- Email: [email protected]
Accelerated formation of sodium depletion layer on soda lime glass surface by corona discharge treatment in hydrogen atmosphere
Applied Surface Science 300 (2014) 149–153
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Accelerated formation of sodium depletion layer on soda lime glass surface by corona discharge treatment in hydrogen atmosphere Keiga Kawaguchi a , Hiroshi Ikeda a , Daisuke Sakai a , Shiro Funatsu b , Keiichiro Uraji b , Kiyoshi Yamamoto c , Toshio Suzuki c , Kenji Harada d , Junji Nishii a,∗ a
Research Institute for Electronic Science, Hokkaido University, N20 W10, Kita-ku, Sapporo, Hokkaido 001-0020, Japan Production Technology Center, Asahi Glass Co., Ltd. , 1-1 Suehiro-cyo, Tsurumiku, Yokohama, Kanagawa, 230-0045, Japan c Research Center, Asahi Glass Co., Ltd., 1150 Hazawa-cho, Kanagawa-ku, Yokohama, Kanagawa, 221-8755, Japan d Department of Computer Science, Kitami Institute of Technology, 165 Koen-cho, Kitami, Hokkaido 090-8507, Japan b
a r t i c l e
i n f o
Article history: Received 18 September 2013 Received in revised form 4 February 2014 Accepted 7 February 2014 Available online 17 February 2014 Keywords: Corona discharge Silicate glass Alkali Proton Hydrogen
a b s t r a c t Formation of a sodium depletion layer on a soda lime glass surface was accelerated efficiently using a corona discharge treatment in H2 atmosphere. One origin of such acceleration was the preferential generation of H+ with a larger mobility at an anode needle end with a lower applied voltage than that in air. The second origin was the applied voltage across the glass plate during the corona discharge treatment, which was estimated theoretically as 2.7 times higher than that in air. These two effects doubled the depletion layer thickness compared with that in air.
1. Introduction A high direct current (DC) voltage applied across an alkalicontaining glass induces the migration of alkali ions from the anode side to the cathode side. Subsequently, the alkali depletion layer with a negative charge is formed at the anode side glass surface [1,2]. Such negative charge should be compensated by the penetration of H+ provided by water molecules in air. When the glass surface is coated with a metal electrode, the negative charge remains in the surface layer [3−6], which was used for the secondharmonic generation (SHG) [7,8]. Okada et al. reported SHG in a glass waveguide [9] and a silica glass film [10] after corona discharge treatment, which is known as a non-contact method to apply a high DC voltage to a polymer [11] or a glass [12] without coating any electrode on those materials. The origin of SHG was also regarded as formation of charged defects in the material surface layer. In our previous study, the corona discharge was used for the formation of surface relief gratings on a soda lime glass surface [12,13]. The glass surface was treated using the corona discharge through a
∗ Corresponding author. Tel.: +81 11 706 9377; fax: +81 11 706 9377. E-mail address: [email protected] (J. Nishii). http://dx.doi.org/10.1016/j.apsusc.2014.02.024 0169-4332/© 2014 Elsevier B.V. All rights reserved.
© 2014 Elsevier B.V. All rights reserved.
template of self-assembled hexagonally close-packed polystyrene particles [13]. The diffraction efficiency increased with the treatment time immediately before the damage of polystyrene particles. Additionally, we recently revealed the formation of sodium depletion layer by the injection of H+ to the Na+ sites by the corona discharge treatment [14], which is expected to be closely related to the formation of the alkali depletion pattern with a low refractive index according to the polystyrene template. The diffraction efficiency might increase with the alkali depletion layer thickness. This report describes the effect of H2 gas on accelerating the formation of alkali depletion layer on the glass surface by the corona discharge treatment. Characteristics of the H+ injection from the anode side and the alkali migration to the cathode side in a H2 atmosphere were investigated and discussed quantitatively. 2. Experimental Commercially available soda lime glass plate of 1 mm thickness (AS glass, Tg = 555 ◦ C; Asahi Glass Co. Ltd.) was used for the corona discharge treatment. The main glass components were SiO2 , Na2 O, CaO, with inclusion of small amounts of other alkali and alkali earth metal oxides. The Na2 O concentration is 13 mol%. The corona discharge treatment was performed in an atmosphere of ambient air or 5 N purity N2 and/or H2 . Fig. 1 shows the experimental set up. The
150
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
Fig. 1. Experimental setup for corona discharge treatment in air and N2 −H2 atmospheres.
glass was placed on the cathode of a carbon plate. The anode was a Pt coated steel needle. The end point of needle was located 5 mm above the glass surface. The anode was connected to a DC voltage supplier (AKTB-010K1PN/S, Touwakeisoku Co. Ltd.) and the cathode was grounded. The treatment temperature was kept at 200 ◦ C. The DC voltage applied to the anode needle was increased up to 4 kV with an increasing rate of 0.5 kV/min. The current in the circuit during the corona discharge treatment was measured using a data logger (midi Logger GL220, Graphtec Corp.). The glass surface after the corona discharge treatment was analyzed using an inductive coupled plasma atomic emission spectrometer (ICP-AES; ICPE-9000, Shimadzu Corp.) and a Raman spectrometer (in Via Reflex, Renishaw Plc.). The cross-section of the glass was observed using an optical microscope (BX60F5, OLYMPUS Corp.). The chemical composition of the treated glass surface was investigated using an energy dispersive X-ray spectroscopy (EDS; JED-2300, JEOL Ltd.). The OH concentration of the glass was determined from the infrared (IR) absorption spectra measured by a Fourier transform IR spectrometer (FT-IR; Affinity-1, Shimadzu Corp.). 3. Results Fig. 2 portrays the relation between the applied voltage and the current in the early stage of corona discharge treatment in air and N2 −H2 atmospheres. The current increased gradually with the voltage after the corona discharge occurred at the anode needle point. The current−voltage slope increased steeply with the volume fraction of H2 , which reflects the efficient replacement of Na+ by H+
Fig. 2. Relation between applied voltage and current during corona discharge treatment to soda lime glass in air and N2 −H2 atmospheres at 200 ◦ C.
at the anode side surface and their migration to the cathode side. The current measured during the treatment in air is also shown in the figure, which is much lower than that in the H2 -containing atmosphere. The corona discharge in the H2 atmosphere is therefore effective to accelerate the formation of sodium depletion layer on the anode side glass surface. Fig. 3 shows the current variation against the corona discharge treatment time in (a) air and (b) H2 . The profiles of applied voltage were mutually identical. In the former case, the current decreased slightly during the treatment. By contrast, the current for the latter case increased steeply up to 53 A, which was approximately 10 times higher than that in air. Subsequently it decreased gradually under the constant applied voltage. According to our previous study [14], H+ is the predominant charge carrier in the glass during the corona discharge treatment of the soda lime glass. The mobility of H+ in the glass is smaller than that of Na+ . Hence, the decrease in current indicates the formation of a highly resistive sodium depletion layer by the H+ injection. Rather low and stable current was observed during the treatment in air, which can be attributable to the lower amount of injected H+ in air compared with that in H2 atmosphere. Fig. 4 depicts photographs of the specimens after the treatment for 540 min. White precipitates were observed on the cathode side glass surface. The precipitates were identified as Na2 CO3 using the
Fig. 3. Relation between corona discharge treatment time and current for the soda lime glass at 4 kV, 200 ◦ C in (a) air and (b) H2 atmosphere.
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
151
Fig. 4. Photographs of the cathode side glass surfaces after the corona discharge treatment at 4 kV, 200 ◦ C for 540 min in (a) air and (b) H2 atmosphere.
ICP-AES and the Raman spectrometer [14]. Larger amounts of precipitates were recognized for the specimen treated in H2 than in air. The treated area diameters were 9 mm in air and 16 mm in H2 . Fig. 5 portrays cross-sectional views and concentration profiles of representative cations measured using EDS before and after the corona discharge treatment at 4 kV, 200 ◦ C for 540 min. The crosssectional view revealed the existence of alteration layers of 5 m and 10 m thickness, respectively, for the glasses treated in air and in H2 . The EDS profiles exhibited the exclusive decrease in sodium concentration in the anode side glass surfaces, although no change was observed for any elements in the cathode side glass surface. These results indicate that the Na+ migrated preferentially to the cathode side surface as a predominant charge carrier. Then they were discharged at the cathode side glass surface by capturing an electron supplied from the ground through the cathode carbon electrode. After the treatment, the atomic sodium reacted with an ambient H2 O and CO2 gases. The sodium carbonate should be precipitated. The H+ incorporated in the anode side glass surface is expected to exist as OH groups. Then the increase in OH concentration in the glass was estimated using the two-band method, which gives the H2 O concentration from the OH absorption peaks [15,16]. Fig. 6 shows the treatment time dependence of the H2 O concentration. The amount of migrated Na+ is also shown as Na2 O in the figure, as estimated from the glass composition, the depletion layer thickness and the density of the pristine glass. From these results, it is
evident that the concentrations of both incorporated H+ and the discharged Na+ exhibited almost identical dependence against the corona discharge treatment time, which indicates that the H+ were injected into the glass surface to compensate the charge balance after the migration of Na+ to the cathode side. Furthermore, this charge balance was always guaranteed to be independent from the atmosphere, and the treatment time. 4. Discussion Fig. 7 portrays the time dependences of the sodium depletion layer thickness after the corona discharge treatment in air and in H2 atmosphere. The depletion layer thickness was estimated from the cross-sectional view and the EDS analysis shown in Fig. 6, which mutually agreed rather well. The formation rate of the resistive sodium depletion layer corresponds to the variation of the current in Fig. 3. The variations of sodium depletion layer thickness were simulated using a theoretical model, which was advocated for the description of field-assisted ion-exchange processes to fabricate planar optical waveguides on a glass [17]. The time-dependence of the sodium depletion layer thickness l(t) can be expressed by the following equation. l(t) =
1 ˛NH
2˛N V H H
−1
(1 − ˛NH )2
t
+ L2
−L
(1)
Fig. 5. Cross-sectional EDS profiles (lower) and optical microscope photographs (upper) of the anode side glass surface: (a) untreated, (b) and (c) treated respectively at 4 kV, 200 ◦ C for 540 min in air and H2 atmosphere.
152
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
Fig. 6. Treatment time dependences of H2 O and Na2 O concentrations at 4 kV and 200 ◦ C in (a) air and (b) H2 atmosphere. The former was obtained using the two-band method. The latter was estimated from the glass composition, the depletion layer thickness and the density of the pristine glass.
Therein, H and Na respectively denote the mobilities of H+ and Na+ . Also, L stands for the glass thickness, V is the voltage across the glass, and ˛ = 1 − H /Na represents the ion-mismatched param0 , eter. Normalized concentration NH is expressed as NH = CH /CNa 0 where CH and CNa respectively represent the concentrations of H+ and intrinsic Na+ in the depletion layer. Here, NH is assumed to be 1, because the Na+ in the sodium depletion layer is exchanged completely to the H+ . To estimate the voltage across the glass, the sodium depletion layer thickness shown in Fig. 7 was fitted using Eq. (1) substituting the parameters L = 1 mm and (Na /H ) = 103 [18]. The broken lines in the figure exemplify the calculated results, which well accorded with the experimentally obtained results. Based on these calculations, the voltage across the glass during the corona discharge treatment was respectively evaluated as 920 V in air and 2520 V in H2 . This is the first reason why the injection efficiency of H+ into the glass in H2 was so much more efficient in forming the sodium depletion layer on the glass surface compared with that in air. The second reason is ascribable to the characteristics of ionized atoms and/or molecules at the anode needle top. Fig. 8 represents the voltage distribution between the needle and plate electrodes. The applied voltage was consumed mainly at two regions [19]. First was the ionization area (grow region), where the molecules around the needle electrode are ionized by the high electric field. Second is the drift region, where the ionized atoms and/or molecules are drifted toward to the glass surface. Table 1 represents the principal factors affecting the H+ injection into the glass surface.
Fig. 8. Schematic distribution of voltage during corona discharge treatment estimated using Eq. (1).
Table 1 Principal factors affecting the H+ injection by corona discharge treatment. Atmosphere
Electric strength (0 ◦ C, 1 atm) [20]
Generated cation
Air Hydrogen
35.5 kV cm−1 15.5 kV cm−1
H3 O+ (H2 O)n [21] H+ , H3 + , H5 + [22]
H2 exhibits lower electric strength than that of air [20], resulting higher ionization efficiency of H2 molecules compared with that of air. Furthermore, the dominantly generated ion in air is H3 O+ (H2 O)n [21]. In H2 atmosphere, on the other hand, small size ions such as H+ , H3+ and H5+ are generated [22] and preferentially reached to the glass surface because of their large mobility. According to these reasons on the applied voltage across the glass and the characteristics of the atmosphere, the formation rate of the sodium depletion layer in H2 was faster than that in air. The low-temperature and large-scale surface treatment using corona discharge is particularly beneficial for the formation of alkali-free glass surfaces applicable to productions of information and medical devices. The treatment though a fine pattern is also expected to be attractive for the optical functionalization of glass surfaces. 5. Conclusion
Fig. 7. Sodium depletion layer thickness vs. corona discharge treatment time at 4 kV and 200 ◦ C. Closed symbols and solid lines were obtained, respectively using EDS measurement and by the calculation using Eq. (1).
Efficient generation of sodium depletion layer was realized using the corona discharge treatment in H2 atmosphere. The H+ was injected into the anode side glass surface. The Na+ migrated to the cathode side. The sodium depletion layer thickness was about twice that formed by the treatment in air. The theoretical model
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
used for field-assisted ion-exchange revealed that the applied voltage across the glass during the corona discharge treatment in H2 was about 2.7 times higher than that in air. Furthermore, the corona discharge in H2 occurred at a lower applied voltage than that in air, which suggests the preferential generation of H+ -related ions with a large mobility at the anode needle. Therefore, the treatment in H2 atmosphere is attractive for the rapid fabrication of an alkali-free or H+ -injected thick layer on a glass surface. References [1] U.K. Krieger, W.A. Lanford, Field assisted transport of Na+ ions, Ca2+ ions and electrons in commercial soda-lime glass. 1. Experimental, J. Non-Cryst. Solids 102 (1988) 50–61. [2] E.C. Ziemath, V.D. Araujo, C.A. Escanhoela Jr., Compositional and Structural Changes at the Anodic Surface of Thermally Poled Soda-Lime Float Glass, J. Appl. Phys. 104 (2008) 054912. [3] M. Dussauze, V. Rodriguez, A. Lipovskii, M. Petrov, C. Smith, K. Richardson, T. Cardinal, E. Fargin, E.I. Kamitsos, How does thermal poling affect the structure of soda-lime glass? J. Phys. Chem. C 114 (2010) 12754–12759. [4] C.M. Lepienski, J.A. Giacometti, G.F.L. Ferreira, F.L. Freire Jr., C.A. Achete, Electric field distribution and near-surface modifications in soda-lime glass submitted to a DC potential, J. Non-Cryst. Solids 159 (1993) 204–212. [5] X.M. Liu, M.D. Zhang, Theoretical study for thermal/electric field poling of fused silica, Jpn. J. Appl. Phys. 40 (1) (2001) 4069–4076. [6] V. Pruneri, F. Samoggia, G. Bonfrate, P.G. Kazansky, G.M. Yang, Thermal poling of silica in air and under vacuum: the influence of charge transport on second harmonic generation, Appl. Phys. Lett. 74 (1999) 2423–2425. [7] T.G. Alley, S.R.J. Brueck, R.A. Myers, Space charge dynamics in thermally poled fused silica, J. Non-Cryst. Solids 242 (1998) 165–176. [8] A.L. Moura, M.T. de Araujo, E.A. Gouveia, M.V.D. Vermelho, J.S. Aitchison, Deep and shallow trap contributions to the ionic current in the thermal-electric field poling in soda-lime glasses, Opt. Express 15 (2007) 143–149.
153
[9] A. Okada, K. Ishii, K. Mito, K. Sasaki, Phase-matched second-harmonic generation in novel corona poled glass waveguides, Appl. Phys. Lett. 60 (1992) 2853–2855. [10] A. Okada, K. Ishii, K. Mito, K. Sasaki, Second-order optical nonlinearity in coronapoled glass films, J. Appl. Phys. 74 (1993) 531–535. [11] J.A. Giacometti, P.A. Ribeiro, M. Raposo, J.N. Marat-Mendes, J.S.C. Campos, A.S. DeReggi, Study of poling behavior of biaxially stretched poly(vinylidene fluoride) films using the constant-current corona triode, J. Appl. Phys. 78 (1995) 5597–5603. [12] D. Sakai, K. Harada, S. Kamemaru, T. Fukuda, Hologram replication technique in glass plates using corona charging, Appl. Phys. Lett. 90 (2007) 06112. [13] D. Sakai, D. Miho, K. Harada, D. Barada, T. Fukuda, Replicative fabrication of diffractive structure from self-assembled particles onto a glass substrate using corona-charging treatment, Opt. Rev. 17 (2010) 187–190. [14] H. Ikeda, D. Sakai, S. Funatsu, K. Yamamoto, T. Suzuki, K. Harada, J. Nishii, Generation of alkali-free and high-proton concentration layer in a soda lime glass using non-contact corona discharge, J. Appl. Phys. 114 (2013) 063303. [15] H. Behrens, A. Stuke, Quantification of H2 O contents in silicate glasses using IR spectroscopy—a calibration based on hydrous glasses analyzed by Karl-Fischer titration, Glass Sci. Technol. 76 (2003) 176–189. [16] U. Harder, H. Geissler, IR spectroscopic investigations on the determination of the water content in glasses, Fresenius J. Anal. Chem. 361 (1998) 585–586. [17] X. Prieto, J. Linares, Increasing resistivity effects in field-assisted ion exchange for planar optical waveguide fabrication, Opt. Lett. 21 (1996) 1363–1365. [18] G. Hetherin, K.H. Jack, M.W. Ramsay, High-temperature electrolysis of vitreous silica I, Oxidation ultra-violet induced fluorescence and irradiation color, Phys. Chem. Glasses 6 (1965) 6–15. [19] J.A. Giacometti, O.N. Oliveira, Corona charging of polymers, IEEE. T. Electr. Insul. 27 (1992) 924–943. [20] W.M. Thornton, The electric strength of gases, measured by corona discharge, Phil. Mag. 28 (1939) 666–678. [21] M. Pavlik, J.D. Skalny, Generation of [H3 O]+ .(H2 O)n clusters by positive corona discharge in air, Rapid Commun. Mass Spectrom. 11 (1997) 1757–1766. [22] D.J. Mellor, J.R. Travis, Determination of ion species in a hydrogen corona discharge between concentric hemispheres, J. Phys. D Appl. Phys. 5 (1972) 1117–1119.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Accelerated formation of sodium depletion layer on soda lime glass surface by corona discharge treatment in hydrogen atmosphere Keiga Kawaguchi a , Hiroshi Ikeda a , Daisuke Sakai a , Shiro Funatsu b , Keiichiro Uraji b , Kiyoshi Yamamoto c , Toshio Suzuki c , Kenji Harada d , Junji Nishii a,∗ a
Research Institute for Electronic Science, Hokkaido University, N20 W10, Kita-ku, Sapporo, Hokkaido 001-0020, Japan Production Technology Center, Asahi Glass Co., Ltd. , 1-1 Suehiro-cyo, Tsurumiku, Yokohama, Kanagawa, 230-0045, Japan c Research Center, Asahi Glass Co., Ltd., 1150 Hazawa-cho, Kanagawa-ku, Yokohama, Kanagawa, 221-8755, Japan d Department of Computer Science, Kitami Institute of Technology, 165 Koen-cho, Kitami, Hokkaido 090-8507, Japan b
a r t i c l e
i n f o
Article history: Received 18 September 2013 Received in revised form 4 February 2014 Accepted 7 February 2014 Available online 17 February 2014 Keywords: Corona discharge Silicate glass Alkali Proton Hydrogen
a b s t r a c t Formation of a sodium depletion layer on a soda lime glass surface was accelerated efficiently using a corona discharge treatment in H2 atmosphere. One origin of such acceleration was the preferential generation of H+ with a larger mobility at an anode needle end with a lower applied voltage than that in air. The second origin was the applied voltage across the glass plate during the corona discharge treatment, which was estimated theoretically as 2.7 times higher than that in air. These two effects doubled the depletion layer thickness compared with that in air.
1. Introduction A high direct current (DC) voltage applied across an alkalicontaining glass induces the migration of alkali ions from the anode side to the cathode side. Subsequently, the alkali depletion layer with a negative charge is formed at the anode side glass surface [1,2]. Such negative charge should be compensated by the penetration of H+ provided by water molecules in air. When the glass surface is coated with a metal electrode, the negative charge remains in the surface layer [3−6], which was used for the secondharmonic generation (SHG) [7,8]. Okada et al. reported SHG in a glass waveguide [9] and a silica glass film [10] after corona discharge treatment, which is known as a non-contact method to apply a high DC voltage to a polymer [11] or a glass [12] without coating any electrode on those materials. The origin of SHG was also regarded as formation of charged defects in the material surface layer. In our previous study, the corona discharge was used for the formation of surface relief gratings on a soda lime glass surface [12,13]. The glass surface was treated using the corona discharge through a
∗ Corresponding author. Tel.: +81 11 706 9377; fax: +81 11 706 9377. E-mail address: [email protected] (J. Nishii). http://dx.doi.org/10.1016/j.apsusc.2014.02.024 0169-4332/© 2014 Elsevier B.V. All rights reserved.
© 2014 Elsevier B.V. All rights reserved.
template of self-assembled hexagonally close-packed polystyrene particles [13]. The diffraction efficiency increased with the treatment time immediately before the damage of polystyrene particles. Additionally, we recently revealed the formation of sodium depletion layer by the injection of H+ to the Na+ sites by the corona discharge treatment [14], which is expected to be closely related to the formation of the alkali depletion pattern with a low refractive index according to the polystyrene template. The diffraction efficiency might increase with the alkali depletion layer thickness. This report describes the effect of H2 gas on accelerating the formation of alkali depletion layer on the glass surface by the corona discharge treatment. Characteristics of the H+ injection from the anode side and the alkali migration to the cathode side in a H2 atmosphere were investigated and discussed quantitatively. 2. Experimental Commercially available soda lime glass plate of 1 mm thickness (AS glass, Tg = 555 ◦ C; Asahi Glass Co. Ltd.) was used for the corona discharge treatment. The main glass components were SiO2 , Na2 O, CaO, with inclusion of small amounts of other alkali and alkali earth metal oxides. The Na2 O concentration is 13 mol%. The corona discharge treatment was performed in an atmosphere of ambient air or 5 N purity N2 and/or H2 . Fig. 1 shows the experimental set up. The
150
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
Fig. 1. Experimental setup for corona discharge treatment in air and N2 −H2 atmospheres.
glass was placed on the cathode of a carbon plate. The anode was a Pt coated steel needle. The end point of needle was located 5 mm above the glass surface. The anode was connected to a DC voltage supplier (AKTB-010K1PN/S, Touwakeisoku Co. Ltd.) and the cathode was grounded. The treatment temperature was kept at 200 ◦ C. The DC voltage applied to the anode needle was increased up to 4 kV with an increasing rate of 0.5 kV/min. The current in the circuit during the corona discharge treatment was measured using a data logger (midi Logger GL220, Graphtec Corp.). The glass surface after the corona discharge treatment was analyzed using an inductive coupled plasma atomic emission spectrometer (ICP-AES; ICPE-9000, Shimadzu Corp.) and a Raman spectrometer (in Via Reflex, Renishaw Plc.). The cross-section of the glass was observed using an optical microscope (BX60F5, OLYMPUS Corp.). The chemical composition of the treated glass surface was investigated using an energy dispersive X-ray spectroscopy (EDS; JED-2300, JEOL Ltd.). The OH concentration of the glass was determined from the infrared (IR) absorption spectra measured by a Fourier transform IR spectrometer (FT-IR; Affinity-1, Shimadzu Corp.). 3. Results Fig. 2 portrays the relation between the applied voltage and the current in the early stage of corona discharge treatment in air and N2 −H2 atmospheres. The current increased gradually with the voltage after the corona discharge occurred at the anode needle point. The current−voltage slope increased steeply with the volume fraction of H2 , which reflects the efficient replacement of Na+ by H+
Fig. 2. Relation between applied voltage and current during corona discharge treatment to soda lime glass in air and N2 −H2 atmospheres at 200 ◦ C.
at the anode side surface and their migration to the cathode side. The current measured during the treatment in air is also shown in the figure, which is much lower than that in the H2 -containing atmosphere. The corona discharge in the H2 atmosphere is therefore effective to accelerate the formation of sodium depletion layer on the anode side glass surface. Fig. 3 shows the current variation against the corona discharge treatment time in (a) air and (b) H2 . The profiles of applied voltage were mutually identical. In the former case, the current decreased slightly during the treatment. By contrast, the current for the latter case increased steeply up to 53 A, which was approximately 10 times higher than that in air. Subsequently it decreased gradually under the constant applied voltage. According to our previous study [14], H+ is the predominant charge carrier in the glass during the corona discharge treatment of the soda lime glass. The mobility of H+ in the glass is smaller than that of Na+ . Hence, the decrease in current indicates the formation of a highly resistive sodium depletion layer by the H+ injection. Rather low and stable current was observed during the treatment in air, which can be attributable to the lower amount of injected H+ in air compared with that in H2 atmosphere. Fig. 4 depicts photographs of the specimens after the treatment for 540 min. White precipitates were observed on the cathode side glass surface. The precipitates were identified as Na2 CO3 using the
Fig. 3. Relation between corona discharge treatment time and current for the soda lime glass at 4 kV, 200 ◦ C in (a) air and (b) H2 atmosphere.
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
151
Fig. 4. Photographs of the cathode side glass surfaces after the corona discharge treatment at 4 kV, 200 ◦ C for 540 min in (a) air and (b) H2 atmosphere.
ICP-AES and the Raman spectrometer [14]. Larger amounts of precipitates were recognized for the specimen treated in H2 than in air. The treated area diameters were 9 mm in air and 16 mm in H2 . Fig. 5 portrays cross-sectional views and concentration profiles of representative cations measured using EDS before and after the corona discharge treatment at 4 kV, 200 ◦ C for 540 min. The crosssectional view revealed the existence of alteration layers of 5 m and 10 m thickness, respectively, for the glasses treated in air and in H2 . The EDS profiles exhibited the exclusive decrease in sodium concentration in the anode side glass surfaces, although no change was observed for any elements in the cathode side glass surface. These results indicate that the Na+ migrated preferentially to the cathode side surface as a predominant charge carrier. Then they were discharged at the cathode side glass surface by capturing an electron supplied from the ground through the cathode carbon electrode. After the treatment, the atomic sodium reacted with an ambient H2 O and CO2 gases. The sodium carbonate should be precipitated. The H+ incorporated in the anode side glass surface is expected to exist as OH groups. Then the increase in OH concentration in the glass was estimated using the two-band method, which gives the H2 O concentration from the OH absorption peaks [15,16]. Fig. 6 shows the treatment time dependence of the H2 O concentration. The amount of migrated Na+ is also shown as Na2 O in the figure, as estimated from the glass composition, the depletion layer thickness and the density of the pristine glass. From these results, it is
evident that the concentrations of both incorporated H+ and the discharged Na+ exhibited almost identical dependence against the corona discharge treatment time, which indicates that the H+ were injected into the glass surface to compensate the charge balance after the migration of Na+ to the cathode side. Furthermore, this charge balance was always guaranteed to be independent from the atmosphere, and the treatment time. 4. Discussion Fig. 7 portrays the time dependences of the sodium depletion layer thickness after the corona discharge treatment in air and in H2 atmosphere. The depletion layer thickness was estimated from the cross-sectional view and the EDS analysis shown in Fig. 6, which mutually agreed rather well. The formation rate of the resistive sodium depletion layer corresponds to the variation of the current in Fig. 3. The variations of sodium depletion layer thickness were simulated using a theoretical model, which was advocated for the description of field-assisted ion-exchange processes to fabricate planar optical waveguides on a glass [17]. The time-dependence of the sodium depletion layer thickness l(t) can be expressed by the following equation. l(t) =
1 ˛NH
2˛N V H H
−1
(1 − ˛NH )2
t
+ L2
−L
(1)
Fig. 5. Cross-sectional EDS profiles (lower) and optical microscope photographs (upper) of the anode side glass surface: (a) untreated, (b) and (c) treated respectively at 4 kV, 200 ◦ C for 540 min in air and H2 atmosphere.
152
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
Fig. 6. Treatment time dependences of H2 O and Na2 O concentrations at 4 kV and 200 ◦ C in (a) air and (b) H2 atmosphere. The former was obtained using the two-band method. The latter was estimated from the glass composition, the depletion layer thickness and the density of the pristine glass.
Therein, H and Na respectively denote the mobilities of H+ and Na+ . Also, L stands for the glass thickness, V is the voltage across the glass, and ˛ = 1 − H /Na represents the ion-mismatched param0 , eter. Normalized concentration NH is expressed as NH = CH /CNa 0 where CH and CNa respectively represent the concentrations of H+ and intrinsic Na+ in the depletion layer. Here, NH is assumed to be 1, because the Na+ in the sodium depletion layer is exchanged completely to the H+ . To estimate the voltage across the glass, the sodium depletion layer thickness shown in Fig. 7 was fitted using Eq. (1) substituting the parameters L = 1 mm and (Na /H ) = 103 [18]. The broken lines in the figure exemplify the calculated results, which well accorded with the experimentally obtained results. Based on these calculations, the voltage across the glass during the corona discharge treatment was respectively evaluated as 920 V in air and 2520 V in H2 . This is the first reason why the injection efficiency of H+ into the glass in H2 was so much more efficient in forming the sodium depletion layer on the glass surface compared with that in air. The second reason is ascribable to the characteristics of ionized atoms and/or molecules at the anode needle top. Fig. 8 represents the voltage distribution between the needle and plate electrodes. The applied voltage was consumed mainly at two regions [19]. First was the ionization area (grow region), where the molecules around the needle electrode are ionized by the high electric field. Second is the drift region, where the ionized atoms and/or molecules are drifted toward to the glass surface. Table 1 represents the principal factors affecting the H+ injection into the glass surface.
Fig. 8. Schematic distribution of voltage during corona discharge treatment estimated using Eq. (1).
Table 1 Principal factors affecting the H+ injection by corona discharge treatment. Atmosphere
Electric strength (0 ◦ C, 1 atm) [20]
Generated cation
Air Hydrogen
35.5 kV cm−1 15.5 kV cm−1
H3 O+ (H2 O)n [21] H+ , H3 + , H5 + [22]
H2 exhibits lower electric strength than that of air [20], resulting higher ionization efficiency of H2 molecules compared with that of air. Furthermore, the dominantly generated ion in air is H3 O+ (H2 O)n [21]. In H2 atmosphere, on the other hand, small size ions such as H+ , H3+ and H5+ are generated [22] and preferentially reached to the glass surface because of their large mobility. According to these reasons on the applied voltage across the glass and the characteristics of the atmosphere, the formation rate of the sodium depletion layer in H2 was faster than that in air. The low-temperature and large-scale surface treatment using corona discharge is particularly beneficial for the formation of alkali-free glass surfaces applicable to productions of information and medical devices. The treatment though a fine pattern is also expected to be attractive for the optical functionalization of glass surfaces. 5. Conclusion
Fig. 7. Sodium depletion layer thickness vs. corona discharge treatment time at 4 kV and 200 ◦ C. Closed symbols and solid lines were obtained, respectively using EDS measurement and by the calculation using Eq. (1).
Efficient generation of sodium depletion layer was realized using the corona discharge treatment in H2 atmosphere. The H+ was injected into the anode side glass surface. The Na+ migrated to the cathode side. The sodium depletion layer thickness was about twice that formed by the treatment in air. The theoretical model
K. Kawaguchi et al. / Applied Surface Science 300 (2014) 149–153
used for field-assisted ion-exchange revealed that the applied voltage across the glass during the corona discharge treatment in H2 was about 2.7 times higher than that in air. Furthermore, the corona discharge in H2 occurred at a lower applied voltage than that in air, which suggests the preferential generation of H+ -related ions with a large mobility at the anode needle. Therefore, the treatment in H2 atmosphere is attractive for the rapid fabrication of an alkali-free or H+ -injected thick layer on a glass surface. References [1] U.K. Krieger, W.A. Lanford, Field assisted transport of Na+ ions, Ca2+ ions and electrons in commercial soda-lime glass. 1. Experimental, J. Non-Cryst. Solids 102 (1988) 50–61. [2] E.C. Ziemath, V.D. Araujo, C.A. Escanhoela Jr., Compositional and Structural Changes at the Anodic Surface of Thermally Poled Soda-Lime Float Glass, J. Appl. Phys. 104 (2008) 054912. [3] M. Dussauze, V. Rodriguez, A. Lipovskii, M. Petrov, C. Smith, K. Richardson, T. Cardinal, E. Fargin, E.I. Kamitsos, How does thermal poling affect the structure of soda-lime glass? J. Phys. Chem. C 114 (2010) 12754–12759. [4] C.M. Lepienski, J.A. Giacometti, G.F.L. Ferreira, F.L. Freire Jr., C.A. Achete, Electric field distribution and near-surface modifications in soda-lime glass submitted to a DC potential, J. Non-Cryst. Solids 159 (1993) 204–212. [5] X.M. Liu, M.D. Zhang, Theoretical study for thermal/electric field poling of fused silica, Jpn. J. Appl. Phys. 40 (1) (2001) 4069–4076. [6] V. Pruneri, F. Samoggia, G. Bonfrate, P.G. Kazansky, G.M. Yang, Thermal poling of silica in air and under vacuum: the influence of charge transport on second harmonic generation, Appl. Phys. Lett. 74 (1999) 2423–2425. [7] T.G. Alley, S.R.J. Brueck, R.A. Myers, Space charge dynamics in thermally poled fused silica, J. Non-Cryst. Solids 242 (1998) 165–176. [8] A.L. Moura, M.T. de Araujo, E.A. Gouveia, M.V.D. Vermelho, J.S. Aitchison, Deep and shallow trap contributions to the ionic current in the thermal-electric field poling in soda-lime glasses, Opt. Express 15 (2007) 143–149.
153
[9] A. Okada, K. Ishii, K. Mito, K. Sasaki, Phase-matched second-harmonic generation in novel corona poled glass waveguides, Appl. Phys. Lett. 60 (1992) 2853–2855. [10] A. Okada, K. Ishii, K. Mito, K. Sasaki, Second-order optical nonlinearity in coronapoled glass films, J. Appl. Phys. 74 (1993) 531–535. [11] J.A. Giacometti, P.A. Ribeiro, M. Raposo, J.N. Marat-Mendes, J.S.C. Campos, A.S. DeReggi, Study of poling behavior of biaxially stretched poly(vinylidene fluoride) films using the constant-current corona triode, J. Appl. Phys. 78 (1995) 5597–5603. [12] D. Sakai, K. Harada, S. Kamemaru, T. Fukuda, Hologram replication technique in glass plates using corona charging, Appl. Phys. Lett. 90 (2007) 06112. [13] D. Sakai, D. Miho, K. Harada, D. Barada, T. Fukuda, Replicative fabrication of diffractive structure from self-assembled particles onto a glass substrate using corona-charging treatment, Opt. Rev. 17 (2010) 187–190. [14] H. Ikeda, D. Sakai, S. Funatsu, K. Yamamoto, T. Suzuki, K. Harada, J. Nishii, Generation of alkali-free and high-proton concentration layer in a soda lime glass using non-contact corona discharge, J. Appl. Phys. 114 (2013) 063303. [15] H. Behrens, A. Stuke, Quantification of H2 O contents in silicate glasses using IR spectroscopy—a calibration based on hydrous glasses analyzed by Karl-Fischer titration, Glass Sci. Technol. 76 (2003) 176–189. [16] U. Harder, H. Geissler, IR spectroscopic investigations on the determination of the water content in glasses, Fresenius J. Anal. Chem. 361 (1998) 585–586. [17] X. Prieto, J. Linares, Increasing resistivity effects in field-assisted ion exchange for planar optical waveguide fabrication, Opt. Lett. 21 (1996) 1363–1365. [18] G. Hetherin, K.H. Jack, M.W. Ramsay, High-temperature electrolysis of vitreous silica I, Oxidation ultra-violet induced fluorescence and irradiation color, Phys. Chem. Glasses 6 (1965) 6–15. [19] J.A. Giacometti, O.N. Oliveira, Corona charging of polymers, IEEE. T. Electr. Insul. 27 (1992) 924–943. [20] W.M. Thornton, The electric strength of gases, measured by corona discharge, Phil. Mag. 28 (1939) 666–678. [21] M. Pavlik, J.D. Skalny, Generation of [H3 O]+ .(H2 O)n clusters by positive corona discharge in air, Rapid Commun. Mass Spectrom. 11 (1997) 1757–1766. [22] D.J. Mellor, J.R. Travis, Determination of ion species in a hydrogen corona discharge between concentric hemispheres, J. Phys. D Appl. Phys. 5 (1972) 1117–1119.