- Email: [email protected]
Material removal function of the capacitive coupled hollow cathode plasma source for plasma polishing
Available online at www.sciencedirect.com
Physics Procedia 19 (2011) 408–411
International Conference on Optics in Precision Engineering and Nanotechnology
Material removal function of the capacitive coupled hollow cathode plasma source for plasma polishing Dasen Wanga, Weiguo Liua,b 1*, Yilong Wub, Lingxia Hangb, Huadong Yua, Na Jinb a
College of Mechanical and Electric Engineering, Changchun University of Science and Technology, Changchun 130022, China b Micro-optoelectronic Systems Labs, Xi’an Technological University, Xi’an 710032, China
Abstract A novel plasma polishing process has been developed. In the process, highly stable SF6 and Ar/O2 plasmas were generated by using capacitive coupled hollow cathode (CCHC) RF discharge method. The influences of the plasma source operational parameters such as gas flow rate, gas flow rate ratio of the mixed gases, pressure and discharge power on the material removal function of the plasma polishing process were investigated. Material used in the polishing process was fused silica, which was pre-polished to an initial surface roughness of about 1.4nm rms.
© PublishedbybyElsevier Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of © 2011 2010 Published B.V. the ICOPEN 2011 conference Keywords: plasma polishing; RF discharge plasma; fused silica; removal function; roughness.
1. Introduction Supersmooth optics found important applications in ICF, EUV lithography and high resolution imaging. To fabricate high quality supersmooth optics, different polishing methods have been developed[1,2], including magnetorheological figuring (MRF), elastic emission machining (EEM), ion beam figuring (IBF) and plasma assisted chemical etching (PACE). In late 1980s, Kodak[3] developed the PACE method for the figuring and polishing of large scale mirrors for telescopes with size up to 2.5m. It has high etching rate up to 100ȝm/min. CVM is a technology developed by Nikon and Osaka University for the fabrication of EUV lithography optics[4-6]. The etching rate of CVM method can be as high as 200ȝm/min. Based on the same idea, IOM developed an atmosphere plasma etching method called Plasma Jet Chemical Etching (PJCE)[7]. The PACE, CVM and PJCE share similar working principle. PACE is a sub-aperture polishing method, in which, a plasma source is used to generate plasma tool for the polishing. Scanning of the plasma over the optics to be polished is essential for the polishing process. A control algorithm has to be developed to get a well controlled polishing result. To develop the scanning algorithm, knowledge about the residual error of the polished surface, the scanning path selected and the material removing characteristics of the plasma source must be obtained in advance. * Corresponding author. Tel.: +86-29-83208008; fax: +86-83208210. E-mail address: [email protected].
1875-3892 © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the ICOPEN 2011 conference doi:10.1016/j.phpro.2011.06.183
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
409
A novel vacuum plasma polishing apparatus was established in house[8]. A capacitive coupled hollow cathode (CCHC) RF vacuum discharge process was utilized to generate high density plasma in vacuum to maintain high etching rate and to keep a clean processing environment. In this paper, the influences of the plasma source operational parameters such as gas flow rate, gas flow rate ratio of the mixed gases, pressure and discharge power on the material removal function of the plasma polishing process were investigated. Material used in the polishing process was fused silica, which was pre-polished to an initial surface roughness of about 1.4nm rms. 2. Material removing function for plasma polishing control algorithm[9] The surface residual error can be determined by using interferometer, which is denoted as 'hij. The scanning pathway of the plasma polishing can be represented as that in figure 1(a), which is a meander pattern. The removal function of the plasma etching process can be described as a Gaussian function as that in figure 1(b). There are two major parameters to describe the Gaussian function: the amplitude in terms of maximum etching rate in Pm/s and FWHM in mm.
(a) Figure 1. Scanning pathway (a) and removal function (b)
(b)
The cross-section of the Gaussian function can be described as a two- dimensional function rij. The control algorithm is actually to find the local dwell time matrix tij by knowing the 'hij and rij. (1) Various deconvolution algorithms including Lucy-Richardson, modified Van Cittert, Wiener, and regular filter can be used to find the dwell time matrix tij. After the dwell time matrix tij is found, the residual error topology can be derived: (2) In both equation (1) and (2), the determination of the removal function rij is essential. 3. Experimental The developed CCHC RF discharge facility for the plasma polishing process is shown in Fig. 2(a). In the system, a vacuum chamber is used to facilitate all the parts. The plasma is generated by using a 100MHz RF power source. There are three gas sources in the system, SF6 as the active etching gas, Ar and O2 or their mixture as carrying gases. The pumping system and gas flow controllers are used in the system to keep the working pressure constant. CCHC plasma source is used in the apparatus as shown in figure 2. Appropriate geometric parameter of the plasma source has been investigated to optimize its discharge characteristics to get stable discharge under various conditions including different working pressure, discharge power and the type of working gas. The total gas flow rate of the SF6 and Ar/O2 gases was kept at 200 SCCM, and the gas flow rate of the Ar was kept at 140 SCCM, and O2 gas flow rate was selected between 0 and 60 SCCM with the remaining gas being SF6. The working pressure of the vacuum chamber was kept at 5Pa, the RF power was kept at 60W.
410
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
Figure 2: The plasma polishing apparatus(left), the plasma source (right) 1. plasma source, 2. motion controller, 3. workpiece holder, 4. gas distributor, 5. vacuum chamber, 6. vacuum pump and pressure controller , 7. exhaust scriber, 8. 9. 10. reactive and carrying gases, 11. 12. 13. mass flow controllers, 14. RF feed through, 15. 16. RF power and match box 4. Results and discussions To determine the removal rate under the conditions given above, a thin layer of MgO was deposited onto the fused silica to be polished. The MgO layer was then patterned to expose the underneath silica to the plasma. TaylorHobson CCI2000 white light interferometer (WLI) was used to measure the etched depth, and the removal rate was calculated accordingly. In figure 3, it can be seen that the removal rate is strongly affected by the gas flow rate ratio between the O2 and the SF6. A maximum removal rate was observed when the gas flow rate ratio between the O2 and the SF6 is around 1:1. 4 .0
R e m o v a l R a te
Removal Rate (nm/min)
3 .5 3 .0 2 .5 2 .0 1 .5 1 .0 0 .5
0
10
20
30
40
50
60
S F 6 F lo w R a te (s c c m )
Figure 3. Removal rate of silica at different gas flow rate According to Zou’s experiment[10], the presence of O will promote the dissociation of SF6 into SF5 and F*, the O will associate with the SF5 and prevent the recombination of F* with SF5. So it can be seen that with the increase of the O2 gas flow rate, the removal rate increases. With the further increase of the O2 gas flow rate, the O* generated will react with fused silica surface to form SiO=SiO aggregates, which will reduce the etching rate of the silica. So the removal rate decreases with the increase of the O2 gas flow rate. The plasma source was kept 5mm away from the surface of the polished silica, and the Taylor-Hobson Form Talysurf PGI-1240 surface profiler was used to measure the surface profile of the polished surface. To investigate the geometric parameters of the cathode on the FWHM of the removal function, two cathodes with different diameters were used. The diameter selected was 10mm and 25mm respectively. Figure 4 shows the profile of the etched spots for different cathodes with the O2 gas flow rate kept at 30 SCCM. If the input plasma power was kept constant at 60W, decreases in cathode diameter will result a much higher relative etch rate and much smaller FWHM. This can be easily explained by considering that the active F* radical density increases with the decrease of the cathode diameter.
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
411
10mm cathode FWHM ~6mm
25mm cathode FWHM ~15mm
Figure 4. Etch spots for different cathodes From the result in figure 5, it can be seen that, with the increase of RF power, the polished surface showed an increased roughness. When the RF power is higher than 100W, the surface roughness will be higher than 1.1nm rms. The minimum surface roughness was obtained when the RF power was 60W, which was the reason to keep the RF power to be 60W for other experiments. Ƶbefore polish ƶafter polish
Power /W
Figure 5. Influence of RF power on surface roughness 5. Conclusions The etching spot profile of the plasma polishing can be well described by a Gaussian function. The material removal rate was strongly affected by the gas flow rate, especially the O2 and SF6 gas flow rate ratio. Maximum removal rate was obtained when the O2 and SF6 gas flow rate ratio is 1:1. The removal rate can be further increased by increasing the input plasma power, but the surface roughness increases when the power is higher than 100W. Acknowledgement This work was supported by China Defense Industrial Technology Development Program. References 1. S. D. Jacobs. Convergence, 11(1) (2003) 1. 2. S. D. Jacobs. Convergence, 11(2) (2003) 1. 3. D. Bollinger, G. Gallatin, J. Samuels et al. Proc. SPIE, 1333 (1990)44. 4. Y. Mori, K. Yamamura, K. Yamauchi et al. Nanotechnology, 4(1993)225. 5. H. Takino, N. Shibata, H. Itoh et al. Applied Optics, 37(1998)5198. 6. H. Takino, N. Shibata, H. Itoh et al. Applied Optics, 41(2002)3971. 7. A. Schindler, G. Boehm, T. Hansel et al. Proc. SPIE, 4451(2001)242. 8. W. G. Liu, D. S. Wang, M. D. Hu et al. Proc. SPIE, 7282(2009)72822T. 9. A. Schindler, T. Hansel, F. Frost, et al. Talks on International 21st century COE symposium on atomistic fabrication technology, Osaka, Japan, 2007. 10. H. Zou, Microsystem Technologies, 10(2004) 603.
Physics Procedia 19 (2011) 408–411
International Conference on Optics in Precision Engineering and Nanotechnology
Material removal function of the capacitive coupled hollow cathode plasma source for plasma polishing Dasen Wanga, Weiguo Liua,b 1*, Yilong Wub, Lingxia Hangb, Huadong Yua, Na Jinb a
College of Mechanical and Electric Engineering, Changchun University of Science and Technology, Changchun 130022, China b Micro-optoelectronic Systems Labs, Xi’an Technological University, Xi’an 710032, China
Abstract A novel plasma polishing process has been developed. In the process, highly stable SF6 and Ar/O2 plasmas were generated by using capacitive coupled hollow cathode (CCHC) RF discharge method. The influences of the plasma source operational parameters such as gas flow rate, gas flow rate ratio of the mixed gases, pressure and discharge power on the material removal function of the plasma polishing process were investigated. Material used in the polishing process was fused silica, which was pre-polished to an initial surface roughness of about 1.4nm rms.
© PublishedbybyElsevier Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of © 2011 2010 Published B.V. the ICOPEN 2011 conference Keywords: plasma polishing; RF discharge plasma; fused silica; removal function; roughness.
1. Introduction Supersmooth optics found important applications in ICF, EUV lithography and high resolution imaging. To fabricate high quality supersmooth optics, different polishing methods have been developed[1,2], including magnetorheological figuring (MRF), elastic emission machining (EEM), ion beam figuring (IBF) and plasma assisted chemical etching (PACE). In late 1980s, Kodak[3] developed the PACE method for the figuring and polishing of large scale mirrors for telescopes with size up to 2.5m. It has high etching rate up to 100ȝm/min. CVM is a technology developed by Nikon and Osaka University for the fabrication of EUV lithography optics[4-6]. The etching rate of CVM method can be as high as 200ȝm/min. Based on the same idea, IOM developed an atmosphere plasma etching method called Plasma Jet Chemical Etching (PJCE)[7]. The PACE, CVM and PJCE share similar working principle. PACE is a sub-aperture polishing method, in which, a plasma source is used to generate plasma tool for the polishing. Scanning of the plasma over the optics to be polished is essential for the polishing process. A control algorithm has to be developed to get a well controlled polishing result. To develop the scanning algorithm, knowledge about the residual error of the polished surface, the scanning path selected and the material removing characteristics of the plasma source must be obtained in advance. * Corresponding author. Tel.: +86-29-83208008; fax: +86-83208210. E-mail address: [email protected].
1875-3892 © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the ICOPEN 2011 conference doi:10.1016/j.phpro.2011.06.183
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
409
A novel vacuum plasma polishing apparatus was established in house[8]. A capacitive coupled hollow cathode (CCHC) RF vacuum discharge process was utilized to generate high density plasma in vacuum to maintain high etching rate and to keep a clean processing environment. In this paper, the influences of the plasma source operational parameters such as gas flow rate, gas flow rate ratio of the mixed gases, pressure and discharge power on the material removal function of the plasma polishing process were investigated. Material used in the polishing process was fused silica, which was pre-polished to an initial surface roughness of about 1.4nm rms. 2. Material removing function for plasma polishing control algorithm[9] The surface residual error can be determined by using interferometer, which is denoted as 'hij. The scanning pathway of the plasma polishing can be represented as that in figure 1(a), which is a meander pattern. The removal function of the plasma etching process can be described as a Gaussian function as that in figure 1(b). There are two major parameters to describe the Gaussian function: the amplitude in terms of maximum etching rate in Pm/s and FWHM in mm.
(a) Figure 1. Scanning pathway (a) and removal function (b)
(b)
The cross-section of the Gaussian function can be described as a two- dimensional function rij. The control algorithm is actually to find the local dwell time matrix tij by knowing the 'hij and rij. (1) Various deconvolution algorithms including Lucy-Richardson, modified Van Cittert, Wiener, and regular filter can be used to find the dwell time matrix tij. After the dwell time matrix tij is found, the residual error topology can be derived: (2) In both equation (1) and (2), the determination of the removal function rij is essential. 3. Experimental The developed CCHC RF discharge facility for the plasma polishing process is shown in Fig. 2(a). In the system, a vacuum chamber is used to facilitate all the parts. The plasma is generated by using a 100MHz RF power source. There are three gas sources in the system, SF6 as the active etching gas, Ar and O2 or their mixture as carrying gases. The pumping system and gas flow controllers are used in the system to keep the working pressure constant. CCHC plasma source is used in the apparatus as shown in figure 2. Appropriate geometric parameter of the plasma source has been investigated to optimize its discharge characteristics to get stable discharge under various conditions including different working pressure, discharge power and the type of working gas. The total gas flow rate of the SF6 and Ar/O2 gases was kept at 200 SCCM, and the gas flow rate of the Ar was kept at 140 SCCM, and O2 gas flow rate was selected between 0 and 60 SCCM with the remaining gas being SF6. The working pressure of the vacuum chamber was kept at 5Pa, the RF power was kept at 60W.
410
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
Figure 2: The plasma polishing apparatus(left), the plasma source (right) 1. plasma source, 2. motion controller, 3. workpiece holder, 4. gas distributor, 5. vacuum chamber, 6. vacuum pump and pressure controller , 7. exhaust scriber, 8. 9. 10. reactive and carrying gases, 11. 12. 13. mass flow controllers, 14. RF feed through, 15. 16. RF power and match box 4. Results and discussions To determine the removal rate under the conditions given above, a thin layer of MgO was deposited onto the fused silica to be polished. The MgO layer was then patterned to expose the underneath silica to the plasma. TaylorHobson CCI2000 white light interferometer (WLI) was used to measure the etched depth, and the removal rate was calculated accordingly. In figure 3, it can be seen that the removal rate is strongly affected by the gas flow rate ratio between the O2 and the SF6. A maximum removal rate was observed when the gas flow rate ratio between the O2 and the SF6 is around 1:1. 4 .0
R e m o v a l R a te
Removal Rate (nm/min)
3 .5 3 .0 2 .5 2 .0 1 .5 1 .0 0 .5
0
10
20
30
40
50
60
S F 6 F lo w R a te (s c c m )
Figure 3. Removal rate of silica at different gas flow rate According to Zou’s experiment[10], the presence of O will promote the dissociation of SF6 into SF5 and F*, the O will associate with the SF5 and prevent the recombination of F* with SF5. So it can be seen that with the increase of the O2 gas flow rate, the removal rate increases. With the further increase of the O2 gas flow rate, the O* generated will react with fused silica surface to form SiO=SiO aggregates, which will reduce the etching rate of the silica. So the removal rate decreases with the increase of the O2 gas flow rate. The plasma source was kept 5mm away from the surface of the polished silica, and the Taylor-Hobson Form Talysurf PGI-1240 surface profiler was used to measure the surface profile of the polished surface. To investigate the geometric parameters of the cathode on the FWHM of the removal function, two cathodes with different diameters were used. The diameter selected was 10mm and 25mm respectively. Figure 4 shows the profile of the etched spots for different cathodes with the O2 gas flow rate kept at 30 SCCM. If the input plasma power was kept constant at 60W, decreases in cathode diameter will result a much higher relative etch rate and much smaller FWHM. This can be easily explained by considering that the active F* radical density increases with the decrease of the cathode diameter.
Dasen Wang et al. / Physics Procedia 19 (2011) 408–411
411
10mm cathode FWHM ~6mm
25mm cathode FWHM ~15mm
Figure 4. Etch spots for different cathodes From the result in figure 5, it can be seen that, with the increase of RF power, the polished surface showed an increased roughness. When the RF power is higher than 100W, the surface roughness will be higher than 1.1nm rms. The minimum surface roughness was obtained when the RF power was 60W, which was the reason to keep the RF power to be 60W for other experiments. Ƶbefore polish ƶafter polish
Power /W
Figure 5. Influence of RF power on surface roughness 5. Conclusions The etching spot profile of the plasma polishing can be well described by a Gaussian function. The material removal rate was strongly affected by the gas flow rate, especially the O2 and SF6 gas flow rate ratio. Maximum removal rate was obtained when the O2 and SF6 gas flow rate ratio is 1:1. The removal rate can be further increased by increasing the input plasma power, but the surface roughness increases when the power is higher than 100W. Acknowledgement This work was supported by China Defense Industrial Technology Development Program. References 1. S. D. Jacobs. Convergence, 11(1) (2003) 1. 2. S. D. Jacobs. Convergence, 11(2) (2003) 1. 3. D. Bollinger, G. Gallatin, J. Samuels et al. Proc. SPIE, 1333 (1990)44. 4. Y. Mori, K. Yamamura, K. Yamauchi et al. Nanotechnology, 4(1993)225. 5. H. Takino, N. Shibata, H. Itoh et al. Applied Optics, 37(1998)5198. 6. H. Takino, N. Shibata, H. Itoh et al. Applied Optics, 41(2002)3971. 7. A. Schindler, G. Boehm, T. Hansel et al. Proc. SPIE, 4451(2001)242. 8. W. G. Liu, D. S. Wang, M. D. Hu et al. Proc. SPIE, 7282(2009)72822T. 9. A. Schindler, T. Hansel, F. Frost, et al. Talks on International 21st century COE symposium on atomistic fabrication technology, Osaka, Japan, 2007. 10. H. Zou, Microsystem Technologies, 10(2004) 603.