Open Access
Add to BibSonomyAbstract
The effectiveness of ion beam planarization (IBP) to reduce surface roughness of diamond turned NiP surfaces was investigated. The surfaces with various spatial wavelengths and depths of turning marks were spray-coated and planarized with broad ion beam. The ion beam planarization was performed at a special angle where the etching rate of photoresist is closely similar to NiP. It is found that the combined process of spray-coating and ion-beam-planarization can effectively reduce the surface roughness of diamond turned NiP. The spatial wavelength and depth of turning marks have limited influence on surface roughness reduction rate. The final surface roughness after ion beam planarization is 30%~40% of the original roughness, irrespective of spatial wavelength and depth of turning marks. Extending planarization time does not alter surface quality after photoresist is etched away. These results show that the IBP is applicable to roughness minimization of diamond turned surfaces.
© 2017 Optical Society of America
1. Introduction
Nickel phosphor (NiP) has been employed to protect the surfaces of many metallic alloys because NiP possesses unique properties of corrosion and wear resistance and relative ease of machining [1]. In extreme ultraviolet lithography (EUVL) systems, NiP plated metal optics are used due to relatively less manufacturing cost and easier machinability [2,3]. NiP plated metal aspheric mirrors are also utilized in neutron scattering facilities [4]. These systems require that the surfaces of mirrors are extremely smooth to reduce scattering loss as much as possible. For examples, surface roughness of <0.3nm (RMS) and 50nm PV surface figure are necessary for next-generation X-ray telescopes [5,6]. The metal optics are usually machined with an ultra-precision diamond turning machine, which can guarantee the required surface figure for various surface shapes like flat, aspheric as well as free-form surfaces. Owing to the mechanics of diamond turning, regular patterns (turning marks) are generally left on machined surface. The marks must be removed in subsequent processing steps because they strongly affect the performance of the optics and increase the scattering of surfaces [4]. Several polishing techniques have been applied to removing the turning marks, including bonnet polishing, fluid jet polishing, mechanical polishing, ion beam etching, etc [4–10]. For instance, Beaucamp et al processed diamond turned NiP with fluid jet containing 4μm and 2μm alumina to reduce down surface roughness and then successively used 0.2μm and 7nm silica to finalize surface roughness to <0.3nm [5].
Ion beam etching is high-precision manufacturing technology and can be made use of to polish a great number of materials and coatings, inclusive of glass, semiconductors, ceramics, metals, and so on [11–19]. Since it is characterized by atomic level removal, ion beam etching is frequently employed to finalize surface roughness of optics [9,16–19]. The material removal in ion beam etching process is influenced by numerous factors, such as ion energy, incident angle of ion beam, kind of ions, etc [20,21]. On the other hand, ion beam etching is a highly deterministic process and thus can be adopted as the final step to correct surface figures of optical components, which is referred to as ion beam figuring [22,23]. As a precise tool, ion beam etching has immense potential in smoothing the surface of dielectrics, semiconductors and metals for applications of X-ray and EUVL. Johnson et al has proposed using a layer of photoresist to aid the polishing of fused silica [24,25], where a planarization angle was adopted and at this angle of incidence the etching rates of coated resist and workpiece are nearly the same and thus ion beam etching can transfer the smoothness of resist into workpiece. This way, the workpiece can be readily planarized which it is otherwise difficult to planarize directly with ion beam. Frost et al have successfully smoothed diamond turned copper and electroless plated NiP surface using ion beam processing with the aid of an auxiliary coating (named ion beam planarization) [7–9]. The surface roughness of copper and NiP has shown to be decreased significantly to <1nm and ~0.3nm, respectively. Although it has been shown that ion beam planarization can smooth diamond turned surfaces, the effects of different spatial frequency (wavelength) and depths of turning marks remain elusive. Therefore, the present work concentrates on the smoothing effectiveness of ion beam planarization on the diamond turned surfaces with varied spatial wavelengths and depth of turning marks. NiP samples with different spatial frequencies and depths of turning marks were planarized under various conditions, including different incident angle of ion beam, etching time, etc. The results suggest that surface roughness of all samples can be satisfactorily decreased after ion beam planarization. The figure of merit (defined to be the ratio of surface roughness after processing to that before processing) is 30%~40%, regardless of the spatial frequency and initial surface roughness of samples. Moreover, surface roughness is hardly altered for long time planarization up to 60min. The experimental procedures and results are detailed in the following sections.
2. Experimental
In order that NiP surface can be smoothed with the aid of photoresist, a common incident angle of ion beam should be searched out first, at which angle the NiP can be etched at a rate nearly the same as photoresist [7–9]. Since the etching rate is changed with respect to the incident angle of ion beam, the etching rate for NiP and photoresist was tested prior to planarizing the diamond-turned NiP surfaces. The experimental results show that incident angle of 30°~40°can yield similar etching rates for both materials. Therefore, the incident angle was set to be 35°in the article unless otherwise specified.
Six electroless plated NiP sampels were diamond-turned to generate different spatial frequencies and depths of turning marks. The initial thickness of NiP layer was ~100 μm and the total amount of material cut away was approximately 20~30 μm for each sample. The diameter of the samples was 53 mm. In general, the profile of the turning marks is determined by the feed rate, the cut depth, and the tip radius of the tool. Thus, to produce the NiP surfaces with the different spatial frequencies of turning marks, the surfaces were turned by varying the feed rate. On the other hand, to produce them with the different depths of turning marks, the surfaces were turned by varying the cut depth using the single crystal diamond tools with different tip radii. The sample was then spray-coated with photoresist and baked after measuring the morphology of diamond turned surface with an AFM (Brucker Dimension Icon, Germany). Prior to etching the samples with an ion beam machine (NTG ISA-200, Germany), the thicknesses of coatings were scanned with thin film analyzing system (Mikropack NanoCalc 2000, Germany) to determine the thickness distribution on each sample. The thickness of coatings is 170~300nm to level off the turning marks (depth of turning marks <100nm). For the spray coating a specially developed resist, optimized for the deposition of thin resist layers (≤ 500 nm) has been used (Allresist GmbH, Germany). For the AFM measurements scan sizes of 20µm × 20µm and 80µm × 80µm were used. The AFM raw data were processed by a line-wise levelling (flatten) using a polynomial of 1st order in order to realize a sample tilt/offset compensation and to remove noise along the slow scan direction. Finally, from the processed images the root mean square roughness (Rq) and the power spectral density (PSD) were calculated. Overall etching time depends on coating thickness, basically 20min~25min, but each sample was etched 5min after etching away the photoresist with ion beam. The flowchart is illustrated in Fig. 1. Sample surface was covered with a mask with a hole (10mm in diameter) during ion beam planarization processes so that diamond turned surface can be etched under various conditions to compare the potential different effects of ion beam etching conditions on the identical original diamond turned surfaces.
Fig. 1 Preparation steps of NiP samples. The diamond turned samples were coated, baked and ion-etched successively. The NiP surface was covered with a mask with a hole of 10mm in diameter during ion beam planarization. This way, the diamond surfaces can be etched under various conditions. (Not to scale)
As to etching conditions, the ion used in our experiments was Ar+ and the flow rate of the noble gas during the etching process was set to 4sccm. The beam voltage was 700V, so the ion energy was ~700eV and the ion beam current was kept constant at 70mA in the experiments. The ion beam current density profile was stabilized for >2h and Faraday scans (to determine the spatial distribution of ion beam) were performed immediately before and after each ion etching run. With the given aperture of the ion source (180mm), the FWHM of the beam profile was between 120mm and 150mm as determined from the Faraday measurements and controlled by footprint measurements.
3. Results and discussion
3.1 Etching rate of NiP and photoresist as well as conditions of diamond turned surfaces
The etching rates of NiP and photoresist depend on the incident angle of ion beam. It is clear that both etching rates of NiP and photoresist basically increase and then drop down with the incident angle of ion beam, as shown in Fig. 2. The rate of NiP is generally greater than photoresist as the angle is <30°while rate of NiP is much lower than that of resist for angles of >40°. Thus there exists an incident angle of ion beam (~35°) at which the rates for both NiP and resist are nearly the same and the corresponding etching rate is 10~12nm/min.
Fig. 2 Etching rate of photoresist and NiP versus the incident angle of ion beam. The rates for NiP and photoresist are almost the same as the incident angle is in between 30°~40°.
The diamond turned surfaces were scanned with AFM to acquire the knowledge of the spatial wavelengths and depths of turning marks as well as surface roughness before and after being spray-coated. For one set of the samples, the spatial wavelengths of diamond turned surface are 1.5µm, 3.5µm, and 6µm while the depth of turning marks is a constant of 20nm illustrated in Fig. 3. For the second set of samples, different depths of turning marks 10nm, 20nm, and 60nm were imprinted into the NiP with the sample spatial wavelength of 6µm. The spatial wavelength of sample F was increased to 25µm while keeping depth of turning marks on the level of 60nm. The characteristics of the diamond turned samples are summarized in Fig. 3.
Fig. 3 Diamond turned surfaces with different spatial wavelengths and depths of turning marks (sample A~F).
3.2 Surface roughness and Figure of Merit (FOM)
The samples were coated to reduce surface roughness before being etched with ion beam. After spray coating, the roughness of original diamond turned surfaces can be significantly reduced by 60%~70%. Then using ion beam to etch away the photoresist and further 5min etching into NiP will generate smooth surface with final surface roughness improved by 60%~70%. Presented in Fig. 4 is the surface morphology of sample A. The RMS surface roughness decreases from >6nm to ~2nm after spray coating and is almost unchanged after undergoing ion beam planarization. The detailed results of all samples are tabulated in Table 1.
Fig. 4 Surface morphology of sample A after spray coating and ion beam planarization. The surface roughness dramatically decreased from >6nm to ~2nm. The sizes of the images are all 80μm × 20μm.
Table 1. Surface roughness of samples with various spatial wavelength before and after ion beam planarization @ 35°.
We define the figure of merit (FOM) to be the ratio of surface roughness prior to and after being processed. Therefore, the smaller the FOM is, the better a processing technique is in our cases. The FOM of a combined process is the product of the FOM of each individual process constituting the combined process. The FOM and surface roughness with respect to different depths of turning marks are plotted in Fig. 5. It is apparent that the depth of turning marks makes trivial difference to the FOM. The surface roughness has been reduced to 30%~40% of original surface roughness regardless of the depth of turning marks on samples B(60nm), D(20nm), E(10nm) as shown in Fig. 5. It is also clear from Fig. 5 that ion beam planarization can reserve the surface roughness throughout the course of etching process at the planarization angle of 35°because the FOMi of ion beam planarization is roughly equal to 1 and the etching behavior of ion beam on photoresist and NiP is similar. The surface roughness of resist is transferred into NiP surface. Similarly, the FOMp is 0.3~0.4 for different spatial wavelengths (1.5µm, 3.5µm, 6µm) when depth of turning marks is 20nm and is irrelevant to spatial wavelength from Fig. 6. Increasing depth of turning marks and spatial wavelength (sample B, F) has little influence on the efficiency of ion beam planarization. The surface roughness after ion beam planarization is decreased to 30%~40% of diamond turned surface.
Fig. 5 The effect of depth of turning marks. (a) FOM for samples B, D, E with varied depth of turning mark after spray coating and ion beam planarization; (b) surface roughness after diamond turning, spray coating and planarization.
Fig. 6 The effect of spatial wavelength. a) FOM for samples A, C, D with different spatial wavelength of turning mark after spray coating and ion beam planarization; (b) surface roughness after diamond turning, spray coating and ion beam planarization.
The surface roughness evolution of NiP under long-time etching (planarization) conditions was also investigated to find out the possible effects of over-etching on surface roughness of NiP. NiP surface is mostly etched for 5min after resist is completely removed in our experiments. However, NiP surface was etched for 1hour in over-etching process. The evolution of NiP surface roughness was investigated by etching sample B for 60min in total after removing photoresist. The surface roughness was recorded after 5min, 10min, 20min, 40min, 60min etching (planarization). The surface morphology after 5min is almost the same as that after 60min etching. The surface roughness remains ~6.30nm (RMS) during the etching process even though the etching time was prolonged to 60min, as shown in Fig. 7. We did not find any obvious change in surface roughness with etching time, indicating that ion beam etching was much stable under experimented conditions. From the results, it can be understood that ion beam etching can retain shapes of original surface in our experiments and the pre-processing (e.g. spray coating) is necessary so as to smooth diamond turned surface while the thickness uniformity of resist is not very important to smoothing effect at the planarization angle as long as the surface of photoresist is sufficiently flat (power spectral density analysis also suggests similar results).
Fig. 7 Surface morphology of diamond-turned NiP sample B after different etching time. The surface roughness kept almost unchanged with increasing etching time up to 60min. The sizes of the images are 80μm × 20μm.
3.3 Power spectral density analysis
The smoothing capability of ion beam planarization over wide spatial wavelength can be evaluated in terms of power spectral density (PSD). The PSD’s of samples C, E, F after diamond turning and ion beam planarization are compared to ascertain the effects of different spatial wavelength on the smoothing ability of ion beam planarization in Fig. 8. All samples can be smoothed over the measured spatial frequency after planarization. Sample E (different area on the same surface) was also etched at incident angle 0°.The roughness of diamond turned surface (Rq = 5.0nm,) can be reduced at both 0°(Rq = 2.4nm) and 35°(Rq = 1.5nm), although the surface seems smoother at the angle of 35°. The PSD plot in Fig. 9 shows that the smoothing effect at 0° is not so effective as the angle 35°, which is consistent with surface roughness measurements. Accordingly, the surface roughness at 0°is inferior to other cases in that the etching rate of NiP is much different from photoresist at 0°. However, the PSD’s have been ameliorated at all incident angles of ion beam as compared to diamond turned surface.
Fig. 8 PSD of samples C,E,F after ion beam planarization at 35°. All the samples can be smoothed over the measured band of spatial frequency, in particular the samples of large spatial wavelength. The diamond turned surfaces were scanned at 85μm × 85μm. The discrepancy between scan size 20μm × 20μm and 80μm × 80μm for sample F after ion beam planarization lies in the great spatial wavelength (25μm) of the sample.
Fig. 9 PSD of sample E after ion beam planarization at different incident angles. The diamond turned surfaces were scanned at 85μm × 85μm and 35μm × 35μm. Both planarization angles at 0°and 35°can improve PSD, but the improvement at 35°appears more significant.
As previously stated, over-etching has ignorable effects on surface roughness. Here the influence of over etching on PSD is examined. The experimental results show that over-etching up to 60min does not alter the PSD of NiP surface. 5min etching is as effective as 60min in improving PSD and reducing surface roughness. The PSD can be hardly improved by means of long-time etching and surface roughness keeps almost constant with etching time in our experiments, 6.29nm after 5min vs. 6.30nm after 60min as shown in Fig. 7. Thus extra etching time is not necessary in order to achieve smoother surface or better PSD.
4. Conclusions
In this article, ion beam planarization of diamond turned NiP surface was studied. We examined the influence of spatial wavelength and depth of diamond turning marks on surface roughness and power spectral density (PSD). The results indicate that surface roughness of diamond-turned NiP can be significantly reduced by combined process of spray coating and ion beam planarization. Spray coating can reduce surface roughness significantly and ion beam planarization retains surface roughness under certain conditions that the etching rates of photoresist and NiP are much similar. The surface roughness drops to 30%~40% of initial diamond turned surface after ion beam planarization, irrespective of spatial wavelength and depth of diamond turning marks on NiP surfaces. Long time planarization of NiP (up to 60min) will have trivial effects on surface roughness. The PSD of NiP surface can also be greatly improved over the measured spatial frequency by ion beam planarization. Long time planarization again has insignificant influence on PSD. The process of ion beam planarization is thus insensitive to an unintentional over etching, at least for the investigated NiP. Therefore, ion beam planarization can be used as a final processing step to reduce the final surface roughness of diamond surfaces for a wide range of spatial wavelength and depth of tool marks.
Funding
Yaguo Li gratefully acknowledges the funding of the Deutscher Akademischer Austauschdienst (DAAD No. 57191174) and the National Natural Science Foundation of China (NSFC No. 51505444).
Acknowledgments
The assistance in this work from Mr. Toni Liebeskind and Mrs. Katrin Ohndorf of the IOM is appreciated.
References and links
1. J. Sudagar, J. Lian, and W. Sha, “Electroless nickel, alloy, composite and nano coatings – A critical review,” J. Alloys Compd. 571, 183–204 (2013). [CrossRef]
2. R. Soufli, E. Spiller, M. A. Schmidt, J. C. Robinson, S. L. Baker, S. Ratti, M. A. Johnson, and E. M. Gullikson, “Smoothing of diamond-turned substrates for extreme ultraviolet illuminators,” Opt. Eng. 43(12), 3089–3095 (2004). [CrossRef]
3. R. Soufli, E. Spiller, M. A. Schmidt, J. C. Robinson, S. L. Baker, S. Ratti, M. A. Johnson, and E. M. Gullikson, “Smoothing of diamond-turned substrates for extreme ultraviolet lithography illuminators,” Proc. SPIE 5193, 98–104 (2004). [CrossRef]
4. S. Takeda, J. Guo, S. Morita, H. Ono, T. Oda, J. Kato, H. Sato, M. Hino, Y. Yamagata, and M. Furusaka, “Development of highly - mechanically polished metal - substrate for neutron supermirrors,” J. Phys. Conf. Ser. 528, 012011 (2014). [CrossRef]
5. A. Beaucamp and Y. Namba, “Super-smooth finishing of diamond turned hard X-ray molding dies by combined fluid jet and bonnet polishing,” CIRP Ann. - Manuf. Technol. 62(1), 315–318 (2013). [CrossRef]
6. Y. Namba, T. Shimomura, A. Fushiki, A. Beaucamp, I. Inasaki, H. Kunieda, Y. Ogasaka, and K. Yamashita, “Ultra-precision polishing of electroless nickel molding dies for shorter wavelength applications,” CIRP Ann. - Manuf. Technol. 57(1), 337–340 (2008). [CrossRef]
7. F. Frost, H. Takino, R. Fechner, A. Schindler, N. Ohi, and K. Nomura, “Sub-nanometer smoothing of diamond-turned metal surfaces using ion beams,” in Towards Synthesis of Micro-/Nano-systems (Springer London), The 11th International Conference on Precision Engineering (ICPE),August 16–18,2006, Tokyo, Japan, pp. 239–242.
8. F. Frost, H. Takino, R. Fechner, A. Schindler, N. Ohi, and K. Nomura, “Smoothing of diamond-turned copper surfaces using ion beams with aid of planarizing film,” Jpn. J. Appl. Phys. 46(9A), 6071–6073 (2007). [CrossRef]
9. F. Frost, R. Fechner, B. Ziberi, J. Völlner, D. Flamm, and A. Schindler, “Large area smoothing of surfaces by ion bombardment: fundamentals and applications,” J. Phys. Condens. Matter 21(22), 224026 (2009). [CrossRef] [PubMed]
10. A. Beaucamp, R. Freeman, R. Morton, K. Ponudurai, and D. D. Walker, “Removal of diamond-turning signatures on x-ray mandrels and metal optics by fluid-jet polishing,” Proc. SPIE 7018, 701835 (2008). [CrossRef]
11. O. C. Yonts, C. E. Normand, and D. E. Harrison Jr., “High-energy sputtering,” J. Appl. Phys. 31(3), 447–450 (1960). [CrossRef]
12. D. Rosenberg and G. K. Wehner, “Sputtering yields for low energy He+-, Kr+-, and Xe+-ion bombardment,” J. Appl. Phys. 33(5), 1842–1845 (1962). [CrossRef]
13. N. Laegreid and G. K. Wehner, “Sputtering yields of metals for Ar+ and Ne+ ions with energies from 50 to 600eV,” J. Appl. Phys. 32(3), 365–369 (1961). [CrossRef]
14. A. B. Meinel, S. Bashkin, and D. A. Loomis, “Controlled figuring of optical surfaces by energetic ionic beams,” Appl. Opt. 4(12), 1674 (1965). [CrossRef]
15. D. M. Allen, P. Shore, R. W. Evans, C. Fanara, W. O’Brien, S. Marson, and W. O’Neill, “Ion beam, focused ion beam, and plasma discharge machining,” CIRP Ann. - Manuf. Technol. 58(2), 647–662 (2009). [CrossRef]
16. T. Arnold, G. Bohm, R. Fechner, J. Meister, A. Nickel, F. Frost, T. Hansel, and A. Schindler, “Ultra-precision surface finishing by ion beam and plasma jet techniques — status and outlook,” Nucl. Instrum. Methods Phys. Res. A 616(2-3), 147–156 (2010). [CrossRef]
17. R. Fechner, D. Flamm, W. Frank, A. Schindler, F. Frost, and B. Ziberi, “Ion beam assisted smoothing of optical surfaces,” Appl. Phys., A Mater. Sci. Process. 78(5), 651–654 (2004). [CrossRef]
18. F. Frost, R. Fechner, B. Ziberi, D. Flamm, and A. Schindler, “Large area smoothing of optical surfaces by low-energy ion beams,” Thin Solid Films 459(1-2), 100–105 (2004). [CrossRef]
19. F. Frost, B. Ziberi, A. Schindler, and B. Rauschenbach, “Surface engineering with ion beams: from self-organized nanostructures to ultra-smooth surfaces,” Appl. Phys., A Mater. Sci. Process. 91(4), 551–559 (2008). [CrossRef]
20. R. G. Wilson, “Ionic polishing of fused silica and glass,” Opt. Technol. 2(1), 19–26 (1970). [CrossRef]
21. S. Somekh, “Introduction to ion and plasma etching,” J. Vac. Sci. Technol. 13(5), 1003–1007 (1976). [CrossRef]
22. P. Gailly and K. Fleury-Frenette, “Ion beam figuring for precision optics,” SPIE Newsroom (2008).
23. T. Hänsel, F. Frost, A. Nickel, and A. Schindler, “Ultra-precision surface finishing by ion beam techniques,” Vakuu. Forsch. Prax. 19(5), 24–30 (2007). [CrossRef]
24. L. F. Johnson and K. A. Ingersoll, “Ion polishing with the aid of a planarizing film,” Appl. Opt. 22(8), 1165–1167 (1983). [CrossRef] [PubMed]
25. L. F. Johnson, K. A. Ingersoll, and D. Kahng, “Planarization of patterned surfaces by ion beam erosion,” Appl. Phys. Lett. 40(7), 636–638 (1982). [CrossRef]
