposition
The chromium buffer layer and W/B4C multilayers were both deposited on Si wafers with <100 > orientation by direct-current (DC) magnetron sputtering at room temperature. The sputtering targets used in this study were rectangular (5 × 20 cm) with a purity of 99.9999%. All films were deposited in dynamic mode, where the substrate moved back and forth in front of the sputter sources. The distance between the target and substrate was 9 cm. The movement speeds for the deposition of Cr, W, and B4C were 10, 10, and 8 mm/s, respectively. The base pressure prior to the deposition was 3 × 10−4 Pa. The sputtering gas was argon with a constant pressure of 0.3 Pa and 0.15 Pa for the Cr buffer layer and W/B4C multilayer, respectively. The period number of the W/B4C multilayers was 10. The sputtering powers for the Cr, W, and B4C targets were 40, 50, and 300 W, respectively. The re-coating process of the samples after etching at different times was the same as that described above.
2.2 Stripping
The stripping process of the single Cr film and W/B4C multilayers with the Cr buffer layer was performed using a commercial etchant (Chromium etch 1020AC, Transene Company) in a teflon beaker at room temperature. The etchant was a mixture of 5–10% acetic acid, 10–15% ceric ammonium nitrate (CAN), and the rest (75–85%) water by weight. The samples were placed at the bottom of the beaker, and the liquid level of the etching solution was approximately 1 cm above the surface of the samples. The specimens were removed from the beaker at various intervals, ultrasonically cleaned with deionized water followed by alcohol, and dried in static air.
2.3 Characterization
All samples were measured by grazing incidence X-ray reflection (GIXRR) at the Cu Kα line in a 2θ–ω scanning pattern. The curve-fitting method was used to process the data from GIXRR to assess the film thickness and surface/interface roughness. The surface morphologies of the samples were also investigated using a non-contact 3D optical surface profiler [24]. Several positions were measured for each sample to ensure reproducibility of the results.
3. Results and discussion
3.1 Prepared samples
Figure 1 shows the GIXRR spectra (black lines) and fitted curves (red lines) of the original Si wafer (Fig. 1(a)), as-deposited single Cr film (Fig. 1(b)), and W/B4C multilayers with the Cr buffer layer (Fig. 1(c)). The curves of GIXRR present a series of Bragg peaks, which mean the different diffraction orders of multilayer structures. The structural parameters of each layer could be analyzed by fitting the measured GIXRR data based on the Parratt formalism [25]. The fitted results show that the surface roughness of the Si substrate is 4.1 Å, and that the surface roughness of the single Cr film on Si is 5.0 Å. The phenomenon that the roughness of the Cr film is greater than that of the Si substrate could be explained by stress in the Cr film [23,26]. The thickness of the Cr buffer layer is 19.3 Å. The thicknesses of the W and B4C layers in the W/B4C multilayer sample with the Cr buffer layer are 1.30 and 1.51 nm, respectively. The interface roughness values of B4C on W and W on B4C are both 3.3 Å. Although the roughness of the Cr film was slightly higher than that of the original Si substrate, the roughness of the W/B4C multilayers was still 3.3 Å. This is in good agreement with the GIXRR results reported by Morawe et al. [23].
Fig. 1. GIXRR spectra (black lines) and fitted curves (red lines) of (a) the original Si wafer, (b) the as-deposited single Cr film and (c) the W/B4C multilayers with the Cr buffer layer.
3.2 Stripping
3.2.1 Etching the single Cr film
The GIXRR spectra of the samples with a single Cr film on a Si substrate etched at room temperature at different times are shown in Fig. 2. The chemical reaction that occurs for etching Cr is as follows [27]:
Fig. 2. GIXRR spectra of samples with the single Cr film on Si substrate etched at room temperature at different time.
3.2.2 Stripping the W/B4C multilayers with Cr buffer layer
Figure 3 shows the GIXRR spectra of the W/B4C multilayer samples with a Cr buffer layer between the W/B4C multilayers and the Si substrate after static etching at 20–25°C at different times. The curve of the sample etched for an hour was different from the curves of the multilayer sample and the original Si substrate. It is clear that one hour was too short to peel off all the films on the Si substrate for the W/B4C ML samples. This indicated that the etching process of the W/B4C multilayer samples was slower than that of the single Cr film sample. For the single Cr film sample, the etching interface was a solid-solution interface, and the etching process occurred simultaneously on the entire sample surface. However, for the W/B4C multilayer samples, the etchant attacked chromium from the edges of the samples or through the multilayers. The Cr buffer layer was approximately 2 nm thick, thus the direct contact interface for the etching of Cr at the edge of each sample was relatively small. To pass through the W/B4C multilayers, the etchant mainly diffused through the grain boundaries in the multilayers. Because of the small contact interface and difficult grain boundary diffusion, the etching process of the W/B4C ML samples was more time-consuming than that of the single Cr film samples. Even the spectrum of the W/B4C ML sample after etching for 2 h differed greatly from the curve of the original Si substrate. Two hours were still insufficient for the stripping of the W/B4C multilayer samples. The two curves of the W/B4C multilayer samples after etching for 4 h and 24 h were very close to those of the original Si substrate. This indicates that 4 h is probably sufficient for etching the W/B4C multilayer samples.
Fig. 3. GIXRR spectra of W/B4C multilayer samples with the Cr buffer layer between the multilayers and the Si substrates after static etching at room temperature at different time.
Figure 4 displays the surface morphology measured by a 3D optical surface profiler of the W/B4C multilayer samples after static etching at room temperature at different times. The window dimensions were 0.1 × 0.1 mm. Many holes could be observed in Figs. 4(b) and 4(c). Average diameters of holes in Figs. 4(b) and 4(c) were 5.4 and 2.2 µm, respectively. Although the diameter of defects in the sample etched for an hour was larger than that in the sample etched for 2 h, the defect density of the latter was much greater than that of the former. No holes could be observed in Figs. 4(d) and 4(e). The surface morphology of the samples after etching for 4 h (Fig. 4(d)) and 24 h (Fig. 4(e)) was similar to that of the original silicon wafer (Fig. 4(a)). It can be clearly observed from Fig. 4 that the films gradually peeled off as the etching time increased and completely peeled off at 4 h, which is consistent with the results in Fig. 3.
Fig. 4. Surface morphology of samples measured by 3D optical surface profiler: (a) the original Si wafer, the W/B4C multilayer samples with the Cr buffer layer between the multilayers and the Si substrates after static etching at room temperature for (b) 1 h, (c) 2 h, (d) 4 h, and (e) 24 h.
As shown in Fig. 3, although the GIXRR curves of the W/B4C multilayer samples after etching for 4 and 24 h were close to those of the original silicon wafer, there were still some differences between the two samples and the original Si wafer. According to the product introduction of the etchant [31], the surface of the Si substrate could still be oxidized, although the acidic environment provided for CAN was rather mild to minimize the damage to the substrate. The GIXRR patterns of the samples etched for 4 and 24 h could be fitted well by introducing a layer of silicon dioxide (Fig. 5). The fitted results, shown in Fig. 5, are listed in Table 1. With an increase in the etching time, the thickness of the oxide film formed on the surface of the sample increased slightly, and the surface roughness increased from 4.1 to 5.2 Å. Oxidation of the substrate surface occurred simultaneously with the dissolution of chromium during the etching process. The oxidation process of the substrate surface continued with an increase in etching time. A 4-h etching time is sufficient to strip the films with less oxidation and better roughness on the sample surface. The roughness was the same as that of the original Si wafer (4.1 Å).
Fig. 5. GIXRR spectra (black lines) and fitted curves (red lines) of W/B4C multilayer samples left in the etchant solution for 4 h (a) and 24 h (b).
Table 1. Fitted results of the GIXRR measurements shown in Fig. 5
3.3 Re-coating
To explore whether the etched sample could be used directly for re-coating, W/B4C multilayers with a Cr buffer layer were recoated on the etched samples and a new original silicon wafer simultaneously. Figure 6 shows the GIXRR curves of the recoated samples. The black curve represents the GIXRR curve of the W/B4C multilayers with a Cr buffer layer on the original Si wafer. The GIXRR curves of the samples that have been etched at different times and then recoated are indicated by red lines. As shown in Fig. 6(a), the multilayer film failed to form on the surface of the sample that had been etched for 1 h owing to the large number of remains. Figure 6(b) shows that W/B4C multilayers with a Cr buffer layer could be formed again on the surface of the sample that had been etched for 2 h, although there were still some films on the sample surface after etching. The newly formed coating (Fig. 6(b)) differed significantly from that of the control sample. The W/B4C multilayers with a Cr buffer layer grew well on the samples that had been etched for 4 h (Fig. 6(c)) and 24 h (Fig. 6(d)), and the GIXRR curves of the two samples were nearly identical and very close to that of the contrast sample.
Fig. 6. GIXRR curves of samples that have been etched at different times and then recoated (red lines): (a) 1 h, (b) 2 h, (c) 4 h, and (d) 24 h. The black curve represents the GIXRR data of the multilayers on a brand-new substrate which was coated together with those etched substrates for comparison.
Figure 7 shows the fitted curves of the recoated samples, and the oxide layer was included during the fitting of the etched samples. The fitted results are listed in Table 2. For the sample which had been etched for 2 h, the fitted curve of the recoated sample did not match the experimental data very well because the etching process failed to peel off the film completely. The period thickness was 27.1 Å, which is the same as that of the contract sample. The interface roughness values were approximately 6 Å, almost twice those of the contract sample. For the samples that had been etched for 4 h and 24 h, the curves fit well with the experimental data. The period thickness was 27.4 Å and 27.3 Å for the samples etched for 4 h and 24 h, respectively. The interface roughness values of W and B4C for the sample after etching for 4 h were 3.4 Å and 3.1 Å, respectively. For the sample after etching for 24 h, the interface roughness values of W and B4C were 3.5 Å and 3.2 Å, respectively. The thickness and roughness of the samples after etching for 4 h and 24 h are both quite close to those of the contract sample. These results indicate that the existing oxide film formed during the etching process on the substrate did not affect the subsequent re-coating process, and the slight increase in the substrate roughness from 4.1 Å to 5.2 Å of the newly formed oxide film did not affect the interface roughness of the W/B4C multilayer coatings. For W/B4C multilayer coatings with a two-nanometer chromium buffer layer, when the etching time was more than 4 h, the substrate that could be used to re-coat the W/B4C multilayer coatings could be obtained.
Table 2. Fitted results of the GIXRR measurements shown in Fig. 7
4. Conclusions
A W/B4C multilayer coating with the two-nanometer Cr buffer layer was prepared on a Si substrate and then successfully peeled off from the Si substrate using an etchant to dissolve the Cr buffer layer. The films gradually peeled off as the etching time increased, and completely peeled off after 4 h. The etching process of the Cr buffer layer in the W/B4C multilayer samples was much slower than that in the single Cr film sample, which required less than 1 h. The etching process was slowed down owing to the thinness of the chromium buffer layer and the existence of W/B4C multilayers. The roughness values of the samples after etching for 4 and 24 h were 4.1 Å and 5.2 Å, respectively. The former was the same as that of the original Si wafer. A 4-h etching time was sufficient to strip the films with less oxidation and better roughness on the substrate surface. The W/B4C multilayers with a Cr buffer layer were recoated on the etched samples. The interface roughness values of the W/B4C multilayer coatings recoated on the samples which had been etched for 4 and 24 h were both quite close to those on the original Si substrate. A slight increase in the substrate roughness from 4.1 Å to 5.2 Å of the newly formed oxide film on the Si substrate did not affect the interface roughness of the recoated W/B4C multilayer coatings. The method of dissolving the Cr buffer layer using the etchant could be used to refurbish the W/B4C multilayer coatings on Si.
Funding
National Natural Science Foundation of China (12005250); International Partnership Program of Chinese Academy of Sciences (113111KYSB20160021).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. A. Rack, L. Assoufid, W. K. Lee, B. Shi, C. Liu, C. Morawe, R. Kluender, R. Conley, and N. Bouet, “Hard X-ray multilayer mirror round-robin on the wavefront preservation capabilities of W/B4C coatings,” Radiat. Phys. Chem. 81(11), 1696–1702 (2012). [CrossRef]
2. J. P. Sutter, P. A. Chater, R. Signorato, D. S. Keeble, M. R. Hillman, M. G. Tucker, S. G. Alcock, I. T. Nistea, and H. Wilhelm, “1 m long multilayer-coated deformable piezoelectric bimorph mirror for adjustable focusing of high-energy X-rays,” Opt. Express 27(11), 16121–16142 (2019). [CrossRef]
3. Z. Wang, J. Zhu, R. Chen, J. Xu, F. Wang, Z. Zhang, W. Wu, L. Liu, H. Zhang, D. Xu, H. Jiang, L. Chen, H. Zhou, T. Huo, M. Cui, and Y. Zhao, “Extreme ultraviolet multilayer mirrors for astronomical observation,” Proc. SPIE 6984, 698433 (2008). [CrossRef]
4. H. Jiang, S. Yan, J. Zhu, Z. Dong, Y. Zheng, Y. He, and A. Li, “Investigation of defects in sputtered W/B4C multilayers,” Appl. Surf. Sci. 357, 1180–1186 (2015). [CrossRef]
5. E. Louis, A. E. Yakshin, T. Tsarfati, and F. Bijkerk, “Nanometer interface and materials control for multilayer EUV-optical applications,” Prog. Surf. Sci. 86(11-12), 255–294 (2011). [CrossRef]
6. S. Schroder, T. Feigl, A. Duparre, and A. Tunnermann, “EUV reflectance and scattering of Mo/Si multilayers on differently polished substrates,” Opt. Express 15(21), 13997–14012 (2007). [CrossRef]
7. P. C. Pradhan, A. Majhi, and M. Nayak, “Optical performance of W/B4C multilayer mirror in the soft x-ray region,” J. Appl. Phys. 123(9), 095302 (2018). [CrossRef]
8. S. S. Panini, M. Nayak, K. C. Shyama Narendranath, P. C. Pradhan, P. S. Athiray, and P. Sreekumar, “Development of multilayer mirrors for space-based astronomical X-ray optics,” J. Opt. 47(1), 91–95 (2018). [CrossRef]
9. H. Jiang, Z. Wang, and J. Zhu, “Interface characterization of B4C-based multilayers by X-ray grazing-incidence reflectivity and diffuse scattering,” J Synchrotron Radiat 20(3), 449–454 (2013). [CrossRef]
10. D. L. Windt, E. M. Gullikson, and C. C. Walton, “Normal-incidence reflectance of optimized W/B4C x-ray multilayers in the range 1.4 nm < λ < 2.4 nm,” Opt. Lett. 27(24), 2212–2214 (2002). [CrossRef]
11. M. Dai, Z. Zhang, J. T. Zhu, X. Q. Wang, J. Xu, X. H. Fu, L. Bai, Q. S. Huang, Z. S. Wang, and L. Y. Chen, “Influence of interface roughness on reflectivity of tungsten/boron-carbide multilayers with variable bi-layer number by X-ray reflection and diffuse scattering,” Chin. Opt. Lett. 7(8), 738–740 (2009). [CrossRef]
12. J. F. Seely, G. Gutman, J. Wood, G. S. Herman, M. P. Kowalski, J. C. Rife, and W. R. Hunter, “Normal-incidence reflectance of WB4C multilayer mirrors in the 34–50-Å wavelength region,” Appl. Opt. 32(19), 3541–3543 (1993). [CrossRef]
13. A.E. Yakshin, I.I. Khodos, I.M. Zhelezniak, and A. I. Erko, “Fabrication, structure and reflectivity of W/C and W/B4C multilayers for Hard X-Ray,” Opt. Commun. 118(1-2), 133–142 (1995). [CrossRef]
14. A. F. Jankowski, D. M. Makowiecki, M. A. Wall, and M. A. McKernan, “Subnanometer multilayers for x-ray mirrors: Amorphous crystals,” J. Appl. Phys. 65(11), 4450–4451 (1989). [CrossRef]
15. C. C. Walton, G. Thomas, and J. B. Kortright, “X-ray optical multilayers: Microstructure limits on reflectivity at ultra-short periods,” Acta Mater. 46(11), 3767–3775 (1998). [CrossRef]
16. P. N. Rao, S. K. Rai, M. Nayak, and G. S. Lodha, “Stability and normal incidence reflectivity of W/B4C multilayer mirror near the boron K absorption edge,” Appl. Opt. 52(25), 6126–6130 (2013). [CrossRef]
17. P. C. Pradhan, M. Nayak, P. Mondal, and G. S. Lodha, “Interface study and performance of large layer pair ultra-short period W/B4C X-ray multilayer,” AIP Conf. Proc. 1667, 080030 (2015). [CrossRef]
18. L. Assoufid, A. Khounsary, J. Qian, A. Macrander, and D. Mills, “Surface quality of x-ray mirrors evaluated at the advanced photon source,” Proc. SPIE 4501, 54 (2001). [CrossRef]
19. P. K. Yadav, M. H. Modi, M. K. Swami, and P. J. Singh, “Ex-situ characterization of synchrotron radiation induced carbon contamination on LiF window,” J. Electron. Spectrosc. Relat. Phenom. 211, 64–69 (2016). [CrossRef]
20. P. K. Yadav, M. Kumar, R. K. Gupta, M. Sinha, J. A. Chakera, and M. H. Modi, “Refurbishment of an Au-coated toroidal mirror by capacitively coupled RF plasma discharge,” J Synchrotron Radiat 26(4), 1152–1160 (2019). [CrossRef]
21. A. Khounsary, P.J. Eng, L. Assoufid, A. Macrander, and J. Qian, “Recoating mirrors having a chromium underlayer,” in: Advances in Mirror Technology for X-Ray, Euv Lithography, Laser, and Other Applications, Proc. SPIE5193177–181 (2004).
22. H. Moreno Fernandez, D. Rogler, G. Sauthier, M. Thomasset, R. Dietsch, V. Carlino, and E. Pellegrin, “Characterization of Carbon-Contaminated B4C-Coated Optics after Chemically Selective Cleaning with Low-Pressure RF Plasma,” Sci Rep 8(1), 1293 (2018). [CrossRef]
23. C. Morawe, S. Labouré, J.-C. Peffen, F. Perrin, A. Vivo, R. Barrett, C. Morawe, A. M. Khounsary, and S. Goto, “Differential deposition for figure correction of x-ray mirrors,” Proc. SPIE 11108, 1110807 (2019). [CrossRef]
24. C. Santusa, P. Neria, L. Romolib, A. Luteyb, S. Raghavendrac, and M. Benedetti, “Fatigue fracture surface investigations with a 3D optical profiler,” Procedia Structural Integrity 39, 450–459 (2022). [CrossRef]
25. C. Morawe, D. Carau, and J. C. Peffen, “Double multilayer monochromators for upgraded ESRF beamlines,” in Advances in X-Ray/EUV Optics and Components XII (2017).
26. A. Misra and M. Nastasi, “Evolution of tensile residual stress in thin metal films during energetic particle deposition,” J. Mater. Res. 14(12), 4466–4469 (1999). [CrossRef]
27. K. R. Williams, K. Gupta, and M. Wasilik, “Etch rates for micromachining processing - Part II,” J. Microelectromech. Syst. 12(6), 761–778 (2003). [CrossRef]
28. S. Lai, P. Lorenz, M. Ehrhardt, B. Han, D. Hirsch, I. Zagoranskiy, J. Lu, and K. Zimmer, “Laser-induced frontside etching of silicon using 1550-nm nanosecond laser pulses,” Optics and Lasers in Engineering 122, 245–253 (2019). [CrossRef]
29. J. Schmitt, W. Nelissen, U. Wallrabe, and F. Völklein, “Implementation of smooth nanocrystalline diamond microstructures by combining reactive ion etching and ion beam etching,” Diamond Relat. Mater. 79, 164–172 (2017). [CrossRef]
30. C. Romeo and P. Cantù, “Lithography Masks and Pattern Transfer,” Encyclopedia of Condensed Matter Physics 7, 136–145 (2005). [CrossRef]
31. Transene Corp., “Etchant/Metal Compatibility Chart,” https://transene.com/etch-compatibility/ (accessed 25 August 2003).
