本論文主要探討鎂釔鋅合金在鋯基電解液進行電漿電解氧化(Plasma Electrolytic Oxidation)中，弱工作(火花區)(soft sparking)發生的參數及弱工作發生的機制，在回顧弱工作現象的論文，可以知道弱工作現象伴隨的現象為電壓的下降、聲音及亮度的減弱及能夠得到較為緻密且厚實的氧化膜，大幅提升電漿電解氧化應用的可能性。然而以往研究，並沒有對發生弱工作的參數，有完整的了解，較多提及的條件為:負電量需大於正電量，或負電流大於正電流。而在以下實驗將進行電流及電量的討論，分別定義為電流比(IR)及電荷比(CR)，使用的IR值為0.86(A+/A-, 0.6/0.7)及1.17(A+/A-, 0.7/0.6)，而CR值為(Q+/Q-) 0.33至1.71
實驗結果顯示，在不同電性參數下能得到不同的弱工作行為，在電流比(IR)0.86時，電荷比(CR)越高，弱工作發生時間越早發生，亦即正電量越多，越早達到弱工作發生的條件，而在電流比(IR)為1.17時，如果正電量多於負電量，未發生弱工作現象，而在負電量大於正電量，亦即CR值0.33, 0.46, 0.58時，弱工作現象發生且CR值越低，弱工作發生時間越早，於此可看見兩個趨勢，當陰極電流大於陽極電流時，正電量越大越快達到弱工作發生的條件，而當陽極電流大於陰極電流時，則是負電量越大越快達到弱工作發生條件，而在IR 1.17 條件下的電漿強度較強，弱工作現象發生後，電壓會再次的攀升，爾後再次的下降，如此反覆，當IR值為1.17時，電場強度較強的狀態下，所產生的膜層，氧化鋯的吸附較多，處理時間45分鐘時，含鋯量高的膜層厚度約60um，整體厚度約為160um而在IR0.86 條件下氧化膜，含鋯量高的膜層厚度約20um，整體厚度約為120um，且IR值1.17膜層厚但膜層中鮮少或沒有電漿通道，而在IR值為0.86有少許的電漿通道，推測為在IR值為0.86時，在還沒發生弱工作現象時電漿所造成的電漿通道，弱工作現象後沒有足夠數量的電漿產生熔融填補通道，而在IR值為1.17

時，電壓會反覆上升，卻只有些許的光亮及聲響，推測此時的電漿數量多且密集，但為微小電漿，可以熔融後填補為發生弱工作現象時所造成的大電漿通道。
為了將添加相鍍製到膜層中，除了使用氟鋯酸鉀，也添加二氧化鋯的顆粒，實驗中使用微米粒徑及奈米粒徑的顆粒，在相同電性參數條件下IR1.17、CR0.33，由X-RAY及EDS結果看出不同顆粒，所形成的相成分及含鋯量差異性大，使用奈米顆粒含鋯量最高可達48.26%，結晶性也高於氧化鎂，而使用微米顆粒和鋯量最高為12.26，膜層含鋯的結晶性亞於氧化鎂，不同粒徑電解液的PH值及導電度皆無太大差異，卻表現出不同的弱工作發生時間或下降幅度，由結果推斷，此原因為試片氧化膜成分有關，氧化鋯成分含量高的氧化膜，在IR值0.86時，表現出弱工作發生時間較晚，且下降斜率緩的電壓。
由於弱工作伴隨現象之一，電壓的下降，思考電壓下降的原因，在實驗過程中電流維持固定，電壓與電阻成正比，電壓下降亦即電阻下降，而造成氧化物導電度提高的原因，可能為氫缺陷，電漿電解技術中，試片處於陰極時產生氫，且在陽極時，陽離子與氫及氧，產生氫缺陷的氧化物，當導電高的氧化物，連通試片氧化膜時，電壓下降，在傅立葉紅外光譜中也看到於弱工作處理的試片有明顯的O-H峰值，更證實了氫於電獎電解氧化時，與氧化物的反應，也在Mott–Schottky量測結果顯示，弱工作處理過後的試片載子濃度的提高，可能的原因亦為氫缺陷的存在，使其載子濃度提升。

This paper aims to define the comprehensive parameters of soft sparking and mechanism in the plasma electrolytic oxidation (PEO). Soft sparking denotes an abrupt transition in the micro arc state of seemingly routine PEO. It is manifested by the declines in light and acoustic emission intensities. This allows hard, thick and dense coatings. Currently, there is a large amount of dispersed experimental data describing specific features of PEO treatment under the soft sparking mode. Unfortunately, the absence of a fundamental understanding of the soft sparking is a main limiting factor for its industrial application. Now we define the current ratio (IR) and charge ratio (CR) to discuss the experiments.
Result shows that there are two kinds of soft sparking under different value of IR. With IR value 0.86, soft sparking happen earlier when CR value increased. In contrast, the soft sparking wouldn’t happen under IR 1.17 with CR bigger than one. With IR value 1.17, soft sparking only happen with CR value less than one. Most interesting is that the voltage would rise after soft sparking and voltage decline again under IR 1.17. Although voltage rise but there are no obvious light or sound of discharge even when the voltage reach to 600 volt but fragmentary light and acoustic. It is indicate that electric field is stronger of sparking mode of IR 1.17 than sparking mode of IR 0.86. Coatings with soft sparking mode of IR 1.17 is rich in zirconia and thicker than the coatings with soft sparking mode of IR 0.86. Function layer which contain zirconia in soft sparking mode of IR1.17 is 60 um and total oxide thickness is 160 um. In the other hand, thickness of coatings with soft sparking mode of IR 0.86 is 120 um and function layer is 20 um. Also it is hard to observe discharge channels in coatings under soft sparking mode of IR1.17 but a few channel in the coatings with soft sparking mode of IR 0.86. We assume that discharge channels formed in arcing regime and it is hard to fill with melting in soft sparking mode of IR 0.86 but it is filled after soft sparking mode of IR 1.17. Because of the stronger electric field after soft sparking, there are numerous and small discharge would modify and fill the channels due to higher conductivity oxide.
In order to obtain the zirconia in the coatings, we add the zirconia powder in the electrolyte. We use micro size and nano-size of zirconia particle with same electric parameters in experiment and they show different behavior of soft sparking behavior and oxide composition. The atomic percentage of zirconia in the coatings with nano-particle electrolyte shows the 48.6% by EDS analysis. In contrast, the coatings with micro-particle electrolyte only perform 12.26%. from the X-ray analysis, the zirconia crystallization is higher than the MgO with nano-particle electrolyte. Specially, we observe that the density and complete coating can be formed with nano-particle electrolyte under soft sparking mode of IR 1.17. or it would form a porous and drastic destroyed coating under soft sparking mode of IR1.17 with micro-particle electrolyte. Also we observe that soft sparking would happen later with IR 0.86 with nano-particle electrolyte. It reveal that the oxide composition and soft sparking correlate to each other. The coatings with high zirconia result in later transition of soft sparking and the reason might be the oxide with zirconia is not easy influenced
One of phenomenon accompanied with the soft sparking is drop of voltage. By the rule of V=IR, we thought the reason is drop of resistance because current is constant during the process. We assume the hydrogen defect is produced when specimen is in cathodic period and hydrogen is produced in anodic period. With the proof that the carrier density is increasing after soft sparking process. It is opposite to other paper said that the carrier density would decrease after PEO. The specimen would show the O-H bonding peak in the Infrared spectroscopy but specimen without soft sparking.

Keyword: Plasma electrolytic oxidation, Magnesium alloy, Zirconia, soft sparking, current ratio, charge ratio

[1] R.H.J. Hannink, "Microstructural Development of Sub-Eutectoid Aged MgO-ZrO2 Alloys" J.Mater. Sci., 18, 457-470 (1983)
[2] R. Chaim and D.G. Brandon, "Microstructure Evolution and Ordering in Commercial Mg-PSZ", J. Mater. Sci., 19, 2934-2942 (1984).
[3]R. Ambat, Naing Naing Aung and W.zhou, "Evaluation of microstructural effects on corrosion behavior of AZ91D magnesium alloy", Corrosion Science, Vol.42, p.1433, 2000.
[4] M.M. Lohrengel, Thin anodic oxide layers on aluminum and other valve metals, Mater. Sci. Eng. R11 (1993) 243-294.
[5]C.K. Dyer, Electrolytic rectification and cathodic charge reversibility of some valve metals, Electrocomp. Sci. Technol. 1 (1974) 121-127.
[6]A. Middelhoek, The mechanism of electrolytic rectification, J. Electrochem. Soc. 111 (1964) 379-380.
[7] W. Gebarowski, S. Pietrzyk, Influence of the cathodic pulse on the formation and morphology of oxide coatings on aluminum produced by plasma electrolytic oxidation, Arch. Metall. Mater. 58 (2013) 241-245.
[8] F. Jaspard-Mécuson, T. Czerwiec, G. Henrion, T. Belmonte, L. Dujardin , A. Viola, J. Beauvir, Tailored aluminum oxide layers by bipolar current adjustment in the plasma electrolytic
[9] C. Blawert, W. Dietzel, E. Ghali, G. Song, Anodizing treatments for magnesium alloys and their effect on corrosion resistance in various environments, Adv. Eng. Mater. 8 (2006) 511-533.
[10] J.W. Diggle, T.C. Downie, C.W. Goulding, Anodic oxide films on aluminum, Chem. Rev. 69 (1969) 365-405.
[11] H.A. Samir, H.C. Qian, Effects of cathodic component of current on porosity and hardness characteristics of micro plasma oxidation coatings on aluminum alloy, Trans. Nonferrous Met. Soc. China 15 (2005) 113-118.
[12] W. Xue, Z. Deng, R. Chen, T. Zhang, Growth regularity of ceramic coatings formed by micro arc oxidation on Al-Cu-Mg alloy, Thin Solid Films 372 (2000) 114-117.
[13] W. Xue, Z. Deng, Y. Lai, R. Chen, Analysis of phase distribution for ceramic coatings formed by microarc oxidation on aluminum alloy, J. Am. Ceram. Soc. 81 (1998) 1365-1368.
[14] V.N. Malyshev, G.A. Markov, V.A. Fedorov, A.A. Petrosyants, G.P. Terleeva, Features of the structure and properties of coatings applied by the method of micro arc oxidation, Chem. Pet. Eng. 20 (1984) 41-43.
[15] A.V. Timoshenko, Y.V. Magurova, Application of oxide coatings to metals in electrolytic solutions by microplasma methods, Rev. Metal. (Madrid) 36 (2000) 323-330.
[16] H. Duan, Y. Li, Y. Xie, S. Chen, Transient voltage-current characteristics: new insights into PEO process of aluminum alloy, Int. J. Electrochem. Sci. 7 (2012) 7619-7630.
[17] J. Martin, A. Nomine, F. Brochard, J.-L. Briancon, C. Noel, T. Belmonte, T. Czerwiec, G. Henrion, Delays in micro-discharges appearance during PEO of Al: Evidence of a mechanism of charge accumulation at the oxide-electrolyte interface, Appl. Surf. Sci. 410 (2017) 29-41.
[18] S.C. Farmer and L.H. Schoenlein, "Precipitation of Mg2Zr5O12 in MgO-Partially Stabilized ZrO2", J.Am. Ceram. Soc., 66, 104-9 (1987).
[19] P. Bala Srinivasan, J.Liang, C.Blawert, M. stormer, W. Dietzel, "Effect of current density on the microstructure and corrosion behavior of plasma electrolytic oxidation treated AM50 magnesium alloy", Applied Surface Science 255 (2009) 4212-4218.
[20] R. Chaim and D.G. Brandon, "Microstructure Evolution and Ordering in Commercial Mg-PSZ", J. Mater. Sci., 19, 2934-2942 (1984).
[21] S. C. Farmer and L. H. Schoenlein, “Precipitation of Mg2Zr5O12 in MgO-Partially-Stabilized ZrO2., J. Am. Ceram. Soc., 66, 107-9(1987)
[22] K. M. Lee, K. R. Shin, S. Namgung, B. Yoo and D. H. Shin, "Electrochemical response of ZrO2-incorporated oxide layer on AZ91 Mg alloy processed by plasma electrolytic oxidation." Surface and Coating Technology 205 (2011) 3779-3784
[23] Hongping Duan, Chuanwei Yan, Fuhui Wang, “Effect of electrolyte additives on performance of plasma electrolytic oxidation films formed on magnesium alloy AZ91D” H. Duan et al. / Electrochimica Acta 52 (2007) 3785–3793
[24] Hongping Duan, Chuanwei Yan, Fuhui Wang, “Growth process of plasma electrolytic oxidation films formed on magnesium alloy AZ91D in silicate solution” H. Duan et al. / Electrochimica Acta 52 (2007) 5002–5009
[25] S. K. MOHAPATRA, S. K. TIKU, and F. A. KROGER, ”The Defect Structure of Unintentionally Doped a-AI203 Crystals” Journal of The American Ceramic Society-Mohpatm et al. Vol. 62, No. 1-2
[26] etin Kılıc¸ Alex Zungerb “n-type doping of oxides by hydrogen” Appl. Phys. Lett., Vol. 81, No. 1, 1 July 2002
[27] Truls Norby, Marius Widerøe, Ronny Glöckner and Yngve Larring, “Hydrogen in oxides” D a l t o n T r a n s . , 2 0 0 4 , 3 0 1 2 – 3 0 1 8
[28] Chris G. Van de Walle, “Hydrogen as a Cause of Doping in Zinc Oxide” 0031-9007_00_85(5)_1012(4)$15.00 © 2000 The American Physical Society
[29] Matthew D. McCluskey, Marianne C. Tarun, and Samuel T. Teklemichael, “Hydrogen in oxide semiconductors” J. Mater. Res., Vol. 27, No. 17, Sep 14, 2012 _ Materials Research Society 2012
[30] Hefeng Cheng, Meicheng Wen, Xiangchao Ma, Yasutaka Kuwahara,
Kohsuke Mori, Ying Dai, Baibiao Huang, and Hiromi Yamashita Hydrogen Doped “Metal Oxide Semiconductors with Exceptional and Tunable Localized Surface Plasmon Resonances” Journal of the American Chemical Society
[31] M. Kodu, M. Aints, T. Avarmaa, V. Denks, E. Feldbach, R. Jaaniso, M. Kirm, A. Maaroos, J. Raud, “Hydrogen doping of MgO thin films prepared by pulsed laser deposition” M. Kodu et al. / Applied Surface Science 257 (2011) 5328–5331
[32] Vijayalakshmi V1, Dhanasekaran “Synthesis and structural properties characterization of Ap/MgO nanocomposites for biomedical applications.
” Biol Med Case Rep 2017 Volume 1 Issue 1
[33] Heekyoung Kwon and Dong Gon Park ”Infra-Red Study of Surface Carbonation on Polycrystalline Magnesium Hydroxide” Bull. Korean Chem. Soc. 2009, Vol. 30, No. 11 2569
[34] ERNEST0 L. RODRIGUEZ, The BF Goodrich Company, Avon Lake “On the Thickening Mechanism of Unsaturated Polyesters by Alkaline Earth Oxides and Hydroxides” Journal of Applied Polymer Science, Vol. 40, 1847-1855 (1990)
[35] 張筑婷, 周振嘉, “正負電荷比對鎂合金微弧氧化鍍製含鋯膜層之影響與腐蝕特性分析”, 國立台灣科技大學機械系, P124-125, 2014.
[36] Yoshihito Kawamura, Kentaro Hayashi, Akihisa Inoue, Tsuyoshi Masumoto "Rapidly Solidified Powder Metallurgy Mg97Zn1Y2Alloys with Excellent Tensile Yield Strength above 600 MPa" Materials Transactions, Vol.42, No.7 (2001) pp.1172 to 1176