z-切鈮酸鋰局部表面導電改質製程的開發與特性研究

Abstract

鈮酸鋰(LiNbO3)為積體光學的關鍵基板，然其寬能隙與高電阻率特性，限制了其與主動電子元件的整合。受寬能隙氧化物缺陷工程啟發，本論文開發出一種質子交換輔助的兩階段製程，將共融態生長鈮酸鋰內產生的鋰空缺與反位鈮缺陷，活化為局部導電通道層。製程首先以鎢光罩定義區域並進行質子交換（熔融態苯甲酸/己二酸，235 oC，2小時）以預活化目標區域；隨後進行四階段氨氣（NH3）退火（升降溫速率限制5 oC/min）以誘發氧空缺與氮相關缺陷，其中鎢薄膜兼作光罩與還原保護層。本研究亦排除了早期兩條製程路線：NH2Na輔助還原（因表面殘留副產物）與二氧化鋯(ZrO2)擴散阻擋層，因無法有效阻擋氨氣擴散；惟該次失敗二氧化鋯樣品的XPS量測意外偵測到N 1s訊號，進而導向最終的製程配方，此排除歷程本身即構成核心工程貢獻。經圖案轉移、表面/化學分析（原子力顯微鏡 AFM／掃描式電子顯微鏡 SEM/X光光電子能譜 XPS）及傳輸線模型（transmission line model, TLM）分析驗證，鈮酸鋰之改質區與非鈮酸鋰之改質區的電阻對比達五至六個數量級，片電阻介於數百至104 ohms/sq。霍爾量測顯示，載子以n型為主且局部區域並存p型，於部分樣品中觀察到異常高的遷移率，且整體n/p型載子尚無法穩定控制。此外，本製程存在導電性與光學穿透率的本質性取捨：提高質子交換時間雖有利於提升導電性，卻會加深Nb4+相關的還原著色。

Lithium niobate (LiNbO3) is a key substrate for integrated optics, yet its wide bandgap and high resistivity (10¹³–10¹⁴ Ω·cm) limit its integration with active electronic devices. Inspired by defect engineering in wide-bandgap oxides,this thesis develops a proton-exchange-assisted two-stage process that activates the intrinsic lithium vacancies and antisite niobium defects of congruent LiNbO3 into localized conductive channels. A tungsten mask first defines the target regions, which are pre-activated by proton exchange in a benzoic/adipic acid melt (235 °C, 2 h). A subsequent four-stage NH₃ anneal (ramp rate limited to 5 °C/min) induces oxygen vacancies and nitrogen-related defects, with the tungsten film serving simultaneously as a mask and as a reduction-protective layer. Two earlier routes were ruled out: NaNH2-assisted reduction, which left residual by-products on the surface, and a ZrO2 diffusion barrier, which failed to block NH3 diffusion. An N 1s signal unexpectedly detected by XPS on the failed ZrO2 sample nevertheless pointed to the final recipe, and this process-elimination history constitutes the core engineering contribution of this work. Pattern transfer, surface and chemical characterization (atomic force microscopy, AFM; scanning electron microscopy, SEM; X-ray photoelectron spectroscopy, XPS) and transmission line model (TLM) analysis confirm a resistance contrast of five to six orders of magnitude between modified and unmodified regions, with sheet resistances ranging from several hundred to 104 Ω/sq. Hall measurements show predominantly n-type conduction with locally coexisting p-type behavior, anomalously high mobility in some samples, and n/p carrier type that is not yet stably controlled. The process further exhibits an intrinsic trade-off between conductivity and optical transmittance: a longer proton-exchange time improves conductivity but deepens the Nb4+-related reduction coloration.