200844648 九、發明說明： 【發明所屬之技術領域】 本發明係有關於在液晶顯示裝置（Liquid Crystal Display:以下稱爲LCD)之製造等所使用的灰階光罩之缺陷 修正方法、灰階光罩之製法及灰階光罩，尤其係有關於適 合用於製造在薄膜電晶體液晶顯示裝置的製造等所使用之 薄膜電晶體基板（TFT基板）的灰階光罩之缺陷修正方法、灰 階光罩之製法及灰階光罩。 【先前技術】 現在，在LCD領域，薄膜電晶體液晶顯示裝置（Thin Film Transistor Liquid Crystal Display:以下稱爲 TFT — LCD) 和CRT(陰極線管）相比，由於易製成薄型且耗電力低之優 點，正急速地進行商品化。TFT - LCD具有在液晶相之介入 下，將TFT排列於排成陣列狀之各像素的構造之TFT基 板，和對應於各像素，排列紅、綠以及藍之像素圖案的彩 色濾光片重疊之槪略構造。TFT - LCD之製造步驟數多，僅 TFT基板，便使用5〜6片光罩製造。在這種狀況下，提議 使用4片光罩製造TFT基板之方法（例如非專利文獻1:「月 刊 FPD Intelligence」，1 999 年 5 月，p.31— 35)。 此方法’係藉由使用具有遮光部、透光部以及半透光 部（灰階部）之光罩（以下稱爲灰階光罩），而減少所使用之光 罩片數的方法。在此’半透光部意指在使用光罩將圖案轉 印於被轉印體時，使透過之曝光光之透過量減少既定量， 以控制被轉印體上之光阻膜在顯像後的殘膜量之部分。將 和遮光部、透光部一起具備半透光部之光罩稱爲灰階光罩。 200844648 在第1(a)圖〜第1(c)圖及第2(a)圖〜第2(c)圖，表示使 用灰階光罩之TFT基板的製程之一例。第2(a)圖〜第2(c) 圖係表示第1(a)圖〜第1(c)之製程的後續。 在玻璃基板1上，形成閘極用金屬膜，並利用使用光 罩之光蝕刻製程形成閘極2。然後，依序形成閘極絕緣膜 3、第1半導體膜4(a— Si:非晶形矽）、第2半導體膜5(N + a 一 Si)、源極汲極用金屬膜6以及正型光阻膜7(第1(a)圖）。 接著，使用具有遮光部1 1、透光部1 2以及半透光部1 3之 φ 灰階光罩1 〇，藉由將正型光阻膜7進行曝光、顯像，而覆 蓋TFT通道部形成區域、源極/汲極形成區域、資料線形成 區域，而且以TFT通道部形成區域變成比源極/汲極形成區 域更薄的方式形成第1阻劑圖案7a(第1(b)圖）。 然後，將第1阻劑圖案7a作爲光罩，並將源極汲極 用金屬膜6、第2、第1半導體膜5、4進行飩刻（第1(c)圖）。 接著，利用由氧氣所產生之灰化（ashing)除去TFT通道部形 成區域的薄光阻膜，而形成第2阻劑圖案7b(第2(a)圖）。 φ 然後，將第2阻劑圖案7b作爲光罩，並將源極汲極用金屬 膜6進行蝕刻，而形成源極/汲極6a、6b，接著將第2半導 體膜5進行飩刻（第2(b)圖），而將最後所殘留的第2阻劑圖 案7b剝離（第2(c)圖）。 作爲在上述之製程所使用的灰階光罩，已知以微細圖 案形成半透光部之構造。灰階光罩例如如第3圖所示，具 有對應於源極/汲極之遮光部1 1 a、1 1 b、透光部1 2以及對 應於TFT通道部的半透光部（灰階部）13。半透光部13係形 成由使用灰階光罩之LCD用曝光機的解析度界限以下之微 200844648 細圖案所構成的遮光圖案13a。遮光部lla、lib和遮光圖 案1 3 a，一般都由鉻或鉻化合物等之相同的材料所構成之厚 度相冋的膜形成。使用灰階光罩之LCD用曝光機的解析度 界限，在大部分的情況，在步進方式之曝光機係約3 /z m， 而在鏡投影方式的曝光機係約4 /z m。因而，例如在第3圖 可將在半透光部1 3之透過部1 3b的間隔寬度設爲未滿3 μ m，並將遮光圖案13a之線寬設爲曝光機之解析度界限以下 的未滿3 // m。 上述之微細圖案型式的半透光部，在灰階部分之設 計，具體而言，在用以使具有遮光部和透光部之中間的灰 階效果之微細圖案具有採用線和間隔（line and space)型式 或網點（dot)型式、或者其他的圖案的選擇備案。此外，在 微細圖案採用線和間隔型式的情況，考慮將線寬設爲多 少，或將光透過之部分和遮光的部分之比率設爲多少，或 整體的透過率設計成多大等，而進行設計。 另一方面，提議將想進行灰階曝光之部分作成半透過 性的灰階膜（半透光膜）（例如專利文獻1:特開2002 — 1 89280 號公報）。能藉由使用此灰階膜，使灰階部分的曝光量變少 而進行灰階曝光。使用灰階膜的情況，在設計上檢討整體 之透過率需要多少，而在光罩上藉由選擇係灰階膜的膜種 (材料）或膜厚，而可生產光罩。在光罩製造進行灰階膜的 膜厚控制。在以灰階光罩之灰階部形成TFT通道部的情 況，若係灰階膜，因爲容易利用光微影步驟而產生圖案， 所以具有即使TFT通道部係複雜的圖案形狀，亦可形成之 優點。 200844648 又，在專利文獻2(特開2004 - 309515號公報）記載， 在具有遮光部、透光部以及灰階部之灰階光罩，於修正灰 階部的缺陷時’爲了使灰階部的膜變成可得到正常之灰階 效果之膜厚’而藉由使用FIB(Focused Ion Beam Deposition) 之飩刻來減少膜厚或形成膜。 【發明內容】 在如上述所示之專利文獻1記載的灰階光罩，無法避 免在由半透光膜所構成之灰階部發生缺陷。 Φ 另一方面，若依據該專利文獻2，在灰階部具有微細 圖案的灰階光罩，對在該微細圖案部分所產生之缺陷，可 比較易於進行修正度高的修正。即，因爲正常圖案係微細， 雖然在發生缺陷時難將該微細圖案復原成相同的形狀，但 是在專利文獻2，解決了如下之課題，例如在使用雷射CVD 裝置將遮光膜形成於白缺陷的方法、或除去黑缺陷部分並 重新形成遮光膜的方法，難進行用以得到適當之灰階效果 的透過率控制。 φ 此外，在此，將因爲膜圖案的過剩或遮光膜成分之附 著、或異物，而使透過率變成比既定値低的缺陷稱爲黑缺 陷，將因爲膜圖案的不足，而透過率變成比既定値高的缺 P曰稱爲白缺陷。 可是，即使如該專利文獻2使用FIB，亦未必易於對 缺陷部分實施如確保既定之曝光光之透過量的修正。例 如5在半透光部使用半透光膜並控制曝光光之透過量之型 式的灰階光罩，在發生該半透光膜之欠缺所引起的白缺陷 之情況，理想上，在決定變成所預定之光透過量的修正膜 200844648 的膜材料、膜厚以及用以將其進行成膜的條件等後，將修 正膜形成於該缺陷部分即可。另一方面，在黑缺陷之情況， 在除去黑缺陷部分的膜後，一樣地，將修正膜形成於該除 去部分即可。可是，若依據本發明者的檢討，實際上難進 行那樣的修正。 即，若依據本發明者的檢討，雖然FIB裝置在對局部 性之部位的成膜係有效之手段，但是即使用同一成膜材 料，並應用同一成膜條件（相當於每單位面積之劑量的電流 φ 値），亦若修正膜的成膜面積相異，可能產生修正膜之成膜 膜厚的變動（光透過率亦隨著變動）。例如，在修正膜之形 成區域存在大的部分和小的部分時，尺寸比較小之修正膜 的膜厚可能變成比尺寸比較大之修正膜的膜厚還大（變 厚）。 使用第4(a)圖〜第4(c)圖說明，設定在由形成於透明 基板24上之半透光膜26所構成的半透光部，因該半透光 膜的欠缺而發生小尺寸的白缺陷60和大尺寸的白缺陷 φ 61(第4(a)圖）。在此情況，若使用FIB裝置，並應用同一成 膜材料、同一成膜條件，形成大小各自和各缺陷的尺寸相 稱的修正膜28a、28b(第4(b)圖），則尺寸比較小之修正膜 28a的膜厚會成爲比尺寸比較大之修正膜28b的膜厚還厚 (參照第4(c)圖的剖面圖）。 此外，若依據本發明者之專心的檢討，發現如上述所 示之成膜面積所引起的修正膜厚變動之現象，係由於無法 避免在FIB裝置之成膜時的掃描速度和對成膜部位之成膜 材料的供給量之關係發生變動。又，本發明者發現對該變 -10- 200844648 動之影響要素係複雜，若成膜面積小，相對於掃描速度， 成膜材料的供給量過度地減少，若進一步地成膜面積變 小’反之成膜材料的供給量變成過度等，成膜面積和膜厚 之關係未必清楚地相關，因此，即使想要對具有某面積之 所要的部位進行成膜，實際上進行成膜之膜厚的控制及預 測亦不容易。順便地，第5圖係表示FIB裝置之膜厚的成 膜面積相依性之關係的一例，係表示將碳修正膜形成於所 要之部位的情況之成膜面積和膜厚變動的相關聯之一例。 φ 雖然亦有藉由 FIB裝置之條件變更而消除這種變動 的方法，但是係成膜對象之灰階光罩的半透光部之修正膜 所需的膜厚變動（即光透過率變動）之容許値係極嚴，僅靠 FIB裝置的參數調整，經常無法得到所要値之膜厚。 又，以往之缺陷修正方法的問題點係不僅如上述所示 之成膜面積所引起的膜厚變動之問題。實際之白缺陷未限 定爲截面欠缺成在膜厚方向係大致垂直的形狀。例如如第 6(a)圖所示，設定在半透光膜26發生截面欠缺成在朝向透 0 明基板24側變窄的形狀。在此情況，若在該缺陷部位形成 膜厚均勻的修正膜28c(第6(b)圖），因爲將修正膜形成於部 分殘留於白缺陷之部分的半透光膜26之上，所以發生在半 透光膜和修正膜重疊之區域的光透過量變成比所要値小之 問題（參照第6(c)圖的剖面圖）。此外，符號25係遮光膜。 又，如第7(a)圖所示，在因爲遮光膜成分或異物附著 於半透光膜26上而發生黑缺陷63的情況，難僅除去黑缺 陷63部分，並對半透光膜26無影響。例如以FIB裝置僅 除去黑缺陷63部分的情況’若除去的方法不充分，黑缺陷 -11- 200844648 成分就殘留。若想要完全地除去黑缺陷，就連其下面之半 透光膜的一部分都除去，而變成在半透光膜26發生新的缺 陷64(參照第7(b)、（c)圖）。此外，若想以修正膜修正該新 的缺陷64,發生和在上述之第6(a)圖〜第6(c)圖所說明的一 樣的問題。 本發明係鑑於上述之問題點而開發者，其第1目的在 於提供可適當地修正在半透光部所發生的缺陷之灰階光罩 的缺陷修正方法。 φ 本發明之第2目的在於提供具有應用上述之缺陷修 正方法的缺陷修正步驟之灰階光罩的製法。 本發明之第3目的在於提供經適當地修正在半透光 部所發生的缺陷之灰階光罩。 爲了解決上述之課題，本發明具有以下之構造。 (構造1) 一種灰階光罩之缺陷修正方法，灰階光罩係藉由將半 透光膜和遮光膜形成於透明基板上，並施加既定的圖案 | 化，而具有遮光部、透光部以及將使用光罩的所使用的曝 光光之透過量減少既定量之半透光部，用以將膜厚階段或 連續地相異之阻劑圖案形成於被轉印體上之灰階光罩的缺 陷修正方法，其特徵爲：該半透光部係利用該半透光膜形 成；在該半透光部發生缺陷時特定該缺陷部分；決定用以 將修正膜形成於該經特定之缺陷部分的成膜手段和成膜材 料；在應用該經決定之成膜手段和成膜材料時，決定使該 決 經 該 成 形 積 面 膜 成 的 內 圍 範。 定膜 既正 爲修 成該 量的 過積 透面 之膜 光成 光之 曝定 -12- 200844648 (構造2) 在構造1所記載之灰階光罩的缺陷修正方法，亦可作 成該成膜面積，係在應用該經決定之成膜手段和成膜材料 時，設定使該曝光光之透過量成爲既定範圍內的成膜膜 厚，並將該經設定之成膜膜厚應用於預先求得之成膜膜厚 和成膜面積的相關關係而決定。 - (構造3) 在構造1或2所記載之灰階光罩的缺陷修正方法，較 φ 佳爲在該修正膜的形成之前具有一個步驟，其對於包含有 該缺陷部分之和該經決定的成膜面積大致相等之面積的區 域，使該透明基板露出。 (構造4) 在構造1至3的任一種所記載之灰階光罩的缺陷修正 方法，該缺陷部分和正常之半透光部相比，因爲具有半透 光膜之膜厚小或欠缺半透光膜的部位，所以係曝光光之透 過量比該正常之半透光部大的部分。 φ (構造5) 在構造1至3的任一種所記載之灰階光罩的缺陷修正 方法，該缺陷部分，因爲在半透光部附著有半透光膜以外 之成分，所以係曝光光之透過量比正常之半透光部小的部 分。 ^ (構造6) 在構造1至5的任一種所記載之灰階光罩的缺陷修正 方法，較佳爲在半透光部發生複數個缺陷部分時，對該複 數個缺陷部分，各自形成大致相同之成膜面積的修正膜。 -13- 200844648 (構造7) 在構造1至6的任一種所記載之灰階光罩的缺陷修正 方法，亦可作成對該經決定之成膜面積的整數倍之區域， 將形成該經決定之成膜面積的修正膜之步驟，重複地進行 僅該整數倍的次數。 (構造8) 在構造1至7的任一種所記載之灰階光罩的缺陷修正 方法，該修正膜之成膜手段係應用聚焦離子束法。 φ (構造9) 本發明之灰階光罩的製法，其特徵爲包含有藉由構造 1至8的任一種所記載之缺陷修正方法的缺陷修正步驟。 (構造10) 本發明之灰階光罩，其藉由將半透光膜和遮光膜形成 於透明基板上，並施加既定的圖案化，而具有遮光部、透 光部以及將使用光罩時所使用的曝光光之透過量減少既定 量之半透光部，並用以將膜厚階段或連續地相異之阻劑圖 φ 案形成於被轉印體上，該灰階光罩之特徵爲：將複數個大 致固定面積或其整數倍之面積的修正膜形成於該半透光 部。 (構造1 1) 在構造1 0所記載之灰階光罩，亦可預先形成於該半 透光部之半透光膜和該修正膜具有相異的組成。 若依據本發明之灰階光罩的缺陷修正方法，因爲可將 半透光部所要求之曝光光之透過量設定於所要的範圍，並 可高再現性地形成變成該所要之曝光光之透過量的成膜面 -14- 200844648 積之修正膜，所以可實施在半透光部之曝光光之透過量的 精度高、安定的修正。結果，已修正缺陷的區域變成可得 到和在半透光部之正常的灰階部分同等之灰階效果，並可 適當地修正在半透光部所發生的缺陷。又，在半透光部所 發生之缺陷的修正膜之安定性、控制性提高，而灰階光罩 的良率大幅度地提高。 又，若依據本發明之灰階光罩的製法，藉由具有應用 這種本發明之缺陷修正方法的缺陷修正步驟，而可得到已 Φ 適當地修正在半透光部所發生之缺陷的灰階光罩。 又，若依據本發明之灰階光罩，在半透光部，形成複 數個大致同一面積的修正膜，並對複數個缺陷部分，高重 現性地形成變成所要之曝光光之透過量的成膜面積之修正 膜’而可得到已實施在半透光部之曝光光之透過量的精度 高、安定之適當的修正之灰階光罩。 【實施方式】 以下，根據圖面說明用以實施本發明之最佳形態。 φ [第1實施形態] 第8(a)圖〜第8(d)圖係表示本發明之灰階光罩的缺陷 修正方法之第1實施形態，第8(a)圖〜第8(c)圖各自係用以 按照步驟順序說明缺陷修正方法的立體圖，第8 (d)圖係沿 著在第8 (c)圖之L 一 L線的側剖面圖。又，第9圖係用以說 明使用本發明之灰階光罩的圖案轉印方法之剖面圖。此 外，在第9圖未表示已修正的缺陷部分。 第9圖所示之本發明的灰階光罩20,係用以製造例如 液晶顯示裝置（LCD)之薄膜電晶體（TFT)或彩色濾光片、或 -15- 200844648 者電漿顯示面板（PDP)等，並係在被轉印體30上形成膜厚 階段或連續地相異之阻劑圖案3 3。此外，在第9圖，符號 32A、32B係表示在被轉印體30疊層於基板31上的膜。 具體而言，灰階光罩20具有：遮光部21，係在使用 該灰階光罩20時將曝光光遮光（透過率約0%);透光部22， 係藉由透明基板24的表面露出而使曝光光約100%透過； 以及半透光部23，係使曝光光之透過率降低至約20〜60%。 半透光部23係將光半透過性之半透光膜26形成於玻璃基 板等之透明基板24上而構成。又，遮光部2 1係將該半透 光膜26及遮光性的遮光膜25依序形成於透明基板24上而 構成。此外，此遮光部2 1有因應於光罩製造所使用之光罩 半成品（mask blank)的構造及製程，將遮光膜25、半透光膜 26依序形成於透明基板24上的情況，或僅將遮光膜25形 成於透明基板24上的情況。又，第9圖所示之遮光部2 1、 透光部22以及半透光部23之圖案形狀完全是代表例，當 然不是將本發明限定爲此的主旨。 作爲半透光膜26，可舉出鉻化合物、MoSi、Si、W、 A卜其中，在鉻化合物，有氧化鉻（Cr〇x) '氮化鉻（CrNx)、 氮氧化鉻（CrOxN)、氟化鉻（CrFx)、或這些包含有碳或氫者。 又，作爲遮光膜25，可舉出Cr、Si、W' A1等。遮光部21 的透過率係藉由遮光膜25 (或遮光膜25和半透光膜26)之膜 材質和膜厚的選定而設定。又，半透光部23的透過率係藉 由半透光膜26之膜材質和膜厚的選定而設定。 在使用如上述所示之灰階光罩20時，因爲在遮光部 21，曝光光實質上無法透過，在半透光部23,曝光光減少’ -16 - 200844648 所以塗布於被轉印體30上之光阻膜（正型光阻膜）， 於遮光部2 1之部分，膜厚變厚，而在對應於半透： 的部分，膜厚變薄，在對應於透光部22之部分形成 阻劑圖案33 (參照第9圖）。在該阻劑圖案33，將在 半透光部23的部分，膜厚變薄之效果稱爲灰階效 外，在使用負型光阻的情況，雖然需要進行考慮到 部及透光部對應之光阻膜厚相反的設計，但是在 況，亦可充分地得到本發明之效果。 而，在第9圖所示之阻劑圖案33的無膜之部 在被轉印體30之例如膜32A及32B實施第1蝕刻， 用灰化等除去阻劑圖案3 3之膜的薄部分之此部分， 轉印體30之例如膜32B實施第2飩刻。依此方式 片灰階光罩20實施以往之2片光罩分量的步驟，而 罩片數。 其次，說明本實施形態之缺陷修正方法。在：Φ 態，使用TFT基板製造用的灰階光罩，其係在透明^ 依序將包含有矽化鉬（Mo Si)之半透光膜（曝光光 5 0%)、以鉻（CO爲主成分的遮光膜進行成膜，並1 之圖案化，而藉以具備有遮光部、透光部以及半透 在本實施形態，說明在該半透光部所產生之 修正方法。 (1)關於所製造之灰階光罩，使用缺陷檢查 行光罩圖案的缺陷檢查。而，在半透光部存在缺 定該缺陷部分的位置資訊和形狀資訊。此情況之 爲相對於正常之半透光部，半透光膜的膜厚小， 在對應 3 2 3 無膜的 對應於 果。此 和遮光 這種情 分，對 而在利 對在被 ，使用1 減少光 實施形 板上， 透過率 加既定 光部。 缺陷的 置，進 時，特 陷，因 t具有欠 -17- 200844648 缺半透光膜的部位，所以係如曝光光之透過量比正常之半 透光部大的部分之所謂的白缺陷。 缺陷檢查的結果如第8(a)圖所示，在由半透光膜26 所構成之半透光部中，存在有小尺寸的白缺陷5 0及大尺寸 的白缺陷5 1。又，雖然實際上發生於光罩的缺陷多爲不規 則形狀者，但爲了方便起見在此表示爲矩形狀。 (2)其次，決定用以在上述經特定之缺陷部分形成修 正膜的成膜手段和成膜材料。在本實施形態，應用FIB， • 作爲成膜手段。又，成膜材料採用適合藉FIB之成膜的碳。 當然，未限定爲碳，亦可使用和半透光膜一樣之包含有矽 化鉬的材料。 在此，說明FIB裝置。FIB裝置不僅成膜，亦可用於 膜之除去。 如第10圖所示，FIB裝置40具有：離子源41，係產 生Ga +離子；電磁光學系統42 ;電子槍43，係放出用以將 Ga +離子中和的電子；蝕刻用氣體槍49，係放出々氣體； φ 以及氣體槍44，係放出芘氣體。電磁光學系統42係將從離 子源4 1所產生之G a +離子作爲離子束4 7，並利用掃瞄放大 器46掃瞄該離子束47。 而，在XY工作台45上，放置係被修正對象物的灰 階光罩20，藉由使XY工作台45移動，而將在灰階光罩20 施加修正的缺陷區域移至離子束照射區域。接著，使離子 束47掃瞄施加修正的缺陷區域，藉由檢測此時所產生之二 次離子的二次離子檢測器4 8的作用，而檢測施加修正之缺 陷區域的位置。藉由離子束47經由電磁光學系統42，照射 -18- 200844648 施加灰階光罩2 0之修正的缺陷區域，而實施修正膜的形成 或膜的除去（例如黑缺陷區域之半透光膜的除去）。此外， 離子束之束徑係0.1 μ πιφ以下。 在形成修正膜的情況，一面經由電磁光學系統42放 出離子束47, 一面利用氣體槍44放出芘氣體。因而，芘氣 體接觸離子束47而聚合（化學反應），修正膜堆積於離子束 47的照射區域並進行成膜。 又，例如在除去黑缺陷區域之半透光膜的情況，利用 φ 蝕刻用氣體槍4 9放出/3氣體，藉由在此狀態經由電磁光學 系統42照射離子束47，而除去該半透光膜。 (3) 接著，在應用上述經決定之成膜手段和成膜材料 時’決定可得到既定之曝光光之透過量之修正膜的成膜面 積。作爲既定之曝光光之透過量，在此設爲40%的透過率， 而作爲滿足該透過率之碳膜的膜厚，設定45 urn。爲了得到 這種修正膜，要使用FIB裝置並以既定之成膜條件進行成 膜，根據在預先求得之和上述一樣的成膜條件之成膜膜厚 φ 和成膜面積的相關關係，決定將成膜面積設爲400 /z m2係 可最安定地進行成膜。如此經決定之成膜面積係包含有位 於半透光部之該白缺陷50、5 1的大小。在此，此成膜膜厚 和成膜面積的相關關係例如係如上述之第5圖所示的相關 關係。 (4) 接著，對於各自包含有位於半透光部之該白缺陷 50、5 1並比該白缺陷大的區域，除去相當於在上述（3)經決 定之面積（成膜面積）（相等或大致相等）的半透光膜26。作 爲半透光膜26之除去手段，雖然使用FIB裝置，但是亦可 -19- 200844648 使用其他之例如雷射裝置。結果，如第8 (b)圖所示，對大 小相異之白缺陷5 0、5 1，僅除去同一大小及形狀，即同一 面積的半透先膜26(26a、26b)’在所除去之部分26a、26b， 透明基板24露出。 (5)將已除去該半透光膜之部分26a、26b作爲修正膜 的成膜區域，並將所需之位置資訊等輸入FIB裝置，而且 輸入成膜膜厚或其他的成膜條件，在已除去該半透光膜之 部分26a、26b，形成同一大小及形狀（即同一面積）的修正 φ 膜27a、27b(參照第8(c)圖、第8(d)圖）。所形成之修正膜 27a、27b的膜厚，在使用AFM(原子間力顯微鏡）量測時， 最大高低差爲1.26nm，在面內係定値，而修正膜27a和27b 的膜厚無變動。因此，形成可得到預先所設定之所要的曝 光光之透過量之修正膜27a、27b。 此外，在第8(d)圖，雖然將半透光膜26和修正膜27a、 27b之厚度畫成大致相同，但是因爲只要控制成半透光部具 有既定之曝光光之透過量即可，所以在半透光膜26和修正 φ 膜27a、27b之膜材料相異的情況，亦有膜厚相異的情況。 若依據以上所說明之第1實施形態，可得到如下的效 果。 1. 因爲可將半透光部所要求之曝光光之透過量設定於所要 的範圍，並可高再現性地形成變成該所要之曝光光之透過 量的成膜面積之修正膜，所以可實施在半透光部之曝光光 之透過量的精度高並安定之修正。 2. 因此，已修正缺陷的區域變成可得到和在半透光部之正 常的灰階部分同等的灰階效果，而可適當地修正在半透光 -20- 200844648 部所產生之缺陷。· 3.又，在半透光部所產生之缺陷的修正膜之安定性、控制 性提高，灰階光罩之良率大幅度提高。 在本實施形態，爲了對複數個缺陷部分形成同一成膜 面積的修正膜，雖然形成包含有缺陷部分之同一大小、形 狀（矩形）的修正膜，但是形狀未限定爲矩形，亦可係例如 圓形之其他的形狀。又，只要係同一成膜面積，亦可複數 個修正膜的形狀彼此相異。此外，即使複數個修正膜之面 φ 積嚴格上不是相同，只要未發生成膜面積所引起的膜厚變 動，亦可稍微相異。 又，在本實施形態，作爲修正膜之成膜手段，雖然應 用 FIB，但是成膜手段未限定爲FIB，例如亦可應用雷射 CVD等其他的成膜手段。 [第2實施形態] 第1 1(a)圖〜第1 1(d)圖係表示本發明之灰階光罩的缺 陷修正方法之第2實施形態，第11(a)圖〜第11(c)圖各自係 φ 用以按照步驟順序說明缺陷修正方法的立體圖，第11(d) 圖係沿著在第11(c)圖之L一 L線的側剖面圖。 雖然在第6(a)圖〜第6(c)圖亦說明，但是實際之白缺 陷未限定爲欠缺成截面在膜厚方向係大致垂直的形狀。例 如如第1 1(a)圖所示，在半透光膜26,有發生欠缺成截面向 透明基板24側變窄之硏鉢形的白缺陷52之情況。在此情 況，若在該缺陷部位形成膜厚均勻的修正膜，因爲在部分 殘留於白缺陷部分的半透光膜26之上形成修正膜，所以在 半透光膜之一部分和修正膜重疊的區域，曝光光之透過量 -21- 200844648 變成比所要値小。這種形狀以外之複雜的形狀之白缺陷的 情況亦一樣。 在本第2實施形態，亦在應用和上述之第1實施形態 一樣地決定之成膜手段和成膜材料時，決定變成既定之曝 光光之透過量的修正膜之成膜面積。然後，對於包含有位 於半透光部之該白缺陷52，並比該白缺陷大的區域，除去 相當於在上述經決定之面積（成膜面積）的半透光膜26(參 照第1 1(b)圖）。 φ 在已除去該半透光膜的部分26c，形成在上述經決定 之成膜面積的修正膜27c(參照第11(c)圖、第11(d)圖）。因 而，形成可得到預先所設定之所要的曝光光之透過量之具 有均勻的膜厚之修正膜。 因此，在第2實施形態，亦可同樣地得到上述之第1 實施形態的1〜3之效果。 [第3實施形態] 第12(a)圖、第12(b)圖係表示本發明之灰階光罩的缺 φ 陷修正方法之第3實施形態，係用以按照步驟順序說明缺 陷修正方法的立體圖。 在產生於灰階光罩之半透光部的自缺陷爲超過固定 面積，例如與前述第1實施形態一樣地進行而決定之修正 膜的成膜面積之大小的情況下，將此一定面積的整數倍區 域設定爲缺陷修正區域。 首先，在第3實施形態，亦在應用和上述之第1實施 形態一樣地決定之成膜手段和成膜材料時，決定變成既定 之曝光光之透過量的修正膜之成膜面積。 -22- 200844648 然後，在半透光部所產生之白缺陷係超過在上述所決 定的成膜面積之大小的情況，對於包含有位於半透光部之 該白缺陷，並比該白缺陷大的區域，除去相當於在上述經 決定之面積（成膜面積）的整數位（例如2倍）之半透光膜 26(參照第12(a)圖）。 對已除去該半透光膜的部分26d，將形成在上述經決 定之成膜面積的修正膜之步驟，以所形成的修正膜相鄰之 方式重複地進行僅該整數倍的次數。因而，在已除去該半 φ 透光膜之部分26d，形成可得到預先所設定之所要的曝光光 之透過量之具有均勻的膜厚之修正膜27d(參照第12(b)圖）。 因此，利用第3實施形態，亦可一樣地得到上述之第 1實施形態的1〜3之效果。 [第4實施形態] 第13U)圖〜第13(c)圖係表示本發明之灰階光罩的缺 陷修正方法之第4實施形態，係用以按照步驟順序說明缺 陷修正方法的立體圖。 φ 在包含有白缺陷之半透光部的面積小，並接近遮光膜 2 5的情況’亦可將固定之面積（例如和上述之第1實施形態 一樣地決定之修正膜的成膜面積）作爲缺陷修正區域，並有 效地應用本發明。 即’在第4實施形態，亦在應用和上述之第1實施形 態一樣地決定之成膜手段和成膜材料時，決定變成既定之 曝光光之透過量的修正膜之成膜面積。 然後，如第1 3 (a)圖所示，因爲發生白缺陷5 3之半透 光部的面積和周圍之遮光膜2 5接近，而未達到在上述所決 •23- 200844648 定的成膜面積之情況，對於包含有位於半透光部之該白缺 陷53,無遮光膜25之露出的半透光膜區域，除去半透光膜 26(參照第13(b)圖）。 形成包含有已除去該半透光膜的部分26e之在上述所 決定的成膜面積之修正膜27e。因而，在已除去該半透光膜 的部分26e，形成可得到預先所設定之所要的曝光光之透過 量之具有均句的膜厚之修正膜27e(參照第13(c)圖）。此外， 因爲在上述經決定之成膜面積（修正膜之實際的成膜面積） Φ 比發生白缺陷53之半透光部的面積還大，雖然在遮光膜25 上亦形成修正膜27e的一部分，但是因爲係遮光部，所以 不會產生不良。亦可後來除去遮光膜25上之不要的遮光膜 27e ° 因此，利用第4實施形態，亦可一樣地得到上述之第 1實施形態的1〜3之效果。 [第5實施形態] 第14(a)圖〜第14(c)圖係表示本發明之灰階光罩的缺 φ 陷修正方法之第5實施形態，係用以按照步驟順序說明缺 陷修正方法的立體圖。 例如，在製程欠缺微小之半透光膜的情況等，係透明 基板上孤立的半透光部並包含有白缺陷之半透光膜的面積 小，而未達到固定之面積（例如和上述之第1實施形態一樣 地決定之修正膜的成膜面積）的情況，即使本來僅對小區域 進行成膜即可，亦可將該固定面積作爲缺陷修正區域，並 有效地應用本發明。 即，在第5實施形態，亦在應用和上述之第1實施形 -24- 200844648 態一樣地決定之成膜手段和成膜材料時，決定變成既定之 曝光光之透過量的修正膜之成膜面積。 然後，如第14(a)圖所示，成爲白缺陷54之半透光部 的面積未達到在上述所決定的成膜面積之情況，對於包含 有該白缺陷54之透明基板24上的區域，形成在上述經決 疋之成膜面積的修正膜27f(篸照第14(b)圖）。因而，在包 含有該自缺陷5 4之區域，形成可得到預先所設定之所要的 曝光光之透過量之具有均勻的膜厚之修正膜27f。在修正膜 鲁 27f之成膜後’可除去不要之區域55的修正膜。 因此，在第5實施形態，亦可一樣地得到上述之第i 實施形態的1〜3之效果。 [第6實施形態] 第15(a)圖〜第15(c)圖係表示本發明之灰階光罩的缺 陷修正方法之第6實施形態，係用以按照步驟順序說明缺 陷修正方法的立體圖。 亦如在上述之第7(a)圖〜第7(c)圖的說明所示，因遮 φ 光膜成分或異物附著於半透光膜上而發生黑缺陷時，要僅 除去黑缺陷部分，並對半透光膜無影響，這在習知方法係 困難。如以下之說明所示，對這種黑缺陷的修正亦可有效 地應用本發明。 首先，在第6實施形態，亦在應用和上述之第1實施 形態一樣地決定之成膜手段和成膜材料時’決定變成既定 之曝光光之透過量的修正膜之成膜面積。 然後，對於如第1 5 (a)圖所示之包含有在半透光部所 產生之黑缺陷5 6，並比該黑缺陷區域5 6大的區域’將相當 -25- 200844648 於在上述經決定之面積（成膜面積）的半透光膜26和黑缺陷 56 —起除去（參照第16(b)圖）。 在已除去該半透光膜的部分26g，形成在上述經決定 之成膜面積的修正膜27g(參照第15(c)圖）。因而，除去在 半透光部所產生之黑缺陷56，並在已除去該半透光膜之部 分26g，形成可得到預先所設定之所要的曝光光之透過量之 • 具有均勻的膜厚之修正膜27g。 因此，在和黑缺陷之修正相關的第6實施形態，亦可 φ 一樣地得到上述之第1實施形態的1〜3之效果。 【圖式簡單說明】 第1(a)圖〜第1(c)圖係表示使用灰階光罩的TFT基板 之根據相關技術的製程之示意剖面圖。 第2(a)圖〜第2(c)圖係表示第1(a)圖〜第1(c)所示之製 程的後續之示意剖面圖。 第3圖係表示微細圖案型式之灰階光罩的一例之平 面圖。 φ 第4(a)圖〜第4(c)圖係用以說明以往之缺陷修正方法 的問題點之圖，第4(a)圖和第4(b)圖係立體圖，第4(c)圖 係剖面圖。 第5圖係表示使用FIB裝置進行成膜之情況的成膜面 積和膜厚之關係的圖。 第6(a)圖〜第6(c)圖係用以說明以往之缺陷修正方法 的問題點之圖，第6(a)圖和第6(b)圖係立體圖，第6(c)圖 係剖面圖。 第7(a)圖〜第7(c)圖係用以說明以往之缺陷修正方法. -26- 200844648 的問題點之圖，第7(a)圖和第7(b)圖係立體圖，第7(c)圖 係剖面圖。 第8(a)圖〜第8(d)圖係表示本發明之灰階光罩的缺陷 修正方法之第1實施形態，第8(a)圖〜第8(c)圖各自係用以 按照步驟順序說明缺陷修正方法的立體圖，第8(d)圖係沿 著在第8(c)圖之L一 L線的側剖面圖。 第9圖係用以說明使用本發明之灰階光罩的圖案轉 印方法之剖面圖。 • 第10圖係表示在本發明所使用之FIB裝置的構造之 示意側視圖。 第11(a)圖〜第11(d)圖係表示本發明之灰階光罩的缺 陷修正方法之第2實施形態，第11(a)圖〜第11(c)圖各自係 用以按照步驟順序說明缺陷修正方法的立體圖，第1丨（d) 圖係沿著在第1 1 (c)圖之L 一 L線的側剖面圖。 第12(a)圖、第12(b)圖係表示本發明之灰階光罩的缺 陷修正方法之第3實施形態，係用以按照步驟順序說明缺 φ 陷修正方法的立體圖。 第13(a)圖〜第13(c)圖係表示本發明之灰階光罩的缺 陷修正方法之第4實施形態’係用以按照步驟順序說明缺 陷修正方法的立體圖。 第14(a)圖〜第14(c)圖係表示本發明之灰階光罩的缺 陷修正方法之第5實施形態，係用以按照步驟順序說明缺 陷修正方法的立體圖。 第15(a)圖〜第15(c)圖係表示本發明之灰階光罩的缺 陷修正方法之第6實施形態’係用以按照步驟順序說明缺 27- 200844648 陷修正方法的立體圖。 【主要元件符號說明】 1 玻璃基板 2 閘極[Technical Field] The present invention relates to a method for correcting a defect of a gray scale mask used in the manufacture of a liquid crystal display device (hereinafter referred to as LCD), and gray scale light. A mask manufacturing method and a gray scale mask, in particular, a method for correcting a defect of a gray scale mask suitable for manufacturing a thin film transistor substrate (TFT substrate) used in the manufacture of a thin film transistor liquid crystal display device, and a gray scale The method of making the mask and the gray scale mask. [Prior Art] Now, in the field of LCDs, Thin Film Transistor Liquid Crystal Display (hereinafter referred to as TFT-LCD) is easier to be thin and consumes less power than CRT (Cathode Wireline). Advantages are rapidly becoming commercialized. The TFT-LCD has a TFT substrate in which TFTs are arranged in arrays of pixels in the form of a liquid crystal phase, and color filters in which pixel patterns of red, green, and blue are arranged corresponding to respective pixels are overlapped. Strategic structure. The number of manufacturing steps of the TFT-LCD is large, and only a TFT substrate is used, and a 5 to 6 photomask is used. Under such circumstances, a method of manufacturing a TFT substrate using four masks is proposed (for example, Non-Patent Document 1: "Flip Intelligence", May 999, p. 31—35). This method is a method of reducing the number of masks used by using a photomask having a light-shielding portion, a light-transmitting portion, and a semi-transmissive portion (gray-scale portion) (hereinafter referred to as a gray scale mask). Here, the 'semi-transmissive portion means that when the pattern is transferred to the object to be transferred by using the photomask, the amount of transmission of the transmitted exposure light is reduced by a predetermined amount to control the development of the photoresist film on the transfer target. Part of the amount of residual film after. A mask having a semi-transmissive portion together with a light-shielding portion and a light-transmitting portion is referred to as a gray scale mask. 200844648 An example of a process for using a TFT substrate using a gray scale mask is shown in Figs. 1(a) to 1(c) and Figs. 2(a) to 2(c). Fig. 2(a) to Fig. 2(c) show the subsequent steps of the processes of the first (a) to the first (c). On the glass substrate 1, a metal film for a gate electrode is formed, and a gate electrode 2 is formed by a photolithography process using a photomask. Then, the gate insulating film 3, the first semiconductor film 4 (a-Si: amorphous germanium), the second semiconductor film 5 (N + a-Si), the source drain metal film 6, and the positive type are sequentially formed. Photoresist film 7 (Fig. 1(a)). Next, the φ gray scale mask 1 having the light shielding portion 1 1 , the light transmitting portion 1 2 and the semi-light transmitting portion 13 is used to cover the TFT channel portion by exposing and developing the positive resist film 7 The formation region, the source/drain formation region, and the data line formation region, and forming the first resist pattern 7a such that the TFT channel portion formation region becomes thinner than the source/drain formation region (Fig. 1(b) ). Then, the first resist pattern 7a is used as a mask, and the source drain metal film 6, the second and first semiconductor films 5 and 4 are etched (Fig. 1(c)). Next, the thin resist film of the TFT channel portion forming region is removed by ashing by oxygen to form the second resist pattern 7b (Fig. 2(a)). φ Then, the second resist pattern 7b is used as a mask, and the source drain + metal film 6 is etched to form source/drain electrodes 6a and 6b, and then the second semiconductor film 5 is etched (first) 2(b))), and the second resist pattern 7b remaining last is peeled off (Fig. 2(c)). As the gray scale mask used in the above process, a structure in which a semi-transmissive portion is formed in a fine pattern is known. The gray scale mask has, as shown in FIG. 3, a light shielding portion 1 1 a, 1 1 b corresponding to the source/drain, a light transmitting portion 12, and a semi-transmissive portion corresponding to the TFT channel portion (gray scale) Department) 13. The semi-transmissive portion 13 is formed with a light-shielding pattern 13a composed of a micro-200844648 fine pattern of a resolution limit of an LCD exposure machine using a gray scale mask. The light-shielding portions 11a, 11b and the light-shielding pattern 13a are generally formed of a film having a thickness corresponding to the same material such as chromium or a chromium compound. The resolution limit of the exposure machine for the LCD using a gray scale mask is, in most cases, about 3 /z m in the stepwise exposure system and about 4 /z m in the mirror projection mode. Therefore, for example, in FIG. 3, the interval width of the transmissive portion 13b of the semi-transmissive portion 13 can be set to less than 3 μm, and the line width of the light-shielding pattern 13a can be set to be equal to or less than the resolution limit of the exposure machine. Less than 3 // m. The semi-transmissive portion of the fine pattern type described above is designed in the gray-scale portion, specifically, the fine pattern for the gray-scale effect having the light-shielding portion and the light-transmitting portion has a line and a space (line and Space) Type or dot type, or other pattern selection. Further, in the case where the fine pattern is in the line and space type, the design is considered as to what the line width is, or the ratio of the portion through which the light is transmitted and the portion where the light is blocked, or the overall transmittance is designed. . On the other hand, it is proposed to form a half-transmissive gray-scale film (semi-transmissive film) in which the gray-scale exposure is desired (for example, Patent Document 1: JP-A-2002-108280). Gray scale exposure can be performed by using this gray scale film to reduce the exposure amount of the gray scale portion. In the case of using a gray scale film, it is necessary to review the overall transmittance of the design, and a mask can be produced by selecting a film type (material) or film thickness of the gray scale film on the photomask. The film thickness control of the gray scale film is performed in the reticle manufacturing. In the case where the TFT channel portion is formed by the gray scale portion of the gray scale mask, if the gray scale film is formed, since the pattern is easily formed by the photolithography step, it is possible to form the pattern even if the TFT channel portion is complicated. advantage. In the case of a gray scale mask having a light-shielding portion, a light-transmitting portion, and a gray-scale portion, in order to correct a defect in the gray-scale portion, in order to make the gray-scale portion, the method of the present invention is disclosed in Japanese Laid-Open Patent Publication No. 2004-309515. The film becomes a film thickness that can obtain a normal gray-scale effect, and the film thickness is reduced or formed by using a FIB (Focused Ion Beam Deposition). According to the gray scale mask described in Patent Document 1, as described above, it is inevitable that defects occur in the gray scale portion composed of the semi-transmissive film. Φ On the other hand, according to Patent Document 2, a gray scale mask having a fine pattern in the gray scale portion can relatively easily correct the defects generated in the fine pattern portion. In other words, since the normal pattern is fine, it is difficult to restore the fine pattern to the same shape when a defect occurs. However, Patent Document 2 solves the problem that, for example, a light-shielding film is formed on a white defect using a laser CVD apparatus. The method, or the method of removing the black defect portion and reforming the light shielding film, makes it difficult to perform transmittance control for obtaining an appropriate gray scale effect. φ In addition, a defect in which the transmittance is lower than a predetermined value due to excessive film pattern or adhesion of a light-shielding film component or foreign matter is referred to as a black defect, and the transmittance is changed due to the shortage of the film pattern. The lack of P, which is determined to be high, is called white defect. However, even if FIB is used as in Patent Document 2, it is not always easy to perform correction such as ensuring a predetermined amount of exposure light to the defective portion. For example, in the case of a gray scale mask in which a semi-transmissive film is used in a semi-transmissive portion and a permeation amount of exposure light is controlled, in the case where a white defect caused by the lack of the semi-transmissive film occurs, it is desirable to become a decision. The film material of the correction film 200844648 having a predetermined light transmission amount, the film thickness, the conditions for forming the film, and the like may be formed on the defective portion. On the other hand, in the case of the black defect, after the film of the black defect portion is removed, the correction film may be formed in the removed portion. However, according to the review by the inventors, it is actually difficult to carry out such correction. That is, according to the review by the present inventors, although the FIB device is effective as a means for forming a film at a localized portion, the same film forming material is used, and the same film forming condition (equivalent to a dose per unit area) is applied. The current φ 値), if the film formation area of the correction film is different, may cause variations in the film thickness of the correction film (the light transmittance also varies). For example, when there is a large portion and a small portion in the formation region of the correction film, the film thickness of the correction film having a relatively small size may become larger (thicker) than the film thickness of the correction film having a larger size. As will be described with reference to FIGS. 4(a) to 4(c), the semi-transmissive portion formed by the semi-transmissive film 26 formed on the transparent substrate 24 is small due to the lack of the semi-transmissive film. The white defect 60 of the size and the white defect φ 61 of the large size (Fig. 4(a)). In this case, if the FIB device is used and the same film forming material and the same film forming conditions are applied, and the correction films 28a and 28b each having a size commensurate with the size of each defect are formed (Fig. 4(b)), the size is relatively small. The film thickness of the correction film 28a is thicker than the film thickness of the correction film 28b having a relatively large size (refer to the cross-sectional view of Fig. 4(c)). Further, according to the intensive review by the present inventors, it has been found that the phenomenon of the correction of the film thickness caused by the film formation area as described above is due to the inability to avoid the scanning speed and the film formation site at the time of film formation of the FIB device. The relationship between the supply amount of the film forming material varies. Further, the inventors of the present invention have found that the influence factor of the change to the -10-200844648 is complicated, and if the film formation area is small, the supply amount of the film forming material is excessively reduced with respect to the scanning speed, and if the film formation area is further reduced, On the other hand, the supply amount of the film-forming material becomes excessive, and the relationship between the film formation area and the film thickness is not necessarily clearly related. Therefore, even if it is desired to form a desired portion of a certain area, the film thickness of the film formation is actually performed. Control and forecasting are also not easy. By the way, it is an example of the relationship of the film formation area and the film thickness change of the case where the carbon correction film is formed in a desired part, and shows the relationship of the film formation area dependence of the film thickness of FIB apparatus. . φ Although there is a method of eliminating such variation by the condition change of the FIB device, the film thickness variation (that is, the light transmittance change) required for the correction film of the semi-transmissive portion of the gray scale mask of the film formation target is also obtained. The allowable tanning system is extremely strict, and the film thickness of the desired film is often not obtained by the parameter adjustment of the FIB device. Further, the problem of the conventional defect correction method is not only the problem of variation in film thickness caused by the film formation area as described above. The actual white defect is not limited to a shape in which the cross section lacks a substantially vertical direction in the film thickness direction. For example, as shown in Fig. 6(a), the semi-transmissive film 26 is set to have a shape in which the cross section is narrowed toward the side of the transparent substrate 24. In this case, when the correction film 28c having a uniform film thickness is formed in the defect portion (Fig. 6(b)), since the correction film is formed on the semi-transmissive film 26 partially remaining in the portion of the white defect, it occurs. The amount of light transmitted in the region where the semi-transmissive film and the correction film overlap is smaller than the desired amount (see the cross-sectional view of Fig. 6(c)). Further, the symbol 25 is a light shielding film. Further, as shown in Fig. 7(a), in the case where the black defect 63 occurs due to the light shielding film component or foreign matter adhering to the semi-transmissive film 26, it is difficult to remove only the black defect 63 portion, and the semi-transmissive film 26 is removed. no effect. For example, in the case where only the black defect 63 portion is removed by the FIB device, the black defect -11-200844648 component remains if the removal method is insufficient. If it is desired to completely remove the black defect, a part of the lower semi-transmissive film is removed, and a new defect 64 occurs in the semi-transmissive film 26 (see Figs. 7(b) and (c)). Further, if the new defect 64 is to be corrected by the correction film, the same problem as explained in the above-mentioned sixth (a) to sixth (c) views occurs. The present invention has been made in view of the above problems, and a first object thereof is to provide a defect correcting method for a gray scale mask which can appropriately correct a defect occurring in a semi-transmissive portion. φ A second object of the present invention is to provide a method of manufacturing a gray scale mask having a defect correction step to which the above-described defect correction method is applied. A third object of the present invention is to provide a gray scale mask which is appropriately corrected for defects occurring in the semi-transmissive portion. In order to solve the above problems, the present invention has the following configuration. (Configuration 1) A method for correcting a defect of a gray scale mask, which is formed by forming a semi-transmissive film and a light-shielding film on a transparent substrate, and applying a predetermined pattern to have a light-shielding portion and a light-transmitting portion. And a semi-transmissive portion of the semi-transmissive portion for reducing the amount of exposure light used by the reticle to form a gray-scale light formed on the transfer target by a film thickness step or a continuously different resist pattern A defect correction method for a cover, wherein the semi-transmissive portion is formed by the semi-transmissive film; the defect portion is specified when a defect occurs in the semi-transmissive portion; and the correction film is formed to be formed in the specific portion The film forming means and the film forming material of the defective portion; when applying the determined film forming means and the film forming material, determining the inner circumference of the formed face film. The film is formed by exposing the film to light of the amount of the permeation surface. -12- 200844648 (Structure 2) The defect correction method of the gray scale mask described in the structure 1 can also be used for the film formation. In the case of applying the determined film forming means and film forming material, the film thickness of the exposure light is set to a predetermined film thickness, and the set film thickness is set to be used in advance. It is determined by the correlation between the film thickness and the film formation area. - (Structure 3) In the defect correction method of the gray scale mask described in the configuration 1 or 2, it is preferable that φ has a step before the formation of the correction film, and the determined portion includes the sum of the defective portions. The transparent substrate is exposed in a region of an area where the film formation areas are substantially equal. (Structure 4) The defect correction method of the gray scale mask described in any one of the configurations 1 to 3, the defect portion having a film thickness smaller than that of the normal semi-transmissive portion or having a half thickness The portion of the light-transmissive film is a portion where the amount of exposure light is larger than the normal semi-transmissive portion. φ (Structure 5) The defect correction method of the gray scale mask described in any one of the structures 1 to 3, wherein the defect portion is a component other than the semi-transmissive film attached to the semi-transmissive portion, so that the exposure light is The portion that transmits less than the normal semi-transmissive portion. (Structure 6) In the method of correcting the defect of the gray scale mask of any one of the structures 1 to 5, preferably, when a plurality of defective portions are generated in the semi-transmissive portion, each of the plurality of defective portions is formed substantially A correction film of the same film formation area. -13- 200844648 (Structure 7) The defect correction method of the gray scale mask described in any one of the structures 1 to 6 may be formed as an area which is an integral multiple of the determined film formation area, and the determination is made. The step of correcting the film formation area repeats the number of times only an integral multiple. (Structure 8) The defect correction method of the gray scale mask described in any one of the structures 1 to 7, wherein the film formation means of the correction film is a focused ion beam method. φ (Structure 9) The method of producing the gray scale mask of the present invention is characterized by comprising a defect correcting step of the defect correcting method described in any one of the structures 1 to 8. (Structure 10) The gray scale mask of the present invention has a light-shielding portion, a light-transmitting portion, and a photomask to be used by forming a semi-transmissive film and a light-shielding film on a transparent substrate and applying a predetermined patterning. The amount of exposure light used is reduced by a predetermined amount of the semi-transmissive portion, and is used to form a film thickness step or a continuously different resist pattern φ on the object to be transferred, and the gray scale mask is characterized by A correction film having a plurality of substantially fixed areas or an integral multiple thereof is formed in the semi-transmissive portion. (Structure 1 1) In the gray scale mask described in the structure 10, the semi-transmissive film formed in advance in the semi-transmissive portion and the correction film may have different compositions. According to the defect correction method of the gray scale mask according to the present invention, since the amount of exposure light required for the semi-transmissive portion can be set to a desired range, the transmission of the desired exposure light can be formed with high reproducibility. Since the amount of the film-forming surface is 14-200844648, the correction film is integrated, so that the precision of the transmittance of the exposure light in the semi-transmissive portion can be improved and the stability can be corrected. As a result, the region in which the defect has been corrected becomes a gray-scale effect which is equivalent to the normal gray-scale portion of the semi-transmissive portion, and the defects occurring in the semi-transmissive portion can be appropriately corrected. Further, the stability and controllability of the correction film which is defective in the semi-transmissive portion are improved, and the yield of the gray scale mask is greatly improved. Further, according to the method for producing a gray scale mask of the present invention, by having the defect correcting step of applying the defect correcting method of the present invention, it is possible to obtain a gray which has been Φ appropriately corrected for defects occurring in the semi-transmissive portion. Order mask. Further, according to the gray scale mask of the present invention, a plurality of correction films having substantially the same area are formed in the semi-transmissive portion, and the plurality of defective portions are highly reproducibly formed to have a desired amount of exposure light. The correction film of the film formation area can obtain a gray scale mask which has a high precision and a stable correction for the transmittance of the exposure light in the semi-light-transmitting portion. [Embodiment] Hereinafter, the best mode for carrying out the invention will be described based on the drawings. φ [First Embodiment] Figs. 8(a) to 8(d) are diagrams showing a first embodiment of a defect correction method for a gray scale mask according to the present invention, and Figs. 8(a) to 8(c) The drawings are each used to illustrate a perspective view of the defect correction method in the order of steps, and the eighth (d) is a side cross-sectional view taken along line L-L of the eighth (c) diagram. Further, Fig. 9 is a cross-sectional view for explaining a pattern transfer method using the gray scale mask of the present invention. Further, the corrected defective portion is not shown in Fig. 9. The gray scale mask 20 of the present invention shown in FIG. 9 is used for manufacturing a thin film transistor (TFT) or a color filter such as a liquid crystal display device (LCD), or a plasma display panel of -15-200844648 ( PDP) or the like, and a resist pattern 3 3 having a film thickness stage or a continuous difference is formed on the transfer target body 30. Further, in Fig. 9, reference numerals 32A and 32B denote films laminated on the substrate 31 by the transfer member 30. Specifically, the gray scale mask 20 has a light blocking portion 21 that shields the exposure light when the gray scale mask 20 is used (transmittance is about 0%); and the light transmitting portion 22 passes through the surface of the transparent substrate 24. Exposed to make the exposure light pass through about 100%; and the semi-transmissive portion 23 reduces the transmittance of the exposure light to about 20 to 60%. The semi-transmissive portion 23 is formed by forming a semi-transmissive semi-transmissive film 26 on a transparent substrate 24 such as a glass substrate. Further, the light shielding portion 21 is formed by sequentially forming the semi-transmissive film 26 and the light-shielding light-shielding film 25 on the transparent substrate 24. In addition, the light shielding portion 21 has a structure and a process for mask blank used in the manufacture of the mask, and the light shielding film 25 and the semi-transmissive film 26 are sequentially formed on the transparent substrate 24, or Only the light shielding film 25 is formed on the transparent substrate 24. Further, the pattern shapes of the light shielding portion 2 1 , the light transmitting portion 22 and the semi-light transmitting portion 23 shown in Fig. 9 are completely representative examples, and the present invention is not limited thereto. Examples of the semi-transmissive film 26 include a chromium compound, MoSi, Si, W, and A. Among the chromium compounds, there are chromium oxide (Cr〇x), chromium nitride (CrNx), and chromium oxynitride (CrOxN). Chromium fluoride (CrFx), or those containing carbon or hydrogen. Further, examples of the light shielding film 25 include Cr, Si, and W' A1. The transmittance of the light shielding portion 21 is set by the selection of the film material and film thickness of the light shielding film 25 (or the light shielding film 25 and the semi-transmissive film 26). Further, the transmittance of the semi-transmissive portion 23 is set by the film material and film thickness of the semi-transmissive film 26. When the gray scale mask 20 as described above is used, since the exposure light is substantially impermeable in the light shielding portion 21, the exposure light is reduced by -16 - 200844648 in the semi-transmissive portion 23, so that it is applied to the object to be transferred 30. In the upper photoresist film (positive photoresist film), the film thickness is thicker in the portion of the light shielding portion 21, and the film thickness is thinner in the portion corresponding to the semitransparent: in the portion corresponding to the light transmitting portion 22. A resist pattern 33 is formed (refer to Fig. 9). In the resist pattern 33, the effect of thinning the film thickness in the portion of the semi-transmissive portion 23 is referred to as a gray-scale effect, and in the case of using a negative-type photoresist, it is necessary to consider the portion and the light-transmitting portion. The photoresist film has a reverse thickness, but the effect of the present invention can be sufficiently obtained. On the other hand, in the film-free portion of the resist pattern 33 shown in Fig. 9, for example, the first etching is performed on the films 32A and 32B of the transfer target 30, and the thin portion of the film of the resist pattern 3 3 is removed by ashing or the like. In this portion, for example, the film 32B of the transfer body 30 is subjected to the second engraving. In this manner, the gray scale mask 20 performs the steps of the conventional two mask components, and the number of masks. Next, the defect correction method of this embodiment will be described. In the Φ state, a gray scale mask for TFT substrate fabrication is used, which is a transparent semi-transparent film containing molybdenum molybdenum (Mo Si) (exposed light 50%) and chromium (CO is The light-shielding film of the main component is formed into a film and patterned by one, and the light-shielding portion, the light-transmitting portion, and the semi-transmissive portion are provided in the present embodiment, and a method of correcting the semi-transmissive portion is described. The manufactured gray scale mask uses the defect inspection to check the defect of the mask pattern, and the position information and the shape information of the defective portion are absent in the semi-transmissive portion. This case is relative to the normal semi-transparent light. The film thickness of the semi-transparent film is small, corresponding to the effect of the film without the film of 3 2 3 . This is the effect of shading, and the transmittance of the plate is used to reduce the light. Adding a predetermined light portion. When the defect is placed, it is trapped, because it has a portion where the light is -17-200844648 and the semi-transmissive film is absent, so that the amount of exposure light is larger than that of the normal semi-transmissive portion. The so-called white defect. The result of the defect inspection is shown in Figure 8(a). In the semi-transmissive portion formed by the film 26, there are small-sized white defects 50 and large-sized white defects 51. Further, although the defects actually occurring in the photomask are mostly irregular shapes, for convenience. Here, it is shown as a rectangular shape. (2) Next, a film forming means and a film forming material for forming a correction film on the specific defect portion are determined. In the present embodiment, FIB is used as a film forming means. Further, the film forming material is made of carbon suitable for film formation by FIB. Of course, it is not limited to carbon, and a material containing molybdenum telluride similar to the semi-transmissive film may be used. Here, the FIB device will be described. The film can also be used for the removal of the film. As shown in Fig. 10, the FIB device 40 has an ion source 41 which generates Ga + ions, an electromagnetic optical system 42 and an electron gun 43 which is released for neutralizing Ga + ions. Electron; etching gas gun 49 emits helium gas; φ and gas gun 44 emit helium gas. Electromagnetic system 42 uses G a + ions generated from ion source 41 as ion beam 47 and utilizes Scan amplifier 46 scans the ion beam 4 7. On the XY table 45, the gray scale mask 20 that is the object to be corrected is placed, and by moving the XY table 45, the defect region to which the gray scale mask 20 is applied is moved to the ion beam. Then, the ion beam 47 is scanned to apply the corrected defect region, and the position of the defective region to which the correction is applied is detected by detecting the action of the secondary ion detector 48 of the secondary ions generated at this time. The correction of the formation of the modified film or the removal of the film (for example, the removal of the semi-transmissive film in the black defect region) is performed by the ion beam 47 via the electromagnetic optical system 42 by irradiating -18-200844648 with the modified defect region of the gray scale mask 20. ). In addition, the beam diameter of the ion beam is 0. 1 μ πιφ or less. In the case where the correction film is formed, the ion beam 47 is emitted through the electromagnetic optical system 42, and the helium gas is released by the gas gun 44. Therefore, the xenon gas contacts the ion beam 47 to be polymerized (chemical reaction), and the correction film is deposited on the irradiation region of the ion beam 47 to form a film. Further, for example, in the case where the semi-transmissive film of the black defect region is removed, the gas is blown by the φ etching gas gun 49, and the ion beam 47 is irradiated via the electromagnetic optical system 42 in this state, thereby removing the semi-transparent. membrane. (3) Next, when the above-described determined film forming means and film forming material are applied, the film forming area of the correcting film which can obtain a predetermined amount of exposure light is determined. The transmittance of the predetermined exposure light is set to a transmittance of 40%, and 45 urn is set as the film thickness of the carbon film satisfying the transmittance. In order to obtain such a correction film, it is determined by film formation under a predetermined film formation condition using a FIB apparatus, and it is determined based on the correlation between the film thickness φ and the film formation area of the film formation conditions which are obtained in advance as described above. The film formation area was set to 400 / z m2 to form a film in the most stable manner. The film formation area thus determined includes the size of the white defects 50, 51 which are located in the semi-transmissive portion. Here, the correlation between the film formation film thickness and the film formation area is, for example, the correlation shown in Fig. 5 described above. (4) Next, for each of the regions including the white defects 50 and 51 located in the semi-transmissive portion and larger than the white defects, the area corresponding to the above (3) determined (film formation area) is removed (equal Or substantially equal) semi-transmissive film 26. As the means for removing the semi-transmissive film 26, although a FIB device is used, other laser devices such as -19-200844648 may be used. As a result, as shown in Fig. 8(b), the same size and shape are removed for the white defects 50 and 5 which are different in size, i.e., the semi-transparent film 26 (26a, 26b) of the same area is removed. The portions 26a and 26b are exposed to the transparent substrate 24. (5) The portions 26a and 26b from which the semi-transmissive film has been removed are used as a film formation region of the correction film, and the required position information and the like are input to the FIB device, and the film formation film thickness or other film formation conditions are input. The portions 26a and 26b of the semi-transmissive film are removed, and the modified φ films 27a and 27b having the same size and shape (i.e., the same area) are formed (see Figs. 8(c) and 8(d)). The film thickness of the formed correction films 27a and 27b is 1. When measured by AFM (atomic force microscope), the maximum height difference is 1. At 26 nm, the in-plane system was fixed, and the film thicknesses of the correction films 27a and 27b were not changed. Therefore, the correction films 27a and 27b which can obtain the transmission amount of the desired exposure light set in advance are formed. Further, in the eighth (d) diagram, the thicknesses of the semi-transmissive film 26 and the correction films 27a and 27b are substantially the same, but it is only necessary to control the semi-transmissive portion to have a predetermined amount of exposure light. Therefore, when the film materials of the semi-transmissive film 26 and the modified φ films 27a and 27b are different, the film thickness may be different. According to the first embodiment described above, the following effects can be obtained. 1. Since the amount of exposure light required for the semi-transmissive portion can be set to a desired range, and a correction film that becomes a film-forming area of the desired amount of exposure light can be formed with high reproducibility, it can be implemented in half. The precision of the transmittance of the exposure light of the light transmitting portion is high and the stability is corrected. 2. Therefore, the area where the defect has been corrected becomes a gray-scale effect which is equivalent to the normal gray-scale portion of the semi-transmissive portion, and the defects generated in the semi-transmissive -20-200844648 portion can be appropriately corrected. · 3. Further, the stability and controllability of the correction film of the defect generated in the semi-transmissive portion are improved, and the yield of the gray scale mask is greatly improved. In the present embodiment, in order to form a correction film having the same film formation area for a plurality of defective portions, a correction film having the same size and shape (rectangular shape) including the defective portion is formed, but the shape is not limited to a rectangle, and may be, for example, a circle. Shapes of other shapes. Further, the shape of the plurality of correction films may be different from each other as long as they have the same film formation area. Further, even if the surface φ product of the plurality of correction films is not strictly the same, the film thickness may be slightly different as long as the film formation area does not occur. Further, in the present embodiment, FIB is used as the film forming means of the correction film, but the film forming means is not limited to FIB. For example, other film forming means such as laser CVD may be applied. [Second Embodiment] The first embodiment (1) to the first embodiment (1) show the second embodiment of the defect correction method for the gray scale mask of the present invention, and the eleventh (a) to eleventh (th) c) The figure φ is used to explain the perspective view of the defect correction method in the order of steps, and the 11th (d) figure is a side cross-sectional view along the L-L line in the 11th (c) figure. Although it is also illustrated in Figs. 6(a) to 6(c), the actual white defect is not limited to a shape in which the missing cross section is substantially perpendicular in the film thickness direction. For example, as shown in Fig. 1 (a), in the semi-transmissive film 26, a white defect 52 having a meandering shape which is narrowed to the side of the transparent substrate 24 is formed. In this case, if a correction film having a uniform film thickness is formed at the defect portion, since the correction film is formed on the semi-transmissive film 26 partially remaining on the white defect portion, a portion of the semi-transmissive film overlaps with the correction film. The area, the amount of exposure light - 21 - 200844648 becomes smaller than the original. The same is true for white defects of complicated shapes other than this shape. In the second embodiment, when the film forming means and the film forming material are determined in the same manner as in the above-described first embodiment, the film forming area of the correcting film which is a predetermined amount of the transmitted light is determined. Then, the semi-transmissive film 26 corresponding to the determined area (film formation area) is removed in a region including the white defect 52 located in the semi-light-transmitting portion and larger than the white defect (see the first 1st) (b) Figure). φ In the portion 26c from which the semi-transmissive film has been removed, the correction film 27c having the determined film formation area is formed (see Fig. 11(c) and Fig. 11(d)). Therefore, a correction film having a uniform film thickness capable of obtaining a desired amount of exposure light to be set in advance is formed. Therefore, in the second embodiment, the effects of the first to third embodiments of the first embodiment described above can be obtained in the same manner. [Third Embodiment] Figs. 12(a) and 12(b) are views showing a third embodiment of the method for correcting the gradation mask of the gray scale mask of the present invention, which is for explaining the defect correction method in order of steps. Stereogram. When the self-defect generated in the semi-transmissive portion of the gray scale mask is larger than the fixed area, for example, the film formation area of the correction film determined in the same manner as in the first embodiment, the specific area is The integer multiple area is set as the defect correction area. First, in the third embodiment, when the film forming means and the film forming material are determined in the same manner as in the above-described first embodiment, the film forming area of the correcting film which is a predetermined amount of exposure light is determined. -22- 200844648 Then, when the white defect generated in the semi-transmissive portion exceeds the size of the film formation area determined as described above, the white defect included in the semi-transmissive portion is larger than the white defect The semi-transmissive film 26 corresponding to the above-mentioned determined area (film formation area) is removed (see Fig. 12(a)). The portion 26d from which the semi-transmissive film has been removed is subjected to the step of forming the correction film of the above-mentioned determined film formation area, and the number of times of the integral multiple is repeated repeatedly so that the formed correction film is adjacent to each other. Therefore, in the portion 26d from which the semi-φ transparent film has been removed, a correction film 27d having a uniform film thickness capable of obtaining a desired amount of exposure light to be set in advance (see Fig. 12(b)) is formed. Therefore, according to the third embodiment, the effects of the first to third embodiments of the first embodiment described above can be obtained in the same manner. [Fourth Embodiment] Figs. 13U) to 13(c) are views showing a fourth embodiment of the defect correction method for the gray scale mask of the present invention, and are a perspective view for explaining the defect correction method in order of steps. φ In the case where the area of the semi-transmissive portion including the white defect is small and close to the light-shielding film 25, the fixed area (for example, the film formation area of the correction film determined in the same manner as in the first embodiment described above) may be used. As the defect correction area, the present invention is effectively applied. In the fourth embodiment, when the film forming means and the film forming material are determined in the same manner as in the first embodiment described above, the film forming area of the correcting film which is a predetermined amount of exposure light is determined. Then, as shown in Fig. 1 (a), since the area of the semi-transmissive portion where the white defect 5 3 occurs is close to the surrounding light-shielding film 25, the film formation determined in the above-mentioned decision 23-200844648 is not achieved. In the case of the area, the semi-transmissive film region is removed from the semi-transmissive film region including the white defect 53 located in the semi-transmissive portion without the light-shielding film 25 (see Fig. 13(b)). The correction film 27e having the above-described determined film formation area including the portion 26e from which the semi-transmissive film has been removed is formed. Therefore, in the portion 26e from which the semi-transmissive film has been removed, a correction film 27e having a film thickness of a uniform sentence which can obtain a desired amount of the desired exposure light is formed (see Fig. 13(c)). Further, since the above-mentioned determined film formation area (actual film formation area of the correction film) Φ is larger than the area of the semi-transmissive portion where the white defect 53 is generated, a part of the correction film 27e is also formed on the light shielding film 25. However, because it is a light-shielding part, it does not cause defects. Further, the unnecessary light-shielding film 27e on the light-shielding film 25 can be removed. Therefore, in the fourth embodiment, the effects of the first to third embodiments of the first embodiment can be obtained in the same manner. [Fifth Embodiment] Figs. 14(a) to 14(c) are diagrams showing a fifth embodiment of the method for correcting the gradation mask of the gray scale mask of the present invention, for explaining the defect correction method in the order of steps. Stereogram. For example, in the case where the process lacks a minute semi-transmissive film, the semi-transmissive portion on the transparent substrate and the semi-transmissive film containing the white defect have a small area and do not reach a fixed area (for example, and the above). In the case where the film formation area of the correction film is determined in the same manner as in the first embodiment, the film can be formed only in a small area, and the fixed area can be used as the defect correction area, and the present invention can be effectively applied. In other words, in the fifth embodiment, when the film forming means and the film forming material are determined in the same manner as in the above-described first embodiment, the film forming means and the film forming material are determined, and the correction film which is a predetermined amount of exposure light is determined. Membrane area. Then, as shown in Fig. 14(a), the area of the semi-transmissive portion which becomes the white defect 54 does not reach the above-described determined film formation area, and the area on the transparent substrate 24 including the white defect 54 The correction film 27f is formed on the film formation area of the above-mentioned film (refer to Fig. 14(b)). Therefore, in the region containing the self-defect 5, a correction film 27f having a uniform film thickness capable of obtaining a desired amount of exposure light to be set in advance is formed. After correcting the film formation of the film 27f, the correction film of the unnecessary region 55 can be removed. Therefore, in the fifth embodiment, the effects of 1 to 3 of the above-described i-th embodiment can be obtained in the same manner. [Embodiment 6] Figs. 15(a) to 15(c) are views showing a sixth embodiment of a defect correction method for a gray scale mask according to the present invention, and are for explaining a perspective view of a defect correction method in order of steps. . As also shown in the description of the above-mentioned 7th (a)th to 7th (c), when black defects occur due to the φ film component or foreign matter adhering to the semi-transmissive film, only the black defect portion is removed. It has no effect on the semi-transparent film, which is difficult in the conventional method. The correction of such black defects can also be effectively applied to the present invention as shown in the following description. In the sixth embodiment, when the film forming means and the film forming material are determined in the same manner as in the above-described first embodiment, the film forming area of the correction film which determines the amount of penetration of the predetermined exposure light is determined. Then, for the black defect 5 6 generated in the semi-transmissive portion as shown in Fig. 15 (a), and the region larger than the black defect region 56 will be equivalent to -2544,44648 The semi-transmissive film 26 and the black defect 56 of the determined area (film formation area) are removed together (refer to Fig. 16(b)). In the portion 26g from which the semi-transmissive film has been removed, the correction film 27g having the determined film formation area is formed (see Fig. 15(c)). Therefore, the black defect 56 generated in the semi-transmissive portion is removed, and the portion 26g from which the semi-transmissive film has been removed is formed to have a desired film thickness of the desired exposure light having a uniform film thickness. The film 27g was corrected. Therefore, in the sixth embodiment relating to the correction of the black defect, the effects of the first to third embodiments of the first embodiment described above can be obtained in the same manner as φ. BRIEF DESCRIPTION OF THE DRAWINGS FIGS. 1(a) to 1(c) are schematic cross-sectional views showing a process according to the related art using a TFT substrate of a gray scale mask. Figs. 2(a) to 2(c) are schematic cross-sectional views showing the processes of Figs. 1(a) to 1(c). Fig. 3 is a plan view showing an example of a gray scale mask of a fine pattern type. φ 4(a) to 4(c) are diagrams for explaining the problem of the conventional defect correction method, and 4(a) and 4(b) are perspective views, 4(c) Figure is a sectional view. Fig. 5 is a view showing the relationship between the film formation area and the film thickness in the case where film formation is performed using a FIB apparatus. Fig. 6(a) to Fig. 6(c) are diagrams for explaining the problem of the conventional defect correction method, and Fig. 6(a) and Fig. 6(b) are perspective views, Fig. 6(c) A section view. Figures 7(a) to 7(c) are used to illustrate the previous method of defect correction. -26- 200844648 Figure of the problem point, Figure 7 (a) and Figure 7 (b) are perspective views, and Figure 7 (c) is a sectional view. 8(a) to 8(d) are views showing the first embodiment of the defect correction method of the gray scale mask of the present invention, and the eighth (a) to eighth (c) drawings are respectively used to follow The sequence of steps illustrates a perspective view of the defect correction method, and the eighth (d) diagram is a side cross-sectional view taken along line L-L of the eighth (c) diagram. Figure 9 is a cross-sectional view for explaining a pattern transfer method using the gray scale mask of the present invention. • Fig. 10 is a schematic side view showing the construction of the FIB apparatus used in the present invention. 11(a) to 11(d) are views showing a second embodiment of the defect correction method of the gray scale mask of the present invention, and the 11th (a)th to the 11th (c)th drawings are respectively used to follow The sequence of steps illustrates a perspective view of the defect correction method, and the first 丨(d) diagram is a side cross-sectional view taken along line L-L of the 1st (c) diagram. Figs. 12(a) and 12(b) are views showing a third embodiment of the defect correction method for the gray scale mask of the present invention, which is a perspective view for explaining the method of correcting the missing portion in the order of steps. Figs. 13(a) to 13(c) are views showing a fourth embodiment of the defect correction method for the gray scale mask of the present invention, which is a perspective view for explaining the defect correction method in order of steps. Figs. 14(a) to 14(c) are views showing a fifth embodiment of the defect correction method for the gray scale mask of the present invention, and are perspective views for explaining the defect correction method in order of steps. Figs. 15(a) to 15(c) are views showing a sixth embodiment of the defect correction method for the gray scale mask of the present invention, which is a perspective view for explaining a method of correcting the missing portion in the order of steps. [Main component symbol description] 1 Glass substrate 2 Gate
3 閘極絕緣膜 4 第1半導體膜 5 第2半導體膜 6 源極汲極用金屬膜 7 正型光阻膜 7a 第1阻劑圖案 7b 第2阻劑圖案 10 灰階光罩 11 遮光部 1 la、1 lb 對應於源極/汲極之遮光部 12 透光部 13 半透光部 1 3 a 遮光圖案 13b 透過部 20 灰階光罩 21 遮光部 22 透光部 23 半透光部 24 透明基板 25 遮光膜 26 半透光膜 -28- 2008446483 gate insulating film 4 first semiconductor film 5 second semiconductor film 6 source drain metal film 7 positive resist film 7a first resist pattern 7b second resist pattern 10 gray scale mask 11 light shielding portion 1 La, 1 lb corresponds to the source/drainage light-shielding portion 12, the light-transmitting portion 13, the semi-transmissive portion 1 3 a light-shielding pattern 13b, the transmission portion 20, the gray-scale mask 21, the light-shielding portion 22, the light-transmitting portion 23, the semi-transmissive portion 24, transparent Substrate 25 light shielding film 26 semi-transparent film-28 - 200844648
26a~ 26g 已除去半透光膜之部分 2 7 a 〜2 7 g 修正膜 28a 尺寸比較小之修正膜 28b 尺寸比較大之修正膜 28c 膜厚均勻的修正膜 30 被轉印體 3 1 基板 32A、32B 膜 33 阻劑圖案 40 FIB裝置 41 離子源 42 電磁光學系統 43 電子槍 44 氣體槍 45 XY工作台 46 掃描放大器 47 離子束 48 二次離子檢測器 49 鈾刻用氣體槍 5 0 〜5 4 白缺陷 55 不要之區域 56 黑缺陷 -29-26a to 26g Part of the semi-transparent film removed 2 7 a to 2 7 g Correcting film 28a Correcting film 28b having a relatively small size Correcting film 28c having a relatively large size Correcting film 30 having a uniform film thickness Transfer body 3 1 Substrate 32A , 32B film 33 resist pattern 40 FIB device 41 ion source 42 electromagnetic optical system 43 electron gun 44 gas gun 45 XY table 46 scanning amplifier 47 ion beam 48 secondary ion detector 49 uranium engraving gas gun 5 0 ~ 5 4 white Defect 55 Do not area 56 Black Defect-29-