[具有輔助圖案之孔圖案形成用光罩之設計]
圖1係表示專利文獻1中記載之光罩之轉印用圖案之主要部分者，(a)係俯視模式圖，(b)係(a)之A-A位置之剖視模式圖。
圖示之光罩之轉印用圖案具有包含透光部之主圖案1、及於該主圖案1之周邊包圍主圖案1而配置之輔助圖案2。輔助圖案2係隨附於主圖案1配置於主圖案1之周圍。
又，於透明基板10上形成有相位偏移膜11及低透光膜12。主圖案1包含露出透明基板10之透光部4，輔助圖案2包含露出透明基板10上之相位偏移膜11之相位偏移部5。又，低透光部3分別包圍主圖案1及輔助圖案2。
低透光部3包括形成於透明基板10上之相位偏移膜11與低透光膜12之積層膜。低透光部3亦可包括形成於透明基板10上之低透光膜12之單層膜。亦即，低透光部3包含至少形成有低透光膜12之部分。於圖示之轉印用圖案中，形成有主圖案1及輔助圖案2之區域以外之區域成為低透光部3。
相位偏移膜11具有使處於i線～g線之波長範圍之代表波長之曝光之光偏移大致180度之相位偏移量。即，輔助圖案2具有藉由該相位偏移膜11而將透過光之相位反轉之作用。又，對於上述代表波長之曝光之光，相位偏移膜11具有T1(%)之透過率。
根據專利文獻1，使用具備上述轉印用圖案之光罩進行於被轉印體上形成孔圖案之光學模擬之後，與不具備二元遮罩或輔助圖案之相位偏移遮罩相比，顯然於Eop(將目標尺寸之圖案形成於被轉印體上所需之曝光光量)或DOF(Depth of Focus：焦點深度)等方面，發揮了優異之性能。
本發明者為了於被轉印體上形成更微細之圖案，而對圖2(a)～(c)所示之光罩進行了光學模擬。此處，將圖2(a)之光罩設為參考例1，將圖2(b)所示之光罩設為參考例2，將圖2(c)所示之光罩設為參考例3。繼而，使用具有包含直徑W1為2.0 μm之孔圖案之主圖案之參考例1～3之各光罩，進行於被轉印體(顯示面板基板等)上之正型光阻形成相當於直徑W2(此處為1.5 μm)之孔圖案之轉印圖像之模擬。
再者，如上所述，將光罩上之直徑W1設為相對於被轉印體上之目標直徑W2為W1≧W2(較佳為W1＞W2)。此處，若將遮罩偏差β1(μm)設為β1＝W1－W2，則此處將β1設為0.5(μm)。
模擬之條件如下所述。
(參考例1)
於參考例1中，如圖2(a)所示，使用包含二元遮罩之光罩，將被低透光部(遮光部)3包圍之包含直徑W1＝2 μm之正方形之透光部之孔圖案設為主圖案1。
(參考例2)
於參考例2中，如圖2(b)所示，使用包含半色調式相位偏移遮罩之光罩，將被曝光之光之透過率為5.2%且相位偏移量為180度之相位偏移部5包圍之包含直徑W1＝2 μm之正方形透光部的孔圖案設為主圖案1。
(參考例3)
於參考例3中，如圖2(c)所示，使用包含附有輔助圖案之相位偏移遮罩之光罩，設為將包含直徑W1＝2 μm之正方形透光部之孔圖案設為主圖案1且正八邊形帶之輔助圖案2包圍該圖案1之周邊之構成。又，輔助圖案2包含曝光之光之透過率為45%且相位偏移量為180度之相位偏移部，主圖案1及輔助圖案2以外之區域包含光學密度為OD≧2之低透光部(遮光部)3。主圖案1之中心與輔助圖案2之寬度方向中心位置之距離(L)設為3.25 μm，輔助圖案2之寬度(d)設為1.3 μm。參考例3之光罩係基於專利文獻1中記載之構成設計而成。
使用與上述參考例1～3對應之各個光罩，於被轉印體上形成寬度W2＝1.5 μm之孔圖案。
模擬之曝光條件如下所述。
曝光裝置之光學系統係數值孔徑NA為0.1，同調因子σ為0.5。又，於曝光光源中使用包含所有i線、h線、g線之光源(寬波長光源)，強度比設為g：h：i＝1：1：1。
光罩之光學評價項目如下所述。
(1)焦點深度(DOF)
焦點深度(DOF)係於在曝光時產生散焦之情形時，於被轉印體上用以使相對於目標CD之CD變動成為特定範圍(例如±10%)內之焦點深度，且較理想為其數值較大。若DOF數值較高，則不易受到被轉印體(例如顯示裝置用之面板基板)之平坦度之影響，從而可確實地形成微細之圖案，抑制該CD不均。於本案之模擬中，作為DOF值，將目標CD±10%設為基準。此處，所謂CD係臨界尺寸(Critical Dimension)之簡稱，且以圖案寬度之含義使用。顯示裝置製造用光罩係較半導體裝置製造用光罩，尺寸更大，又，被轉印體(顯示器面板基板等)亦為大尺寸，均難以獲得完全之平坦性，因此，提高DOF數值之光罩之意義較大。
(2)遮罩誤差增強因子(MEEF：Mask Error Enhancement Factor)
遮罩誤差增強因子係表示形成於被轉印體上之圖案之CD誤差相對於光罩上之CD誤差之比率之數值，MEEF越低，則越可減少形成於被轉印體上之圖案之CD誤差。顯示裝置之規格進化，要求圖案之微細化，並且需要具有接近曝光裝置之解像極限之尺寸之圖案之光罩，因此，於顯示裝置製造用光罩中，今後重視MEEF之可能性亦較高。
(3)Eop
於顯示裝置製造用光罩中，特別重要之評價項目中具有Eop(以下亦稱為「Eop劑量」)。Eop係將需要獲得之圖案尺寸形成於被轉印體上所需之曝光光量。用於顯示裝置之製造之光罩的尺寸極大(例如主表面之一邊為300～2000 mm左右之正方形或長方形)。因此，若使用Eop數值較低之光罩，則可提高掃描曝光之速度，生產效率提昇。
將對於上述評價項目之具體之評價結果示於圖3。
首先，若著眼於Eop，則參考例3之光罩相較參考例1及參考例2之光罩，用以獲得目標尺寸之孔圖案之曝光量(Eop數值)較小為30%以上。因此，可知若使用參考例3之光罩，則可獲得較高之生產效率。又，參考例3之光罩相較參考例1及參考例2之光罩，DOF數值變高，MEEF數值變低。因此，可知參考例3之光罩於DOF或MEEF中亦極為有利。
[處於相互接近之位置之孔圖案之設計]
另一方面，若對顯示裝置要求之解像度提高，則因每單位面積之像素數之增加導致積體度提高。因此，作為顯示裝置製造用光罩之轉印用圖案，必須使複數個主圖案(孔圖案)相互接近地配置。以下，對在包含複數個主圖案之密集圖案之形成中適用上述參考例3之光罩之情形進行研究。
圖4表示將適用於上述參考例3之圖案(結合主圖案與形成於其周邊之輔助圖案，於本案說明書中，亦稱為孔形成用圖案)相互接近地配置之情形。此處，將2個主圖案中之一個設為第1主圖案1a，將另一個設為第2主圖案1b。繼而，考察於使第2主圖案1b之位置對於第1主圖案1a自箭頭方向接近時，作為2個主圖案1a、1b之重心間距離之孔間距P及形成於被轉印體上之抗蝕圖案之剖面形狀。
首先，如圖5(a)所示，於第1主圖案1a與第2主圖案1b充分隔開之情形(P＝12 μm)時，如圖5(b)所示，於被轉印體上之抗蝕圖案(此處為包含正型之光阻之圖案)20形成與各個主圖案1a、1b對應之孔圖案21a、21b。
另一方面，如圖6(a)所示，於第2主圖案1b對於第1主圖案1a接近之情形時(P＝8 μm)時，配置於各個主圖案1a、1b之周圍之輔助圖案2a、2b彼此之距離變得極近。此時，若觀察形成於被轉印體上之抗蝕圖案20之剖面，則如圖6(b)所示，除與主圖案1a、1b對應之孔圖案21a、21b以外，於其等之中間部形成凹部22，因此，產生抗蝕膜厚之較大之損耗。此種抗蝕殘膜之局部減少存在對將抗蝕圖案作為蝕刻遮罩進行之顯示裝置基板之加工穩定性造成不良影響之虞。
因此，本發明者對Eop或DOF方面具有有利之特性，但會產生如上所述之抗蝕殘膜之局部減少之類欠佳情況之參考例3之孔形成用圖案，研究了能夠消除該欠佳情況之光罩。
＜光罩之構成＞
其次，對本發明之實施形態之顯示裝置製造用光罩之構成進行說明。
本發明之實施形態之顯示裝置製造用光罩係
於透明基板上具有轉印用圖案者，且
上述轉印用圖案包含：
主圖案，其包含四邊形之透光部；
輔助圖案，其配置於上述主圖案之周邊，且包含相位偏移部；及
低透光部，其形成於上述主圖案及上述輔助圖案以外之區域；
於定義上述主圖案之周邊中包圍上述主圖案之特定寬度之正八邊形帶時，上述輔助圖案構成上述正八邊形帶之至少一部分，
於將上述轉印用圖案所包含之複數個上述主圖案之1個設為第1主圖案時，與上述第1主圖案不同之第2主圖案配置於接近上述第1主圖案之位置，
構成包圍上述第1主圖案之上述正八邊形帶之八區段中之面向上述第2主圖案側之一區段中具有缺損之輔助圖案係配置於上述第1主圖案之周邊。
以下，使用圖7具體地進行說明。
圖7係表示本發明之實施形態之顯示裝置製造用光罩所具備之轉印用圖案之主要部分者，(a)係俯視模式圖，(b)係(a)之B-B位置之剖視模式圖。
上述顯示裝置製造用光罩(以下亦簡稱為「光罩」)例如具備藉由使成膜於透明基板10上之相位偏移膜11及低透光膜12分別圖案化而形成之轉印用圖案。該轉印用圖案包含主圖案1(1a、1b)、及配置於主圖案1之周邊之輔助圖案2(2a、2b)。輔助圖案2於定義包圍主圖案1之特定寬度之正八邊形帶時，具有構成該正八邊形帶之至少一部分之形狀。
主圖案1於X方向排列有2個，其中一個成為第1主圖案1a，另一個成為第2主圖案1b。於第1主圖案1a之周邊配置有輔助圖案2a，於第2主圖案1b之周邊配置有輔助圖案2b。再者，於圖7(a)中，將左側之主圖案設為第1主圖案，將右側之主圖案設為第2主圖案，但任一個均可設為第1主圖案。
於本實施形態中，主圖案1包含露出透明基板10之透光部4，輔助圖案2包含露出透明基板10上之相位偏移膜11之相位偏移部5。又，主圖案1及輔助圖案2以外之區域成為於透明基板10上至少形成有低透光膜12之低透光部3。
於本實施形態中，低透光部3係於透明基板10上積層相位偏移膜11及低透光膜12而成。相位偏移膜11具有使處於i線～g線之波長範圍之代表波長之曝光之光偏移大致180度之相位偏移量。又，相位偏移膜11對於上述代表波長之曝光之光之透過率為T1(%)。
低透光膜12可設為對於曝光之光之代表波長具有特定之較低之透過率者。於本實施形態中，低透光膜12可對於處於i線～g線之波長範圍之代表波長之曝光之光具有較相位偏移膜11之透過率T1(%)更低之透過率T2(%)。或者，低透光膜12亦可設為實質上不使曝光之光透過之遮光膜。
此處，於藉由使用光罩之曝光而於被轉印體上形成與光罩之主圖案1對應之微細之圖案(孔圖案)之情形時，若將主圖案1之直徑W1設為4 μm以下，則可於被轉印體上形成直徑W2(μm)(其中，W1≧W2，更佳為W1＞W2)之微細之圖案。
具體而言，主圖案1之直徑W1(μm)較佳為0.8≦W1≦4.0，更佳為1.0≦W1≦3.5。進而，可設為1.2＜W1≦3.0，於必須更微細化之情形時，可設為1.2＜W1＜2.5。
又，作為主圖案1之轉印圖像形成於被轉印體上之孔圖案之直徑W2(μm)較佳為0.8≦W2≦3.0，更佳為0.8≦W2≦2.5，進而較佳為0.8≦W2≦2.0或0.8≦W2≦1.8。或者，可設為0.8＜W2＜2.0或0.8＜W2＜1.8。於必須更微細化之情形時，可設為0.8＜W2＜1.5。
又，本實施形態之光罩可出於形成對顯示裝置之製造有用之微細尺寸之圖案之目的而使用。例如，於主圖案1之直徑W1為3.0(μm)以下時，可獲得更顯著之效果。
然而，於需要在被轉印體上形成孔圖案之光罩(例如上述參考例1～3)中，如上所述賦予遮罩偏差β1較為有利。即，若將光罩上之孔圖案之直徑設為W1，將形成於被轉印體上之孔圖案之直徑設為W2，則亦可設為W1＝W2，但較佳為W1＞W2。例如，若將β1(μm)設為偏差值(W1－W2)，且β1＞0(μm)，則偏差值β1(μm)較佳為0.2≦β1≦1.0，更佳為0.2≦β1≦0.8。如此，藉由將直徑W1與直徑W2之關係規定為偏差值β1，可獲得於被轉印體上可減少抗蝕圖案之膜厚之損耗等有利之效果。
再者，主圖案1包含四邊形之圖案，主圖案1之直徑W1係四邊形之一邊之尺寸。例如，若主圖案1為正方形之圖案，則主圖案1之直徑W1係指正方形之一邊之尺寸，若主圖案1為長方形之圖案，則直徑W1係指長邊之尺寸。又，主圖案1之形狀係指俯視主圖案1時之形狀。進而，形成於被轉印體上之孔圖案之直徑W2係指對向之2邊之間之距離最大之部分之長度。
如上所述之遮罩偏差β1亦可賦予本實施態樣之光罩。於本實施形態中，因2個孔形成用圖案接近特定距離地形成而存在作為遮罩偏差β1而賦予之尺寸較孤立圖案(上述參考例1～3等)變大之情形。
進而，產生本實施態樣之光罩較佳為根據接近地配置之2個孔形成用圖案之位置關係，相對於X方向及與其垂直之Y方向賦予不均等之尺寸之遮罩偏差之情形。因此，將此種遮罩偏差設為β2，將相對於X方向、及與其垂直之Y方向賦予之偏差量分別設為β2(x)、β2(y)。對於遮罩偏差β2之詳情下文敍述。
對於用於具有上述轉印用圖案之本實施形態之光罩之曝光之曝光之光之代表波長，主圖案1與輔助圖案2之相位差ϕ為大致180度。即，透過主圖案1之上述代表波長之光與透過輔助圖案2之上述代表波長之光之相位差ϕ1成為大致180度。所謂大致180度係指120～240度。上述相位差ϕ1較佳為150～210度。
再者，本實施形態之光罩於使用包含i線、h線、及g線之至少1種之曝光之光時效果較為顯著，尤其，較佳為適用包含i線、h線、及g線之寬波長光作為曝光之光。於此情形時，可將處於i線～g線之波長範圍之任一波長設為代表波長。例如，可將h線作為代表波長構成本實施形態之光罩。更佳為對於上述代表波長之相位差ϕ1為ϕ1＝180度。
於本實施形態之光罩中，為實現上述相位差，只要將主圖案1設為透明基板10之主表面露出而成之透光部4，將輔助圖案2設為形成於透明基板10上之相位偏移膜11露出而成之相位偏移部5，將該相位偏移膜11對於上述代表波長之相位偏移量設為大致180度即可。
相位偏移部5所具有之透過率T1可如下所述地設定。即，若將形成於相位偏移部5之相位偏移膜11對於上述代表波長之透過率設為T1(%)，則較佳為20≦T1≦80，更佳為30≦T1≦75，進而較佳為40≦T1≦75。若輔助圖案2之透過率較高，則為獲得特定量之透過光量，可減小輔助圖案寬度(d)，因此，具有獲得可避免密集圖案中之相互之物理干涉而配置之自由度之優點。另一方面，若略微降低透過率T1，擴大輔助圖案寬度(d)，則存在圖案形成之製造上之難易度得以緩和之優點。於此情形時，透過率T1較佳為40～60(%)。再者，此處之透過率T1(%)設為以透明基板10之透過率為基準(100%)時之上述代表波長之透過率。
於本實施形態之光罩中，於形成有主圖案1及輔助圖案2之區域以外之區域形成有低透光部3。此處，主圖案1與輔助圖案2介隔低透光部3分隔。低透光部3可設為如下構成。
低透光部3係曝光之光(處於i線～g線之波長範圍之代表波長之光)實質上不透過之低透光膜(即遮光膜)12，且可設為於透明基板10上形成光學密度OD≧2(較佳為OD≧3)之膜而成者。
又，低透光部3亦可設為形成以特定範圍之透過率使曝光之光透過之低透光膜12而成者。但，於以特定範圍之透過率透過曝光之光之情形時，低透光部3對於上述代表波長之透過率T3(%)設為相較上述相位偏移部5之透過率T1(%)，滿足0＜T3＜T1者，較佳為滿足0＜T3≦20。此處，於相位偏移部5並非包含相位偏移膜11之單層膜而包含相位偏移膜11與低透光膜12之積層膜之情形時，將作為該積層膜之透過率設為T3(%)。此處之透過率T3(%)亦與上述同樣地設為以透明基板10之透過率為基準時之上述代表波長之透過率。
又，於如此般低透光膜12以特定範圍之透過率使曝光之光透過之情形時，相位偏移膜11與低透光膜12之積層狀態下之相位偏移量ϕ3較佳為90(度)以下，更佳為60(度)以下。所謂「90度以下」係指若以弧度表述，則上述相位差為「(2n－1/2)π～(2n＋1/2)π(此處n為整數)」。對於相位差，亦與上述同樣地，設為對於曝光之光中包含之代表波長者。
又，作為用於本實施形態之光罩之低透光膜12之單獨之性質，較佳為實質上不使上述代表波長之光透過(OD≧2，更佳為OD＞3)者，或具有未達30(%)之透過率(T2(%))(即0＜T2＜30)且相位偏移量(ϕ2)為大致180度。所謂大致180度係指120～240度者。較佳為相位偏移量ϕ2為150～210度。
此處之透過率T2(%)亦與上述同樣地設為以透明基板10之透過率為基準時之上述代表波長之透過率。
於本實施形態之轉印用圖案中，若將輔助圖案2之寬度設為d(μm)，則於下述式(1)之關係成立時，可獲得顯著之效果。
0.5≦×d≦1.5・・・(1)
此時，若將主圖案1之中心與輔助圖案2之寬度方向之中心之距離設為狹縫間距L(μm)，則較佳為1.0＜L≦5.0，更佳為1.5＜L≦4.5。但，輔助圖案2構成介隔低透光部3包圍主圖案1之正八邊形帶之區域之至少一部分。因此，可以主圖案1與輔助圖案2不接觸之方式，即以於主圖案1之周圍且與輔助圖案2之間介置低透光部3為條件，決定狹縫間距L、及主圖案之直徑W1。
輔助圖案2之寬度d(μm)係設定為於適用於本實施形態之光罩之曝光條件(使用之曝光裝置)下，具有透過率T1之輔助圖案不進行解像。列舉具體之例，輔助圖案2之寬度d(μm)較佳為d≧0.7，更佳為d≧0.8。又，若與主圖案1之寬度W1(μm)比較，則較佳為d≦W1，更佳為d＜W1。
又，關於輔助圖案2之寬度d(μm)，上述式(1)中表示之關係式更佳為下述式(1)－1，進而較佳為下述式(1)－2。
0.7≦×d≦1.2 ・・・(1)－1
0.75≦×d≦1.0 ・・・(1)－2
本實施形態之轉印用圖案所具備之主圖案1之形狀為四邊形。具體而言，主圖案1之形狀例如較佳為設為正方形或長方形。於主圖案1之形狀為四邊形之情形時，該四邊形之重心位置與輔助圖案2之寬度方向之中心之距離成為狹縫間距L。
於本實施形態中，於定義在主圖案1之周邊包圍主圖案1之特定寬度之正八邊形帶時，輔助圖案2成為構成該正八邊形帶之至少一部分之圖案。所謂正八邊形帶係指外周及內周均為八邊形且大致固定寬度之形狀。如圖7所示，輔助圖案2係角部以外為固定寬度。定義輔助圖案2之正八邊形帶以包圍主圖案1之方式配置於主圖案1之周邊。而且，輔助圖案2之成為正八邊形帶之內輪廓、外輪廓之正八邊形之重心位於與主圖案1之重心相同之位置。於本實施形態中，為便於說明，而如圖8所示，將上述正八邊形帶劃分為與外周(或內周)之八邊形之各邊對應之8個區段。此處，將劃分為8個之區段中之圖8之右端之區段設為區段A，將與其鄰接之上側之區段設為區段H，將下側之區段設為區段B。亦即，自區段A起以順時針設為區段B、C、D、E、F、G、H。
如上述圖4所示，於對於第1主圖案1a將第2主圖案1b配置於接近之位置之情形時，於本實施形態中，於第1主圖案1a之周邊配置構成上述正八邊形帶之8個區段A～H中之面向第2主圖案1b側之一區段中具有缺損之形狀之輔助圖案2a。具體而言，如圖7(a)所示，於第1輔助圖案1a之周邊配置八邊形帶之一部分區段缺損之輔助圖案2a。如圖示般，隨附於第1主圖案1a之輔助圖案2a成為於面向第2主圖案1b之側、即於右端之區段(與上述區段A對應)具有缺損之形狀。又，隨附於第2主圖案1b之輔助圖案2b亦成為於面向第1主圖案1a側之一區段(與上述區段E對應)具有缺損之形狀。亦即，2個主圖案具有於位於相互對向之側之一區段分別具有缺損之八邊形帶形狀之輔助圖案。
即，於2個孔形成用圖案以特定距離以下之接近距離配置之情形時，各個主圖案所具備之輔助圖案於上述2個主圖案之間缺損。因此，若以直線將2個主圖案之重心連結(未圖示)，則該直線完全不穿過輔助圖案。
再者，較佳為，於被2個主圖案之相互對向之邊夾持之區域S(圖7(a)中虛線所示)內實質上不配置輔助圖案。但，於輔助圖案之一部分進入區域S之情形時，較佳為該部分不具有與2個主圖案之相互對向之邊平行之邊。於圖7所示之態樣中，輔助圖案之端部進入區域S內，但該端部僅具有對於2個主圖案之相互對向之邊傾斜之邊。
又，較佳為，於區域S內不存在島狀(被閉合之直線或曲線包圍之形狀)之輔助圖案。
進而，較佳為，上述區域S內之面積之90%以上包含低透光部。藉此，形成於被轉印體上之2個孔圖案之抗蝕圖案形狀變得良好，從而可抑制抗蝕殘膜厚度之損耗。
若以此方式配置一部分區段缺損之輔助圖案2a、2b，則因上述圖6(b)所示之凹部22導致之抗蝕膜厚之損耗如圖9所示般消除，成為具有良好之輪廓之抗蝕圖案形狀。再者，圖6(b)及圖9均為孔間距P為8 μm之情形。
圖6(b)中觀察之抗蝕膜厚之損耗之容許範圍可根據欲使用光罩獲得之顯示裝置之製造條件等而決定。若將抗蝕膜之初期膜厚(塗佈膜厚)設為100%，則其損耗之容許範圍設為初期膜厚之10%以下更佳為5%以下可謂良好之條件。
因此，於本實施形態中，是否使隨附於相互接近之主圖案1a、1b之輔助圖案2a、2b一部分缺損，只要預先藉由實驗或模擬等掌握抗蝕膜厚之損耗量，並基於其結果進行判斷即可。具體而言，以如下方式進行判斷較為有用，即，於抗蝕膜厚之損耗量例如超過10%之情形時，使輔助圖案2a、2b一部分缺損，於成為10%以下之情形時，不使輔助圖案2a、2b缺損。
進而，於對於第1主圖案1a將第2主圖案1b配置於接近之位置之情形時，較理想為於輔助圖案2a、2b彼此之距離D(參照圖4)成為1.0 μm以下之情形時，使配置於第1主圖案1a之周邊之輔助圖案2a之面向第2主圖案1b側之區段(上文所述之區段A)缺損。同樣地，較佳為於第2主圖案1b中亦形成使面向第1主圖案1a側之區段(上文所述之區段E)缺損之輔助圖案2b。進而，更佳為上述區段之缺損於輔助圖案2a、2b彼此之距離D成為1.5 μm以下時適用。
上述情況係鑒於顯示裝置用曝光裝置所具有之解像性能而例示者。
再者，所謂輔助圖案彼此之距離如圖4所示般係指對向之區段彼此之距離(垂線之長度)。因此，於如圖示般，2個主圖案1a、1b於某一方向相鄰地排列，且各個主圖案1a、1b分別由對應(隨附)之輔助圖案2a、2b包圍之情形時，於2個主圖案1a、1b之排列方向上，輔助圖案2a、2b之相對向之(相互面向之)邊彼此之距離成為輔助圖案彼此之距離D。
又，本實施形態之光罩於作為主圖案1a、1b之重心間距離之孔間距P有1.6 μm以上較佳為3 μm以上之情形時，可獲得本發明之顯著之效果。若孔間距P之值過小，則存在產生如下欠佳情況之風險，即，產生無法充分獲得形成於與兩主圖案間對應之位置之抗蝕圖案之殘膜量之情形，或下述遮罩偏差β2之數值變得過大而圖案設計變得困難等。
以下對本發明之實施例及參考例之光學模擬進行說明。
圖10係表示參考例之光罩所具備之轉印用圖案之主要部分之俯視模式圖，(a)表示參考例4，(b)表示參考例5，(c)表示參考例6，(d)表示參考例7，(e)表示參考例8。圖11係表示本發明之實施例之光罩所具備之轉印用圖案之主要部分之俯視模式圖，(f)表示實施例1，(g)表示實施例2，(h)表示實施例3，(i)表示實施例4。
又，圖10(a)表示主圖案1a、1b之孔間距P(參照圖5～圖7)為16 μm且正八邊形帶之輔助圖案2a、2b無區段之缺損之情形，圖10(b)表示主圖案1a、1b之孔間距P為12 μm且輔助圖案2a、2b無區段之缺損之情形。又，圖10(c)表示主圖案1a、1b之孔間距P為9 μm且輔助圖案2a、2b無區段之缺損之情形，圖10(d)表示主圖案1a、1b之孔間距P為8.75 μm且輔助圖案2a、2b無區段之缺損之情形。圖10(e)表示2個主圖案1a、1b之孔間距P為8.75 μm且將各個輔助圖案2a、2b之一區段結合使之共有之情形。
另一方面，圖11(f)表示主圖案1a、1b之孔間距P為8.75 μm且使輔助圖案2a、2b缺損一區段之情形，圖11(g)表示主圖案1a、1b之孔間距P為8 μm且使輔助圖案2a、2b缺損一區段之情形。又，圖11(h)表示主圖案1a、1b之孔間距P為7.5 μm且使輔助圖案2a、2b缺損一區段之情形，圖11(i)表示主圖案1a、1b之孔間距P為7 μm且使輔助圖案2a、2b缺損3區段之情形。
又，於該模擬中，將使用上述圖2(a)所示之光罩(二元遮罩)之情形設為參考例1，將使用上述圖2(b)所示之光罩(半色調式相位偏移遮罩)之情形設為參考例2。適用於參考例1及參考例2之主圖案均為無輔助圖案之孤立圖案。
對上述各個光罩之圖案進行光學模擬，結果獲得了如圖12所示之結果。於該模擬中，除參考例1及參考例2以外，以作為用於上述圖2(c)所示之主圖案(孤立圖案)之轉印之曝光能量之80 mJ/cm2
之劑量(Eop劑量)為基準，對適用該劑量時形成於被轉印體上之抗蝕圖案進行評價。再者，「panel X－CD」及「panel Y－CD」係與光罩之主圖案對應地形成於被轉印體上之孔圖案之X方向及Y方向之尺寸。再者，於各參考例及各實施例中，將X－CD及Y－CD之目標尺寸設定為1.5 μm。
首先，於圖10(a)之參考例4中，2個主圖案1a、1b充分地隔開，因此，各個主圖案1a、1b實質上發揮作為孤立圖案之光學性能。該方面於圖10(b)之參考例5或圖10(c)之參考例6中亦相同。另一方面，若如圖10(d)之參考例7般，2個主圖案1a、1b接近且孔間距P變窄為8.75 μm，則輔助圖案2a、2b彼此之距離D成為0.95 μm。此時，抗蝕膜厚之損耗超過12%。
此處，於如圖10(e)之參考例8般，將相互接近之主圖案1a、1b之輔助圖案2a、2b彼此局部地接合為一條(1區段)而使之共有之情形時，抗蝕膜厚之損耗亦接近12%，幾乎未見改善。
根據本發明者之研究，認為該問題與適用於顯示裝置製造之光阻相關。具體而言，以用於顯示裝置之製造之光阻(正型光阻)與半導體裝置製造用之光阻不同而具有較高之感度之方式進行設計。因此，於相對較低之劑量中亦未避免相應之減膜，容易於意外之部分產生局部之膜厚之損耗。
再者，於參考例3之光罩中產生之主圖案1與輔助圖案2之相互作用係利用透過輔助圖案2之反轉相位之光所形成之光學影像提高因主圖案1之透過光產生之光強度峰值之現象。而且，獲得了圖3所示之優異之轉印性能(DOF、MEEF)。另一方面，因透過輔助圖案之光而產生之光強度亦藉由與主圖案1之相互作用而略有增加。可認為由於用於顯示裝置製造用之光阻之感度較高，故而，除輔助圖案2彼此接近之情形以外， 若輔助圖案2被複數個主圖案1所共有，則會產生輔助圖案2之透過光使被轉印體上之抗蝕劑厚度減少之風險。
另一方面，於圖11(f)之實施例1中，使構成包圍第1主圖案1a之正八邊形帶之八區段中之面向第2主圖案1b側之一區段缺損，將具有剩餘之7區段之輔助圖案2a配置於第1主圖案1a之周邊。即，設為使包圍第1主圖案1a之周邊之正八邊形帶中之與包圍第2主圖案1b之周邊之正八邊形帶最接近且相向之部分之區段缺損之輔助圖案2a之形狀。又，對於第2主圖案1b亦同樣地，使面向第1主圖案1a側之一區段缺損，將具有剩餘之7區段之輔助圖案2b配置於第2主圖案1b之周邊。
再者，於使輔助圖案之區段缺損時，無需按照圖8所示之區段之交界線切取，例如，如圖7(a)所示般，只要至少使該區段之主要部分缺損即可。關於缺損部分之面積，較佳為，例如，以若係使八區段中之一區段缺損之情形，則為一區段量之面積之80%以上方式，設為使之缺損之區段個數量之面積之80%以上。此時，於2個主圖案之間產生僅介置低透光部3之部分，連結兩主圖案1a、1b之重心之直線不會穿過輔助圖案。
如此，可知於使接近之兩主圖案之輔助圖案中之相互對面之區段缺損之情形時，於所形成之抗蝕圖案中，抗蝕膜厚之損耗變為零等，大幅地減少損耗，而可獲得顯著之效果。又，可知與其同樣之效果於如圖11(g)之實施例2或圖11(h)之實施例3般使兩主圖案1a、1b進而接近之情形(孔間距P＝8 μm、P＝7.5 μm之情形)時亦可獲得。該情形之抗蝕圖案之剖面構造與圖9相同。
然而，於圖11(f)～(i)之實施例中，儘管適用與上述參考例4～8相同之曝光劑量，轉印於抗蝕膜之孔圖案之直徑亦較初期之目標尺寸略微變小。具體而言，就實施例1而言，相對於作為初期之目標尺寸之1.5(μm)，X－CD(X方向之直徑)成為1.39(μm)，Y－CD(Y方向之直徑)成為1.37(μm)。因此，為形成孔圖案，以使被轉印體上之X－CD、Y－CD均接近目標尺寸(1.5 μm)，較理想為使上述遮罩偏差β1大於0.5 μm。
進而，為了將被轉印體上之X－CD及Y－CD同等地設為目標尺寸(1.5 μm)，較佳為於X方向及Y方向分別賦予適當之遮罩偏差β2。
因此，如圖12所示，於圖11(f)～(h)之實施例中，藉由對於遮罩上之CD賦予適當之遮罩偏差β2(x)、β2(y)，而於被轉印體上，X－CD、及Y－CD均取得了作為目標尺寸之1.5 μm。
又，其結果，可知於圖11(f)～(h)之實施例之光罩之情形時，DOF(焦點深度)數值(24)較參考例1之二元遮罩(圖2(a))或參考例2之半色調式相位偏移遮罩(圖2(b))大，而可於被轉印體上穩定地形成所需之直徑之孔圖案。又，由於曝光所需之Eop劑量相較於參考例1之二元遮罩或參考例2之半色調式相位偏移遮罩之情形小，故而可有效率地進行曝光步驟。
再者，於圖11(f)～(h)之實施例中，使隨附於第1主圖案1a之輔助圖案2a之八區段與隨附於第2主圖案1b之輔助圖案2b之八區段中之相互對向之一區段缺損。亦即，對於1個主圖案1，將輔助圖案2之區段數設為7。另一方面，於使兩主圖案1a、1b進而接近配置之情形時，亦能以已使之缺損之區段為中心，使位於其兩側之區段進而缺損。例如，於圖11(i)之實施例4中，使隨附於第1主圖案1a之輔助圖案2a之八區段中之區段A、及位於其兩側之區段B、H缺損，並且使隨附於第2主圖案1b之輔助圖案2b之八區段中之區段E、及位於其兩側之區段D、F缺損，藉此，將抗蝕膜厚之損耗設為零。結果，此處，對1個主圖案1使3個區段缺損，藉此，將輔助圖案2之區段數設為5。進而，相互接近配置之主圖案之個數並不限定於2，可將更多數個孔形成圖案配置於接近之距離，於此情形時，亦可與上述同樣地設計區段之缺損。
於包含接近配置之複數個主圖案(分別至少與其他任一個接近)之圖案群中，於將所包含之主圖案之個數設為N，將輔助圖案之區段總數設為K時，
可設為K≦(8－1)N。
再者，於圖11(f)～(i)中，僅例示了第2主圖案1b對於第1主圖案1a於X方向(圖之左右方向)接近之情形，但於在Y方向上接近之情形時，亦可與上述同樣地配置使八區段中之任一塊缺損之輔助圖案2。
進而，於第2主圖案1b對於第1主圖案1a自主圖案1之對角線方向接近之情形時等自傾斜方向接近之情形時，亦可有效地適用本發明。於此情形時，亦只要使隨附於第1主圖案1a之輔助圖案2a之八區段中之面向第2主圖案1b側之至少一區段缺損即可。
又，於使第3主圖案、進而第4主圖案對於第1主圖案接近配置之情形時，亦可與上述同樣地，使隨附於各個主圖案之輔助圖案之八區段中之相互對向之側之區段缺損。於此情形時，根據複數個主圖案之相對之配置，除具有七區段之輔助圖案之主圖案、具有五區段之輔助圖案之主圖案以外，亦可存在具有六區段、四區段、三區段、二區段、進而一區段之輔助圖案之主圖案。
例如，可設為2個主圖案相互具有逐個缺損1區段之輔助圖案之圖案群、或相互具有逐個缺損3區段之輔助圖案之圖案群。
或者，可設為3個主圖案根據其排列(X方向、Y方向、或X及Y方向、以下同樣)分別具有缺損1或2區段之輔助圖案之圖案群。進而，亦可設為4個主圖案根據其排列分別具有缺損1～5區段之輔助圖案之圖案群。
以下，亦可設為5個或6個或7個主圖案根據其排列分別具有缺損1～6區段之輔助圖案之圖案群，或設為8個主圖案根據其排列分別具有缺損1～7區段之輔助圖案之圖案群。
另一方面，1個主圖案所具有之輔助圖案可設為除面向另一主圖案之側之區段、或其兩側之區段以外無缺損者。
又，有時根據複數個孔形成用圖案之接近方向，形成於被轉印體上之孔圖案之直徑於X方向及Y方向上成為不同數值。其原因在於：對於X方向及Y方向不均等地產生因輔助圖案之一部分區段之缺損而產生之光學影像之變化。因此，出於補償對X方向及Y方向之不均等之光學影像之影響之目的，對圖案之描繪資料賦予在X方向及Y方向不同之尺寸之遮罩偏差β2較為有用。
例如，如上述圖7(a)所示，於第1主圖案1a與第2主圖案1b排列於X方向，且使隨附於兩主圖案1a、1b之輔助圖案2a、2b中之相互面向之區段(即，於Y方向延伸之區段)缺損之情形時，於關於對第1主圖案1a及第2主圖案1b之尺寸賦予之遮罩偏差β2(μm)，將X方向之賦予量設為β2(x)、將Y方向之賦予量設為β2(y)時，可設為β2(y)＞β2(x)。
具體而言，自主圖案觀察，於與存在輔助圖案之缺損部之方向(圖7(a)中為X方向)垂直之方向(圖7(a)中為Y方向)賦予正值之偏差β2(y)。進而，可視需要，於存在輔助圖案之缺損部之方向(圖7(a)中為X方向)賦予負值之偏差β2(x)。
將包含如此般賦予有遮罩偏差β2之主圖案1a、1b之光罩之轉印用圖案例示於圖13。此處，賦予上述遮罩偏差β2之結果，主圖案1a、1b之X方向之尺寸小於Y方向之尺寸，主圖案1a、1b之形狀成為縱長之長方形。
於圖11(g)之實施例中，作為正方形之主圖案1a、1b之轉印圖像形成於被轉印體上之孔圖案成為橫長之長方形(X－CD＝1.40 μm、Y－CD＝1.37 μm)。因此，若藉由遮罩偏差之賦予而將主圖案1a、1b之形狀設為縱長之長方形，則可消除因輔助圖案2a、2b之一部分區段之缺損導致之圖案尺寸之誤差。其結果，可於被轉印體上形成X方向之尺寸與Y方向之尺寸相等之孔圖案。
因此，藉由光學模擬，對X方向、Y方向之各者求出用於X方向與Y方向之尺寸在被轉印體上成為相等之偏差量β2，並將其反映至圖案描繪資料即可。
因此，於光罩所具有之轉印用圖案中，賦予了偏差β2之主圖案成為長方形。即，第1主圖案成為於面向處於接近之位置之第2主圖案之側具有長邊之長方形。
就如上述圖13之排列例而言，主圖案之長邊W3(y)成為W3(y)＝W1＋β2(y)，短邊W3(x)成為W3(x)＝W1＋β2(x)。而且，W3(x)、及W3(y)較佳為滿足下式。
對於長邊
為0.8≦W3(y)≦4.0，更佳為1.0≦W3(y)＜3.5，
對於短邊，
為0.8≦W3(x)≦4.0，更佳為1.0≦W3(x)≦3.0。
如上所述，於將本實施形態之光罩用作顯示裝置製造用光罩之情形時，即，於將本實施形態之光罩與顯示裝置製造用光阻組合使用之情形時，可大幅地減少於被轉印體上與輔助圖案對應之部分之抗蝕膜厚之損耗。
＜光罩之製造方法＞
其次，以下，參照圖14(a)～(f)對能夠適用於本發明之實施形態之光罩之製造方法之一例進行說明。再者，於圖14(a)～(f)中，左側表示剖視圖，右側表示俯視圖，並且為了簡化，光罩圖案形狀僅示出第1主圖案及隨附於第1主圖案之輔助圖案。
首先，如圖14(a)所示，準備光罩基底30。於該光罩基底30中，於包含玻璃等之透明基板10上依序形成有相位偏移膜11及低透光膜12，進而塗佈有第1光阻膜13。
相位偏移膜11係形成於透明基板10之主表面上。相位偏移膜11於將i線、h線、g線之任一者設為曝光之光之代表波長時，對於該代表波長之透過率T1(%)較佳為20～80(%)，更佳為30～75(%)，進而較佳為40～75(%)。又，相位偏移膜11對於上述代表波長之相位偏移量為大致180度。藉由此種相位偏移膜11，可將包含透光部之主圖案與包含相位偏移部之輔助圖案之間之透過光之相位差設為大致180度。此種相位偏移膜11使處於i線～g線之波長範圍之代表波長之光之相位偏移大致180度。作為相位偏移膜11之成膜方法，可適用濺鍍法等公知之方法。
較理想為相位偏移膜11滿足上述透過率及相位差，且如下所述，由能夠進行濕式蝕刻之材料形成。但，若於濕式蝕刻時產生之側蝕刻之量變得過大，則會產生CD精度之劣化或因底切導致之上層膜之破壞等欠佳情況。因此，相位偏移膜11之膜厚較佳為設為2000 Å以下，較佳為300～2000 Å，更佳為300～1800 Å。
又，為了滿足該等條件，相位偏移膜11之材料之曝光之光所包含之代表波長(例如h線)之折射率較佳為1.5～2.9，更佳為1.8～2.4。
進而，為了充分地發揮相位偏移效果，較佳為，利用濕式蝕刻所得之圖案剖面(被蝕刻面)相對於透明基板10之主表面接近垂直。
於考慮上述性質時，作為相位偏移膜11之材料，可使用包含Zr、Nb、Hf、Ta、Mo、Ti之任一種及Si之材料或包含該等材料之氧化物、氮化物、氮氧化物、碳化物、或碳氮氧化物之材料。
於相位偏移膜11上形成低透光膜12。作為低透光膜12之成膜方法，與相位偏移膜11之情形同樣地，可適用濺鍍法等公知之方法。又，較佳為相位偏移膜11所具有之相位偏移量之波長依存性對於i線、h線、及g線，變動幅度為40度以內。
低透光膜12可包含實質上不使曝光之光透過之遮光膜。又，除此以外，亦可由對於曝光之光之代表波長具有特定之較低之透過率之膜構成低透光膜12。用於本實施形態之光罩之製造之低透光膜12對於處於i線～g線之波長範圍之代表波長之光具有較相位偏移膜11之透過率T1(%)低之透過率T2(%)。
於低透光膜12以較低之透過率使曝光之光透過之情形時，要求低透光膜12對於曝光之光之透過率及相位偏移量可達成本實施形態之光罩之低透光部之透過率及相位偏移量。較佳為，於相位偏移膜11與低透光膜12之積層狀態下，對於曝光之光之代表波長之光之透過率T3(%)為T3≦20，且相位偏移量ϕ3較佳為90(度)以下，更佳為60(度)以下。
作為低透光膜12之單獨之性質，較佳為實質上不使上述代表波長之光透過者或具有未達30(%)之透過率(T2(%))(即，0＜T2＜30)，且相位偏移量(ϕ2)為大致180度。所謂大致180度，係指120～240度。較佳為相位偏移量ϕ2為150～210(度)。
低透光膜12之材料可為Cr或其化合物(氧化物、氮化物、碳化物、氮氧化物、或碳氮氧化物)，或者，亦可為含有Mo、W、Ta、Ti之金屬之矽化物或該矽化物之上述化合物。但，低透光膜12之材料較佳為能夠與相位偏移膜11同樣地進行濕式蝕刻且對於相位偏移膜11之材料具有蝕刻選擇性之材料。即，較理想為低透光膜12對於相位偏移膜11之蝕刻劑具有耐性，相位偏移膜11對於低透光膜12之蝕刻劑具有耐性。
於低透光膜12上塗佈第1光阻膜13。本實施形態之光罩較佳為藉由雷射描繪裝置而描繪，因此，設為適於雷射描繪裝置之光阻。構成第1光阻膜13之光阻可為正型亦可為負型，但以下設為正型之光阻進行說明。
其次，如圖14(b)所示，對於第1光阻膜13，使用描繪裝置，進行利用基於轉印用圖案之描繪資料之描繪(第1描繪)。繼而，將藉由顯影而獲得之第1抗蝕圖案13p作為遮罩，對低透光膜12進行濕式蝕刻，藉此，形成低透光膜圖案12p。於此階段，劃定成為低透光部之區域，並且劃定由低透光部包圍之輔助圖案(低透光膜圖案12p)之區域。用於濕式蝕刻之蝕刻液(濕式蝕刻劑)可使用適合所使用之低透光膜12之組成之公知者。例如，若低透光膜12係含有Cr之膜，則作為濕式蝕刻劑，可使用硝酸鈰銨等。
其次，如圖14(c)所示，將第1抗蝕圖案13p剝離。藉此，使低透光膜圖案12p、及相位偏移膜11之一部分露出。
其次，如圖14(d)所示，於包含低透光膜圖案12p之整面塗佈第2光阻膜14。
其次，如圖14(e)所示，於對第2光阻膜14進行第2描繪之後，藉由顯影而形成第2抗蝕圖案14p。其次，將第2抗蝕圖案14p及低透光膜圖案12p作為遮罩，對相位偏移膜11進行濕式蝕刻。藉由該蝕刻(顯影)，透明基板10之主表面作為透光部露出，藉此，劃定包含透光部之主圖案之區域。
再者，第2抗蝕圖案14p係覆蓋成為輔助圖案之區域，且於包含透光部之主圖案之區域具有開口者。於此情形時，較佳為以於較第2抗蝕圖案14p之開口緣更靠內側，露出低透光膜圖案12p之邊緣部分之方式，對第2描繪之描繪資料進行定尺寸。如此一來，可吸收於第1描繪與第2描繪之間產生之對準偏移，而防止轉印用圖案之CD精度之劣化。
即，若於第2描繪時進行第2抗蝕圖案14p之定尺寸，則於欲在被轉印體上形成孤立之孔圖案之情形時，不會於相位偏移膜11與低透光膜12之圖案化時產生位置偏移。因此，於如圖1所例示之轉印用圖案中，可使主圖案1及輔助圖案2之重心精密地一致。
用於相位偏移膜11之蝕刻之濕式蝕刻劑係根據相位偏移膜11之組成而適當選擇。
其次，如圖14(f)所示，將第2抗蝕圖案14p剝離。藉此，完成具備轉印用圖案之光罩。再者，圖14中示出了形成無區段之缺損之正八邊形之輔助圖案之情形，但於使八區段中之任一者缺損之情形時，只要於在圖14(b)中劃定輔助圖案之區域時，根據使之缺損之區段之位置及大小變更描繪資料即可。
於上述光罩之製造中，於能夠在使相位偏移膜11或低透光膜12等光學膜圖案化時適用之蝕刻中存在乾式蝕刻或濕式蝕刻。可採用其等之任一個，但於本發明中，濕式蝕刻尤其有利。其原因在於：顯示裝置製造用光罩之尺寸相對較大，進而，存在多種尺寸。於製造此種光罩時，若適用需要真空腔室之乾式蝕刻，則會導致乾式蝕刻裝置之大型化或製造步驟之效率降低。
但，亦存在伴隨著於製造此種光罩時適用濕式蝕刻之課題。濕式蝕刻具有各向同性蝕刻之性質，因此，於在深度方向蝕刻特定之膜而使之溶出時，於對於深度方向垂直之方向亦進行蝕刻。例如，於蝕刻膜厚為F(nm)之相位偏移膜11而形成狹縫時，成為蝕刻遮罩之抗蝕圖案之開口較所需之狹縫寬度小2F(nm)(即，單側F(nm))，但越成為微細寬度之狹縫，則越不易維持抗蝕圖案開口之尺寸精度。因此，將輔助圖案之寬度d設為1 μm以上、較佳為1.3 μm以上較為有用。
又，於上述膜厚F(nm)較大之情形時，側蝕刻量亦變大，因此，使用即便膜厚較小亦具有大致180度之相位偏移量之膜材料較有利。因此，較理想為對於曝光之光之代表波長，相位偏移膜11之折射率較高。具體而言，較佳為使用如對於上述代表波長之折射率較佳為1.5～2.9、更佳為1.8～2.4之材料，形成相位偏移膜11。
本發明包含顯示裝置之製造方法，該顯示裝置之製造方法包含使用本實施形態之光罩藉由曝光裝置而進行曝光，而於被轉印體上轉印上述轉印用圖案之步驟。
本發明之顯示裝置之製造方法係首先準備本實施形態之光罩。其次，使用數值孔徑(NA)為0.08～0.15且具有包含i線、h線、g線之曝光光源之曝光裝置，對上述轉印用圖案進行曝光，而於被轉印體上形成直徑W2為0.8～3.0(μm)之孔圖案。通常於曝光時適用等倍曝光，而較有利。
於使用本實施形態之光罩轉印轉印用圖案時，亦可使用縮小曝光，但作為用於顯示裝置製造用光罩之曝光機，係進行等倍之投影曝光之方式，較佳為以下者。
例如，光學系統之數值孔徑(NA)較理想為0.08≦NA＜0.20，更佳為0.08≦NA≦0.15，進而較理想為0.08＜NA＜0.15。又，同調因子σ為0.4≦σ≦0.9，更佳為0.4＜σ＜0.7，進而較佳為0.4＜σ＜0.6。
曝光光源係使用曝光之光中包含i線、h線及g線之至少一種之光源。於適用單一波長之曝光之光之情形時，較佳為使用i線。另一方面，使用i線、h線、g線均包含之光源(亦稱為寬波長光源)於確保足夠之光量方面較有用。
又，所使用之曝光裝置之光源亦可使用斜光照明(環狀照明等)，但藉由使用不適用斜光照明之通常照明，可充分獲得本發明之優異之效果。
根據本發明，於使用顯示裝置製造用遮罩之顯示裝置之製造方法中，即便係微細之密集圖案，亦可穩定地進行向被轉印體上之轉印。具體而言，可一面確保DOF或MEEF等製造時之製程之裕度(Process Margin)，一面精密地形成孔圖案。進而，於形成密集圖案時，可充分地確保形成於被轉印體上之抗蝕圖案之厚度。其於顯示裝置生產上提高CD精度且提供穩定生產、高良率等產業上之利益。[Design of Mask for Forming Hole Pattern with Auxiliary Pattern] FIG. 1 shows the main part of the pattern for transfer of the mask described in Patent Document 1, (a) is a schematic plan view, (b) is (a) ) is a schematic cross-sectional view of the AA position. The pattern for transfer of the photomask shown in the figure has a main pattern 1 including a light-transmitting portion, and an auxiliary pattern 2 arranged around the main pattern 1 around the main pattern 1 . The auxiliary pattern 2 is arranged around the main pattern 1 along with the main pattern 1 . Moreover, the phase shift film 11 and the low light transmission film 12 are formed on the transparent substrate 10 . The main pattern 1 includes a light-transmitting portion 4 exposing the transparent substrate 10 , and the auxiliary pattern 2 includes a phase shifting portion 5 exposing the phase shifting film 11 on the transparent substrate 10 . In addition, the low light-transmitting portion 3 surrounds the main pattern 1 and the auxiliary pattern 2, respectively. The low light transmission portion 3 includes a lamination film of the phase shift film 11 and the low light transmission film 12 formed on the transparent substrate 10 . The low light transmittance portion 3 may also include a single-layer film of the low light transmittance film 12 formed on the transparent substrate 10 . That is, the low-light-transmitting portion 3 includes at least a portion where the low-light-transmitting film 12 is formed. In the pattern for transfer shown in the figure, the region other than the region in which the main pattern 1 and the auxiliary pattern 2 are formed becomes the low light transmittance portion 3 . The phase shift film 11 has a phase shift amount that shifts the exposure light of the representative wavelength in the wavelength range of the i-line to the g-line by approximately 180 degrees. That is, the auxiliary pattern 2 has a function of inverting the phase of the transmitted light by the phase shift film 11 . In addition, the phase shift film 11 has a transmittance of T1 (%) with respect to the exposure light of the above-mentioned representative wavelength. According to Patent Document 1, after an optical simulation of forming a hole pattern on a transfer target body using a photomask having the above-mentioned pattern for transfer, compared with a phase shift mask that does not include a binary mask or an auxiliary pattern, it is obvious that It has excellent performance in Eop (exposure light amount required to form a pattern of target size on the transferred body) or DOF (Depth of Focus: Depth of Focus). In order to form a finer pattern on a to-be-transferred body, the present inventors carried out optical simulation of the photomask shown in FIG.2(a)-(c). Here, the photomask shown in Fig. 2(a) is referred to as reference example 1, the photomask shown in Fig. 2(b) is referred to as reference example 2, and the photomask shown in Fig. 2(c) is referred to as reference example 3. Next, using each photomask of Reference Examples 1 to 3 having a main pattern including a hole pattern with a diameter W1 of 2.0 μm, a positive-type photoresist was formed on a transfer target body (display panel substrate, etc.) corresponding to the diameter W2. Simulation of a transferred image of a hole pattern (here 1.5 μm). Further, as described above, the diameter W1 on the mask is set to be W1≧W2 (preferably W1>W2) with respect to the target diameter W2 on the transfer target body. Here, if the mask deviation β1 (μm) is assumed to be β1=W1−W2, β1 is assumed to be 0.5 (μm) here. The conditions of the simulation are as follows. (Reference Example 1) In Reference Example 1, as shown in FIG. 2( a ), a photomask including a binary mask is used, and a photomask with a diameter W1 = 2 μm surrounded by a low-transmittance portion (light-shielding portion) 3 is used. The hole pattern of the square light-transmitting part is set as the main pattern 1 . (Reference Example 2) In Reference Example 2, as shown in Figure 2(b), a photomask including a halftone phase shift mask was used, and the transmittance of the light to be exposed was 5.2% and the phase shift amount was 5.2%. A hole pattern including a square light-transmitting portion with a diameter W1=2 μm surrounded by the phase shift portion 5 of 180 degrees is set as the main pattern 1 . (Reference Example 3) In Reference Example 3, as shown in FIG. 2( c ), a photomask including a phase shift mask with an auxiliary pattern is used, and a square light-transmitting portion with a diameter of W1 = 2 μm is used. The hole pattern is set as the main pattern 1 and the auxiliary pattern 2 of the regular octagonal band surrounds the periphery of the pattern 1. In addition, the auxiliary pattern 2 includes a phase shift portion with an exposure light transmittance of 45% and a phase shift of 180 degrees, and the area other than the main pattern 1 and the auxiliary pattern 2 includes a low transmittance with an optical density of OD≧2. part (shading part) 3. The distance (L) between the center of the main pattern 1 and the center position in the width direction of the auxiliary pattern 2 was set to 3.25 μm, and the width (d) of the auxiliary pattern 2 was set to 1.3 μm. The photomask of Reference Example 3 was designed based on the structure described in Patent Document 1. Using each of the masks corresponding to the above-mentioned Reference Examples 1 to 3, a hole pattern having a width of W2 = 1.5 μm was formed on the transfer target body. The simulated exposure conditions are as follows. The optical system coefficient value aperture NA of the exposure device is 0.1, and the coherence factor σ is 0.5. In addition, as the exposure light source, a light source including all i-lines, h-lines, and g-lines (broad-wavelength light sources) was used, and the intensity ratio was set to g:h:i=1:1:1. The optical evaluation items of the photomask are as follows. (1) Depth of focus (DOF) Depth of focus (DOF) is used on the transferred body to make the CD variation relative to the target CD within a specific range (eg ±10%) when defocus occurs during exposure. The depth of focus within, and ideally its larger value. If the DOF value is high, the flatness of the transfer object (eg, a panel substrate for a display device) is not easily affected, so that a fine pattern can be reliably formed, and the CD unevenness can be suppressed. In the simulation of this case, as the DOF value, the target CD±10% was set as the standard. Here, the so-called CD is an abbreviation of critical dimension (Critical Dimension), and is used in the meaning of pattern width. The photomask for display device manufacturing is larger in size than the photomask for semiconductor device manufacturing, and the transferred body (display panel substrate, etc.) is also large in size, so it is difficult to obtain complete flatness. Therefore, it is difficult to increase the DOF value. The mask is of great significance. (2) Mask Error Enhancement Factor (MEEF: Mask Error Enhancement Factor) The mask error enhancement factor is a numerical value representing the ratio of the CD error of the pattern formed on the transferred body to the CD error on the mask. The lower it is, the more the CD error of the pattern formed on the transferred body can be reduced. As the specifications of display devices evolve, pattern miniaturization is required, and a mask with a pattern size close to the resolution limit of an exposure device is required. Therefore, in the mask for display device manufacturing, the possibility of paying more attention to MEEF in the future is also high. . (3) Eop Eop (hereinafter, also referred to as "Eop dose") is a particularly important evaluation item in a photomask for display device manufacturing. Eop is the amount of exposure light required to form the desired pattern size on the transferred body. The size of the photomask used in the manufacture of the display device is extremely large (for example, one side of the main surface is a square or a rectangle of about 300-2000 mm). Therefore, if a photomask with a lower Eop value is used, the speed of scanning exposure can be increased, and the production efficiency can be improved. The specific evaluation results for the above evaluation items are shown in FIG. 3 . First, in terms of Eop, the photomask of Reference Example 3 has a smaller exposure (Eop value) than the photomasks of Reference Example 1 and Reference Example 2 to obtain a hole pattern of the target size (Eop value) by more than 30%. Therefore, it can be seen that if the photomask of Reference Example 3 is used, higher production efficiency can be obtained. Moreover, the photomask of Reference Example 3 had higher DOF value and lower MEEF value than the photomasks of Reference Example 1 and Reference Example 2. Therefore, it can be seen that the photomask of Reference Example 3 is also extremely advantageous in DOF or MEEF. [Design of hole patterns at positions close to each other] On the other hand, if the resolution required for the display device is increased, the integration degree is increased due to the increase in the number of pixels per unit area. Therefore, it is necessary to arrange a plurality of main patterns (hole patterns) in close proximity to each other as a pattern for transfer of a mask for manufacturing a display device. Hereinafter, the case where the photomask of the above-mentioned Reference Example 3 is applied to the formation of a dense pattern including a plurality of main patterns will be examined. FIG. 4 shows a situation in which the patterns applied to the above-mentioned Reference Example 3 (the combined main pattern and the auxiliary pattern formed on the periphery thereof, also referred to as the pattern for hole formation in this specification) are arranged close to each other. Here, one of the two main patterns is referred to as the first main pattern 1a, and the other is referred to as the second main pattern 1b. Next, when the position of the second main pattern 1b is made to approach the first main pattern 1a from the direction of the arrow, the hole pitch P, which is the distance between the centers of gravity of the two main patterns 1a, 1b, and the resistance formed on the transfer object are examined. The cross-sectional shape of the etched pattern. First, as shown in FIG. 5( a ), when the first main pattern 1 a and the second main pattern 1 b are sufficiently spaced apart (P=12 μm), as shown in FIG. The upper resist pattern (here, a pattern including a positive-type photoresist) 20 forms hole patterns 21a, 21b corresponding to the respective main patterns 1a, 1b. On the other hand, as shown in FIG. 6( a ), when the second main pattern 1b is close to the first main pattern 1a (P=8 μm), the auxiliary patterns arranged around the main patterns 1a and 1b 2a, 2b become extremely close to each other. At this time, if the cross section of the resist pattern 20 formed on the transfer target body is observed, as shown in FIG. 6(b), except for the hole patterns 21a, 21b corresponding to the main patterns 1a, 1b, the same The concave portion 22 is formed in the middle portion, so that a large loss of the resist film thickness occurs. Such local reduction of the resist residue may adversely affect the processing stability of the display device substrate using the resist pattern as an etching mask. Therefore, the present inventors have studied the pattern for hole formation of Reference Example 3, which has favorable characteristics in terms of Eop and DOF, but has the problem of local reduction of the resist residue as described above, and has studied that the defect can be eliminated. A mask of good conditions. <Structure of a photomask> Next, the structure of the photomask for display device manufacture which concerns on embodiment of this invention is demonstrated. A photomask for manufacturing a display device according to an embodiment of the present invention has a pattern for transfer on a transparent substrate, and the pattern for transfer includes: a main pattern including a quadrangular light-transmitting portion; an auxiliary pattern arranged on the above-mentioned The periphery of the main pattern, including a phase shift portion; and a low-transmittance portion, which is formed in an area other than the main pattern and the auxiliary pattern; a regular eight sides surrounding the specific width of the main pattern in the periphery of the main pattern that defines the main pattern When forming a belt, the auxiliary pattern constitutes at least a part of the regular octagonal belt, and is different from the first main pattern when one of the plurality of main patterns included in the transfer pattern is used as the first main pattern The second main pattern is arranged at a position close to the first main pattern, and one of the eight sections of the regular octagonal band that constitutes the above-mentioned regular octagonal band surrounding the first main pattern has a defect in the section facing the second main pattern side. The auxiliary pattern is arranged on the periphery of the above-mentioned first main pattern. Hereinafter, it demonstrates concretely using FIG. 7. FIG. 7 shows the main part of the pattern for transfer provided in the photomask for manufacturing a display device according to the embodiment of the present invention, (a) is a schematic plan view, and (b) is a schematic cross-sectional view at the BB position of (a) picture. The above-mentioned photomask for display device manufacturing (hereinafter also simply referred to as a "mask") includes, for example, a transfer film formed by patterning the phase shift film 11 and the low light transmittance film 12 formed on the transparent substrate 10, respectively. pattern. The pattern for transfer includes a main pattern 1 ( 1 a , 1 b ) and an auxiliary pattern 2 ( 2 a, 2 b ) arranged around the main pattern 1 . The auxiliary pattern 2 has a shape that constitutes at least a part of the regular octagonal strip when defining a regular octagonal strip of a specific width surrounding the main pattern 1 . Two main patterns 1 are arranged in the X direction, and one of them is the first main pattern 1a, and the other is the second main pattern 1b. The auxiliary pattern 2a is arrange|positioned at the periphery of the 1st main pattern 1a, and the auxiliary pattern 2b is arrange|positioned at the periphery of the 2nd main pattern 1b. In addition, in FIG.7(a), although the main pattern on the left side is set as a 1st main pattern, and the main pattern on the right side is set as a 2nd main pattern, any one may be set as a 1st main pattern. In this embodiment, the main pattern 1 includes a light-transmitting portion 4 exposing the transparent substrate 10 , and the auxiliary pattern 2 includes a phase shifting portion 5 exposing the phase shifting film 11 on the transparent substrate 10 . In addition, the region other than the main pattern 1 and the auxiliary pattern 2 becomes the low-light-transmitting portion 3 in which at least the low-light-transmitting film 12 is formed on the transparent substrate 10 . In the present embodiment, the low light transmission portion 3 is formed by laminating the phase shift film 11 and the low light transmission film 12 on the transparent substrate 10 . The phase shift film 11 has a phase shift amount that shifts the exposure light of the representative wavelength in the wavelength range of the i-line to the g-line by approximately 180 degrees. Moreover, the transmittance|permeability of the phase shift film 11 with respect to the exposure light of the said representative wavelength is T1 (%). The low-transmittance film 12 can be set to have a specific low transmittance for a representative wavelength of exposure light. In this embodiment, the low transmittance film 12 can have a lower transmittance T2 ( %). Alternatively, the low-light-transmitting film 12 may be a light-shielding film that does not substantially transmit the exposure light. Here, when a fine pattern (a hole pattern) corresponding to the main pattern 1 of the mask is formed on the transfer target body by exposure using a mask, the diameter W1 of the main pattern 1 is set to 4 A fine pattern with a diameter of W2 (μm) (wherein W1≧W2, more preferably W1>W2) can be formed on the transfer target body under μm. Specifically, the diameter W1 (μm) of the main pattern 1 is preferably 0.8≦W1≦4.0, more preferably 1.0≦W1≦3.5. Furthermore, 1.2<W1≦3.0 can be set, and when it is necessary to make it finer, 1.2<W1<2.5 can be set. Moreover, the diameter W2 (μm) of the hole pattern formed on the transfer target body as the transfer image of the main pattern 1 is preferably 0.8≦W2≦3.0, more preferably 0.8≦W2≦2.5, and still more preferably 0.8 ≦W2≦2.0 or 0.8≦W2≦1.8. Alternatively, it may be set to 0.8<W2<2.0 or 0.8<W2<1.8. When it is necessary to make it finer, it can be set to 0.8<W2<1.5. In addition, the photomask of this embodiment can be used for the purpose of forming a pattern of a fine size useful for the manufacture of a display device. For example, when the diameter W1 of the main pattern 1 is 3.0 (μm) or less, a more significant effect can be obtained. However, in the mask (for example, the above-mentioned Reference Examples 1-3) which needs to form a hole pattern in the to-be-transferred body, it is advantageous to give the mask deviation β1 as described above. That is, if the diameter of the hole pattern on the photomask is W1 and the diameter of the hole pattern formed on the transfer target body is W2, W1=W2, but preferably W1>W2. For example, if β1 (μm) is the deviation value (W1−W2), and β1>0 (μm), the deviation value β1 (μm) is preferably 0.2≦β1≦1.0, more preferably 0.2≦β1≦0.8 . In this way, by specifying the relationship between the diameter W1 and the diameter W2 as the deviation value β1, advantageous effects such as reducing the loss of the film thickness of the resist pattern on the transfer target body can be obtained. Furthermore, the main pattern 1 includes a quadrilateral pattern, and the diameter W1 of the main pattern 1 is the size of one side of the quadrilateral. For example, if the main pattern 1 is a square pattern, the diameter W1 of the main pattern 1 refers to the size of one side of the square, and if the main pattern 1 is a rectangular pattern, the diameter W1 refers to the size of the long side. In addition, the shape of the main pattern 1 means the shape when the main pattern 1 is planarly viewed. Furthermore, the diameter W2 of the hole pattern formed on the body to be transferred refers to the length of the part where the distance between the two opposing sides is the largest. The mask deviation β1 described above can also be applied to the mask of this embodiment. In this embodiment, since the two hole forming patterns are formed close to a specific distance, the size given as the mask deviation β1 may become larger than the isolated patterns (reference examples 1 to 3 mentioned above). Furthermore, it is preferable that the mask of the present embodiment is produced in a case where mask deviations of unequal size are given with respect to the X direction and the Y direction perpendicular thereto according to the positional relationship of the two closely arranged hole forming patterns. Therefore, let this mask deviation be β2, and let the amount of deviation given with respect to the X direction and the Y direction perpendicular to it be β2(x) and β2(y), respectively. Details of the mask deviation β2 are described below. The phase difference ϕ between the main pattern 1 and the auxiliary pattern 2 is approximately 180 degrees with respect to the representative wavelength of the exposure light used for the exposure of the photomask of the present embodiment having the above-described pattern for transfer. That is, the phase difference ϕ1 of the light of the representative wavelength transmitted through the main pattern 1 and the light of the representative wavelength transmitted through the auxiliary pattern 2 is approximately 180 degrees. The so-called approximately 180 degrees means 120 to 240 degrees. The aforementioned phase difference ϕ1 is preferably 150 to 210 degrees. Furthermore, the photomask of this embodiment is more effective when using exposure light including at least one of i-line, h-line, and g-line, especially, it is preferably suitable for use with i-line, h-line, and g-line The broad wavelength light is used as the exposure light. In this case, any wavelength in the wavelength range of the i-line to the g-line can be set as the representative wavelength. For example, the mask of the present embodiment can be constructed with the h-line as the representative wavelength. More preferably, the phase difference ϕ1 with respect to the above-mentioned representative wavelength is ϕ1 = 180 degrees. In the photomask of the present embodiment, in order to realize the above-mentioned retardation, the main pattern 1 only needs to be the light-transmitting portion 4 formed by exposing the main surface of the transparent substrate 10 , and the auxiliary pattern 2 should be formed on the transparent substrate 10 . For the phase shift portion 5 in which the phase shift film 11 is exposed, the phase shift amount of the phase shift film 11 with respect to the above-mentioned representative wavelength may be set to approximately 180 degrees. The transmittance T1 of the phase shift portion 5 can be set as follows. That is, if the transmittance of the phase shift film 11 formed on the phase shift portion 5 with respect to the above-mentioned representative wavelength is set to T1 (%), it is preferably 20≦T1≦80, more preferably 30≦T1≦75, More preferably, it is 40≦T1≦75. If the transmittance of the auxiliary pattern 2 is high, the width (d) of the auxiliary pattern can be reduced in order to obtain a specific amount of transmitted light. Therefore, there is an advantage of obtaining a degree of freedom of arrangement that can avoid mutual physical interference in dense patterns. . On the other hand, if the transmittance T1 is slightly lowered and the width (d) of the auxiliary pattern is enlarged, there is an advantage that the difficulty in manufacturing the pattern formation is alleviated. In this case, the transmittance T1 is preferably 40 to 60 (%). In addition, the transmittance T1 (%) here is the transmittance of the above-mentioned representative wavelength when the transmittance of the transparent substrate 10 is based on the standard (100%). In the photomask of the present embodiment, the low light-transmitting portion 3 is formed in the region other than the region where the main pattern 1 and the auxiliary pattern 2 are formed. Here, the main pattern 1 and the auxiliary pattern 2 are separated by a low light transmission portion 3 . The low light-transmitting portion 3 can be configured as follows. The low light transmittance portion 3 is a low light transmittance film (ie, a light shielding film) 12 that is substantially impermeable to exposure light (light with a representative wavelength in the wavelength range of i-line to g line), and can be provided on the transparent substrate 10 A film having an optical density of OD≧2 (preferably OD≧3) is formed. In addition, the low-light-transmitting portion 3 may be formed by forming the low-light-transmitting film 12 that transmits the exposure light with a transmittance of a specific range. However, in the case of transmitting the exposure light with a transmittance of a specific range, the transmittance T3 (%) of the low light transmittance portion 3 to the representative wavelength is set to be compared with the transmittance T1 (%) of the phase shift portion 5 , which satisfies 0<T3<T1, preferably 0<T3≦20. Here, when the phase shift portion 5 is not a single-layer film including the phase shift film 11 but includes a laminated film of the phase shift film 11 and the low light transmittance film 12, the transmittance of the laminated film is set to be T3(%). The transmittance T3 (%) here is also set to the transmittance of the above-mentioned representative wavelength based on the transmittance of the transparent substrate 10 as described above. In addition, when the low light transmission film 12 transmits the exposure light with a transmittance of a specific range, the phase shift amount ϕ3 in the laminated state of the phase shift film 11 and the low light transmission film 12 is preferably 90 (degree) or less, more preferably 60 (degree) or less. "90 degrees or less" means that the above-mentioned phase difference is "(2n-1/2)π to (2n+1/2)π (where n is an integer)" when expressed in radians. The retardation is also set to the representative wavelength included in the exposure light in the same manner as described above. In addition, as a single property of the low-light-transmitting film 12 used in the photomask of the present embodiment, it is preferable that the light of the above-mentioned representative wavelength is not substantially transmitted (OD≧2, more preferably OD>3), or It has a transmittance (T2(%)) of less than 30(%) (ie, 0<T2<30) and the phase shift amount (ϕ2) is approximately 180 degrees. The so-called approximately 180 degrees refers to those of 120 to 240 degrees. Preferably, the phase shift amount ϕ2 is 150 to 210 degrees. The transmittance T2 (%) here is also set to the transmittance of the above-mentioned representative wavelength based on the transmittance of the transparent substrate 10 as described above. In the pattern for transfer of the present embodiment, if the width of the auxiliary pattern 2 is d (μm), when the relationship of the following formula (1) is established, a remarkable effect can be obtained. 0.5≦ ×d≦1.5・・・(1) In this case, if the distance between the center of the main pattern 1 and the center in the width direction of the auxiliary pattern 2 is the slit pitch L (μm), it is preferably 1.0<L≦5.0 , more preferably 1.5<L≦4.5. However, the auxiliary pattern 2 constitutes at least a part of the region of the regular octagonal band surrounding the main pattern 1 through the low light transmittance portion 3 . Therefore, the slit pitch L and the distance between the main patterns can be determined in such a way that the main pattern 1 and the auxiliary pattern 2 are not in contact, that is, on the condition that the low transmittance portion 3 is interposed around the main pattern 1 and between the main pattern 1 and the auxiliary pattern 2. Diameter W1. The width d (μm) of the auxiliary pattern 2 is set so that the auxiliary pattern having the transmittance T1 is not resolved under the exposure conditions (exposure device used) suitable for the mask of the present embodiment. As a specific example, the width d (μm) of the auxiliary pattern 2 is preferably d≧0.7, more preferably d≧0.8. Moreover, when compared with the width W1 (μm) of the main pattern 1, d≦W1 is preferable, and d<W1 is more preferable. Moreover, as for the width d (μm) of the auxiliary pattern 2 , the relational expression represented by the above-mentioned formula (1) is more preferably the following formula (1)-1, and still more preferably the following formula (1)-2. 0.7≦ ×d≦1.2 ・・・(1)－1 0.75≦ ×d≦1.0 ・・・(1)−2 The shape of the main pattern 1 included in the transfer pattern of the present embodiment is a quadrangle. Specifically, the shape of the main pattern 1 is preferably, for example, a square or a rectangle. When the shape of the main pattern 1 is a quadrangle, the distance between the center of gravity of the quadrangle and the center in the width direction of the auxiliary pattern 2 is the slit pitch L. In the present embodiment, when defining a regular octagonal strip with a specific width surrounding the main pattern 1 at the periphery of the main pattern 1, the auxiliary pattern 2 becomes a pattern constituting at least a part of the regular octagonal strip. The so-called regular octagonal belt refers to a shape in which both the outer periphery and the inner periphery are octagonal and have a substantially constant width. As shown in FIG. 7 , the auxiliary pattern 2 has a fixed width other than the corners. The regular octagonal band defining the auxiliary pattern 2 is arranged on the periphery of the main pattern 1 so as to surround the main pattern 1 . In addition, the center of gravity of the regular octagon which becomes the inner contour and the outer contour of the regular octagon belt of the auxiliary pattern 2 is located at the same position as the center of gravity of the main pattern 1 . In this embodiment, for convenience of description, as shown in FIG. 8 , the above-mentioned regular octagonal belt is divided into 8 sections corresponding to each side of the octagon of the outer circumference (or inner circumference). Here, among the segments divided into eight segments, the segment at the right end of FIG. 8 is called segment A, the segment on the upper side adjacent to it is called segment H, and the segment on the lower side is called segment B. That is, the sections B, C, D, E, F, G, and H are set clockwise from section A. As shown in FIG. 4 described above, when the second main pattern 1b is arranged at a position close to the first main pattern 1a, in the present embodiment, the regular octagonal belt is arranged around the first main pattern 1a and constitutes the above-mentioned regular octagonal belt. Among the eight sections A to H, there is an auxiliary pattern 2a having a defective shape in a section facing the side of the second main pattern 1b. Specifically, as shown in FIG. 7( a ), an auxiliary pattern 2 a in which a partial segment of the octagonal band is missing is arranged on the periphery of the first auxiliary pattern 1 a. As shown in the figure, the auxiliary pattern 2a attached to the first main pattern 1a has a shape having a defect on the side facing the second main pattern 1b, that is, in the segment at the right end (corresponding to the segment A described above). Moreover, the auxiliary pattern 2b attached to the 2nd main pattern 1b also has the shape which has a defect in the segment (corresponding to the said segment E) which faces the 1st main pattern 1a side. That is, the two main patterns have auxiliary patterns in the shape of octagonal strips each having a defect in a section located on the sides opposite to each other. That is, when two hole formation patterns are arrange|positioned by the close distance below a specific distance, the auxiliary pattern with which each main pattern is equipped is missing between the said two main patterns. Therefore, when the centers of gravity of the two main patterns are connected by a straight line (not shown), the straight line does not pass through the auxiliary pattern at all. Furthermore, it is preferable that the auxiliary pattern is not substantially arranged in the region S (shown by the dotted line in FIG. 7( a )) sandwiched by the mutually opposing sides of the two main patterns. However, when a part of the auxiliary pattern enters the region S, it is preferable that the part does not have a side parallel to the mutually opposite sides of the two main patterns. In the aspect shown in FIG. 7 , the end of the auxiliary pattern enters the area S, but the end has only a side inclined to the mutually opposing sides of the two main patterns. Moreover, it is preferable that the auxiliary pattern in the shape of an island (a shape surrounded by a closed straight line or a curved line) does not exist in the region S. As shown in FIG. Furthermore, it is preferable that 90% or more of the area in the said area|region S contains a low light transmittance part. Thereby, the shape of the resist pattern of the two hole patterns formed on the to-be-transferred body becomes favorable, and the loss of the thickness of a resist residual film can be suppressed. By arranging the auxiliary patterns 2a and 2b with partial segmental defects in this way, the loss of resist film thickness due to the concave portion 22 shown in FIG. 6(b) is eliminated as shown in FIG. 9, and a favorable profile is obtained. The shape of the resist pattern. In addition, FIG. 6(b) and FIG. 9 are the case where the hole pitch P is 8 μm. The permissible range of the loss of the resist film thickness observed in FIG. 6( b ) can be determined according to the manufacturing conditions and the like of the display device to be obtained by using the photomask. If the initial film thickness (coating film thickness) of the resist film is 100%, it is a good condition that the allowable range of the loss is 10% or less of the initial film thickness, more preferably 5% or less. Therefore, in this embodiment, whether or not to partially destroy the auxiliary patterns 2a and 2b attached to the main patterns 1a and 1b that are close to each other, it is only necessary to grasp the loss amount of the resist film thickness in advance through experiments, simulations, etc. The result can be judged. Specifically, it is useful to judge that when the loss amount of the resist film thickness exceeds, for example, 10%, a part of the auxiliary patterns 2a and 2b is damaged, and when it is 10% or less, it is not used. The auxiliary patterns 2a and 2b are missing. Furthermore, when the second main pattern 1b is arranged at a position close to the first main pattern 1a, it is preferable that the distance D (see FIG. 4 ) between the auxiliary patterns 2a and 2b is 1.0 μm or less. The section (section A described above) facing the second main pattern 1b side of the auxiliary pattern 2a arranged on the periphery of the first main pattern 1a is missing. Similarly, it is preferable to also form an auxiliary pattern 2b in the second main pattern 1b so that the segment facing the side of the first main pattern 1a (the segment E described above) is missing. Furthermore, it is more preferable to apply the defect of the segment when the distance D between the auxiliary patterns 2a and 2b is 1.5 μm or less. The above case is exemplified in view of the resolution performance of the exposure device for display devices. Furthermore, the so-called distance between the auxiliary patterns as shown in FIG. 4 refers to the distance (the length of the vertical line) between the opposing sections. Therefore, when the two main patterns 1a, 1b are arranged adjacent to each other in a certain direction as shown in the figure, and each main pattern 1a, 1b is surrounded by the corresponding (supplied) auxiliary patterns 2a, 2b, respectively, in In the arrangement direction of the two main patterns 1a and 1b, the distance between the opposite sides (which face each other) of the auxiliary patterns 2a and 2b becomes the distance D between the auxiliary patterns. Further, in the mask of the present embodiment, when the hole pitch P, which is the distance between the centers of gravity of the main patterns 1a and 1b, is 1.6 μm or more, preferably 3 μm or more, the remarkable effect of the present invention can be obtained. If the value of the hole pitch P is too small, there is a risk of causing a situation where the residual film amount of the resist pattern formed at the position corresponding to the two main patterns cannot be sufficiently obtained, or the following masking may occur. The numerical value of the deviation β2 becomes too large, and the pattern design becomes difficult, for example. The optical simulations of the examples and reference examples of the present invention will be described below. 10 is a schematic plan view showing the main part of the transfer pattern included in the mask of the reference example, (a) shows the reference example 4, (b) shows the reference example 5, (c) shows the reference example 6, (d) shows the reference example ) represents Reference Example 7, and (e) represents Reference Example 8. 11 is a schematic plan view showing the main part of the transfer pattern included in the photomask according to the embodiment of the present invention, (f) shows Example 1, (g) shows Example 2, and (h) shows Example 3 , (i) represents Example 4. 10(a) shows the case where the hole pitch P of the main patterns 1a, 1b (refer to FIGS. 5 to 7) is 16 μm, and the auxiliary patterns 2a, 2b of the regular octagonal belt have no segment defect, and FIG. 10( b) represents the case where the hole pitch P of the main patterns 1a, 1b is 12 μm and the auxiliary patterns 2a, 2b have no segment defect. 10(c) shows the case where the hole pitch P of the main patterns 1a, 1b is 9 μm and the auxiliary patterns 2a, 2b have no segment defect, and FIG. 10(d) shows that the hole pitch P of the main patterns 1a, 1b is 8.75 μm and the auxiliary patterns 2a and 2b have no segmental defect. FIG. 10(e) shows the case where the hole pitch P of the two main patterns 1a, 1b is 8.75 μm, and a section of each auxiliary pattern 2a, 2b is combined and shared. On the other hand, FIG. 11(f) shows the case where the hole pitch P of the main patterns 1a, 1b is 8.75 μm and a section of the auxiliary patterns 2a, 2b is missing, and FIG. 11(g) shows the hole pitch of the main patterns 1a, 1b A case where P is 8 μm and a section of the auxiliary patterns 2a and 2b is missing. 11(h) shows the case where the hole pitch P of the main patterns 1a, 1b is 7.5 μm and a section of the auxiliary patterns 2a, 2b is missing, and FIG. 11(i) shows that the hole pitch P of the main patterns 1a, 1b is 7 μm and the auxiliary patterns 2a and 2b are deficient in 3 sections. In this simulation, the case where the photomask (binary mask) shown in FIG. 2( a ) is used is referred to as reference example 1, and the photomask (halftone mask) shown in FIG. 2( b ) is used as reference example 1. The case of the type phase shift mask) is set as Reference Example 2. The main patterns applied to Reference Example 1 and Reference Example 2 are all isolated patterns without auxiliary patterns. Optical simulations were performed on the patterns of each of the above-mentioned masks, and as a result, the results shown in FIG. 12 were obtained. In this simulation, except for Reference Example 1 and Reference Example 2 , a dose (Eop A dose) was used as a reference, and the resist pattern formed on the transfer object when the dose was applied was evaluated. In addition, "panel X-CD" and "panel Y-CD" are the dimensions of the X direction and the Y direction of the hole pattern formed on the transferred body corresponding to the main pattern of the mask. In addition, in each reference example and each Example, the target size of X-CD and Y-CD was set to 1.5 micrometers. First, in Reference Example 4 of FIG. 10( a ), since the two main patterns 1 a and 1 b are sufficiently spaced apart, each of the main patterns 1 a and 1 b substantially exhibits the optical performance as an isolated pattern. This aspect is also the same in Reference Example 5 of FIG. 10( b ) or Reference Example 6 of FIG. 10( c ). On the other hand, when the two main patterns 1a and 1b are close to each other and the hole pitch P is narrowed to 8.75 μm as in Reference Example 7 of FIG. 10( d ), the distance D between the auxiliary patterns 2a and 2b becomes 0.95 μm. At this time, the loss of resist film thickness exceeds 12%. Here, as in Reference Example 8 of FIG. 10(e), when the auxiliary patterns 2a, 2b of the main patterns 1a, 1b that are close to each other are partially joined to each other into one (one segment) and shared, The loss of resist film thickness is also close to 12%, almost no improvement. According to the research of the present inventors, it is believed that this problem is related to photoresists suitable for the manufacture of display devices. Specifically, the photoresist (positive type photoresist) used for the manufacture of the display device is designed in such a way that it has a higher sensitivity than the photoresist used for the manufacture of the semiconductor device. Therefore, the corresponding film reduction is not avoided even in a relatively low dose, and it is easy to cause local loss of film thickness in unexpected parts. Furthermore, the interaction between the main pattern 1 and the auxiliary pattern 2 generated in the mask of Reference Example 3 utilizes the optical image formed by the phase-reversed light transmitted through the auxiliary pattern 2 to improve the optical image produced by the transmitted light of the main pattern 1. The phenomenon of peak light intensity. Furthermore, excellent transfer performance (DOF, MEEF) as shown in FIG. 3 was obtained. On the other hand, the light intensity due to the light transmitted through the auxiliary pattern is also slightly increased by the interaction with the main pattern 1 . It can be considered that due to the high sensitivity of the photoresist used in the manufacture of display devices, unless the auxiliary patterns 2 are close to each other, if the auxiliary patterns 2 are shared by a plurality of main patterns 1, the transmission of the auxiliary patterns 2 will occur. Risk of reducing the thickness of the resist on the transferred body by light. On the other hand, in Example 1 of FIG. 11( f ), a segment facing the side of the second main pattern 1b among the eight segments constituting the regular octagonal band surrounding the first main pattern 1a is missing, and there will be a The auxiliary patterns 2a of the remaining 7 sections are arranged on the periphery of the first main pattern 1a. That is, the auxiliary pattern 2a has the shape of the auxiliary pattern 2a in which the segment of the part of the regular octagonal band surrounding the periphery of the first main pattern 1a that is closest to and opposite to the regular octagonal band surrounding the periphery of the second main pattern 1b is missing. . Similarly, for the second main pattern 1b, a segment facing the first main pattern 1a side is missing, and the auxiliary pattern 2b having the remaining seven segments is arranged around the second main pattern 1b. Furthermore, when the section of the auxiliary pattern is to be damaged, it is not necessary to cut it according to the boundary line of the section shown in FIG. 8 , for example, as shown in FIG. Can. The area of the defective portion is preferably, for example, a segment to be deficient so that it is 80% or more of the area of one segment if one of the eight segments is deficient. More than 80% of the area of each number. At this time, a portion where only the low light-transmitting portion 3 is interposed is generated between the two main patterns, and a straight line connecting the centers of gravity of the two main patterns 1a and 1b does not pass through the auxiliary patterns. In this way, it can be seen that in the case where the sections facing each other in the auxiliary patterns of the two adjacent main patterns are deficient, in the formed resist pattern, the loss of the resist film thickness becomes zero, etc., and the loss is greatly reduced. And can get a significant effect. In addition, it can be seen that the same effect is obtained in the case where the two main patterns 1a and 1b are brought closer to each other as in Example 2 of FIG. 11(g) or Example 3 of FIG. 11(h) (the hole spacing P=8 μm, P= 7.5 μm case) can also be obtained. The cross-sectional structure of the resist pattern in this case is the same as that of FIG. 9 . However, in the examples of FIGS. 11( f ) to ( i ), although the same exposure dose as in the above-mentioned Reference Examples 4 to 8 was applied, the diameter of the hole pattern transferred to the resist film was slightly changed from the initial target size Small. Specifically, in Example 1, X-CD (diameter in X direction) was 1.39 (μm), and Y-CD (diameter in Y direction) was 1.37 with respect to 1.5 (μm) as the initial target size (μm). Therefore, in order to form a hole pattern so that both X-CD and Y-CD on the transfer target body are close to the target size (1.5 μm), it is preferable to make the above-mentioned mask deviation β1 larger than 0.5 μm. Furthermore, in order to make X-CD and Y-CD on the to-be-transferred body the same target size (1.5 μm), it is preferable to give an appropriate mask deviation β2 in the X direction and the Y direction, respectively. Therefore, as shown in FIG. 12, in the embodiments of FIGS. 11(f)-(h), by assigning appropriate mask deviations β2(x) and β2(y) to the CD on the mask, On the transfer body, both X-CD and Y-CD had a target size of 1.5 μm. Furthermore, as a result, it can be seen that in the case of the photomask of the embodiment shown in FIGS. 11( f ) to ( h ), the DOF (depth of focus) value (24) is higher than that of the binary mask of Reference Example 1 ( FIG. 2( a ) ) or the halftone phase shift mask of Reference Example 2 (Fig. 2(b)) is large, and can stably form a hole pattern with a desired diameter on the transferred body. Also, since the dose of Eop required for exposure is smaller than that of the binary mask of Reference Example 1 or the halftone-type phase shift mask of Reference Example 2, the exposure step can be efficiently performed. Furthermore, in the embodiment of FIGS. 11(f)-(h), the eighth section of the auxiliary pattern 2a attached to the first main pattern 1a and the eighth section of the auxiliary pattern 2b attached to the second main pattern 1b are One of the segments opposite to each other is missing. That is, for one main pattern 1, the number of segments of the auxiliary pattern 2 is set to seven. On the other hand, when the two main patterns 1a and 1b are further arranged close to each other, the segments located on both sides of the segment can be further damaged by centering on the segment that has been damaged. For example, in Example 4 of FIG. 11(i), the segment A among the eight segments of the auxiliary pattern 2a attached to the first main pattern 1a, and the segments B and H on both sides thereof are deficient, In addition, the segment E among the eight segments of the auxiliary pattern 2b attached to the second main pattern 1b and the segments D and F located on both sides of the segment E and the segments D and F located on both sides thereof are deficient, thereby setting the loss of the resist film thickness to zero. . As a result, here, the number of segments of the auxiliary pattern 2 is set to five by missing three segments for one main pattern 1 . Furthermore, the number of main patterns arranged close to each other is not limited to 2, and more hole forming patterns may be arranged at a close distance. In this case, the segmental defect can be designed in the same manner as described above. In a pattern group including a plurality of main patterns arranged in close proximity (respectively at least close to any other one), when the number of included main patterns is set as N, and the total number of sections of auxiliary patterns is set as K, the Set K≦(8-1)N. 11( f ) to ( i ), only the case where the second main pattern 1b approaches the first main pattern 1a in the X direction (the left and right direction in the figure) is illustrated, but in the case where the second main pattern 1b approaches in the Y direction. In this case, the auxiliary pattern 2 for missing any one of the eight segments may be arranged in the same manner as described above. Furthermore, the present invention can also be effectively applied when the second main pattern 1b approaches the first main pattern 1a in the diagonal direction of the main pattern 1, such as when approaching from an oblique direction. In this case, at least one section facing the side of the second main pattern 1b among the eight sections of the auxiliary pattern 2a attached to the first main pattern 1a may be damaged. Furthermore, when the third main pattern and then the fourth main pattern are arranged close to the first main pattern, in the same manner as described above, the eight segments of the auxiliary patterns attached to each main pattern may be arranged to face each other. The segment on the opposite side is defective. In this case, according to the relative arrangement of the plurality of main patterns, in addition to the main pattern with seven-segment auxiliary patterns and the main pattern with five-segment auxiliary patterns, there may also be six-segment and four-segment main patterns. , the main pattern of the auxiliary patterns of three sections, two sections, and one section. For example, two main patterns can be set as a pattern group in which auxiliary patterns each have 1 section of each other missing, or a pattern group in which auxiliary patterns each have 3 sections of each other missing. Alternatively, the three main patterns can be set as a pattern group in which the auxiliary patterns of each of the three main patterns are missing 1 or 2 segments according to their arrangement (X direction, Y direction, or X and Y directions, the same applies hereinafter). Furthermore, it is also possible to set the four main patterns as a pattern group each having auxiliary patterns in which 1 to 5 segments are missing according to the arrangement thereof. Hereinafter, 5, 6, or 7 main patterns may be set as a pattern group of auxiliary patterns with 1 to 6 sections of defects, respectively, according to their arrangement, or 8 main patterns may be set to have 1 to 7 of defects according to their arrangement, respectively. Pattern group of auxiliary patterns of the segment. On the other hand, the auxiliary pattern of one main pattern can be set to be free of defects except for the segment facing the side of the other main pattern, or the segment on both sides thereof. Moreover, the diameter of the hole pattern formed in the to-be-transferred body may become a different value in the X direction and the Y direction depending on the approach direction of a plurality of hole forming patterns. The reason for this is that the variation of the optical image caused by the defect of a partial section of the auxiliary pattern is unequally generated for the X direction and the Y direction. Therefore, for the purpose of compensating for the influence of the unequal optical image in the X direction and the Y direction, it is useful to assign the mask deviation β2 of different sizes in the X direction and the Y direction to the drawing data of the pattern. For example, as shown in FIG. 7( a ), the first main pattern 1a and the second main pattern 1b are arranged in the X direction, and the auxiliary patterns 2a and 2b attached to both main patterns 1a and 1b are arranged to face each other. When the segment (that is, the segment extending in the Y direction) is missing, regarding the mask deviation β2 (μm) assigned to the size of the first main pattern 1a and the second main pattern 1b, the X direction is assigned When the amount is β2(x) and the amount given in the Y direction is β2(y), β2(y)>β2(x). Specifically, in the observation of the autonomous pattern, a positive deviation β2 ( y). Furthermore, if necessary, a negative deviation β2(x) is given to the direction (X direction in FIG. 7( a )) in which the defect portion of the auxiliary pattern exists. FIG. 13 shows an example of a transfer pattern including a mask including the main patterns 1 a and 1 b provided with the mask deviation β2 in this way. Here, as a result of giving the above-mentioned mask deviation β2, the dimensions in the X direction of the main patterns 1a and 1b are smaller than those in the Y direction, and the shapes of the main patterns 1a and 1b are vertically long rectangles. In the example of FIG. 11( g ), the hole patterns formed on the transfer target body as the transfer images of the square main patterns 1a and 1b are horizontally long rectangles (X-CD=1.40 μm, Y-CD = 1.37 μm). Therefore, if the shapes of the main patterns 1a and 1b are made to be vertically long rectangles by applying the mask deviation, the error of the pattern size caused by the defect of a partial section of the auxiliary patterns 2a and 2b can be eliminated. As a result, a hole pattern whose size in the X direction and the size in the Y direction are equal can be formed on the transfer object. Therefore, by optical simulation, the deviation amount β2 for making the dimensions in the X direction and the Y direction equal to the transfer target body can be obtained for each of the X direction and the Y direction, and this can be reflected in the pattern drawing data. . Therefore, among the patterns for transfer which the mask has, the main pattern to which the deviation β2 is given becomes a rectangle. That is, the 1st main pattern becomes the rectangle which has a long side on the side which faces the 2nd main pattern in the position close|similar. 13, the long side W3(y) of the main pattern becomes W3(y)=W1+β2(y), and the short side W3(x) becomes W3(x)=W1+β2(x). Furthermore, W3(x) and W3(y) preferably satisfy the following formula. For the long side, 0.8≦W3(y)≦4.0, more preferably 1.0≦W3(y)<3.5, for the short side, 0.8≦W3(x)≦4.0, more preferably 1.0≦W3(x)≦3.0 . As described above, when the photomask of the present embodiment is used as a photomask for manufacturing a display device, that is, when the photomask of the present embodiment is used in combination with a photoresist for manufacturing a display device, a large Reduce the loss of resist film thickness on the part corresponding to the auxiliary pattern on the transferred body. <The manufacturing method of a mask> Next, an example of the manufacturing method of the mask which can be applied to the embodiment of this invention is demonstrated with reference to FIGS. 14(a)-(f). 14( a ) to ( f ), the left side shows a cross-sectional view and the right side shows a plan view, and for simplicity, the mask pattern shape only shows the first main pattern and the auxiliary pattern attached to the first main pattern. First, as shown in FIG. 14( a ), a mask base 30 is prepared. In the mask base 30 , a phase shift film 11 and a low light transmittance film 12 are sequentially formed on the transparent substrate 10 including glass or the like, and the first photoresist film 13 is further coated. The phase shift film 11 is formed on the main surface of the transparent substrate 10 . When any of the i-line, h-line, and g-line is set as the representative wavelength of the exposure light, the transmittance T1(%) of the representative wavelength is preferably 20-80(%) of the phase shift film 11, More preferably, it is 30-75 (%), More preferably, it is 40-75 (%). In addition, the phase shift amount of the phase shift film 11 with respect to the above-mentioned representative wavelength is approximately 180 degrees. With such a phase shift film 11, the phase difference of the transmitted light between the main pattern including the light-transmitting portion and the auxiliary pattern including the phase shift portion can be set to approximately 180 degrees. Such a phase shift film 11 shifts the phase of light having a representative wavelength in the wavelength range of i-line to g-line by approximately 180 degrees. As a film formation method of the phase shift film 11, a well-known method, such as a sputtering method, can be applied. Preferably, the phase shift film 11 satisfies the above-mentioned transmittance and retardation, and is formed of a material capable of wet etching as described below. However, if the amount of side etching generated during wet etching becomes too large, there may be undesirable situations such as deterioration of CD accuracy or destruction of the upper layer film due to undercutting. Therefore, the film thickness of the phase shift film 11 is preferably set to 2000 Å or less, preferably 300 to 2000 Å, more preferably 300 to 1800 Å. Moreover, in order to satisfy these conditions, the refractive index of the representative wavelength (for example, h line) included in the exposure light of the material of the phase shift film 11 is preferably 1.5-2.9, more preferably 1.8-2.4. Furthermore, in order to sufficiently exert the phase shift effect, it is preferable that the cross section (etched surface) of the pattern obtained by wet etching is nearly perpendicular to the main surface of the transparent substrate 10 . In consideration of the above properties, as the material of the phase shift film 11, a material including any one of Zr, Nb, Hf, Ta, Mo, Ti and Si, or an oxide, nitride, oxynitride including these materials can be used material, carbide, or oxycarbonitride. A low light transmission film 12 is formed on the phase shift film 11 . As a film formation method of the low light transmission film 12, like the case of the phase shift film 11, a well-known method, such as a sputtering method, can be applied. In addition, it is preferable that the wavelength dependence of the phase shift amount of the phase shift film 11 is within 40 degrees for the i-line, h-line, and g-line. The low light-transmitting film 12 may include a light-shielding film that does not substantially transmit the exposure light. Moreover, in addition to this, the low light transmission film 12 may be comprised by the film which has a specific low transmittance|permeability with respect to the representative wavelength of exposure light. The low transmittance film 12 used in the manufacture of the photomask of the present embodiment has a transmittance T2 lower than the transmittance T1 (%) of the phase shift film 11 for light with a representative wavelength in the wavelength range of i-line to g-line (%). When the low transmittance film 12 transmits the exposure light with a lower transmittance, the transmittance and phase shift of the low light transmittance film 12 for the exposure light are required to reach the low transmittance of the mask of the present embodiment. Transmittance and phase shift of the light part. Preferably, in the laminated state of the phase shift film 11 and the low light transmittance film 12, the transmittance T3 (%) of the light with the representative wavelength of the exposure light is T3≦20, and the phase shift amount ϕ3 is preferably It is 90 (degrees) or less, more preferably 60 (degrees) or less. As a separate property of the low light transmittance film 12, it is preferable that it does not substantially transmit the light of the representative wavelength mentioned above or has a transmittance (T2(%)) of less than 30(%) (that is, 0<T2<30 ), and the phase shift amount (ϕ2) is approximately 180 degrees. The so-called approximately 180 degrees means 120 to 240 degrees. Preferably, the phase shift amount ϕ2 is 150 to 210 (degrees). The material of the low light transmission film 12 can be Cr or its compound (oxide, nitride, carbide, oxynitride, or oxycarbonitride), or it can also be a metal containing Mo, W, Ta, Ti. A silicide or the aforementioned compounds of the silicide. However, the material of the low light transmittance film 12 is preferably a material that can perform wet etching in the same manner as the phase shift film 11 and has etching selectivity to the material of the phase shift film 11 . That is, it is preferable that the low light transmission film 12 has resistance to the etchant of the phase shift film 11 , and the phase shift film 11 has resistance to the etchant of the low light transmission film 12 . A first photoresist film 13 is coated on the low light transmittance film 12 . The photomask of the present embodiment is preferably drawn by a laser drawing device, and therefore, it is set as a photoresist suitable for a laser drawing device. The photoresist constituting the first photoresist film 13 may be either positive or negative, but the following description is made as a positive photoresist. Next, as shown in FIG.14(b), with respect to the 1st photoresist film 13, using a drawing apparatus, drawing (1st drawing) using drawing data based on the pattern for transfer is performed. Next, using the first resist pattern 13p obtained by the development as a mask, the low-light-transmitting film 12 is wet-etched, thereby forming the low-light-transmitting film pattern 12p. At this stage, the region to be the low-light-transmitting portion is defined, and the region of the auxiliary pattern (low-light-transmitting film pattern 12p) surrounded by the low-light-transmitting portion is defined. As the etchant (wet etchant) used for the wet etching, a known one suitable for the composition of the low light transmittance film 12 to be used can be used. For example, if the low light transmittance film 12 is a film containing Cr, cerium ammonium nitrate or the like can be used as the wet etchant. Next, as shown in FIG.14(c), the 1st resist pattern 13p is peeled off. Thereby, the low light-transmitting film pattern 12p and a part of the phase shift film 11 are exposed. Next, as shown in FIG. 14( d ), the second photoresist film 14 is coated on the entire surface including the low light transmittance film pattern 12p. Next, as shown in FIG.14(e), after performing the 2nd drawing on the 2nd photoresist film 14, the 2nd resist pattern 14p is formed by development. Next, the phase shift film 11 is wet-etched using the second resist pattern 14p and the low light transmittance film pattern 12p as masks. By this etching (development), the main surface of the transparent substrate 10 is exposed as a light-transmitting portion, thereby defining a region of the main pattern including the light-transmitting portion. Furthermore, the second resist pattern 14p covers the region that becomes the auxiliary pattern, and has an opening in the region of the main pattern including the light-transmitting portion. In this case, it is preferable to dimension the drawing data of the second drawing so that the edge portion of the low light-transmitting film pattern 12p is exposed on the inner side of the opening edge of the second resist pattern 14p. Thereby, the misalignment which arises between 1st drawing and 2nd drawing can be absorbed, and the deterioration of the CD precision of the pattern for transfer can be prevented. That is, if the size of the second resist pattern 14p is determined during the second drawing, when an isolated hole pattern is to be formed on the transfer target body, the phase shift film 11 and the low light transmittance film will not be affected. 12. A position shift occurs during patterning. Therefore, in the pattern for transfer as illustrated in FIG. 1, the center of gravity of the main pattern 1 and the auxiliary pattern 2 can be precisely aligned. The wet etchant used for the etching of the phase shift film 11 is appropriately selected according to the composition of the phase shift film 11 . Next, as shown in FIG.14(f), the 2nd resist pattern 14p is peeled off. Thereby, the photomask provided with the pattern for transfer is completed. Furthermore, FIG. 14 shows the case where an auxiliary pattern of a regular octagon without segment defects is formed, but in the case where any one of the eight segments is deficient, only in FIG. 14(b) When demarcating the area of the auxiliary pattern, it is sufficient to change the drawing data according to the position and size of the segment to be missing. In the manufacture of the above-described photomask, dry etching or wet etching can be used in etching that can be applied when patterning optical films such as the phase shift film 11 or the low light transmittance film 12 . Any of these may be employed, but in the present invention, wet etching is particularly advantageous. The reason for this is that the size of the photomask for manufacturing a display device is relatively large, and further, there are various sizes. In the manufacture of such a photomask, if dry etching that requires a vacuum chamber is applied, the size of the dry etching apparatus will be increased or the efficiency of the manufacturing process will be reduced. However, there is also a problem associated with applying wet etching when manufacturing such a mask. Since wet etching has the property of isotropic etching, when a specific film is etched in the depth direction to dissolve it, the etching is also performed in the direction perpendicular to the depth direction. For example, when the phase shift film 11 with a film thickness of F (nm) is etched to form a slit, the opening of the resist pattern that becomes the etching mask is 2F (nm) smaller than the required slit width (that is, one side F (nm)), but it is difficult to maintain the dimensional accuracy of the opening of the resist pattern as the slit is as fine as possible. Therefore, it is useful to set the width d of the auxiliary pattern to be 1 μm or more, preferably 1.3 μm or more. In addition, when the above-mentioned film thickness F (nm) is large, the amount of side etching also becomes large, so it is advantageous to use a film material having a phase shift amount of approximately 180 degrees even if the film thickness is small. Therefore, it is desirable for the refractive index of the phase shift film 11 to be high for the representative wavelength of the exposure light. Specifically, the phase shift film 11 is preferably formed using a material having a refractive index of preferably 1.5 to 2.9, more preferably 1.8 to 2.4, for the above-mentioned representative wavelength. The present invention includes a method of manufacturing a display device including a step of transferring the above-mentioned pattern for transfer onto a transfer target body by exposing the mask using the photomask of the present embodiment by an exposure device. In the manufacturing method of the display apparatus of this invention, the photomask of this embodiment is prepared first. Next, using an exposure device having a numerical aperture (NA) of 0.08 to 0.15 and having exposure light sources including i-line, h-line, and g-line, the above-mentioned pattern for transfer is exposed to form a diameter W2 on the transfer object. 0.8～3.0(μm) hole pattern. It is usually advantageous to apply equal magnification exposure during exposure. When using the photomask of the present embodiment to transfer the pattern for transfer, reduction exposure can also be used, but as an exposure machine used for a photomask for manufacturing a display device, it is a method of performing projection exposure at equal magnification, and the following are preferred: By. For example, the numerical aperture (NA) of the optical system is preferably 0.08≦NA<0.20, more preferably 0.08≦NA≦0.15, and more preferably 0.08<NA<0.15. In addition, the coherence factor σ is 0.4≦σ≦0.9, more preferably 0.4<σ<0.7, and still more preferably 0.4<σ<0.6. The exposure light source is a light source including at least one of i-line, h-line and g-line in the exposure light. In the case where exposure light of a single wavelength is applied, i-line is preferably used. On the other hand, it is useful to use a light source including i-line, h-line, and g-line (also referred to as a broad-wavelength light source) to ensure a sufficient amount of light. In addition, oblique illumination (ring illumination, etc.) may also be used for the light source of the exposure device used, but by using normal illumination not suitable for oblique illumination, the excellent effects of the present invention can be sufficiently obtained. According to the present invention, in the method for manufacturing a display device using the mask for manufacturing a display device, even if it is a fine dense pattern, the transfer onto the transfer target body can be performed stably. Specifically, it is possible to precisely form a hole pattern while securing a process margin (Process Margin) at the time of manufacturing DOF, MEEF, or the like. Furthermore, when forming a dense pattern, the thickness of the resist pattern formed on the to-be-transferred body can be ensured sufficiently. It improves the CD precision in the production of display devices and provides industrial benefits such as stable production and high yield.
