TW517306B - Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method - Google Patents
Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method Download PDFInfo
- Publication number
- TW517306B TW517306B TW90121996A TW90121996A TW517306B TW 517306 B TW517306 B TW 517306B TW 90121996 A TW90121996 A TW 90121996A TW 90121996 A TW90121996 A TW 90121996A TW 517306 B TW517306 B TW 517306B
- Authority
- TW
- Taiwan
- Prior art keywords
- processed
- etching
- pattern
- interference light
- amount
- Prior art date
Links
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Drying Of Semiconductors (AREA)
Abstract
Description
517306 A7 B7 五、發明説明(彳) 【發明領域】 本發明係關於半導體元件製程的終點判定方法與裝置 —II------mr, - (請先閲讀背面之注意事項再填寫本頁) 以及使用此方法與裝置的被處理材的處理方法與裝置,特 別是關於藉由發光光譜法判定半導體積體電路裝置的製造 等中的被處理材的蝕刻量或對被處理材的成膜量檢測製程 的終點之半導體元件製程終點判定方法與裝置,以及使用 此方法與裝置的被處理材的處理方法。特別是關於正確地 測定藉由利用電漿(P丨a s m a )放電的蝕刻處理配設在基板 上的各種層的蝕刻量,適合作爲所希望的飩刻深度以及膜 厚的被處理材的深度以及膜厚測定方法以及裝置,與使用 此方法與裝置的被處理材的處理方法以及裝置。 / 【發明背景】 【習知技藝之說明】 線一 經濟部智慧財產局員工消費合作社印製 半導體晶圓的製造對於在晶圓的表面上形成的各種材 料的層以及特別是介電材料的層的除去或圖案形成,廣泛 地使用乾式蝕刻(Dry etching )。對於製程參數(Process parameter)的控制最重要的是在這種層的加工中正確地決 定用以在所希望的蝕刻深度以及膜厚停止蝕刻的蝕刻終點 (End-point) ° 在半導體晶圓的乾式蝕刻處理中,電漿光中的特定波 長的發光強度係伴隨著特定膜的蝕刻進行而變化。因此’ 半導體晶圓的蝕刻終點檢測方法之一,習知以來有在乾式 蝕刻處理中檢測來自電漿的特定波長的發光強度變化’根 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) -4- 517306 經濟部智慧財產局員工消費合作社印製 A7 _____B7 _五、發明説明(2 ) 據此檢測結果檢測特定膜的蝕刻終點的方法。這時需要防 止根據因雜訊(Noise)所造成的檢測波形的搖晃之誤檢測 。用以精度良好地檢測發光強度的變化的方法,例如已知 有日本特開昭6 1 - 5 3 7 2 8號公報、日本特開昭6 3 一 200 5 3 3號公報等。曰本特開昭6 1 — 5 372 8 號公報係藉由移動平均法,而且,日本特開昭6 3 -2 0 0 5 3 3號公報係藉由利用一次最小二乘法的近似處 理進行雜訊的降低。 伴隨著近年的半導體的微細化、高積集化開口率(半 導體晶圓的被蝕刻面積)變小,自光感測器(Sensor)取 入到光檢測器的反應生成物的特定波長的發光強度變的微 弱。其結果來自光檢測器的取樣(Sampling )訊號的等級 (Level )小,終點判定部很難根據來自光檢測器的取樣訊 號確實地檢測蝕刻的終點。 而且,當檢測鈾刻的終點使處理停止時,實際上介電 層的剩餘厚度與預定値相等很重要。習知的製程係使用根 據各層的蝕刻速度爲一定的前提的時間厚度控制技法,監 視全體的製程。蝕刻速度的値例如預先處理取樣晶片( Sample wafer)而求得。此方法藉由時間監視法經過對應 預定的蝕刻膜厚的時間,同時停止鈾刻製程。 但是,實際的膜已知藉由例如L P C V D (低壓化學 氣相沉積,Low Pressure Chemical Vapor Deposition )技 法所形成的S i〇2層其厚度的再現性低。L P C V D中的 製程變動所造成的厚度的容許誤差相當於S i〇2層的初期 本紙張尺度適用中國國家標準(CNS ) A4規格(210 X297公釐) (請先閱讀背面之注意事項再填寫本頁) •^^1. 、11 線 -5- 517306 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明(3) 厚度的約1 0 %。因此,利用時間監視的方法無法正確地 測定殘留於矽基板上的S i〇2層的實際最終厚度。因此, 剩餘的層的實際厚度最終藉由使用標準的光譜干涉儀( Spectral interferometer)的技法而測定,當判定被進行過 度蝕刻(Over etch )時,該晶圓當作不合格而廢棄。 而且,絕緣膜触刻裝置已知隨著重複蝕刻使鈾刻速度 降低等的經時變化。依照情況也有在蝕刻途中停止的情形 ,其解決爲必須。除此之外,爲了製程的穩定運轉,監控 (Monitor)鈾刻速度的經時變動也很重要,惟習知的方法 僅爲終點判定的時間監控,無應付蝕刻速度的經時變化或 變動的適切方法。而且,蝕刻時間爲1 0秒左右短的情形 的終點判定必須是縮短判定準備時間的終點判定方法,以 及也需要充分地縮短判定時間的時刻,惟未必充分。再者 ,絕緣膜被蝕刻面積爲1 %以下的情形很多,伴隨著蝕刻 而產生的來自反應生成物的電漿發光強度變化小。因此, 需要即使是些微的變化也能檢測出的終點判定系統,惟實 用上找不到廉價的系統。 另一方面’半導體晶圓的蝕刻終點檢測方法的其他方 法已知有使用日本特開平5 - 1 7 9 4 6 7號公報、日本 特開平8 — 2 7 4 0 8 2號公報、日本特開平2 0 0 0 — 97648號公報、日本特開平2000—106356 號公報等所揭示的干涉儀的方法。此干涉儀使用方法自雷 射放出的單色放射線係以垂直入射角碰觸包含異種材的疊 層構造的晶圓。例如在S i 3 N 4層上疊層S i〇2層疊層 (請先閲讀背面之注意事項再填寫本頁) -^^1.517306 A7 B7 V. Description of the Invention (彳) [Field of the Invention] The present invention relates to a method and device for determining the end point of a semiconductor device process—II ------ mr,-(Please read the precautions on the back before filling this page ) And processing method and device for processing materials using this method and device, in particular, to determine the etching amount of the processing material or the film formation of the processing material in the manufacturing of semiconductor integrated circuit devices, etc. by light emission spectroscopy Method and device for determining the end point of a semiconductor device process at the end of a mass detection process, and a method for processing a material to be processed using the method and device. In particular, it is suitable for accurately measuring the etching amount of various layers provided on a substrate by an etching process using plasma discharge, and is suitable as a desired etched depth and depth of a material to be processed, and Film thickness measurement method and device, and method and device for processing material to be processed using the method and device. / [Background of the invention] [Explanation of the know-how] The manufacturing of printed semiconductor wafers by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs for the production of layers of various materials formed on the surface of the wafer, especially the layers of dielectric materials For dry removal or pattern formation, dry etching is widely used. For the control of process parameters, the most important thing is to accurately determine the end-point of etching to stop etching at the desired etching depth and film thickness during the processing of this layer. In the dry etching process, the light emission intensity of a specific wavelength in the plasma light changes as the specific film is etched. Therefore, 'One of the methods for detecting the etching end point of semiconductor wafers, there is a conventional method to detect the change in the luminous intensity of a specific wavelength from the plasma in a dry etching process.' The basic paper size applies to the Chinese National Standard (CNS) A4 specification (210X297 mm). ) 517 306 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 _____B7 _V. Description of the Invention (2) A method for detecting the etching end point of a specific film based on the test results. At this time, it is necessary to prevent erroneous detection based on the shaking of the detection waveform caused by noise. As a method for accurately detecting a change in light emission intensity, for example, Japanese Patent Application Laid-Open No. 6 1-5 3 7 2 8 and Japanese Patent Application Laid-Open No. 6 3-200 5 3 3 are known. Japanese Unexamined Patent Publication No. 6 1 — 5 372 8 uses a moving average method, and Japanese Unexamined Patent Publication No. 6 3-2 0 0 5 3 3 performs miscellaneous processing by approximation using a least square method. Reduced information. With the recent miniaturization of semiconductors and a high accumulation aperture ratio (the area to be etched on semiconductor wafers), light emitted from a photodetector (Sensor) to a specific wavelength of a reaction product is emitted. The intensity becomes weak. As a result, the level of the sampling signal from the photodetector is small, and it is difficult for the endpoint determination unit to reliably detect the end of the etching based on the sampling signal from the photodetector. Furthermore, when detecting the end of the uranium engraving to stop processing, it is actually important that the remaining thickness of the dielectric layer is equal to the predetermined plutonium. The conventional process uses a time-thickness control technique based on the premise that the etching rate of each layer is constant, and monitors the entire process. The value of the etching rate is obtained, for example, by processing a sample wafer in advance. This method uses a time monitoring method to elapse a time corresponding to a predetermined etching film thickness, and simultaneously stops the uranium etching process. However, the actual film is known to have a low reproducibility of the thickness of the Si 102 layer formed by, for example, a Low Pressure Chemical Vapor Deposition (LPC VD) technique. The tolerance of the thickness caused by the process variation in LPCVD is equivalent to the initial stage of the Si02 layer. The paper size applies the Chinese National Standard (CNS) A4 specification (210 X297 mm) (Please read the precautions on the back before filling in this Page) • ^^ 1., 11 lines -5- 517306 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (3) About 10% thick. Therefore, the actual final thickness of the Si 102 layer remaining on the silicon substrate cannot be accurately measured by the time monitoring method. Therefore, the actual thickness of the remaining layers is finally determined by using a standard spectral interferometer (Spectral interferometer) technique. When it is judged that the wafer is over etched, the wafer is rejected as rejected and discarded. In addition, it is known that the insulating film etching device changes with time, such as a decrease in the uranium etching speed with repeated etching. Depending on the situation, it may stop during the etching, and its solution is necessary. In addition, for the stable operation of the process, it is also important to monitor the change over time of the uranium engraving speed. However, the conventional method is only for the time monitoring of the end point determination. Appropriate method. In addition, the end point determination in the case where the etching time is about 10 seconds is short. The end point determination method for shortening the determination preparation time and the time for which the determination time needs to be shortened sufficiently are not necessarily sufficient. In addition, there are many cases where the area of the insulating film to be etched is 1% or less, and the plasma luminous intensity change from the reaction product accompanying the etching is small. Therefore, there is a need for an endpoint determination system that can detect even slight changes, but in practice, no inexpensive system can be found. On the other hand, other methods for detecting the etching endpoint of a semiconductor wafer are known from Japanese Patent Application Laid-Open No. 5-1 7 9 4 6 7; Japanese Patent Application Laid-Open No. 8-2 7 4 0 8 2; The method of the interferometer disclosed in 2000-97648, Japanese Patent Application Laid-Open No. 2000-106356, and the like. This interferometer uses a monochromatic radiation emitted from a laser to contact a wafer of a stacked structure containing dissimilar materials at a normal incidence angle. For example, the S i〇2 layer is stacked on the S i 3 N 4 layer (please read the precautions on the back before filling this page)-^^ 1.
、1T 線一 本紙張尺度適用中國國家標準(CNS )八4規格(210X297公釐) -6 - 517306 A7 B7 五、發明説明(4) 中,藉由S i 〇2層頂面所反射的放射光與形成於S i〇2 層與S 1 3 N 4層之間的邊界面所反射的放射光形成干涉條 紋。反射的放射光被照射到適當的檢測器,此乃因蝕刻中 的S i〇2層的厚度而生成強度變化的訊號。蝕刻製程中若 露出S i〇2層的頂面,則立即可連續正確地監視鈾刻速度 與現行蝕刻厚。取代雷射已知也有利用光譜儀計測由電漿 所放出的預定放射光的方法。 【發明槪要】 歸納上述各習知文獻的內容的話如以下所示。 在曰本特開平5 — 1 79 46 7號公報中,使用紅、 綠、藍三種類的濾色器(Color filter)檢測干涉光(電漿 光),進行蝕刻的終點檢測。 而且,日本特開平8 — 274082號公報(USP 5 6 5 8 4 1 8 )使用兩個波長的干涉波形的時間變化與 其微分波形,計算干涉波形的極値(波形的最大、最小: 微分波形的零通過點)。藉由計測計算値(Count value ) 到達預定値爲止的時間,算出蝕刻速度,根據算出的蝕刻 速度求出到達預定膜厚爲止的剩餘蝕刻時間,根據此時間 進行蝕刻製程的停止。 而且,日本特開平2 000 - 97648號公報求出 處理前的干涉光的光強度圖案(Pattern )(以波長爲參數 )與處理後或處理中的干涉光的光強度參數的差的波形( 以波長爲參數),藉由比較該差波形與被資料庫(Data 本紙張尺度適用中國國家標準(CNS ) A4規格(210 X 297公釐) (請先閲讀背面之注意事項再填寫本頁) 訂 -線 經濟部智慧財產局員工消費合作社印製 517306 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明(5) base)化的差波形測定層差(膜厚)。 而且,日本特開平2000—106356號公報係 關於旋轉塗佈裝置,測定遍及多波長的干涉光的時間變化 求出膜厚。 而且,U S P 6 0 8 1 3 3 4藉由測定干涉光的時 間變化的特徵動作而求出資料庫化,藉由該資料庫與所測 定的干涉波形的比較進行蝕刻的終了判定。藉由此判定促 進蝕刻製程條件的變更。 以上的公知例產生以下的問題點。 (1 )、若進行使用罩幕(Mask )材(例如光阻( Resist)、氮化膜、氧化膜)的蝕刻,則來自被融刻的材料 的干涉光被來自罩幕材的干涉光重疊。 (2 )、被進行製程處理的被處理材的材料(例如矽 以及配設於其上的罩幕材)的蝕刻處理,因罩幕材與矽一 起被蝕刻,故有僅謀求被處理材的鈾刻量(鈾刻深度)無 法正確地測定矽的蝕刻量之虞。 (3 )、量產製程的加工用晶圓因起因於裝置( Device)構造罩幕材的初期厚度或被蝕刻材的初期厚度在 晶圓面內具有分布,故來自不同膜厚的干涉光被重疊。 由以上的點很難以要求的測定精度正確地測定/控制 被處理層(成爲半導體製程處理的對象層),特別是電漿 蝕刻處理中的被處理層的蝕刻深度或殘膜量。 本發明的目的係提供可解消上述習知技術的問題點的 半導體元件製程的終點判定方法與裝置以及使用此方法與 (請先閲讀背面之注意事項再填寫本頁) %1. 訂 線一 本紙張尺度適用中國國家標準(CNS ) A4規格(210 X297公釐) -8- 517306 A7 B7 五、發明説明(6) 裝置的被處理材的處理方法與裝置。 本發明的其他目的係提供在電漿處理特別是在電漿鈾 刻處理中,可在線上(〇 η -1 i n e )正確地測定被處理層的實 際蝕刻深度或殘膜量的半導體元件製程的終點判定方法與 裝置以及使用此方法與裝置的被處理材的處理方法與裝置 〇 本發明的再其他目的係提供可在線上高精度地控制半 導體裝置的各層於預定的蝕刻深度以及膜厚的蝕刻製程。 本發明的再其他目的係提供可在線上正確地測定被處 理層的實際蝕刻深度以及膜厚的被處理材的蝕刻深度或膜 厚測定裝置。 本案發明者等對於應解消上述習知技術的問題點而且 應達成上述本案發明的目的之一例,關於複數波長的每一 個求出其干涉波形的時間微分的波形,根據此波形求出顯 示干涉波形的微分値的波長依存性的圖案(即以波長爲參 數的干涉波形的微分値的圖案),以使用該圖案進行膜厚 的測定。 在本案發明中使用顯示干涉波形的時間微分値的波長 依存性的圖案之理由如以下所示。 由於以飩刻中的i η - s i t u (即時,R e a I t i m e )測定爲前 提的計測,故被處理膜的膜厚時時刻刻在變化。因此,干 涉波形的時間微分處理爲可能。再者,藉由此微分處理可 進行干涉波形的雜訊除去。 而且,被蝕刻的材料(例如矽與罩幕材的氮化膜)的 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) —I------F, (請先閲讀背面之注意事項再填寫本頁) 訂 •線·- 經濟部智慧財產局員工消費合作社印製 -9- 517306 A7 _ B7_ 五、發明説明(7 ) 屈折率對波長不同。因此,藉由遍及多波長的干涉光計測 可檢測各個物質的特徵變化(膜厚依存)。 如果依照本案發明的一面,測定被處理材的鈾刻量的 蝕刻深度以及膜厚測定方法係具備: a) 、設定對包含罩幕材的第一被處理材的預定蝕刻 量的干涉光的微分値的以波長爲參數的第一被處理材的標 準圖案之步驟; b) 、設定對該第一被處理材的該罩幕材的預定蝕刻 量的干涉光的微分値的以波長爲參數的罩幕材的標準圖案 之步驟; c )、關於複數波長分別測定關於與該第一被處理材 相同構成的第二被處理材的干涉光強度,求出以該被測定 的干涉光強度的微分値的波長爲參數的實圖案之步驟;以 及 d)、根據該第一被處理材的標準圖案與該罩幕材的 標準圖案與該實圖案,求出該第二被處理材的蝕刻量之步 驟。 如果依照本發明,可提供在電漿處理特別是在電漿飩 刻處理中,可在線上(On-line )正確地測定被處理層的實 際蝕刻量的被處理材的蝕刻深度以及膜厚測定方法以及使 用該方法的被處理材的的試樣的處理方法。 而且,可提供在線上高精度地控制半導體裝置的各層 成爲預定的蝕刻量的蝕刻製程。再者,可提供在線上正確 地測定被處理層的實際蝕刻量的被處理材的蝕刻深度或膜 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) ': -10 - (請先閱讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 517306 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明(8) 厚測定裝置。 【圖式之簡單說明】 圖1係顯示具備依照本發明的第一實施例之蝕刻量測 定裝置的半導體晶圓的蝕刻裝置的全體構成之方塊圖。 圖2 A係顯示蝕刻處理途中的被處理材的縱剖面形狀 圖。 圖2 B、2 C係分別顯示不同波長干涉光的波形實圖 案的例子圖。 圖3A、3B係以對應圖2A、2B的A、B、C所 示的矽晶圓的各被蝕刻表面,以及a、b、c所示的罩幕 的各被蝕刻表面之干涉光的微係數値時系列資料的波長爲 參數的圖。 圖4係顯示以圖1的蝕刻量測定裝置求出進行蝕刻處 理時,被處理材的層差以及罩幕殘膜厚度的順序之流程圖 〇 圖5係顯示具備依照本發明的第一實施例的變形例之 蝕刻深度測定裝置的半導體晶圓的蝕刻裝置的全體構成之 方塊圖。 圖6係顯示圖5的實施例的動作之流程圖。 圖7係顯示圖5的實施例的蝕刻深度測定結果圖。 圖8係顯示具備依照本發明的第二實施例之殘膜厚度 測定裝置的半導體晶圓的蝕刻裝置的全體構成之方塊圖。 圖9係顯示飩刻處理途中的被處理材的縱剖面形狀圖 (請先閱讀背面之注意事項再填寫本頁) -^1. 訂 線一 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) -11 - 517306 經濟部智慧財產局員工消費合作社印製 A7 B7五、發明説明(9) 〇 圖1 0係顯示圖8的實施例的動作之流程圖。 圖1 1係顯示圖8的實施例的多晶矽殘膜厚度的測定 結果以及迴歸直線的圖。 圖1 2係顯示具備依照本發明的第三實施例之蝕刻深 度測定裝置的半導體晶圓的蝕刻裝置的全體構成之方塊圖 〇 圖1 3係顯示蝕刻處理途中的被處理材的縱剖面形狀 圖。 圖1 4係顯示圖1 2的實施例的動作之流程圖。 圖1 5係顯示具備依照本發明的第四實施例之蝕刻殘 膜測定裝置的半導體晶圓的蝕刻裝置的全體構成之方塊圖 〇 圖1 6係顯示蝕刻處理途中的被處理材的縱剖面形狀 圖。 圖1 7係顯示圖1 5的實施例的動作之流程圖。 【符號說明】 1 :蝕刻裝置 2 :真空容器 3 :電漿 4 :被處理材 5 :試樣台 8 :光纖 本紙張尺度適用中國國家標準(CNS〉A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) P. 訂 線一 -12- 517306 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明( 9A、90A、90B:反射光 9a、9b、9B:放射光 10:鈾刻量測定裝置 1 1 :光譜儀 1 2、2 2 :第一數位濾波電路 1 3、2 3 :微分器 1 4、2 4 :第二數位濾波電路 1 6、2 6 :微分波形圖案資料庫 1 5、2 5 :微分波形比較器 17:顯示器 1 8、2 8 :資料記錄器 1 9、2 9 :迴歸分析器 3 0、1 3 0、2 3 0、3 3 0 :終點判定器 4 0 :矽基板 4 1、5 1、7 1 :罩幕材 4 2 :罩幕材的殘膜厚 4 4 :層差 5 0 、7 0 :多晶矽 52、72:底層氧化膜 5 3、7 5、7 7 :殘膜厚 6 0 :有機膜 Θ 2 :配線材 7 3 :膜 7 5、7 6 :目標殘膜厚度値 (請先閱讀背面之注意事項再填寫本頁) •mr. 訂 線一 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) -13- 517306 A7 ______B7_____ 五、發明説明(〇 7 8 :閘電極部 7 9 :溝槽部 95A、95B:干涉光強度 9 6 :干涉光 1 0 0 0 :控制裝置 1 0 0 1 :供給裝置 1002:電漿產生裝置 1 0 0 3 :晶圓偏壓電源 【較佳實施例之詳細說明】 以下說明本案發明的各實施例。此外,在以下的各實 施例中具有與第一實施例一樣功能者附加與第一實施例相 同的符號,省略其詳細的說明。在以下的實施例依照本發 明的半導體元件製程的終點判定方法係說明關於測定被處 理材的蝕刻製程中的蝕刻量(蝕刻深度以及膜厚)。但是 ,本發明並未限定於此,也能適用於測定電漿C V D、濺 鍍(Sputtering)等的成膜處理中的成膜量(成膜厚度)等 的方法。 以下以圖1〜圖4說明本發明的第一實施例。 在此實施例中,當對半導體晶圓等的被處理材實施電 發鈾刻時,分別設定顯示對試樣(Sample )用的被處理材 (試樣用晶圓)與該處理材所具有的罩幕材的各蝕刻量的 干涉光的微分値的波長依存性(以波長爲參數)的標準圖 案P S與P Μ。其次,分別測定關於與試樣用被處理材相同 本紙張尺度適用中國國家標準(CNS ) Α4規格(210X297公釐) — (請先閲讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 -14- 517306 A7 B7 五、發明説明(d 構成的實際的被處理材(實際的晶圓)的實際處理中的干 涉光的複數波長的強度,求出顯示該被測定的干涉光強度 的微分値的波長依存性(以波長爲參數)的實圖案,比較 微分値的標準圖案與實圖案,求出實際的被處理材的蝕刻 量(製程的終點)。 首先,使用圖1說明具備本發明的蝕刻量(此處實際 的處理材的蝕刻深度以及膜厚)測定裝置的半導體晶圓的 蝕刻裝置的全體構成。蝕刻裝置1係具備真空容器2,被 導入到其內部的蝕刻氣體藉由微波(Microwave)功率等分 解,變成電漿,藉由此電漿3試樣台5上的半導體晶圓等 的被處理材4被鈾刻。來自蝕刻量(例如蝕刻深度以及膜 厚)測定裝置1 0的光譜儀1 1所具有的測定用光源(例 如鹵素光源)的多波長放射光藉由光纖8被導入真空容器 2內,被以垂直的入射角碰觸被處理材4。被處理材4如 圖2 A所示此處具有被處理材的矽4 0與當作罩幕材的氮 化膜4 1,放射光係藉由被罩幕材反射的放射光9 A與在 無罩幕材的矽表面反射的放射光形成干涉光。即此干涉光 爲罩幕材與矽的層差所造成的干涉成分。來自前述罩幕材 的反射光9 A爲在氮化膜表面反射的放射光9 a以及在氮 化膜與政的邊界面反射的放射光9 b,藉由這些放射光形 成干涉光。即此干涉光係罩幕材的剝離所造成的干涉成分 。這些干涉光係中介光纖8被導入蝕刻量測定裝置1 〇的 光譜儀1 1,根據其狀態進行矽的蝕刻深度以及罩幕材的 膜厚測定或製程(此處爲鈾刻)的終點判定處理。 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) 、τ 經濟部智慧財產局員工消費合作社印製 -15- 517306 經濟部智慧財產局員工消費合作社印製 A7 _________B7 _五、發明説明(12? 蝕刻量測定裝置1 〇具備:光譜儀1 1、第〜數位爐 波(Digital filter)電路 i 2、2 2、微分器 1 3、2 3、 第二數位濾波電路1 4、2 4、微分波形圖案資料庫( Pattern database) 1 6、2 6、微分波形比較器 1 5、 2 5、根據追些比較§5的結果判定餓刻的終點的終點判定 器3 0以及顯示終點判定器3 0的判定結果的顯示器i 7 〇 此外,圖1係顯示蝕刻量測定裝置1 〇的功能的構成 ,除了顯示器1 7與光譜儀1 1的蝕刻量測定裝置1 〇的 實際構成可由保持C P U或蝕刻深度以及膜厚測定處理程 式(Program)或干涉光的微分波形圖案資料庫等的各種資 料的R 0 Μ或測定資料保持用的R A Μ以及外部記憶裝置 等所構成的記憶裝置、資料的輸入輸出裝置以及通訊控制 裝置構成。此點關於圖5等的其他實施例也一樣。 關於光譜儀1 1所取入的被處理材的多波長的發光強 度變成分別根據發光強度的電流檢測訊號被變換成電壓訊 號。藉由光譜儀1 1當作取樣(Sampling)訊號輸出的複 數個特定波長訊號,時系列資料y i ,j 、y ’ i ,j係 被收納於未圖示的R A M等的記憶裝置。此時系列資料 y i ,j 、y ’ i ,j的第一、第二波長頻帶(Wave band zone)分別接著被第一數位濾波電路1 2、2 2平滑 化處理,平滑化時系列資料Y i ,j 、Y ’ i ,j被收納 於R A M等的記憶裝置。以此平滑化時系列資料Y 1 ’ j、Y’ i,j爲基礎,藉由微分器13、23算出各個 本紙張尺度適用中國國家標準(CNS ) A4規格(21〇><297公釐) (請先閲讀背面之注意事項再填寫本頁) P. 訂 -線一 -16 - 517306 A7 B7 五、發明説明(14) (請先閱讀背面之注意事項再填寫本頁) 微係數値(一次微分値或二次微分値)的時系列資料d i ,j 、d ’ i ’ j ,被收納於R A M等的記憶裝置。微係 數値的時系列資料d i ,j 、d ’ i ,j分別被第二數位 濾波電路1 4、2 4平滑化處理,平滑化微係數時系列資 料D i ,j 、D ’ i ,j被收納於R A Μ等的記憶裝置。 然後,由此平滑化微係數時系列資料D i ,j 、D ’ i , j求出顯示干涉光強度的微分値的波長依存性(以波長爲 參數)的實圖案。 關於第一、第二波長頻帶爲了獲得不同的實圖案,圖 1的裝置如以下而構成。即當第一、第二波長頻帶爲相同 時,當作使第一數位濾波電路1 2、2 2的微係數不同的 値,對於此情形使微分器1 3、2 3的値相同或不同,也 使第二數位濾波電路1 4、2 4的微係數相同或不同也可 以。另一方面,當第一、第二波長頻帶爲不同時,當作使 第一數位濾波電路1 2、2 2的微係數相同或不同的値, 也使微分器1 3、2 3的値相同或不同,使第二數位濾波 電路1 4、2 4的微係數相同或不同也可以。 經濟部智慧財產局員工消費合作社印製 此外,在圖1的構成分別關於第一、第二波長頻帶各 別配設由第一數位濾波電路、微分器、第二數位濾波電路 、微分波形圖案資料庫、微分波形比較器所組成的構成。 但是,關於第一、第二波長頻帶僅共通地配設一個這種構 成,每一預定時間切換微係數等,關於第一、第二波長頻 帶使其交互地獲得實圖案也可以。 另一方面,在微分波形圖案資料庫6預先設定對對應 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) •17- 517306 A7 B7 五、發明説明(id 成爲蝕刻量測定對象的被處理材之矽與罩幕材的層差之前 述第一波長頻帶的干涉光強度的微分波形圖案資料値P S j 。而且,在微分波形圖案資料庫2 6預先設定對對應被處 理材的罩幕材的膜厚之前述第二波長頻帶的干涉光強度的 微分波形圖案資料値Ρ μ j 。在微分波形比較器1 5中,對 應層差的實圖案與微分波形圖案資料値P s j被比較,以求 出層差(由罩幕材表面到矽的被蝕刻加工的溝槽底部的深 度)。在微分波形比較器25中,對應罩幕材膜厚的實圖 案與微分波形圖案資料値Ρ M j被比較,以求出罩幕材的膜 厚(殘膜厚度)。其結果被處理材的蝕刻量即蝕刻深度被 求出,由顯示器17顯示。 此外,在本實施例以及以下的各實施例係顯示光譜儀 1 1僅一個的情形,惟對於欲增廣測定控制被處理材的面 內的情形,若配設複數個光譜儀1 1也可以。 圖2 A係顯示蝕刻處理途中的被處理材4的縱剖面形 狀以及圖2 B、2 C係顯示干涉光的波長實圖案的例子。 在圖2A中被處理材(晶圓)4在矽基板4 0上被疊層罩 幕材4 1。此飩刻製程矽基板爲被飩刻材料,這種加工處 理被稱爲例如用以進行元件隔離的S T I (淺渠溝隔離, Shallow Trench Isolation)蝕刻。 自光譜儀11放出的多波長的光係以垂直的入射角碰 觸包含被蝕刻材與罩幕材的疊層構造的被處理材4。被導 入罩幕材4 1的放射光9藉由在罩幕材4 1與矽基板4 0 之間形成的邊界面反射的放射光9 b形成干涉光。被導入 *本紙張尺度適用中國國家標準(CNS ) A4規格(2丨0、〆297公釐) I!______f (請先閲讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 -18- 517306 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明(id 無罩幕材41被蝕刻處理的部分之放射光9藉由在罩幕材 4 1頂面反射的放射光9 a以及在被鈾刻材4 0頂面反射 的放射光9 B與來自前述罩幕材4 1所反射的放射光9 A (由9 a與9 b形成)形成干涉光。放射光9 B與9 a分 別伴隨著蝕刻處理的進行,反射的位置如A ( a ) 、B ( b ) 、C (c)而變化。反射的光被導入光譜儀11 ,藉 由蝕刻中的被蝕刻材4 0與罩幕材層的厚度生成強度變化 的訊號。 如圖2 B所示,長波長區域(第二波長頻帶:例如 7 0 0 nm)的干涉光的原始波形隨著蝕刻處理的進行緩 慢地變化。另一方面,如圖2 C所示,短波長區域(第一 波長頻帶:例如3 0 0 n m )的干涉光的原始波形係周期 長的波被周期短的波重疊而變化。此乃顯示長波長區域的 干涉光因罩幕材的剝離(圖2A的a、b、c面)所造成 的干涉成分的變化,在短波長區域(例如3 0 0 n m )的 干涉光顯現因被蝕刻材的矽基板與罩幕材的層差44(圖 2A的a、b、c面與A、B、C面的各面的差)所造成 的干涉成分的變化。以遍及這些多波長的干涉光的平滑化 時系列資料Y i ,j 、Y ’ i ,j爲基礎,分別算出一次 微分値或二次微分値的微係數値時系列資料d i ,j 、 d ’ i ,j 。圖2 B係顯示波長7 0 0 n m的干涉光的一 次微分値以及二次微分値,圖2 C係顯示波長3 0 0 nm 的干涉光的一次微分値以及二次微分値。 由圖2 B、圖2 C得知藉由這些微分處理,因罩幕材 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) P. 訂 線- -19- 517306 A7 B7 五、發明説明(17) (請先閱讀背面之注意事項再填寫本頁) 的剝離所造成的干涉成分的變化以及因砍基板與罩幕材的 層差所造成的干涉成分的變化變的明確。此乃因被鈾刻的 材料之屈折率(例如矽與罩幕材的氮化膜的屈折率以及溝 槽部分的真空屈折率)對波長不同。本發明係著眼於此事 實’對於可正確地測定藉由短波長區域的干涉光求出矽基 板與罩幕材的層差4 4,藉由長波長區域的干涉光求出罩 幕材的剝離(罩幕材的殘膜厚度4 1 )具有特徵。圖3 A 係顯示干涉光的一次微分波形圖案資料,圖3 B相同係顯 示二次微分波形圖案資料。 圖3A、3B中的PA、PB、PC係顯示圖2A的 A (層差 44 = 300nm) 、B (層差= 400;nm) 經濟部智慧財產局員工消費合作社印製 、C (層差二5 Ο 0 nm)的各鈾刻量中的微分波形圖案 資料。同樣地圖3A、3B中的Pa、Pb、Pc係顯示 圖2A的a (罩幕材的殘膜厚42=95nm) 、b (罩 幕材的殘膜厚= 6 5nm)、(:(罩幕材的殘膜厚二3 5 n m )的各蝕刻量中的微分波形圖案資料。此外,此時的 矽的蝕刻深度4 3在圖2 A的A位置爲2 0 5 n m、在B 位置爲3 3 5 nm、在C位置爲4 6 5 nm。 由圖3 A、3 B得知干涉光的一次微分波形圖案或二 次微分波形圖案,各被處理材的鈾刻量變成特有的圖案。 若被處理材的材料不同的話這些圖案也改變,故關於對處 理必要的種種材料以及鈾刻量的範圍,預先藉由實驗等求 出資料,以一次微分波形圖案或二次微分波形圖案當作標 準圖案保持於記錄裝置(1 6,2 6 )較佳。 $紙^尺度適用中.國國家標準(匚阳)八4規格(210'/297公釐) " -20- 517306 A7 B7 五、發明説明(18) 其次,藉由圖4的流程圖說明關於以圖1的蝕刻量測 定裝置1 0進行蝕刻處理時,求出被處理材的蝕刻量的順 序。 最初進行目標蝕刻量(即目標層差與目標罩幕殘膜厚 )的設定,與藉由圖案資料庫1 6、2 6對應目標層差與 目標罩幕殘膜厚的微分圖案P s j 、Ρ μ j ,與判定値σ s 〇 、σΜ〇的設定(步驟400與420)。即預先被保持於 微分波形圖案資料庫1 6。由關於圖3Α、圖3 Β所示的 複數波長的微分値的標準圖案P A、Ρ Β、P C之中設定 對應目標層差的一個微分圖案。同樣地預先被保持於微分 波形圖案資料庫2 6。由關於圖3A、圖3 B所示的複數 波長的微分値的標準圖案P a、Pb、P c之中設定對應 目標罩幕殘膜厚的一個微分圖案。 在下一個步驟中開始干涉光的取樣(例如每0 · 2 5 〜〇 · 4秒)(步驟4 0 2 )。即隨著蝕刻處理的開始發 出取樣開始命令。依照蝕刻的進行而變化的多波長發光強 度係藉由光譜儀1 1中的光檢測器以依照發光強度的電壓 的光檢測訊號而被檢測出。光譜儀1 1的光檢測訊號被進 行數位變換以取得取樣訊號y i ,j 、y ’ i ,j 。 其次,藉由第一段數位濾波器1 2、2 2將來自光譜 儀1 1的多波長輸出訊號y i ,j 、y ’ i ,j平滑化( Filtering),分別算出時系列資料Y i ,j 、Y ’ i ,j ( 步驟4 0 4、4 2 4 )。即藉由第一段數位濾波器降低雜 訊,求出平滑化時系列資料y i。 本紙張尺度逍用中國國家標準(CNS ) A4規格(210X297公釐) _ (請先閱讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 -21 - 517306 A7 _B7_____ 五、發明説明(d 其次,藉由微分器1 3 、23利用S — G法(1. The paper size of 1T line is applicable to the Chinese National Standard (CNS) 8-4 specification (210X297 mm) -6-517306 A7 B7 5. In the description of the invention (4), the radiation reflected by the top surface of the S i 〇2 layer Light forms interference fringes with the radiated light reflected from the boundary surface formed between the Si 102 layer and the S 1 3 N 4 layer. The reflected radiation is irradiated to a suitable detector. This is due to the thickness of the Si02 layer during the etching, which generates a signal with a change in intensity. If the top surface of the Si02 layer is exposed during the etching process, the uranium etching speed and the current etching thickness can be continuously and correctly monitored immediately. Instead of lasers, it is known to use a spectrometer to measure a predetermined radiated light emitted from a plasma. [Summary of the Invention] The contents of the above-mentioned conventional documents are summarized as follows. In Japanese Patent Application Laid-Open No. 5-1 79 46 7, three types of color filters (red, green, and blue) are used to detect interference light (plasma light), and the end point of the etching is detected. Moreover, Japanese Patent Application Laid-Open No. 8-274082 (USP 5 6 5 8 4 1 8) calculates the extreme amplitude of the interference waveform (the maximum and minimum of the waveform: Zero transit point). The time until 値 (Count value) reaches the predetermined 値 is calculated by measurement, the etching rate is calculated, and the remaining etching time until the predetermined film thickness is obtained based on the calculated etching rate, and the etching process is stopped based on this time. Furthermore, Japanese Patent Application Laid-Open No. 2000-97648 obtains the waveform of the difference between the light intensity pattern (with a wavelength as a parameter) of the interference light before processing and the light intensity parameter of the interference light after or during processing (where The wavelength is a parameter), by comparing the difference waveform with the database (Data This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) (Please read the precautions on the back before filling this page)) -Printed by the Consumers 'Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 517306 A7 B7 Printed by the Consumers' Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. 5. Description of the invention (5) Base waveform measurement of layer difference (film thickness). In addition, Japanese Patent Application Laid-Open No. 2000-106356 relates to a spin coating device, and measures a time change of interference light across a plurality of wavelengths to obtain a film thickness. Furthermore, U S P 6 0 8 1 3 3 4 determines a database by measuring the characteristic action of the time variation of the interference light, and determines the final determination of the etching by comparing the database with the measured interference waveform. Based on this judgment, changes in the etching process conditions are promoted. The above known examples have the following problems. (1) If an etching using a mask material (for example, a resist, a nitride film, and an oxide film) is performed, interference light from the fused material is overlapped with interference light from the mask material . (2) The material of the material to be processed (for example, silicon and the cover material disposed thereon) is etched. Since the cover material is etched together with the silicon, there are only those materials that require the material to be processed. The etched amount of uranium (etched depth of uranium) may not accurately measure the etching amount of silicon. (3) Since the initial thickness of the material used in the production process is due to the initial thickness of the cover material of the device structure or the initial thickness of the material to be etched, the interference light from different film thicknesses is lost. overlapping. From the above points, it is difficult to accurately measure / control the layer to be processed (the target layer for semiconductor processing), especially the etching depth or the amount of residual film of the layer to be processed in the plasma etching process. The object of the present invention is to provide a method and a device for determining the end point of a semiconductor device process which can solve the problems of the conventional technology, and use this method and (please read the precautions on the back before filling this page)% 1. Order a book Paper size applies to Chinese National Standard (CNS) A4 specification (210 X297 mm) -8- 517306 A7 B7 V. Description of the invention (6) Method and device for processing the treated material of the device. Another object of the present invention is to provide a semiconductor device manufacturing process capable of accurately measuring the actual etching depth or the amount of residual film of a layer to be processed on-line (〇η -1 ine) in plasma processing, especially in plasma uranium etching processing. End point determination method and device, and method and device for processing material to be processed using the method and device. Still another object of the present invention is to provide etching capable of controlling each layer of a semiconductor device at a predetermined etching depth and film thickness with high accuracy online Process. Still another object of the present invention is to provide an etching depth of a material to be processed or a film thickness measuring device which can accurately measure the actual etching depth of the processed layer and the film thickness on the line. For example, the inventors of the present invention should solve the problems of the conventional technology and achieve one of the objects of the present invention described above. For each of a plurality of wavelengths, obtain a time-differential waveform of the interference waveform, and obtain a display interference waveform based on the waveform. The pattern of the wavelength dependence of the differential chirp (that is, the differential chirped pattern of the interference waveform with the wavelength as a parameter) is used to measure the thickness of the film. The reason for using the wavelength-dependent pattern showing the time differential chirp of the interference waveform in the present invention is as follows. Since the measurement of i η-s i t u (immediately, R a a t i m e) during the engraving is taken as a pre-measurement, the film thickness of the film to be treated changes from moment to moment. Therefore, the time differential processing of the interference waveform is possible. In addition, the noise of the interference waveform can be removed by the differential processing. In addition, the paper size of the etched materials (such as silicon and the nitride film of the cover material) applies to the Chinese National Standard (CNS) A4 specification (210X297 mm) —I ------ F, (Please read first Note on the back, please fill in this page again) Order • Line ·-Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs -9-517306 A7 _ B7_ V. Description of the invention (7) The inflection rate is different for the wavelength. Therefore, it is possible to detect the characteristic change (film thickness dependence) of each substance by interference light measurement over multiple wavelengths. According to one aspect of the present invention, an etching depth and a film thickness measuring method for measuring the uranium engraved amount of a material to be processed include: a) setting a differential of interference light to a predetermined amount of etching of a first material to be processed including a cover material; A step of the standard pattern of the first processed material with the wavelength as a parameter; b) setting a differential of the interference light of a predetermined etching amount of the cover material of the first processed material with the wavelength of the parameter Steps of the standard pattern of the cover material; c) Measure the interference light intensity of the second processed material having the same structure as the first processed material with respect to a plurality of wavelengths, and obtain a differential value of the measured interference light intensity A step of a solid pattern having a wavelength of chirp as a parameter; and d) obtaining the etched amount of the second material to be processed based on the standard pattern of the first material to be processed and the standard pattern of the cover material and the solid pattern. step. According to the present invention, it is possible to provide an etching depth and a film thickness measurement of a material to be processed that can accurately measure an actual etching amount of a layer to be processed in a plasma process, particularly in a plasma etching process. Method and method for processing a sample of a material to be processed using the method. In addition, an etching process can be provided in which each layer of the semiconductor device is controlled on the line with high accuracy to a predetermined etching amount. In addition, it can provide the etching depth or film of the processed material to accurately measure the actual etching amount of the processed layer on the line. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) ': -10-(Please (Please read the notes on the back before filling this page). Ordering line 1 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 517306 A7 B7 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (8) Thickness measuring device. [Brief Description of the Drawings] Fig. 1 is a block diagram showing the overall configuration of an etching apparatus for a semiconductor wafer including an etching measurement device according to a first embodiment of the present invention. Fig. 2A is a longitudinal sectional view of a material to be processed during the etching process. Figures 2B and 2C show examples of actual waveforms of interference light with different wavelengths. Figs. 3A and 3B show the interference light of each of the etched surfaces of the silicon wafer shown in A, B, and C shown in Figs. 2A and 2B, and the etched surfaces of the mask shown in a, b, and c. A graph of the wavelengths of the series data as parameters. Fig. 4 is a flowchart showing the order of the step of the material to be treated and the thickness of the mask residual film when the etching process is performed by the etching amount measuring device of Fig. 1. Fig. 5 is a diagram showing a first embodiment according to the present invention. A block diagram of the overall configuration of an etching apparatus for a semiconductor wafer according to a modified example of the etching depth measuring apparatus. FIG. 6 is a flowchart showing the operation of the embodiment in FIG. 5. FIG. 7 is a graph showing the measurement results of the etching depth in the example of FIG. 5. Fig. 8 is a block diagram showing the overall configuration of a semiconductor wafer etching apparatus including a residual film thickness measuring device according to a second embodiment of the present invention. Figure 9 shows the longitudinal sectional shape of the material being processed during the engraving process (please read the precautions on the back before filling out this page)-^ 1. Alignment A paper size applies to Chinese National Standard (CNS) A4 specifications ( 210X297 mm) -11-517306 A7 B7 printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of Invention (9) 〇 Figure 10 is a flowchart showing the operation of the embodiment of Figure 8. Fig. 11 is a graph showing the measurement results of the polycrystalline silicon residual film thickness and the regression line in the example of Fig. 8. Fig. 12 is a block diagram showing the overall configuration of an etching apparatus for a semiconductor wafer including an etching depth measuring device according to a third embodiment of the present invention. Fig. 13 is a longitudinal sectional view showing a material to be processed in the course of an etching process. . FIG. 14 is a flowchart showing the operation of the embodiment in FIG. 12. FIG. 15 is a block diagram showing the overall configuration of an etching apparatus for a semiconductor wafer provided with an etching residual film measuring device according to a fourth embodiment of the present invention. FIG. 16 is a longitudinal sectional shape of a material to be processed during the etching process. Illustration. FIG. 17 is a flowchart showing the operation of the embodiment of FIG. 15. [Symbol description] 1: Etching device 2: Vacuum container 3: Plasma 4: Material to be processed 5: Sample stage 8: Optical fiber The paper size applies to Chinese national standards (CNS> A4 specification (210X297 mm) (Please read first Note on the back, please fill out this page again) P. Order 1-12- 517306 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (9A, 90A, 90B: reflected light 9a, 9b, 9B: radiation Light 10: Uranium measurement device 1 1: Spectrometer 1 2, 2 2: First digital filter circuit 1 3, 2 3: Differentiator 1 4, 2 4: Second digital filter circuit 1 6, 2 6: Differential waveform Pattern library 1 5 and 2 5: Differential waveform comparator 17: Display 1 8 and 2 8: Data recorder 1 9 and 2 9: Regression analyzer 3 0, 1 3 0, 2 3 0, 3 3 0: End point Determining device 4 0: Silicon substrate 4 1, 5 1, 7 1: Cover material 4 2: Residual film thickness of cover material 4 4: Steps 5 0, 7 0: Polycrystalline silicon 52, 72: Underlayer oxide film 5 3 , 7 5, 7 7: Residual film thickness 6 0: Organic film Θ 2: Wiring material 7 3: Film 7 5, 7 6: Target residual film thickness 値 (Please read the precautions on the back before filling this page) • mr . Binding a paper The Zhang scale is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) -13- 517306 A7 ______B7_____ 5. Description of the invention (〇7 8: gate electrode section 7 9: groove section 95A, 95B: interference light intensity 9 6: Interfering light 1 0 0 0: control device 1 0 0 1: supply device 1002: plasma generating device 1 0 3: wafer bias power supply [Detailed description of preferred embodiment] The following describes the embodiments of the present invention. In addition, in the following embodiments, those having the same functions as those in the first embodiment are given the same reference numerals as those in the first embodiment, and detailed descriptions thereof are omitted. In the following embodiments, the method for determining the end point of the semiconductor device manufacturing process according to the present invention The description is about measuring the etching amount (etching depth and film thickness) in the etching process of the material to be processed. However, the present invention is not limited to this, and can also be applied to the measurement of film formation such as plasma CVD, sputtering, and the like. The method of the amount of film formation (film thickness), etc. during processing. The first embodiment of the present invention will be described below with reference to Figs. 1 to 4. In this embodiment, when a material to be processed such as a semiconductor wafer is subjected to electrical transmission Uranium carving , And set the wavelength dependence of the differential light of the interference light on each etching amount of the material to be processed (sample wafer) for the sample (Sample) and the cover material that the processing material has (set the wavelength as Parameters) standard patterns PS and PM. Secondly, measure the same paper size as the sample material to be treated. The paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm) — (Please read the precautions on the back before filling this page). Printed by the Consumer Affairs Cooperative of the Property Bureau -14- 517306 A7 B7 V. Description of the Invention (intensity of the complex wavelength of the interference light during the actual processing of the actual processed material (actual wafer) constituted by d) The real pattern of the wavelength dependence (different wavelength) of the differential chirp of the measured interference light intensity is compared with the standard pattern of the differential chirp and the real pattern, and the actual etching amount of the material to be processed (the end point of the process) is determined. The overall configuration of an etching device for a semiconductor wafer including an etching amount (etching depth and film thickness of an actual processing material here) measuring device of the present invention will be described with reference to Fig. 1. The etching device 1 is equipped with a vacuum container 2 and is introduced thereinto. The internal etching gas is decomposed by microwave power or the like to be converted into a plasma. The plasma 3 is used to process materials such as semiconductor wafers on the sample stage 5 4 is etched by uranium. Multi-wavelength emitted light from a measuring light source (for example, a halogen light source) included in the measuring device 10 for the amount of etching (such as etching depth and film thickness) is introduced into the vacuum container 2 through the optical fiber 8 , Is touched by the incident material 4 at a normal incident angle. As shown in FIG. 2A, the treated material 4 has silicon 4 0 as the treated material and a nitride film 41 as the cover material. The radiated light 9 A reflected by the mask material forms interference light with the radiated light reflected on the silicon surface without the mask material. That is, the interference light is an interference component caused by the difference between the mask material and the silicon. From the mask The reflected light 9 A of the curtain material is the radiated light 9 a reflected on the surface of the nitride film and the radiated light 9 b reflected on the boundary surface between the nitride film and the substrate, and these interference lights form interference light. That is, this interference light system Interference components caused by peeling of the cover material. These interference optical intermediary optical fibers 8 are introduced into the spectrometer 11 of the etching amount measuring device 10, and the etching depth of silicon and the film thickness measurement or manufacturing process of the cover material are performed according to the state. (Here is uranium engraving). Paper size applies Chinese National Standard (CNS) A4 specification (210X297 mm) (Please read the precautions on the back before filling out this page), τ Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs and Consumer Cooperatives -15- 517306 Intellectual Property Bureau of the Ministry of Economic Affairs Printed by employee consumer cooperative A7 _________B7 _V. Description of the invention (12? Etching measurement device 1 〇 Equipped with: Spectrometer 1 1st ~ digital filter circuit (Digital filter) i 2, 2 2, differentiator 1 3, 2 3 、 Second digital filter circuit 1 4 、 2 4 、 Differential waveform pattern database (Pattern database) 1 6 、 2 6 、 Differential waveform comparator 1 5 、 2 5 、 Determine the end point of hungry engraving according to the results of §5. The end point determiner 30 and the display i 7 showing the determination result of the end point determiner 30. In addition, FIG. 1 shows a functional configuration of the etching amount measuring device 10, except for the etching amount measurement of the display 17 and the spectrometer 11 The actual configuration of the device 10 can be held by R 0 M or measurement data, which holds various data such as a CPU, an etching depth, a film thickness measurement processing program (Program), or a differential waveform pattern database of interference light. Memory means and an external memory R A Μ constituted devices, data input and output means and communication control means configured. This also applies to the other embodiments of FIG. 5 and the like. The multi-wavelength luminous intensity of the processed material taken by the spectrometer 11 is converted into a voltage signal based on a current detection signal based on the luminous intensity. When the spectrometer 11 outputs a plurality of specific wavelength signals as sampling signals, the time series data y i, j, y ′ i, and j are stored in a memory device such as a RAM (not shown). At this time, the first and second wavelength bands (Wave band zone) of the series data yi, j, y'i, and j are then smoothed by the first digital filter circuit 1, 2, 2 respectively. , J, Y 'i, j are stored in a memory device such as a RAM. Based on this series of smoothed data Y 1 'j, Y' i, j as the basis, the different paper sizes calculated by differentiators 13, 23 are applicable to the Chinese National Standard (CNS) A4 specification (21〇 > < 297) (%) (Please read the notes on the back before filling this page) P. Order-Line 1-16-517306 A7 B7 V. Description of the invention (14) (Please read the notes on the back before filling this page) The time series data di (j-d, i-d) are stored in a memory device such as a RAM. The time series data di, j, d'i, j of the micro coefficient 値 are smoothed by the second digital filter circuits 1 4, 2 4 respectively, and the time series data D i, j, D 'i, j are smoothed by the second digital filter circuit. Stored in a memory device such as RA Μ. Then, the series data D i, j, D ′ i, j when smoothing the micro coefficients are used to obtain a real pattern showing the wavelength dependence (with the wavelength as a parameter) of the differential chirp of the interference light intensity. Regarding the first and second wavelength bands, in order to obtain different real patterns, the device of FIG. 1 is configured as follows. That is, when the first and second wavelength bands are the same, it is regarded as a chirp that makes the differential coefficients of the first digital filter circuits 1 2, 2 2 different. For this case, the chirps of the differentiators 1 3, 2 3 are the same or different. The micro coefficients of the second digital filter circuits 14 and 24 may be the same or different. On the other hand, when the first and second wavelength bands are different, it is considered that the differential coefficients of the first digital filter circuits 1 2, 2 2 are the same or different, and the differential frequencies of the differentiators 1 3, 2 3 are also the same. Alternatively, the micro coefficients of the second digital filter circuits 14 and 24 may be the same or different. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. In addition, the first digital filter circuit, differentiator, second digital filter circuit, and differential waveform pattern data are respectively provided for the first and second wavelength bands in the structure of Figure 1. Library and differential waveform comparator. However, only one such configuration is commonly provided for the first and second wavelength bands, and the micro coefficient is switched every predetermined time. The first and second wavelength bands may alternately obtain a real pattern. On the other hand, the differential waveform pattern database 6 is set in advance to apply the Chinese National Standard (CNS) A4 specification (210X297 mm) corresponding to the paper size. 17- 517306 A7 B7 V. Description of the invention (id becomes the subject of measurement of the amount of etching) The differential waveform pattern data 値 PS j of the interference light intensity of the aforementioned first wavelength band of the difference between the silicon of the processed material and the cover material. In addition, a cover corresponding to the processed material is set in the differential waveform pattern database 26 in advance. The differential waveform pattern data of the interfering light intensity of the aforementioned second wavelength band of the film thickness 値 μ μ. In the differential waveform comparator 15, the real pattern corresponding to the step is compared with the differential waveform pattern data P sj To obtain the layer difference (the depth from the surface of the mask material to the bottom of the etched trench of silicon). In the differential waveform comparator 25, the solid pattern and differential waveform pattern data corresponding to the film thickness of the mask material 値 Ρ M j is compared to obtain the film thickness (residual film thickness) of the cover material. As a result, the etching amount of the material to be treated, that is, the etching depth, is obtained and displayed on the display 17. In addition, in this embodiment, Each of the following examples shows a case where there is only one spectrometer 11, but for the case where the in-plane material to be measured and controlled is to be expanded and controlled, a plurality of spectrometers 1 1 may be provided. Fig. 2 A shows the middle of the etching process 2B and 2C show examples of the actual pattern of the interference light wavelength in FIG. 2B and FIG. 2C. The processed material (wafer) 4 is laminated on a silicon substrate 40 in FIG. 2A. Material 4 1. The silicon substrate in this engraving process is a material to be etched. This processing is called, for example, STI (Shallow Trench Isolation) etching for element isolation. Multi-wavelength emitted from the spectrometer 11 The light system touches the material 4 to be processed including the laminated structure of the material to be etched and the cover material at a normal incident angle. The radiated light 9 introduced into the cover material 4 1 passes through the cover material 41 and the silicon substrate. The radiated light reflected by the boundary surface formed between 4 0 9 b forms interference light. It is introduced * This paper size applies Chinese National Standard (CNS) A4 specifications (2 丨 0, 〆297 mm) I! ______ f (Please read first (Notes on the back, please fill in this page) Printed by the Consumer Cooperative of the Property Bureau-18- 517306 A7 B7 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. 5. Description of the invention (id Unshielded curtain material 41 Emitted light from the etched part 9 1 The radiated light 9 a reflected from the top surface and the radiated light 9 B reflected from the top surface of the uranium carving material 40 and the radiated light 9 A (formed from 9 a and 9 b) reflected from the cover material 4 1 Interfering light. The emitted light 9 B and 9 a are accompanied by the progress of the etching process, respectively, and the reflected positions such as A (a), B (b), and C (c) change. The reflected light is introduced into the spectrometer 11, and a signal of a change in intensity is generated by the thickness of the material to be etched 40 and the cover material layer during the etching. As shown in FIG. 2B, the original waveform of the interference light in the long wavelength region (second wavelength band: 700 nm, for example) gradually changes as the etching process proceeds. On the other hand, as shown in FIG. 2C, the original waveform of the interference light in the short-wavelength region (first wavelength band: 300 nm, for example) has a long period, and a wave having a short period overlaps and changes. This is a display of the interference component caused by the peeling of the mask material in the long wavelength region (a, b, and c surfaces in Figure 2A), and the cause of the interference light manifestation in the short wavelength region (for example, 300 nm). Changes in interference components caused by the step 44 (difference in each of the a, b, and c surfaces and the A, B, and C surfaces of the silicon substrate of the material to be etched) and the cover material. Based on the smoothing time series data Yi, j, Y'i, j across these multi-wavelength interference lights, the differential coefficients 値 time series data di, j, d 'of primary differential 値 or secondary differential 微 are calculated respectively. i, j. Fig. 2 shows the primary differential chirp and the second differential chirp of interference light with a wavelength of 700 nm, and Fig. 2 shows the primary differential chirp and the second differential chirp of interference light with a wavelength of 300 nm. From Figure 2B and Figure 2C, it is known that through these differential processing, the paper size of the cover material applies the Chinese National Standard (CNS) A4 specification (210X297 mm) (Please read the precautions on the back before filling this page) P. Threading--19- 517306 A7 B7 V. Description of the invention (17) (Please read the precautions on the back before filling this page) Changes in interference components caused by peeling and cutting the substrate and the cover material The change in the interference component caused by the difference becomes clear. This is because the inflection rate of the material engraved by uranium (for example, the inflection rate of silicon and the nitride film of the cover material and the vacuum inflection rate of the groove portion) differs with the wavelength. The present invention is based on the fact that the layer difference between the silicon substrate and the cover material can be accurately measured by interference light in a short wavelength region, and the peeling of the cover material is determined by interference light in a long wavelength region. (Residual film thickness 41 of the cover material) is characteristic. Figure 3A shows the data of the primary differential waveform pattern of the interference light, and Figure 3B shows the data of the second differential waveform pattern in the same system. The PA, PB, and PC in Figures 3A and 3B show A (layer difference 44 = 300nm) and B (layer difference = 400; nm) of Figure 2A. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, C (layer difference 2 Differential waveform pattern data in each uranium engraved amount (500 nm). Similarly, Pa, Pb, and Pc in the maps 3A and 3B show a (residual film thickness of the cover material 42 = 95nm), b (residual film thickness of the cover material = 65 nm) in FIG. 2A, (: (cover screen Material's residual film thickness (2.5 nm) in each etching amount. In addition, the silicon etching depth 4 3 at this time is 2 0 5 nm at the A position and 3 at the B position. 3 5 nm and 4 6 5 nm at the position C. From Figures 3 A and 3 B, it is known that the primary differential waveform pattern or the secondary differential waveform pattern of the interference light, the uranium engraved amount of each processed material becomes a unique pattern. These patterns also change if the material of the material to be processed is different. Therefore, for the materials necessary for processing and the range of the uranium engraving amount, data is obtained in advance through experiments and the like, and the primary differential waveform pattern or the secondary differential waveform pattern is used as a standard It is better to keep the pattern on the recording device (16, 2 6). $ Paper size is applicable. National Standard (Liyang) 8 4 specifications (210 '/ 297 mm) " -20- 517306 A7 B7 V. Description of the Invention (18) Next, a description will be given with reference to a flowchart of FIG. 4 when an etching process is performed by the etching amount measuring device 10 of FIG. 1. The order of the etching amount of the material to be processed. Initially set the target etching amount (that is, the target layer difference and the target mask residual film thickness), and correspond to the target layer difference and the target mask residual through the pattern database 16 and 26. The film thickness differential patterns P sj and P μ j and the settings of the determinations σσ s 0 and σM0 (steps 400 and 420). That is, they are held in the differential waveform pattern database 16 in advance. A differential pattern corresponding to the target step is set among the standard patterns PA, PB, and PC of the differential chirp of a complex wavelength shown in B. Similarly, it is held in advance in the differential waveform pattern database 26. About FIG. 3A and FIG. A differential pattern corresponding to the target mask residual film thickness is set among the standard patterns of the differential chirps of the complex wavelengths shown in 3 B, P a, Pb, and P c. In the next step, sampling of interference light is started (for example, every 0 · 2 5 to 0.4 seconds) (step 4 02). That is, a sampling start command is issued as the etching process starts. The multi-wavelength luminous intensity that changes according to the progress of the etching is determined by the photodetector in the spectrometer 11 in accordance with Luminous intensity of voltage light The detection signal is detected. The light detection signal of the spectrometer 11 is digitally transformed to obtain the sampling signals yi, j, y'i, j. Second, the first digital filter 1 2, 2 2 will come from the spectrometer. The 1 multi-wavelength output signals yi, j, y'i, and j are smoothed (Filtering), and the time series data Yi, j, Y'i, and j are calculated respectively (steps 4 0 4 and 4 2 4). That is, the noise is reduced by the first digital filter, and the series data y i during smoothing are obtained. This paper size is in accordance with the Chinese National Standard (CNS) A4 specification (210X297 mm) _ (Please read the precautions on the back before filling this page). _B7_____ V. Description of the invention (d Secondly, the S-G method is used by the differentiators 1 3 and 23 (
Savitzky-Golay method)分別算出微係數 d i ,j 、d’ i ,j (步驟406、426)。即藉由微分處理(S -G法),求出訊號波形的係數(一次或二次)di 。再者 ,分別算出藉由第二段數位濾波器1 4、2 4平滑化( S m ο 〇 t h i n g )微係數時系歹[J D i ,j 、D ’ i ,j (步驟 408、428)。其次,在微分波形比較器15中,關 於層差進行σ = Σ ( D i ,j 一 P s j ) 2値的算出(步驟 4 1 0 )。同樣地在微分波形比較器2 5中,關於罩幕材 的剝離(殘膜厚),也進行cj ’ = Σ ( D ’ i ,j 一 p M j )2値的算出(步驟4 3 0 )。再者,在終點判定器3 0中 分別進行與σ’ $σΜ〇的判定(步驟4 12)。 當 aSas。以及 σ’ $σΜ〇 時,判定層差與罩幕材的殘膜厚分別爲預定値,使蝕 刻處理終了,並且將其結果顯示於顯示器1 7。當σ $ σ"以及σ’ $σΜ。的任一個不滿足時,返回步驟4 0 4 、4 2 4。最後,進行取樣終了的設定(步驟4 1 4 )。 此處,說明關於平滑化微係數時系列資料D i 、d, 丄的算出。數位濾波電路1 2、2 2、1 4、2 4使用例 如二次 Bataworth 型的低通濾波器(Secondary Bataworth low pass filter)。數位濾波電路1 2、2 2係相同構成, 係數b、a在數位濾波電路1 2、2 2間爲同一或不同也 可以。此處,僅說明關於數位濾波電路1 2。藉由二次 Bataworth型的低通濾波器平滑化時系列資料Y i係由式( 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐1 -- (請先閱讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 -22- 517306 A7 B7 五、發明説明(2(j> 1 )求出。Savitzky-Golay method) calculates micro coefficients d i, j, d 'i, j (steps 406 and 426). That is, through the differential processing (S-G method), the coefficient (primary or secondary) di of the signal waveform is obtained. In addition, the smoothing (S m ο 〇 t h i n g) micro coefficients of the second stage digital filters 1, 4 and 2 are respectively calculated as [J D i, j, D ′ i, j (steps 408, 428). Next, the differential waveform comparator 15 calculates σ = Σ (D i, j-P s j) 2 値 with respect to the layer difference (step 4 1 0). Similarly, in the differential waveform comparator 25, cj '= Σ (D' i, j-p M j) 2 値 is calculated for the peeling (residual film thickness) of the cover material (step 4 3 0) . Furthermore, the determination with σ '$ σM0 is performed in the end point determiner 30 (step 4 12). When aSas. And σ ′ $ σΜ〇, it is judged that the layer difference and the residual film thickness of the mask material are respectively predetermined, so that the etching process is finished, and the result is displayed on the display 17. When σ $ σ " and σ ’$ σΜ. If any of them are not satisfied, return to steps 4 0 4 and 4 2 4. Finally, set the end of sampling (step 4 1 4). Here, the calculation of the series data D i, d, 时 when smoothing the micro coefficients will be described. Digital filter circuits 1 2, 2 2, 1 4, 2 4 use examples such as secondary Bataworth low pass filter (Secondary Bataworth low pass filter). The digital filter circuits 1 and 2 and 2 have the same structure, and the coefficients b and a may be the same or different between the digital filter circuits 1 and 2 and 2. Here, only the digital filter circuit 12 will be described. The series data when smoothing by the secondary Bataworth low-pass filter is based on the formula (This paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm 1-(Please read the precautions on the back before (Fill in this page) Line 1 Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs -22- 517306 A7 B7 V. Description of Invention (2 (j > 1)).
Yi=blyi+b2yi-l+b3yi-2-[ a2Yi-l + a3Yi— 2〕…(1 ) 此處,係數b、a因取樣頻率以及截止(Cut off)頻 率數値不同。而且,此數位濾波器的係數値關於層差的波 段(Wave-range )(第一波長頻帶),例如2 7 5 n m到 5 0 〇 n m,以及關於罩幕材的剝離(殘膜厚)的波段( 第二波長頻帶),例如525nm到750nm不同也可 以。例如關於層差的波段的情形a 2 = - 1 · 1 4 3、 a 3 = 0 . 4128、bl = 0 · 067455、乜2 = 〇· 13491、b3 = 0 · 067455 (取樣頻率 10Hz、截止頻率1Hz),關於罩幕材的殘膜厚的波 段的情形 a 2 = — 0 · 0 0 0 7 3 6 1 2、a 3 = 0. 17157、bl = 0- 2927l、b2 = 0 · 58542、b3 = 0 · 29271 (截止頻率 0 · 2 5 Η z )。 二次微係數値的時系列資料d t 、d ’ i係分別藉由 微分器1 3、2 3使用五點時系列資料Y i的多項式適合 平滑化微分法,由式(2 )如以下算出。 j ~ 2 d i = d, i 二 Σ w j Y i + j ··· · ( 2 ) j = 一 2 本紙張尺度適用中國國家標準(CNS )A4規格(210x297公釐) — II------P, (請先閲讀背面之注意事項再填寫本頁) 訂 經濟部智慧財產局員工消費合作社印製 -23- 517306 A7 _B7 _ 五、發明説明(2》 此處,關於權重(Weight) w爲w— 2 = 2、w - 1 =1、w0 = — 2、wl=— 1 〇w2 = 2。 此外,在微分器13、23之間j爲相同或不同也可 以。 使用前述微係數値的時系列資料d i 、d ’ i ’平滑 化微係數時系列資料D i 、D ’ i藉由數位濾波電路1 4 、2 4 (二次Bataworth型的低通濾波器)’由式(3 ) 、(4 )求出。但是,在數位濾波電路1 4、2 4之間係 數a、b的値爲相同或不同也可以。Yi = blyi + b2yi-l + b3yi-2- [a2Yi-l + a3Yi— 2] (1) Here, the coefficients b and a are different depending on the sampling frequency and cut-off frequency. Moreover, the coefficient of this digital filter 値 is related to the wave-range (first wavelength band) of the step, such as 275 nm to 500 nm, and the peeling (residual film thickness) of the cover material. The band (second wavelength band), for example, 525nm to 750nm may be different. For example, regarding the band of the step a 2 =-1 · 1 4 3, a 3 = 0.4128, bl = 0 · 067455, 乜 2 = 〇 · 13491, b3 = 0 · 067455 (sampling frequency 10Hz, cutoff frequency 1Hz), in the case of the band thickness of the residual film thickness of the cover material a 2 = — 0 · 0 0 0 7 3 6 1 2, a 3 = 0. 17157, bl = 0-2927l, b2 = 0 · 58542, b3 = 0 · 29271 (cutoff frequency 0 · 2 5 Η z). The time series data d t and d ′ i of the quadratic differential coefficient 系 are suitable for the smoothing differential method by using the polynomials of the five-point time series data Y i by the differentiators 1 3, 2 3, and calculated from the following formula (2). j ~ 2 di = d, i two Σ wj Y i + j ···· (2) j = one 2 This paper size applies to China National Standard (CNS) A4 (210x297 mm) — II ----- -P, (Please read the notes on the back before filling this page) Order printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs-23- 517306 A7 _B7 _ V. Description of the invention (2) Here, about the weight (Weight) w W-2 = 2, w-1 = 1, w0 = — 2, wl = — 1 〇w2 = 2. In addition, j may be the same or different between the differentiators 13, 23. The aforementioned differential coefficient 値 may be used. The time series data di, d'i 'smooth the micro-coefficient time series data Di, D'i through the digital filter circuits 1 4 and 2 4 (secondary Bataworth-type low-pass filter)' by formula (3) (4), but the coefficients a and b of the digital filter circuits 14 and 24 may be the same or different.
Di=bldi+b2di - l + b3di - 2 —〔 a2Di — l + a3Di— 2〕_"(3) D i=bld’ i + b 2 d , i — l + b3d’ i —2 —〔a2D’ i-l + a3D’ i— 2〕…(4) 如此,如果依照圖1的飩刻量測定裝置,至少設定一 個關於如在圖3A、圖3B以PA、PB、PC、Pa ' P b、P c顯示的複數波長的微分値的標準圖案,分別測 定被處理材的干涉光的複數波長強度,求出該被測定的干 涉光強度的各波長的微分値的實圖案,藉由比較標準圖案 與微分値的實圖案,可求出層差與罩幕材的殘膜厚。例如 當想檢測矽蝕刻深度3 3 5 n m即圖2的B位置時,預先 設定關於對應蝕刻量(層差、罩幕材的殘膜厚)B、b的 複數波長的微分値的標準圖案p B、P b,藉由在複數波 本紙張尺度適财關家縣(CNS ) A4胁(21GX297公釐) " (請先閲讀背面之注意事項再填寫本頁) 訂 線- 經濟部智慧財產局員工消費合作社印製 -24- 517306 A7 B7 五、發明説明( 長中對實圖案的這些標準圖案的一致率分別達到判定値 as。、σΜ〇以內。可檢測出來自罩幕材表面的層差4 4爲 400nm,罩幕材的殘膜厚42爲6 5nm (被處理材 的矽深度43爲335nm)。標準圖案若使用一次微分 値圖案、二次微分値圖案的任一方或兩方皆可。如果依照 本實施例,藉由求出被處理材的蝕刻量(層差、罩幕材的 殘膜厚),可正確地測定矽蝕刻深度例如3 3 5 n m。 其次,藉由顯示該變形例的構成之圖5的方塊圖與圖 6的流程圖說明被計測的干涉光強度包含許多雜訊成分的 情形,用以提高測定的蝕刻量精度的第一實施例的變形例 。此變形例例如每一晶圓其圖案不同,因此適用於因每一 晶圓蝕刻條件(例如放電條件)不同,干涉波每一晶圓不 同的情形。最初設定被處理材(矽、罩幕材)的目標加工 深度(此處目標蝕刻深度:圖2 A的4 3 )(步驟5 5 0 )。其次,由微分波形檲案資料庫1 6、2 6讀出關於預 先保存的層差、罩幕殘膜厚的微分波形圖案與收斂判定値 (層差:Psj 、aSQ,罩幕殘膜厚:pMj 、σΜ。), 分別設定微分波形比較器1 5、2 5。伴隨著蝕刻處理開 始,開始干涉光的取樣(步驟5 0 2 )。其次,與圖4的 步驟4 0 4、410,424 — 430 —樣,實行步驟 504 — 5 10、52 4 - 530。來自光譜儀1 1的短 波段與長波段的光分別中介第一數位濾波器1 2、2 2與 微分器1 3、2 3與第二數位濾波器1 4、2 4,求出平 滑化微係數時系列資料D i ,j 、D ’ i ,j 。分別比較 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) -5'坊 經濟部智慧財產局員工消費合作社印製Di = bldi + b2di-l + b3di-2 — [a2Di — l + a3Di— 2] _ " (3) D i = bld 'i + b 2 d, i — l + b3d' i —2 — [a2D ' il + a3D 'i— 2] ... (4) In this way, if in accordance with the engraved amount measuring device of FIG. 1, at least one set is set as PA, PB, PC, Pa'Pb, Pc as shown in FIG. 3A and FIG. 3B. The standard pattern of the differential chirp of the complex wavelength displayed is used to measure the complex wavelength intensity of the interference light of the material to be processed, and the real pattern of the differential chirp of each wavelength of the measured interference light intensity is obtained, and the standard pattern and the differential are compared. The solid pattern of 値 can be used to determine the layer difference and the residual film thickness of the cover material. For example, when you want to detect a silicon etching depth of 3 3 5 nm, that is, position B in FIG. 2, a standard pattern p of the differential wavelength of the complex wavelength corresponding to the etching amount (layer difference, residual film thickness of the cover material) B, b is set in advance. B, P b, by using the appropriate wavebook paper size for the financial institution of Guancai County (CNS) A4 (21GX297 mm) " (Please read the precautions on the back before filling this page) Alignment-Intellectual Property of the Ministry of Economic Affairs Printed by the Bureau's Consumer Cooperatives-24- 517306 A7 B7 V. Description of the invention (The consistency rate of these standard patterns for the solid pattern in the long and middle reaches the judgment 値 as. And σM0 respectively. The layer from the surface of the cover material can be detected The difference 44 is 400nm, and the residual film thickness 42 of the cover material is 65nm (the silicon depth 43 of the material to be treated is 335nm). If the standard pattern uses one or both of the primary differential pattern and the secondary differential pattern Yes. According to this embodiment, the silicon etching depth (for example, 3 3 5 nm) can be accurately measured by determining the etching amount (layer difference, residual film thickness of the cover material) of the material to be processed. Second, by displaying this The block diagram of FIG. 5 and the flowchart of FIG. 6 of the configuration of the modified example In the case where the measured interference light intensity includes many noise components, a modified example of the first embodiment for improving the accuracy of the measured etching amount. This modified example is different for each wafer, and is therefore suitable for each crystal. The circular etching conditions (such as the discharge conditions) are different, and the interference wave is different for each wafer. The target processing depth of the material to be processed (silicon, cover material) is initially set (here, the target etching depth: 4 3 in Figure 2 A) (Step 5 50). Secondly, the differential waveform pattern database 16 and 2 6 read out the differential waveform pattern and the convergence judgment 关于 (layer difference: Psj, aSQ) about the pre-stored layer difference and the residual thickness of the mask. , Residual film thickness of the mask: pMj, σM.), Differential waveform comparators 15 and 25 are set respectively. As the etching process starts, the interference light sampling is started (step 5 0 2). Secondly, it is the same as the step of FIG. 4 4 0 4, 410, 424 — 430 — like, perform steps 504 — 5 10, 52 4-530. Light from the short-spectrum and long-spectrum bands of the spectrometer 1 1 interpose the first digital filter 1 2, 2 2 and differentiation respectively. Filters 1 3, 2 3 and the second digital filter 1 4, 2 4 The series of data D i, j, D 'i, j when calculating the smoothing coefficient. Compare the paper size to the Chinese National Standard (CNS) A4 specification (210X297 mm). (Please read the precautions on the back before filling in this. Page) -5'Fang Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs
-25- 517306 A7 B7 五、發明説明( 這些平滑化微係數時系列資料D i ,j 、d, i ,j與預 先被設定於微分波形比較器1 5、2 5的微分圖案p s j。 Ρ M j ,算出在該時刻的層差値S i與罩幕材剝離量(殘膜 厚)Mi (步驟 515、535 )。此處,當^7>075〇( 或口>6^〇)時,步驟5 15(或步驟5 3 5)的時刻所 獲得的S i (或M i )不收納,在迴歸分析(Regressi〇n analysis)器1 9的處理’此時刻的殘膜厚度資料不在此限 〇 步驟5 1 5、5 3 5所獲得的層差値與罩幕材剝離量 (殘膜厚)時系列S i、M i分別被收納於資料記錄器 1 8、2 8。使用被收納的過去時系列S j 、M j ,藉由 (請先閲讀背面之注意事項再填寫本頁) 經 •部 智 .慧 財 產 局 員 工 消 費 合 作 社 印 製 迴歸分析器1 X b ( Y :蝕 時間、X a : 迴歸直線算出 Μ )(步驟5 度、初期膜厚 其次,在 出加工深度( 目標加工深度 處理材的蝕刻 於顯示器1 7 步驟5 0 4、 5 14)。 9、2 9求出一次迴歸直線Y = xa * t + 刻量(層差値、罩幕材殘膜厚)、t :蝕刻 絕對値蝕刻速度、X b :初期膜厚),由此 現在的蝕刻量(層差:S、罩幕材殘膜厚: 1 6、5 3 6 )。此處,飩刻時間、蝕刻速 '殘膜厚等爲製程量(此處爲蝕刻量)。 終點判定器3 0中藉由這些飩刻量S、Μ求 =S - Μ )(圖2Α的43),比較此値與 ,若加工深度爲目標加工深度以上的話,被 重爲預定値使餓刻處理終了,將其結果顯不 。當加工深度爲目標加工深度以下時,返回 5 2 4。最後進行取樣終了的設定(步驟 訂 線一 本紙張尺度適用中國國家標準(CNS ) Α4規格(210Χ297公釐) -26 - 517306 A7 B7_____ 五、發明説明(24 圖7係顯示藉由上述實施例所進行的矽深度的測定結 果(在步驟516、536的算出値)。圖中顯示矽加工 深度以及來自罩幕材的層差的時間的變化,S T I加工時 的蝕刻的樣子明確地獲知。此外,此實施例到矽深度 506nm (層差529nm)爲止進行蝕刻處理。 如果依照以上所述的本實施例的蝕刻量測定裝置,可 正確地測定半導體裝置的製程等中的被處理材的蝕刻量。 因此,利用此系統可提供高精度地實施被處理材的蝕刻的 方法。而且,如果依照變形例與第一實施例不同,可測定 以標準圖案設定的被處理材的蝕刻量以外的任意蝕刻量。 此外,圖5的構成分別關於第一、第二波長頻帶配設 由第一數位濾波電路、微分器、第二數位濾波電路、微分 波形圖案資料庫、微分波形比較器、迴歸分析器所組成的 構成。但是,關於第一、第二波長頻帶僅共通地配設一個 這種構成,每一預定時間切換微係數等,關於第一、第二 波長頻帶使其交互地獲得實圖案也可以。 其次,使用圖8、9、10、1 1說明本發明第二實 施例。圖8所示的本實施例的構成與關於圖5的一方的波 長頻帶的構成1 1 — 1 9相同,終點判定器1 3 0的動作 與圖5的終點判定器3 0的動作不同。触刻加工的被處理 材的構造顯示於圖9。在此蝕刻處理中被加工的多晶矽 5 0的區域爲無罩幕材5 1 (例如氮化膜或光阻)的部分 ,被觀測的干涉光係由多晶矽5 0表面的反射光9 0 A與 來自底層氧化膜5 2的反射光9 0 B的干涉而引起。依照 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) ' •27- (請先閲讀背面之注意事項再填寫本頁)-25- 517306 A7 B7 V. Description of the invention (These series of smoothed differential coefficients are D i, j, d, i, j and the differential pattern psj set in the differential waveform comparators 15 and 25 in advance. Ρ M j to calculate the layer difference 値 i at this time and the peeling amount (residual film thickness) Mi of the cover material (steps 515 and 535). Here, when ^ 7 > 075〇 (or mouth > 6 ^ 〇) , S i (or M i) obtained at the time of step 5 15 (or step 5 3 5) is not stored, and is processed by the regression analysis device 19 'The residual film thickness data at this time is not here Limitation: The series S i and M i when the layer difference 步骤 obtained in steps 5 1 5 and 5 3 5 and the peeling amount (residual film thickness) of the cover material are stored in the data loggers 1 and 2 respectively. The past tense series S j and M j are produced by (please read the precautions on the back before filling this page) by the Ministry of Economics. Hui Property Bureau Employee Consumption Cooperative Printed Regression Analyzer 1 X b (Y: Eclipse Time, X a: Calculate M by regression line) (step 5 degrees, initial film thickness, followed by the processing depth (target processing depth of the processed material is etched on the display 1) 7 Step 5 0 4, 5 14). 9, 2 9 Find the one-time regression line Y = xa * t + engraved amount (layer difference 値, residual thickness of the curtain material), t: absolute etching 値 etching speed, X b : Initial film thickness), based on the current etching amount (layer difference: S, residual film thickness of the mask material: 1, 6, 3, 6). Here, the engraving time, etching rate, and residual film thickness are the process amounts. (Here is the etching amount.) The end point determiner 30 uses these engraving amounts S and M to find S = M-(43 in Fig. 2A), and compares this with if the processing depth is greater than the target processing depth. , It is re-determined to make the end of hungry processing, and the results are not displayed. When the machining depth is below the target machining depth, 5 2 4 is returned. Finally, set the end of sampling (steps to order a paper size to apply the Chinese National Standard (CNS) A4 specifications (210 × 297 mm) -26-517306 A7 B7_____ V. Description of the invention (24 Figure 7 shows the use of the above embodiment) The results of the silicon depth measurement (calculated at steps 516 and 536). The graph shows the time variation of the silicon processing depth and the step difference from the cover material, and the state of the etching during STI processing is clearly known. In addition, In this embodiment, an etching process is performed up to a silicon depth of 506 nm (step difference of 529 nm). According to the etching amount measuring device of the present embodiment described above, the etching amount of a material to be processed in a semiconductor device manufacturing process and the like can be accurately measured. Therefore, using this system can provide a method for performing etching of a material to be processed with high accuracy. Moreover, according to the modified example, it is possible to measure an arbitrary amount of etching other than the amount of etching of the material to be processed set in a standard pattern in accordance with the modified example. In addition, the configuration of FIG. 5 is configured by a first digital filter circuit, a differentiator, and a second digital filter for the first and second wavelength bands, respectively. Wave circuit, differential waveform pattern database, differential waveform comparator, regression analyzer. However, only one such configuration is commonly provided for the first and second wavelength bands, and the differential coefficient is switched every predetermined time. It is also possible for the first and second wavelength bands to obtain a real pattern alternately. Next, a second embodiment of the present invention will be described with reference to Figs. 8, 9, 10, and 11. The structure of the present embodiment shown in Fig. 8 and Regarding the configuration of one of the wavelength bands 1 1 to 19 in Fig. 5, the operation of the end point determiner 130 is different from the operation of the end point determiner 30 in Fig. 5. The structure of the material to be processed by the engraving is shown in the figure. 9. The area of polycrystalline silicon 50 processed in this etching process is the part of unshielded curtain material 5 1 (such as a nitride film or a photoresist), and the observed interference light is reflected from the surface of polycrystalline silicon 5 0 9 Caused by the interference of A with the reflected light from the bottom oxide film 5 2 9 B. According to the paper size, the Chinese National Standard (CNS) A4 specification (210X297 mm) is applied. • 27- (Please read the precautions on the back before (Fill in this page)
、1T -線一 經濟部智慧財產局員工消費合作社印製 517306 A7 B7 五、發明説明(23 圖1 0的流程圖說明藉由此干涉光的計測測定多晶矽5 0 的蝕刻量(殘膜厚5 3 :來自底層氧化膜的多晶矽厚度) 的方法。 最初設定關於被處理材(多晶矽)的目標殘膜値與預 先被保存於微分波形圖案資料庫1 6的多晶矽膜厚的所有 標準微分圖案(P z j )與收斂判定値σ z 〇於微分波形比 較器1 5 (步驟6 0 0 )。伴隨著蝕刻處理的開始,開始 干涉光的取樣(步驟6 0 2 )。來自光譜儀1 1的多波長 的光分別中介第一數位濾波器12與微分器13與第二數 位濾波器1 4,與第一實施例的步驟4 0 4 — 4 1 0 —樣 ,求出平滑化微係數時系列資料D i ,j 。比較這些平滑 化微係數時系列資料D i ,j與預先被設定於微分波形比 較器1 5的微分圖案P z j ,算出在該時刻的殘膜値Z i ( 步驟6 1 5 )。此處,σ > σ z 〇時在步驟6 1 5於此時刻 所獲得的Z i値不收納,以迴歸分析器1 9的處理在此時 刻的殘膜厚度資料不在此限。 在步驟6 1 5所獲得的殘膜値係以時系列資料Z i收 納於資料記錄器1 8。使用所收納的過去時系列資料Z i ,藉由迴歸分析器1 9求出一次迴歸直線Y = X a * t + X b ( Y :殘膜量、t :蝕刻時間、X a : X a的絕對値 爲蝕刻速度、X b :初期膜厚),由此迴歸直線算出現在 的殘膜量Z (步驟6 1 6 )。 其次,在終點判定器1 3 〇中比較殘膜量Z與目標殘 膜値,若爲目標殘膜以下的話,被處理材的鈾刻量.爲預定 本紙張尺度適用中.國國家標準(CNS ) A4規格(210X 297公釐) (請先閲讀背面之注意事項再填寫本頁) •^1. 訂 經濟部智慧財產局員工消費合作社印製 -28- 517306 A7 B7 五、發明説明( 値其結果顯示於顯示器1 7。若爲目標殘膜以上的情形, 返回步驟6 0 4。最後進行取樣終了的設定(步驟6 1 4 )° 可是,若目標殘膜値比關於預先保存的多晶矽膜厚的 微分波形圖案資料庫P z j還少的話,在步驟6 1 8判定的 情形進行如以下的處理,使鈾刻處理終了。當殘膜量Z與 關於保存的多晶矽膜厚的微分波形圖案資料庫的最小膜厚 Ym相等時,由上述一次迴歸直線Y = xa * t + xb, 算出目標殘膜値爲γτ的蝕刻時刻(tT二〔Υτ - Xb〕 / X a ),到此時刻t τ爲止繼續進行蝕刻處理。 圖1係顯示由本實施例進行的多晶矽的殘膜測定結果 。此乃關於多晶矽膜厚的微分波形圖案資料庫使用到最小 膜厚Ym=4 5 nm爲止的資料庫,預測目標殘膜値Y t =2 0 nm,由一次迴歸直線目標殘膜値成爲2 0 nm的 蝕刻時刻9 6秒明確地獲知。據此,無微分波形圖案資料 庫的殘膜量的點終點判定爲可能。 而且,一次迴歸直線的斜率的絕對値(二I X a | ) 顯示蝕刻速度,藉由量產管理此蝕刻速度可進行鈾刻裝置 的狀態管理。即若鈾刻速度在某容許値以內的話,可監視 蝕刻裝置正常運轉,且容許値以外的情形爲異常。 再者,一次迴歸直線的切片( = Xb )係顯示被處理 材的初期膜厚,藉由量產管理此初期膜厚可進行蝕刻處理 前的成膜狀態的管理。即若初期膜厚在某容許値以內的話 ,可監視回饋(Feedback)成膜裝置正常運轉,且容許値 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) f 訂 經濟部智慧財產局員工消費合作社印製 -29 - 517306 A7 __ B7 五、發明説明( 以外的情形爲異常。 其次’使用圖1 2、1 3說明本發明的第三實施例。 本實施例在管理蝕刻深度時,因每一晶圓有機膜的膜厚有 誤差,使其由初期膜厚與測定的殘膜厚測定飽刻深度。圖 12所示的本實施例的構成與圖8的構成丄i 目胃 ’終點判定器2 3 0的動作與圖8的終點判定器1 3 〇的 動作不同。蝕刻加工的被處理材的構造顯示於圖1 3。在 此蝕刻處理中被加工的有機膜6 0的區域(溝槽構造)爲 無罩幕材6 1 (例如氮化膜或光阻)的部分,被觀測的+ 涉光係因來自有機膜6 0表面的反射光與配線材6 2 (例 如C u )的反射光的干涉而引起。 依照圖1 4的流程圖說明藉由此干涉光的計測測定有 機膜6 0的蝕刻量(溝槽深度:D與Ε的距離6 5 )的方 法。最初將被處理材(有機膜)的目標深度値與關於預先 保存於微分波形圖案資料庫1 6的有機膜的膜厚之所有的 標準微分圖案(P f j )與收斂判定値(σ F 〇 )設定於微 分波形比較器1 5 (步驟7 0 0 )。伴隨著蝕刻處理的開 始,開始干涉光的取樣(步驟7 0 2 )。來自光譜儀1 1 的多波長的光分別中介第一數位濾波器1 2與微分器1 3 與第二數位濾波器1 4,與第二實施例的步驟6 0 4 -6 1 0 —樣,求出平滑化微係數時系列資料D i ,j °比 較這些平滑化微係數時系列資料D i ,j與預先被設定於 微分波形比較器1 5的微分圖案P f j ’算出在該時刻的殘 膜値F i (步驟7 1 5 )。此處,σ > σ F ◦時於此時刻的 本紙張尺度適用中國國家標準(CNS ) Α4規格(210Χ 297公釐) (請先閲讀背面之注意事項再填寫本頁) 訂 % 經濟部智慧財產局員工消費合作社印製 -30- 517306 A7 ____B7___ 五、發明説明( 殘膜厚度値不收納,以迴歸分析器1 9的處理在此時刻的 殘膜厚度資料不在此限。 在步驟7 1 5所獲得的殘膜値係以時系列資料F i收 納於資料記錄器1 8。使用所收納的過去時系列資料F i ’藉由迴歸分析器1 9求出一次迴歸直線Y = X a * t + X b ( Y :殘膜量、t :蝕刻時間、X a :絕對値爲蝕刻 速度、Xb:初期膜厚),由此迴歸直線算出現在的殘膜 量F與初期膜厚X b (步驟7. 1 6 )。 其次,在終點判定器2 3 0中藉由殘膜量F (圖1 3 的64)與初期膜厚Xb (圖13的63)求出現在的溝 槽深度( = Xb — F)(圖13的65),比較此溝槽深 度與目標深度値,若爲目標深度以上的話,視被處理材的 蝕刻量爲預定値將其結果顯示於顯示器1 7。當爲目標深 度以下時,返回步驟7 0 4。最後進行取樣終了的設定( 步驟7 1 4 )。如此,溝槽加工時的蝕刻深度藉由利用迴 歸分析求出殘膜量F與初期膜厚X b使測定爲可能。 其次,使用圖15、16、17說明本發明的第四實 施例。圖1 5所示的本實施例的構成與圖1 2的構成1 1 一 1 9相同,終點判定器3 3 0的動作與圖1 2的終點判 定器230的動作不同,再者配設控制裝置1〇〇〇。常 常有使用同一個蝕刻裝置1蝕刻處理種種膜質不同的被蝕 刻材。這種情形,根據預先被設定於控制裝置1 〇 〇 〇的 鈾刻條件(例如鈾刻氣體條件或電漿產生功率條件或偏!g (B i a s )條件等),藉由控制裝置1 0 〇 〇氣體時間 本紙張尺度適用中國國家標準(CNS ) A4規格{ 210X297公釐) ' ' -31- (請先閲讀背面之注意事項再填寫本頁)1T-line 1 Printed by the Consumers' Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs 517306 A7 B7 V. Description of the invention (23 Fig. 10 The flow chart illustrates the measurement of polysilicon 5 0 etching (residual film thickness 5 by this interference light measurement) 3: Polycrystalline silicon thickness from the underlying oxide film). Initially, the target residual film of the material to be processed (polycrystalline silicon) and all standard differential patterns of polycrystalline silicon film thickness (P6) previously stored in the differential waveform pattern database 16 are set. zj) and convergence determination 値 σ z 〇 in the differential waveform comparator 15 (step 6 0 0). With the start of the etching process, the interference light sampling is started (step 6 0 2). The multi-wavelength from the spectrometer 11 1 The light mediations the first digital filter 12 and the differentiator 13 and the second digital filter 14 respectively, as in the step 4 0 4 — 4 1 0 — of the first embodiment, and obtains the series of data D i when smoothing the micro coefficients. , J. The series data D i, j when these smoothed differential coefficients are compared with the differential pattern P zj set in the differential waveform comparator 15 in advance, and the residual film 値 Z i at this time is calculated (step 6 1 5). Here, σ > σ z 〇 At this time, the Z i 步骤 obtained at step 6 1 5 is not stored, and the residual film thickness data processed at this time by the regression analyzer 19 is not limited. The residual film 値 system obtained at step 6 1 5 The time series data Z i is stored in the data recorder 18. Using the stored past time series data Z i, a regression line Y = X a * t + X b (Y: residual The film amount, t: etching time, Xa: Xa absolute 値 is the etching rate, Xb: initial film thickness), and the current residual film amount Z is calculated from the regression line (step 6 1 6). Next, at the end point The judging device 130 compares the residual film amount Z with the target residual film 値. If it is less than the target residual film, the uranium engraved amount of the material to be processed. It is planned to be applicable to this paper size. National National Standard (CNS) A4 specification ( 210X 297 mm) (Please read the notes on the back before filling out this page) • ^ 1. Order printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs-28- 517306 A7 B7 V. Description of the invention (値 Results are displayed on the display 1 7. If it is above the target residual film, go back to step 6 0 4. Finally, the sampling is finished. (Step 6 1 4) ° However, if the target residual film is smaller than the differential waveform pattern database P zj regarding the thickness of the polycrystalline silicon film saved in advance, the situation determined in step 6 1 8 is processed as follows, so that The uranium engraving process is finished. When the residual film amount Z is equal to the minimum film thickness Ym of the differential waveform pattern database on the polycrystalline silicon film thickness stored, the target one-time regression line Y = xa * t + xb is used to calculate the target residual film 値 as The etching time of γτ (tT 2 [Υτ-Xb] / X a), and the etching processing is continued until this time t τ. FIG. 1 shows the measurement results of the residual film of polycrystalline silicon in this example. This is a differential waveform pattern database on polycrystalline silicon film thickness. The database with the minimum film thickness Ym = 4 5 nm is used to predict the target residual film 値 Y t = 2 0 nm. The linear target residual film 値 will be changed from 1 to 2 0 The etching time of nm is clearly known at 96 seconds. This makes it possible to determine the end point of the residual film amount in the non-differential waveform pattern database. In addition, the absolute 値 (two I X a |) of the slope of the linear regression line shows the etching rate, and the state of the uranium etching device can be managed by mass production management of this etching rate. That is, if the uranium etching speed is within a certain allowable plutonium, the normal operation of the etching device can be monitored, and the situation other than the allowable plutonium is abnormal. In addition, the slice (= Xb) that returns to a straight line once shows the initial film thickness of the material to be processed, and this initial film thickness can be managed by mass production to manage the film formation state before the etching process. That is, if the initial film thickness is within a certain tolerance, the normal operation of the feedback film-forming device can be monitored, and the paper size is allowed to apply the Chinese National Standard (CNS) A4 specification (210X297 mm) (Please read the back Note: Please fill in this page again.) F Order printed by the Intellectual Property Bureau of the Ministry of Economic Affairs's Consumer Cooperatives. -29-517306 A7 __ B7 V. Explanation of the invention (Except for other cases, it is abnormal. Secondly, use Figure 1 2 and 1 3 The third embodiment. When the etching depth is managed in this embodiment, because the film thickness of the organic film of each wafer has an error, the saturated depth is measured from the initial film thickness and the measured residual film thickness. The structure of the embodiment is different from the structure of FIG. 8 in that the operation of the end point determiner 230 is different from that of the end point determiner 130 in FIG. 8. The structure of the material to be processed by etching is shown in FIG. The area (groove structure) of the organic film 60 processed in this etching process is a portion without a cover material 6 1 (such as a nitride film or a photoresist), and the observed + light-related system originates from the organic film 6 0 Surface reflected light and wiring materials 6 2 ( Such as caused by the interference of the reflected light. The method of measuring the etching amount of the organic film 60 (the depth of the groove: the distance between D and E 6 5) by measuring the interference light will be described in accordance with the flowchart of FIG. 14. Initially, the target depth of the material to be processed (organic film) 値 and all the standard differential patterns (P fj) and the convergence judgment 値 (σ F 〇) regarding the film thickness of the organic film previously stored in the differential waveform pattern database 16 ) Is set to the differential waveform comparator 15 (step 7 0 0). With the start of the etching process, the sampling of the interference light is started (step 7 0 2). The multi-wavelength light from the spectrometer 1 1 respectively intervenes the first digital filter Divider 1 2 and differentiator 1 3 and second digital filter 1 4 are the same as the step 6 0 4 -6 1 0 of the second embodiment. In the same way, the series of data D i when calculating the smoothed coefficients are compared. When smoothing the differential coefficient, the series of data D i, j and the differential pattern P fj ′ set in advance to the differential waveform comparator 15 calculate the residual film 値 F i at this time (step 7 1 5). Here, σ > σ F ◦ At this moment, the paper size applies to the Chinese National Standard (CNS) Α4 Specification (210 × 297 mm) (Please read the notes on the back before filling in this page) Order% Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economics -30- 517306 A7 ____B7___ V. Description of the Invention The residual film thickness data processed by the regression analyzer 19 at this moment is not limited to this. The residual film obtained in step 7 1 5 is stored in the data recorder 18 with the time series data F i. The past time series data F i 'find the regression straight line Y = X a * t + X b (Y: residual film amount, t: etching time, X a: absolute 値 is the etching rate, Xb: initial film thickness), based on the regression line, the current residual film amount F and initial film thickness X b are calculated (step 7. 1). Next, in the end point determiner 230, the depth of the groove (= Xb — F) where the residual film amount F (64 in FIG. 13) and the initial film thickness Xb (63 in FIG. 13) appear is calculated (FIG. 13) 65), compare this groove depth with the target depth, if it is greater than the target depth, consider the etching amount of the material to be treated as predetermined, and display the result on the display 17. When it is less than the target depth, return to step 704. Finally, set the end of sampling (step 7 1 4). As described above, the etching depth during trench processing can be measured by using a regression analysis to obtain the residual film amount F and the initial film thickness Xb. Next, a fourth embodiment of the present invention will be described with reference to Figs. The structure of this embodiment shown in FIG. 15 is the same as the structure 1 1 to 19 of FIG. 12. The operation of the end point determiner 3 3 0 is different from the operation of the end point determiner 230 in FIG. 12. Furthermore, control is provided. Device 1000. The same etching device 1 is often used to etch various etched materials with different film qualities. In this case, according to the uranium engraving conditions (such as the uranium engraving gas conditions or the plasma generation power conditions or partial! (B ias) conditions) set in the control device 1000 in advance, the control device 1 100 〇Gas time This paper size applies the Chinese National Standard (CNS) A4 specification {210X297 mm) '' -31- (Please read the precautions on the back before filling this page)
、1T 線- 經濟部智慧財產局員工消費合作社印製 517306 A7 B7 五、發明説明(2会 (請先閱讀背面之注意事項再填寫本頁) 控制供給裝置1 0 0 1或電漿產生裝置1 0 0 2或晶圓偏 壓(Wafer bias)電源1 〇 〇 3,進行蝕刻處理。但是,蝕 刻加工的被處理材藉由半導體元件構造成爲圖16所示的 疊層構造時,蝕刻處理變的複雜。以利用單純的時間控制 的蝕刻處理無損傷(D a m a g e )的元件加工很困難。 使用圖1 6說明關於這種疊層構造元件的蝕刻加工。在此 蝕刻處理中被加工的多晶矽膜7 0上形成B A R C 7 3 ( 反反反光射塗佈,Back Anti-Reflection Coating)或罩幕 材7 1 (例如氮化膜或光阻),而且在多晶矽膜7 0下形 經濟部智慧財產局員工消費合作社印製 成底層氧化膜7 2。再者,底層氧化膜7 2係電晶體的閘 電極部7 8的厚度(例如約2 n m )與用以隔離電晶體元 件的溝槽部7 9 ( S T I )的厚度(例如約3 0 0 n m ) 大不相同的構造。在此構造的加工中,首先進行 B A R C 7 3的蝕刻處理,接著以同一個蝕刻裝置一貫處 理多晶矽膜7 0的蝕刻處理。這種蝕刻處理若無法適切地 鈾刻各膜的話,會過於剝離閘電極部7 8的底層氧化膜 7 2,帶給元件損傷。因此,首先在B A R C 7 3蝕刻處 理中,必須儘可能地使其不剝離多晶矽7 0而控制蝕刻。 此點在B A R C 7 3蝕刻處理中測定B A R C的殘膜量, 在些微的殘膜量7 5 (例如2 0 n m )的時點將蝕刻條件 變更爲多晶矽難以剝離的條件,以蝕刻剩餘的B A R C材 很重要。其次,在多晶矽7 0的蝕刻處理中測定多晶矽的 殘膜量(殘膜厚),在些微的殘膜量(殘膜厚)7 7 (例 如2 0 n m )的時點將蝕刻條件變更爲底層氧化膜難以剝 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公董) " -32 _ 517306 經濟部智慧財產局員工消費合作社印製 A7 B7五、發明説明(3C> 離的條件,以蝕刻剩餘的多晶矽材很重要。 使用於B A R C的殘膜厚度量測定的光係利用來自 B A R C表面的反射光與來自多晶矽邊界面的反射光的干 涉光9 6。而且,使用於多晶矽的殘膜厚度量測定的光係 利用來自多晶矽表面的反射光與來自底層氧化膜邊界面的 反射光的干涉光9 5 A或9 5 B。此時,底層氧化膜的厚 度因閘電極部7 8的厚度7 7與用以隔離電晶體元件的溝 槽部7 9的厚度7 6不同,故來自各個部分的干涉光強度 9 5A、9 5B不同。來自元件隔離部79的干涉光強度 9 5B比來自閘電極部78的干涉光強度9 5A大。因此 ,在多晶矽的殘膜厚度測定中,以元件隔離部7 9上的多 晶矽爲對象來進行。即多晶矽的蝕刻處理考慮此點使用干 涉光強度9 5 B,進行多晶矽的蝕刻處理到殘膜厚度爲厚 度7 6爲止,然後,進行在底層氧化膜很難剝離的條件下 的多晶矽材的蝕刻處理。 依照圖1 7的流程圖說明進行此蝕刻處理的順序。最 初控制裝置1 0 0 0被設定對疊層膜(例如B A R C 7 3 與多晶矽7 0 )的蝕刻條件(例如氣體條件、放電條件、 壓力條件等)與各膜7 3、7 0的目標殘膜厚度値7 5、 7 6與收斂判定値(步驟8 0 0 )。其次,依照各膜種將 關於預先保存於微分波形圖案資料庫1 6的各膜7 3、 70的膜厚之所有的標準微分圖案(Pzj )與收斂判定値 (σ z 〇 )設定於微分波形比較器1 5 (步驟8 0 1 )。在 下一個步驟中開始鈾刻處理與開始干涉光的取樣(步驟 本紙張尺度適用中國國家標準(CNS ) Α4規格(210Χ297公釐) (請先閲讀背面之注意事項再填寫本頁) -^^1. ,ιτ 線一 • 33 - 517306 A7 經濟部智慧財產局員工消費合作社印製 B7 五、發明説明(31) 802)。其次,藉由來自控制裝置1Q00的指示,將 關於第一被處理的被蝕刻材(例如BAR C材)的標準微 分圖案P z j與收斂判定値σ Z 〇,自微分波形圖案資料庫 1 6設定於微分波形比較器1 5 (步驟8 0 3 )。來自光 譜儀1 1的多波長的光分別由被第一數位濾波器1 2求出 平滑化時系列資料Y i ,j (步驟8 0 4 )。再者,中介 微分器13與第二數位濾波器14與第三實施例的步驟 7 0 4 - 7 1 0 —樣,求出平滑化微係數時系列資料D i ,j (步驟8 0 6、8 0 8 )。比較這些平滑化微係數時 系列資料D i ,j與預先被設定於微分波形比較器1 5的 微分圖案PFj ,求出收斂値σ=Σ(Di , j— P q ) 2對最小收斂値的殘膜厚。此時,當σ S σ z。時, 以求出的殘膜厚度爲在此時刻的殘膜厚度値Z i收納於資 料記錄器18 (步驟810、81 5)。當σ>σζ〇時, 在此時刻的殘膜厚度値不收納,以迴歸分析器1 9的處理 在此時刻的殘膜厚度資料不在此限(步驟8 1 5 )。使用 所收納的過去時系列資料Z i ,藉由迴歸分析器1 9求出 一'次迴歸直線Y = Xa * t + Xb,由此迴歸直線算出現 在的殘膜厚度量Z (步驟8 1 6 )。 其次,在終點判定器3 3 0中比較現在的殘膜厚度量 Z與來自控制裝置1 0 0 〇的目標殘膜厚度値7 5 (例如 BARC殘膜厚度2 0nm),若爲目標殘膜値75以下 話,視被處理材的蝕刻量爲預定値將其結果顯示於顯示器 1 7。同時切換鈾刻條件進行以下的飩刻處理(多晶矽難 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) ' -34- (請先閲讀背面之注意事項再填寫本頁) P. 訂 線一 517306 A7 B7_ 五、發明説明( (請先閱讀背面之注意事項再填寫本頁) 以被剝離的B A R C蝕刻條件)(步驟8 1 8 )。當爲目 標殘膜値以上時,返回步驟8 0 4。切換的下一個蝕刻處 理(多晶矽難以被剝離的B A R C鈾刻條件)例如藉由控 制裝置1 0 0 0控制晶圓偏壓電源1 0 0 3 ,蝕刻殘膜部 ’在預先設定的時間經過後,進行触刻條件的切換(步驟 8 19)。 下一個膜種(例如多晶砍7 0 )的鈾刻在預先設定的 飩刻條件下,控制供給裝置1 0 0 1或電漿產生裝置 1 0 0 2或晶圓偏壓電源1 0 0 3,並且設定關於下一個 膜種(例如多晶矽)的殘膜厚度7 6的標準微分圖案( P z j 、σ 2 〇 )於微分波形比較器1 5 (步驟8 0 3 ), 再度實行步驟8 0 4到8 1 8。在終點判定器3 3 0中, 判定多晶矽的殘膜厚度量Ζ爲目標殘膜厚度値7 6 (例如 經濟部智慧財產局員工消費合作社印製 多晶矽殘膜厚度2 0 n m )(步驟8 1 8 ),其結果顯示 於顯示器1 7的話,同時切換蝕刻條件進行下一個蝕刻處 理(底層氧化膜難以被剝離的多晶矽蝕刻條件)(步驟 8 18)。此蝕刻處理(底層氧化膜難以被剝離的多晶矽 蝕刻條件)藉由時間控制蝕刻多晶矽的殘膜部,在預先設 定的時間經過後,進行蝕刻條件切換的判定(步驟8 1 9 )。當無下一個鈾刻處理時,進行飩刻處理的終了以及取 樣的終了的設定(步驟8 1 4 )。 如以上如果依照第二實施例的上述方法,即使關於被 處理材的殘膜厚度之微分圖案資料庫不充分.,仍可預測殘 膜厚度。再者,在關於殘膜厚度的微分圖案資料庫的作成 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 35 · 517306 A7 B7 五、發明説明(3$ 中,無須完全蝕刻取樣晶圓,可削減取樣晶圓。 而且,因藉由本發明的迴歸分析求出蝕刻速度,藉由 蝕刻速度的每一晶圓的監視可進行量產管理,可防止製品 晶圓的不良發生於未然。 而且,因可藉由本發明的迴歸分析算出被處理材的初 期膜厚,藉由將其結果回饋到成膜裝置,可進行成膜裝置 與蝕刻裝置的一貫處理的量產管理。 如果依照第三實施例的方法,即使被處理材的初期膜 厚有偏差(Variation ),仍可精度良好地算出蝕刻深度, 可進行正確地判定當作目標的飩刻深度之蝕刻處理。 而且,如果依照第四實施例的方法,在具備疊層膜構 造的被處理材的蝕刻處理中,在蝕刻各膜時進行殘膜厚度 測定,在預定的殘膜厚度中藉由切換蝕刻條件,可使底層 膜的蝕刻剝離少。 而且’在閘電極的多晶矽蝕刻中,藉由利用來自元件 隔離部的干涉光測定多晶矽殘膜厚度,可正確地判定預定 殘膜厚度’無過度地蝕刻閘電極部的底層氧化膜,可抑制 晶圓的不良處理片數到最小限。 此外,在上述各實施例中自具有光源的光譜儀放出多 波長的放射光,利用來自被處理材的反射光的干涉光進行 測定。另一方面,利用不具光源的光譜儀使其以藉由電漿 而放出的多波長的放射光爲光源而利用也可以。 此外,本發明更具有以下的特徵。 (1 )、一種蝕刻方法,其特徵爲:儲存在試樣的蝕 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁)Line 1T-printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 517306 A7 B7 V. Description of the invention (2 sessions (please read the notes on the back before filling this page) Control supply device 1 0 0 1 or plasma generation device 1 0 0 2 or wafer bias power supply 003, and the etching process is performed. However, when the material to be processed by the etching process is changed to the stacked structure shown in FIG. 16 by the semiconductor element structure, the etching process is changed. It is complicated. It is difficult to process the element without damage by the simple time-controlled etching process. The etching process of such a laminated structure element will be described using FIG. 16. The polycrystalline silicon film 7 processed in this etching process BARC 7 3 (Back Anti-Reflection Coating) or cover material 7 1 (such as nitride film or photoresist) is formed on 0, and the employee of Intellectual Property Bureau of the Ministry of Economic Affairs is shaped under polycrystalline silicon film 70 The consumer cooperative prints the bottom oxide film 72. Furthermore, the bottom oxide film 72 is a gate electrode portion 7 8 of a transistor (eg, about 2 nm) in thickness and a trench portion 7 9 ( STI) Structure (for example, about 300 nm) is very different. In the processing of this structure, the BARC 7 3 is firstly etched, and then the polysilicon film 70 is etched by the same etching device. This kind of etching If the film cannot be properly etched with uranium, the underlying oxide film 7 2 of the gate electrode portion 7 8 will be peeled off too much, which will damage the device. Therefore, in the BARC 7 3 etching process, it is necessary to prevent the peeling as much as possible. Polysilicon 70 is used to control the etching. At this point, the amount of residual film of BARC is measured during the BARC 7 3 etching process. When the amount of residual film is 7 5 (for example, 20 nm), the etching condition is changed to a condition where the polycrystalline silicon is difficult to peel off. It is important to etch the remaining BARC material. Secondly, the amount of residual film (residual film thickness) of polycrystalline silicon is measured during the etching process of polycrystalline silicon 70, and a slight amount of residual film (residual film thickness) 7 7 (for example, 20 nm) Change the etching conditions to the bottom oxide film. It is difficult to peel off the paper. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297). &Quot; -32 _ 517306 Intellectual Property Bureau, Ministry of Economic Affairs, Consumer Consumption Agency Preparation of A7 B7 V. Description of the invention (3C) It is important to etch the remaining polycrystalline silicon material. The optical system for measuring the thickness of the residual film thickness of BARC uses reflected light from the BARC surface and reflected light from the polycrystalline silicon boundary surface. Interference light 9 6. In addition, the light used for the measurement of the residual film thickness of polycrystalline silicon is the interference light 9 5 A or 9 5 B using the reflected light from the surface of the polycrystalline silicon and the reflected light from the boundary surface of the underlying oxide film. At this time, the thickness of the underlying oxide film is different from the thickness of the gate electrode portion 7 8 and the thickness of the trench portion 7 9 which is used to isolate the transistor element 7 9. Therefore, the intensity of the interference light from each part 9 5A, 9 5B different. The intensity of the interference light 9 5B from the element isolation portion 79 is greater than the intensity of the interference light 9 5A from the gate electrode portion 78. Therefore, the measurement of the residual film thickness of the polycrystalline silicon is performed on the polycrystalline silicon on the element isolation portion 79. That is, the polysilicon etching process considers this point using an interference light intensity of 9 5 B. The polysilicon is etched until the residual film thickness is 7 6 and then the polysilicon material is etched under conditions where the underlying oxide film is difficult to peel off. . The procedure for performing this etching process will be described in accordance with the flowchart of FIG. 17. Initially, the control device 1 0 0 0 is set to the etching conditions (eg, gas conditions, discharge conditions, pressure conditions, etc.) of the laminated film (eg, BARC 7 3 and polycrystalline silicon 7 0) and the target residual film of each film 7 3, 7 0 The thicknesses 値 7, 5, 6 and convergence judgment 値 (step 8 0 0). Next, all standard differential patterns (Pzj) and convergence determinations σ (σ z 〇) for all film thicknesses of the films 7 3, 70 previously stored in the differential waveform pattern database 16 are set to the differential waveforms according to each film type. Comparator 15 (step 8 0 1). In the next step, the uranium engraving process and the sampling of the interference light are started (the paper size of this paper applies the Chinese National Standard (CNS) A4 specification (210 × 297 mm) (please read the precautions on the back before filling this page)-^^ 1 ., ιτ 线 一 • 33-517306 A7 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs B7 V. Invention Description (31) 802). Next, with the instruction from the control device 1Q00, the standard differential pattern P zj and the convergence determination 値 σ Z 〇 on the first material to be etched (for example, BAR C material) are set from the differential waveform pattern database 16 To the differential waveform comparator 15 (step 803). The multi-wavelength light from the spectrometer 11 is obtained from the series data Y i, j at the time of smoothing by the first digital filter 12 (step 8 0 4). Furthermore, the intermediary differentiator 13 and the second digital filter 14 and steps 7 0 4-7 1 0 of the third embodiment are used to obtain a series of data D i, j when smoothing the micro coefficients (step 8 0, 8 0 8). When comparing these smoothed differential coefficients, the series data D i, j and the differential pattern PFj set in the differential waveform comparator 15 in advance, and obtaining the convergence σσ = Σ (Di, j— P q) 2 for the minimum convergence 値Residual film thickness. At this time, when σ S σ z. At this time, the obtained residual film thickness is set as the residual film thickness 値 Z i at this time in the data recorder 18 (steps 810, 815). When σ > σζ0, the residual film thickness 値 at this time is not stored, and is processed by the regression analyzer 19. The residual film thickness data at this time is not limited (step 8 15). Using the stored past-time series data Z i, a regression regression line Y = Xa * t + Xb is obtained by the regression analyzer 19 to calculate the current residual film thickness amount Z from the regression line (step 8 1 6 ). Next, the current residual film thickness amount Z is compared with the target residual film thickness 値 75 (for example, BARC residual film thickness 20nm) from the control device 100 in the end point determiner 3 30. If it is the target residual film 値Below 75, the etching amount of the material to be treated is determined to be predetermined, and the result is displayed on the display 17. Simultaneously switch the uranium engraving conditions for the following engraving treatments (polycrystalline silicon paper size is applicable to Chinese National Standard (CNS) A4 specification (210X297 mm))--34- (Please read the precautions on the back before filling this page) P. Order line 517306 A7 B7_ V. Description of the invention ((Please read the precautions on the back before filling this page) to remove the BARC etching conditions) (step 8 1 8). When the target residual film is more than 値, return to step 8 0 4. Switch to the next etching process (barc uranium etching conditions where polycrystalline silicon is difficult to be stripped). For example, the control device 1 0 0 controls the wafer bias power 1 0 3, and the etching residual film portion is at a preset value. After the time has elapsed, switch the engraving conditions (steps 8 to 19). The uranium of the next film type (for example, polycrystalline cutting 70) is engraved under the preset engraving conditions, and the supply device 1 0 0 1 or plasma is controlled. Generate device 1 0 2 or wafer bias power supply 1 0 3, and set the standard differential pattern (P zj, σ 2 〇) on the residual film thickness 7 6 of the next film type (such as polycrystalline silicon) to compare with the differential waveform Device 15 (step 80 0 3), and then Perform steps 8 0 4 to 8 1 8. In the end point determiner 3 3 0, determine the amount of residual film thickness Z of the polycrystalline silicon as the target residual film thickness 値 7 6 (for example, the polycrystalline silicon residual film printed by the staff consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs). Thickness 20 nm) (step 8 1 8), and if the result is displayed on the display 17, simultaneously switch the etching conditions for the next etching process (polycrystalline silicon etching conditions where the underlying oxide film is difficult to peel off) (step 8 18). This etching The process (polycrystalline silicon etching conditions where the underlying oxide film is difficult to peel off) etches the remaining polycrystalline silicon film by time control. After a preset time has elapsed, the judgment of the switching of the etching conditions is made (step 8 19). When there is no next uranium During the engraving process, the end of the engraving process and the setting of the end of sampling are performed (step 8 1 4). As described above, if the method of the second embodiment is followed, even if the differential pattern database of the residual film thickness of the material to be processed is not Sufficient. The thickness of the residual film can still be predicted. Furthermore, the paper size of the differential pattern database on the thickness of the residual film is subject to the Chinese National Standard (CNS) A4 regulations. (210X297 mm) 35 · 517306 A7 B7 V. Description of the invention (In 3 $, the sampling wafer does not need to be completely etched, and the sampling wafer can be reduced. In addition, the etching rate is obtained by the regression analysis of the present invention, and the etching rate is Monitoring of each wafer enables mass production management, which can prevent defects in the product wafer from occurring in advance. Furthermore, the initial film thickness of the material to be processed can be calculated by the regression analysis of the present invention, and the results can be fed back to The film forming device can perform mass production management of the consistent processing of the film forming device and the etching device. According to the method of the third embodiment, the etching depth can be calculated with high accuracy even if the initial film thickness of the material to be processed varies, and the etching process can be accurately determined as the target etched depth. In addition, according to the method of the fourth embodiment, in the etching process of a material to be processed having a laminated film structure, the residual film thickness is measured when each film is etched, and the etching conditions are switched at a predetermined residual film thickness. The etch peeling of the underlying film can be reduced. In addition, in the polysilicon etching of the gate electrode, the residual film thickness of the polysilicon can be accurately determined by measuring the thickness of the polysilicon residual film using the interference light from the element isolation portion. The number of defective round pieces is the minimum. In addition, in each of the above embodiments, a multi-wavelength radiated light was emitted from a spectrometer having a light source, and the measurement was performed using interference light from reflected light from a material to be processed. On the other hand, a spectrometer without a light source may be used as a light source with multi-wavelength radiation emitted by a plasma as a light source. In addition, the present invention has the following features. (1) An etching method, which is characterized by: the etching stored in the sample. The paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm) (Please read the precautions on the back before filling this page)
、1T 經濟部智慧財產局員工消費合作社印製 -36- 517306 A7 B7 五、發明説明( 刻途中測定的資料’利用該被儲存的過去資料預測前述蝕 刻中的飩刻深度,在預定的蝕刻深度終止前述蝕刻。 (2)、一種鈾刻方法,其特徵爲:在不具底層膜的 被處理材的鈾刻方法中’測定形成於前述被處理材之上的 罩幕的剝離量與來自前述罩幕頂面的前述被處理材的蝕刻 底面爲止的蝕刻層差,管理前述被處理材的蝕刻深度。 (3 )、一種鈾刻方法,其特徵爲:利用試樣的蝕刻 途中的干涉光資料,預測前述試樣的被蝕刻部的殘膜厚度 ,在預定的殘膜厚度終止前述蝕刻。 (4)、一種鈾刻方法,其特徵爲:在金屬鑲嵌蝕刻 (Damascene etching)中,利用前述蝕刻途中的干涉光資 料,預測初期膜厚決定溝槽深度,在預定的溝槽深度終止 前述蝕刻。 (5 )、一種蝕刻方法,其特徵爲:在蝕刻面內厚度 不同的底層膜上所形成的閘極材質的方法中,在前述底層 膜厚度厚的部分測定干涉光,測定前述厚度厚的底層膜上 所形成的閘極材質的殘膜厚度,管理前述閘極材質的膜厚 〇 (6)、一種鈾刻方法,其特徵爲:在被疊層複數個 膜的被處理材的飩刻中,測定來自前述被處理材的干涉光 ,每一前述膜切換數位濾波器處理干涉光的資料,每一前 述膜管理膜厚。 (7 )、一種蝕刻方法,其特徵爲:在B A R C飩刻 中,利用由蝕刻中的被處理材所測定的干涉光管理 本紙張尺度適用中國國家標準(CNS ) A4規格{ 210X297公釐) (請先閲讀背面之注意事項再填寫本頁) 訂 線一 經濟部智慧財產局員工消費合作社印製 -37- 517306 A7 _ B7_ 五、發明説明( B AR C的膜厚,防止底層的被鈾刻材的剝離。 (8)、一種半導體裝置的方法,其特徵爲:在具有 s T I部的半導體元件的製造中,當進行形成前述半導體 元件的一部分之P 〇 1 y - s i的蝕刻時,以形成於則述 STI部上的Po ly — Si的殘膜厚度管理前述 Ρ ο 1 y — S i的蝕刻。 (請先閱讀背面之注意事項再填寫本頁) 訂 經濟部智慧財產局員工消費合作社印製 本紙張尺度適用中國國家標準(CNS ) A4規格(210 X 297公釐) -38 -Printed by 1T Consumer Intellectual Property Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs-36- 517306 A7 B7 V. Description of the invention (data measured during the process of 'using this stored historical data to predict the etch depth in the aforementioned etch, at a predetermined etch depth (2) A method for engraving uranium, characterized in that, in the method for engraving a material to be processed without a base film, 'measure the peeling amount of the mask formed on the material to be processed and from the mask. The etching layer difference between the bottom surface of the material to be processed on the top surface of the curtain and the etching depth of the material to be processed is managed. (3) A method for engraving uranium, characterized by using interference light data during the etching of the sample, The thickness of the residual film of the etched portion of the sample is predicted, and the foregoing etching is terminated at a predetermined thickness of the residual film. (4) A uranium etching method, characterized in that: in damascene etching, the etching process is used According to the interference light data, the initial film thickness is predicted to determine the trench depth, and the aforementioned etching is terminated at a predetermined trench depth. (5) An etching method, characterized in that: In the method of the gate material formed on the underlayer films having different thicknesses, interference light is measured on a thick portion of the underlayer film, the residual film thickness of the gate material formed on the underlayer film with a thick thickness is measured, and the gate is managed. The thickness of the electrode material is 0 (6), and a method for engraving uranium, characterized in that, during the engraving of a material to be processed in which a plurality of films are laminated, interference light from the material to be processed is measured, and each of the foregoing films is switched The digital filter processes the data of the interference light, and each of the aforementioned films manages the film thickness. (7) An etching method characterized in that in the BARC engraving, the interference light measurement book measured by the material to be processed during etching is used. Paper size applies Chinese National Standard (CNS) A4 specification {210X297 mm) (Please read the precautions on the back before filling out this page) Thread 1 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economy -37- 517306 A7 _ B7_ 5 2. Description of the Invention (The thickness of the BARC film prevents the basal layer from being peeled off by the uranium engraved material. (8) A method of a semiconductor device, characterized in that in the manufacture of a semiconductor element having a sTI portion When performing the etching of P 01 y-si forming a part of the semiconductor element, the above-mentioned etching of P ο 1 y-S i is managed by the residual film thickness of Poly — Si formed on the STI portion. (Please (Please read the notes on the back before filling this page) Order the paper printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. The paper size applies to the Chinese National Standard (CNS) A4 (210 X 297 mm) -38-
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90121996A TW517306B (en) | 2001-09-05 | 2001-09-05 | Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90121996A TW517306B (en) | 2001-09-05 | 2001-09-05 | Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method |
Publications (1)
Publication Number | Publication Date |
---|---|
TW517306B true TW517306B (en) | 2003-01-11 |
Family
ID=27801513
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW90121996A TW517306B (en) | 2001-09-05 | 2001-09-05 | Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method |
Country Status (1)
Country | Link |
---|---|
TW (1) | TW517306B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI514465B (en) * | 2012-07-13 | 2015-12-21 | ||
CN106024616A (en) * | 2015-03-26 | 2016-10-12 | 株式会社日立高新技术 | Plasma processing apparatus and plasma processing method |
TWI729254B (en) * | 2017-01-10 | 2021-06-01 | 日商東京威力科創股份有限公司 | Substrate processing method and substrate processing device |
CN113611626A (en) * | 2021-08-04 | 2021-11-05 | 上海信及光子集成技术有限公司 | Method for detecting silicon groove etching depth on line |
-
2001
- 2001-09-05 TW TW90121996A patent/TW517306B/en not_active IP Right Cessation
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI514465B (en) * | 2012-07-13 | 2015-12-21 | ||
CN106024616A (en) * | 2015-03-26 | 2016-10-12 | 株式会社日立高新技术 | Plasma processing apparatus and plasma processing method |
TWI729254B (en) * | 2017-01-10 | 2021-06-01 | 日商東京威力科創股份有限公司 | Substrate processing method and substrate processing device |
CN113611626A (en) * | 2021-08-04 | 2021-11-05 | 上海信及光子集成技术有限公司 | Method for detecting silicon groove etching depth on line |
CN113611626B (en) * | 2021-08-04 | 2024-02-27 | 上海信及光子集成技术有限公司 | Method for detecting etching depth of silicon groove on line |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR100769607B1 (en) | Method and apparatus for processing semiconductor waper | |
US6903826B2 (en) | Method and apparatus for determining endpoint of semiconductor element fabricating process | |
US8088247B2 (en) | Plasma processing apparatus | |
JP3694662B2 (en) | Method and apparatus for measuring film throughput in semiconductor element manufacturing process, method and apparatus for processing material using the same, and method for determining end point of process using the same | |
TWI303090B (en) | Method for in-situ monitoring of patterned substrate processing using reflectometry | |
US20080216956A1 (en) | Plasma processing apparatus | |
TWI593020B (en) | Plasma processing apparatus and plasma processing method | |
JP6504915B2 (en) | Plasma processing apparatus and plasma processing method | |
JP3854810B2 (en) | Method and apparatus for measuring film thickness of material to be processed by emission spectroscopy, and method and apparatus for processing material using the same | |
JPH09283585A (en) | Device manufacturing method | |
JP4943716B2 (en) | Plasma processing equipment | |
JP2018152552A (en) | Measurement system and measurement method of complicated structure | |
Benson et al. | In-situ spectroscopic reflectometry for polycrystalline silicon thin film etch rate determination during reactive ion etchinc | |
JP5411215B2 (en) | Plasma processing equipment | |
JP2014232825A (en) | Plasma processing method | |
TW517306B (en) | Method and device to determine the end point of semiconductor device processing and the processing method and device of the processed material using the method | |
US20050117165A1 (en) | Semiconductor etching process control | |
JP2006119145A (en) | Method and device for processing semiconductor wafer | |
KR100733120B1 (en) | Method and apparatus for detecting processing of semiconductor waper | |
EP3338294B1 (en) | Single-wafer real-time etch rate and uniformity predictor for plasma etch processes | |
KR100438379B1 (en) | Method and apparatus for determining endpoint of semiconductor element fabricating process and method and apparatus for processing member to be processed | |
TW202349492A (en) | Plasma-processing method and plasma-processing device | |
JPH11238723A (en) | Method and apparatus for plasma processing | |
JP2004273872A (en) | Semiconductor manufacturing apparatus | |
JPH05259127A (en) | Method for etching monitoring in plasma etching |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GD4A | Issue of patent certificate for granted invention patent | ||
MK4A | Expiration of patent term of an invention patent |