薄膜沉積反應(yīng)腔室多場建模及輪廓調(diào)控方法研究
本文選題:PECVD + 多場耦合; 參考:《清華大學(xué)》2015年博士論文
【摘要】:薄膜沉積工藝是半導(dǎo)體制造、MEMS、表面工程等領(lǐng)域的基礎(chǔ)工藝。等離子體增強(qiáng)化學(xué)氣相沉積(PECVD)是重要的薄膜沉積工藝方法之一,其工藝系統(tǒng)涉及復(fù)雜多場耦合及化學(xué)反應(yīng)過程。目前PECVD反應(yīng)腔室設(shè)計與工藝調(diào)控的研究普遍依賴經(jīng)驗試錯與大量的工藝參數(shù)試驗,不僅需要大量的時間與物力成本,且復(fù)雜的工藝過程沒有得到揭示。實質(zhì)上,設(shè)備的工藝參數(shù)是影響薄膜沉積工藝性能的表象因素,其深層次的因素是物質(zhì)、能量的空間分布(稱之為工藝因素輪廓)。本文著眼于反應(yīng)腔室的多場特性及工藝?yán)砘^程,從工藝因素輪廓的視角,建立多場耦合工藝系統(tǒng)模型、揭示工藝機(jī)理、研究反應(yīng)腔室設(shè)計與工藝調(diào)控的新思路新方法。針對300 mm豎直噴淋式PECVD反應(yīng)腔室,建立流場、熱場、等離子體子系統(tǒng)模型,為工藝系統(tǒng)建模及反應(yīng)腔室設(shè)計解決子系統(tǒng)建模中的關(guān)鍵結(jié)構(gòu)特征建模的難點問題。基于子系統(tǒng)模型與實驗,研究工藝及結(jié)構(gòu)參數(shù)對子系統(tǒng)物理場特性的影響規(guī)律。針對PECVD工藝過程,研究工藝多場耦合建模方法,建立包含熱流、物質(zhì)輸運(yùn)、化學(xué)反應(yīng)及等離子體放電過程的工藝系統(tǒng)模型,構(gòu)建從工藝參數(shù)輸入到薄膜性能參數(shù)輸出的工藝過程仿真系統(tǒng)。以SiH4/NH3制備氮化硅薄膜工藝為對象,通過片內(nèi)薄膜沉積速率及N/Si分布的對比驗證工藝系統(tǒng)模型的有效性,并探究工藝過程中物質(zhì)及能量過程對薄膜沉積速率及N/Si的影響機(jī)制,為深入研究反應(yīng)腔室結(jié)構(gòu)設(shè)計及工藝調(diào)控提供了模型及理論支撐。針對反應(yīng)腔室設(shè)計與工藝調(diào)控的共性問題——輪廓調(diào)節(jié)問題,建立輪廓誤差反饋模型及其求解方法。基于輪廓誤差構(gòu)造設(shè)計/控制變量序列的反饋迭代算法,求解已知目標(biāo)輪廓確定物理空間多自由度的幾何結(jié)構(gòu)、介質(zhì)屬性、源、邊界條件等逆問題,為多自由度反應(yīng)腔室設(shè)計及控制提供了有效的求解方法。為實現(xiàn)工藝因素空間分布及薄膜性能參數(shù)片內(nèi)分布梯度的精細(xì)化調(diào)控,提出面向輪廓梯度調(diào)控的可控型/阻抗型的反應(yīng)腔室設(shè)計方法,并基于多場模型及輪廓誤差反饋模型方法對設(shè)計方案進(jìn)行求解。以非均勻阻抗模塊為實施案例,實現(xiàn)了對薄膜沉積速率、折射率、溫度輪廓梯度的精細(xì)化調(diào)節(jié),為大面積高一致性薄膜沉積反應(yīng)腔室設(shè)計與工藝調(diào)控提供了一種新的解決方案。
[Abstract]:Thin film deposition is the basic process in semiconductor manufacturing and surface engineering. Plasma enhanced Chemical Vapor deposition (PECVD) is one of the most important process methods for thin film deposition. Its process system involves complex multi-field coupling and chemical reaction processes. At present, PECVD chamber design and process control generally rely on empirical trial and error and a large number of process parameters test, which not only requires a lot of time and material costs, but also the complex process has not been revealed. In essence, the process parameters of the equipment are the apparent factors that affect the properties of the film deposition process, and the deeper factors are the spatial distribution of the material and the energy (which is called the profile of the process factors). This paper focuses on the multi-field characteristics of the reaction chamber and the process of physicochemical process. From the point of view of the outline of the process factors, the model of the multi-field coupling process system is established, the mechanism of the process is revealed, and a new way of thinking about the design and control of the reaction chamber is studied. For the 300mm vertical spray PECVD chamber, the flow field, thermal field and plasma subsystem model are established. The key structural features of the subsystem modeling are solved for the process system modeling and the design of the reaction chamber. Based on the subsystem model and experiment, the influence of process and structure parameters on the physical field characteristics of subsystem is studied. For the PECVD process, the multi-field coupling modeling method is studied, and the process system model including heat flux, mass transport, chemical reaction and plasma discharge process is established. A process simulation system from the input of process parameters to the output of film performance parameters is constructed. Silicon nitride thin films were prepared by SiH _ 4 / NH _ 3. The effectiveness of the system model was verified by comparing the deposition rate and the distribution of N / Si, and the influence mechanism of material and energy process on the deposition rate and N / Si was investigated. It provides a model and theoretical support for the further study of the structure design and process control of the chamber. Aiming at the common problem in the design of reaction chamber and process control, the contour adjustment problem, the contour error feedback model and its solving method are established. A feedback iterative algorithm based on contour error to construct the sequence of design / control variables is used to solve the inverse problems such as geometric structure, medium attribute, source, boundary condition and so on. It provides an effective solution method for the design and control of multi-degree of freedom reaction chamber. In order to realize the fine control of the spatial distribution of process factors and the in-chip distribution gradient of film performance parameters, a controllable / impedance chamber design method for profile gradient control was proposed. The design scheme is solved based on multi-field model and contour error feedback model. Taking the non-uniform impedance module as an example, the fine adjustment of deposition rate, refractive index and temperature profile gradient is realized, which provides a new solution for the design and process control of large area and high consistency film deposition reaction chamber.
【學(xué)位授予單位】:清華大學(xué)
【學(xué)位級別】:博士
【學(xué)位授予年份】:2015
【分類號】:TB306
【相似文獻(xiàn)】
相關(guān)期刊論文 前3條
1 趙前程,鄧善熙,梁發(fā)云;非整圓輪廓的諧波估計方法研究[J];應(yīng)用科學(xué)學(xué)報;2004年03期
2 吳曉君;路超;高斌;許甫茹;;建筑草圖Sketchup下自動生成模型輪廓軌跡的系統(tǒng)開發(fā)[J];機(jī)床與液壓;2013年19期
3 ;[J];;年期
相關(guān)會議論文 前2條
1 王文遠(yuǎn);;一種用于選擇性分割的動態(tài)輪廓模型[A];中國工程物理研究院科技年報(2005)[C];2005年
2 趙向陽;張有為;;視覺語音識別中的函數(shù)變形模板灰度輪廓向量表征法[A];第十屆全國信號處理學(xué)術(shù)年會(CCSP-2001)論文集[C];2001年
相關(guān)博士學(xué)位論文 前6條
1 李由;基于輪廓和邊緣的空間非合作目標(biāo)視覺跟蹤[D];國防科學(xué)技術(shù)大學(xué);2013年
2 夏煥雄;薄膜沉積反應(yīng)腔室多場建模及輪廓調(diào)控方法研究[D];清華大學(xué);2015年
3 李燕;視覺感知中的閉合輪廓提取方法研究[D];北京交通大學(xué);2011年
4 朱張勤;二維光纖光譜數(shù)據(jù)處理中的若干問題研究[D];中國科學(xué)技術(shù)大學(xué);2011年
5 趙前程;輪廓形狀特征的建模與參數(shù)辨識及其應(yīng)用[D];合肥工業(yè)大學(xué);2004年
6 陸承恩;基于形狀的圖像輪廓賦形及目標(biāo)檢測技術(shù)[D];華中科技大學(xué);2009年
相關(guān)碩士學(xué)位論文 前10條
1 周吉;自然圖象中輪廓檢測方法研究與實現(xiàn)[D];華中科技大學(xué);2009年
2 陳敏;基于輪廓模型的細(xì)胞圖像分割與檢測[D];浙江工業(yè)大學(xué);2009年
3 吳曉唯;基于動態(tài)增強(qiáng)核磁共振時間序列腎臟圖像的動態(tài)輪廓分割[D];上海交通大學(xué);2011年
4 郭強(qiáng);基于輪廓編組和形狀指導(dǎo)的目標(biāo)檢測算法研究[D];國防科學(xué)技術(shù)大學(xué);2011年
5 歐陽徠;輪胎斷面輪廓重構(gòu)關(guān)鍵技術(shù)研究[D];華南理工大學(xué);2010年
6 毛曉艷;低信噪比的二維光纖光譜數(shù)據(jù)輪廓恢復(fù)及擬合方法[D];中國科學(xué)技術(shù)大學(xué);2015年
7 王建步;錯覺輪廓及缺失邊界修復(fù)[D];中國石油大學(xué);2008年
8 曹軍;基于動態(tài)輪廓模型的圖像分割算法研究[D];太原理工大學(xué);2010年
9 陳俊;基于輪廓模型的醫(yī)學(xué)圖像分割[D];西安電子科技大學(xué);2012年
10 朱燕平;基于活動感興趣區(qū)域的腦組織自動提取方法[D];南昌航空大學(xué);2012年
,本文編號:2037117
本文鏈接:http://www.wukwdryxk.cn/kejilunwen/cailiaohuaxuelunwen/2037117.html