201003119 六、發明說明： 【發明所屬之技術領域】 本發明係有關於投影光學系、具備該投影光學系的曝 光裝置及使用該曝光裝置來製造元件的元件製造方法。 【先前技術】 平面面板顯示器（以下簡稱FPD )，是藉由光微影工 程所製造。在光微影工程中，是對塗佈有感光劑的基板（ 於FPD的製造中，一般是玻璃板），藉由投影光學系而 將原版的圖案進行投影，使該基板曝光。在用來曝光基板 用的曝光裝置中，降低重合誤差是很重要的。重合誤差， 係例如因爲定位誤差、像差（失真）、倍率誤差所產生。 這些當中，定位誤差係可藉由高精度地調整原版與基板之 相對位置來降低，但像差與倍率誤差是藉由定位仍無法完 全補正。 倍率誤差，係因爲生產製程所產生的熱等因素，導致 玻璃板等基板發生伸縮所引起。因爲基板的縮縮，導致第 二製程以下的重合精度惡化。又，掃描曝光中的掃描方向 與其垂直方向上，會產生不同的伸縮。爲了降低基板伸縮 所帶來的誤差，對於掃描方向的伸縮，係考慮在原版與基 板之間相對性地賦予速度差而進行曝光的方法。又，對於 掃描方向的垂直方向之伸縮，則是考慮將平行平板玻璃配 置在原版或基板的附近，藉由將該平行平板玻璃予以撓曲 而使光線曲折以進行倍率微調之方法。 -5- 201003119 於 Japanese Laid-open No.62-3 5620 號公報、Japanese Patent Ν〇·3 3 4 1 269號公報中，揭露有倍率補正方法。 Japanese Laid-open No.62-3 5 620 號公報中係揭露了，在 投影光學系中使對光軸旋轉對稱的2片透鏡彼此接近配置 ，將其中一方朝光軸方向加以驅動。Japanese Patent No.3 3 4 1 26 9號公報中則揭露，使得對投影光學系的光軸 非旋轉對稱的至少2個圓環型光學構件之至少1者，對光 軸旋轉或在光軸方向上移動。 可是，爲了補正基板的掃描方向上的倍率，而在原版 與基板之間相對性地賦予速度差的方法中，若良像區域是 圓弧狀，則會產生特有的非對稱失真像差。爲了補正之， 必須驅動額外的像差補正用之光學構件。因此會導致機械 機構的複雜化與產生驅動像差殘留。 又，爲了補正與基板的掃描方向垂直之方向上的倍率 ，而將平行平板玻璃撓曲的方法中，要將平行平板玻璃高 精度地彎曲成目標形狀是很困難的，會導致產生非線性像 差誤差成分。而且，爲了對應近年來基板的大型化，平行 平板玻璃也必須要大型化，這也會導致誤差。 在 Japanese Laid-open N〇_62-35620 號公報、Japanese Patent No.334 1 269號公報中所記載之方法中，對掃描方 向的倍率補正與對其垂直方向的倍率補正，無法個別地進 行。 【發明內容】 -6 - 201003119 本發明係有鑑於上記背景而硏發 個正交方向上的倍率補正，能夠個別 本發明的1個側面係爲，有關於 的光路中，依序配置第1凹反射面、 射面，將被配置在前記物體面的物體 用的投影光學系，前記投影光學系係 單元及第2折射光學單元，係被配置 1致動器，係用以驅動前記第1折射 動器，係用以驅動前記第2折射光學 光學單元及前記第2折射光學單元， 的柱形面；前記第1折射光學單元係 方向上的前記投影光學系之投影倍率 單元係被配置成可調整正交於前記第 的前記投影光學系之投影倍率；前記 前記第1折射光學單元的2個以上的 前記第1方向上的投影倍率成爲第1 第2致動器，係變更前記第2折射光 柱形面之間隔，以使前記第2方向上 目標投影倍率。 關於本發明的更多要點，可參照 實施例來理解。 【實施方式】 以下參照添附圖面來說明本發明 ，目的在於，可使2 地進行。 ，在從物體面至像面 凸反射面、第2凹反 的像投影至前記像面 具備：第1折射光學 在前記光路中；和第 光學單元；和第2致 單元；前記第1折射 係分別具有2個以上 被配置成可調整第1 ，前記第2折射光學 1方向之第2方向上 第1致動器，係變更 柱形面之間隔，以使 目標投影倍率；前記 學單元的2個以上的 的投影倍率成爲第2 圖面配合以下說明的 的較佳實施形態。 201003119 [第1實施形態] 圖1係本發明的理想實施形態之曝光裝置EX的槪略 構成之圖示。圖2係圖1所示之曝光裝置EX的投影光學 系PO之構成的模式性圖示。此外，在圖2中，用來彎折 投影光學系PO的光軸AX所需之彎折反射鏡（功率爲零 的反射鏡）MD係被省略。亦可想作是，彎折反射鏡MD ，係將光路予以彎折。彎折反射鏡MD，係被配置成例如 ，在原版Μ所被配置的物體面〇與第1凹反射面Ml之 間，藉由反射面MA而使光軸AX作90度彎折，在第2 凹反射面Μ 3與基板P所被配置的像面I之間，藉由反射 面MB而使光軸ΑΧ作90度彎折。 於現實的構成中，爲了使投影光學系PO小型化而設 置彎折反射鏡MD是有利的，但彎折反射鏡MD並非必須 。在本說明書中，爲了說明空間上的方向，是採用了去除 彎折反射鏡MD之狀態時定義了對光軸AX平行之Z方向 的XYZ座標系。 曝光裝置EX，例如，用來製造液晶顯示器等平面面 板顯示器這類具有大面積的元件時可理想適用，但亦可適 用於製造LSI等半導體元件或其他元件。 曝光裝置EX，係具備：用來照明原版（又稱遮罩或 光罩）Μ的照明光學系IL、用來保持原版Μ的原版平台 MST、投影光學系ΡΟ、用來保持基板（例如玻璃板等平 板）Ρ的基板平台PST。原版Μ係被配置在投影光學系 201003119 PO的物體面0，基板P係被配置在投影光學系PO的像面 I ° 曝光裝置EX，典型而言係被構成爲掃描曝光裝置。 曝光裝置EX，係藉由照明光學系IL，而將原版Μ以被形 成爲帶狀（例如圓弧狀、矩形條紋狀）的光，一邊進行照 明，一邊掃描原版Μ及基板Ρ，而將原版平台MST及基 板平台PST予以掃描驅動。藉此，原版Μ的圖案就會轉 印到基板Ρ之上的感光劑。 投影光學系Ρ0，係在從光軸ΑΧ錯開的位置處，具 有良像區域，該當良像區域是用來將原版Μ之圖案，投 影到基板Ρ時所使用。投影光學系Ρ0，係在從物體面〇 至像面I的光路中，依序配置：第1折射光學單元A、第 1凹反射面Ml、凸反射面M2、第2凹反射面M3、第2 折射光學單元B所構成。在第1折射光學單元A與第1 凹反射面Μ1之間，係配置有彎折反射鏡MD的反射面 ΜΑ。在第2凹反射面M3與第2折射光學單元Β之間， 係配置有彎折反射鏡MD的反射面MB。 第1折射光學單元A及第2折射光學單元B，係不限 於上記的位置，亦可被配置在從物體面0至像面I的光路 之中。第1凹反射面Ml與第2凹反射面M3，典型來說 ，係爲具有同一曲率中心的反射面。第1凹反射面Ml係 具有正功率，凸反射面M2係具有負功率，第2凹反射面 係具有正功率。 第1折射光學單元A及第2折射光學單元B，係用來 -9- 201003119 補正投影光學系p〇之投影倍率用的光學單元。更具體而 言，第1折射光學單元A係被配置成，可以調整第1方 向（當具有彎折反射鏡MD時則是將其去除後之狀態下的 X方向）上的投影光學系PO之投影倍率。又，第2折射 光學單元A係被配置成，可以調整正交於第1方向之第2 方向（當具有彎折反射鏡MD時則是將其去除後之狀態下 的Y方向）上的投影光學系PO之投影倍率。 第1折射光學單元A及第2折射光學單元B，係分別 具有2個以上的柱形面。第1折射光學單元A，係被第1 致動器AA所驅動，以使得第1方向（當具有彎折反射鏡 MD時則是將其去除後之狀態下的X方向）上的投影倍率 會成爲第1目標投影倍率。具體而言，第1致動器AA， 係變更第1折射光學單元A的2個以上的柱形面之間隔 ，以使第1方向上的投影倍率成爲第1目標投影倍率。 第2折射光學單元B，係被第2致動器BA所驅動， 以使得第2方向（當具有彎折反射鏡MD時則是將其去除 後之狀態下的Y方向）上的投影倍率會成爲第2目標投 影倍率。具體而言，第2致動器BA，係變更第2折射光 學單元B的2個以上的柱形面之間隔，以使第2方向上的 投影倍率成爲第2目標投影倍率。 此處，第1目標投影倍率、第2目標投影倍率，係例 如藉由未圖示的定位儀等之計測單元，來偵測出已被形成 在基板P上的定位標記之位置，就可決定之。 第1折射光學單元A，典型而言，係具有2個柱形面 -10- 201003119 。作爲2個柱形面之特性（功率）的組合可考慮有：凸及 凸、凹及凹、凸及凹。此處，爲了使投影倍率對於焦點位 置變動的變化變得遲鈍，將投影光學系P 〇設計成遠心式 構成，較爲理想。因此’ 2個柱形面之特性（功率）的組 合，係爲凸及凹之組合’較爲理想。第1折射光學單元A 中所含之柱形面的母線’典型而言係爲彼此平行。 第2折射光學單元B也是典型而言，具有2個柱形面 ，其特性（功率）的組合’係爲凸及凹之組合’較爲理想 。又，第2折射光學單元B中所含之柱形面的母線’典型 而言係爲彼此平行。 第1折射光學單元A中所含之2個以上的柱形面的 母線係平行於，（當具有彎折反射鏡MD時則是將其去除 後之狀態下的）投影光學系PO之光軸AX所正交的第3 方向（Y方向）。第2折射光學單元B中所含之2個以上 的柱形面的母線係平行於，（當具有彎折反射鏡MD時則 是將其去除後之狀態下的）投影光學系PO之光軸AX及 第3方向所正交的第4方向（X方向）。 圖3、圖4係第1折射光學單元A之構成例的圖示。 此處，圖3係將第1折射光學單元A在XZ面作切斷後的 剖面圖，圖4係將第1折射光學單元A在Y Z面作切斷後 的剖面圖。第1折射光學單元A ’係含有被配置在物體面 〇側的折射光學元件A 1 1、和被配置在像面I側的折射光 學元件A 1 2。折射光學元件A 1 1，係在像面I側具有凸的 柱形面CA11，全體是具有凸的功率。折射光學元件A12 -11 - 201003119 ’係在物體面〇側具有凹的柱形面CA12，全體是具有凹 的功率。第1折射光學單元A，係可具有更多的柱形面。 亦即，第1折射光學單元A，係可具有2個以上的柱形面 。第1折射光學單元A的2個以上的柱形面CA11、CA12 的母線係平行於，（當具有彎折反射鏡MD時則是將其去 除後之狀態下的）投影光學系PO之光軸AX所正交的第 3方向（Y方向）。爲了補正投影光學系PO的第1方向 (當具有彎折反射鏡MD時則是將其去除後之狀態下的X 方向）上的投影倍率，2個以上的柱形面C A 1 1、C A 1 2當 中的至少1個，是被第1致動器AA在光軸ΑΧ方向上進 行驅動。 圖5、圖6係第2折射光學單元Β之構成例的圖示。 此處，圖5係將第2折射光學單元Β在ΧΖ面作切斷後的 剖面圖，圖6係將第2折射光學單元Β在ΥΖ面作切斷後 的剖面圖。第2折射光學單元Β ’係含有被配置在物體面 〇側的折射光學元件Β 1 1、和被配置在像面I側的折射光 學元件Β 1 2。折射光學元件Β 1 1 ’係在物體面I側具有凹 的柱形面CB11，全體是具有凹的功率。折射光學元件 Β 1 2，係在物體面〇側具有凸的柱形面CB 1 2，全體是具 有凸的功率。第2折射光學單兀Β’係可具有更多的柱形 面。亦即，第2折射光學單元Β’係可具有2個以上的柱 形面。第2折射光學單元Β的2個以上的柱形面CB11、 CΒ 1 2的母線係平行於’（當具有彎折反射鏡MD時則是 將其去除後之狀態下的）投影光學系P0之光軸AX所正 -12- 201003119 交的第4方向（X方向）。爲了補正投影光 2方向（當具有彎折反射鏡M D時則是將其 下的Υ方向）上的投影倍率，2個以上的杉 CB12當中的至少1個，是被第2致動器ΑΒ 向上進行驅動。 若依據此實施形態，則2個正交的方向 向及正交於其之方向上的投影倍率之補正， 行。藉此，例如，可隨應於已形成在基板之 補正在掃描方向及正交於其之方向的投影倍 製程以後的像的重疊精度提升。 [第2實施形態] 以下，說明本發明的第2實施形態。此 及的事項，係和第1實施形態相同。 圖7、圖8係第2實施形態中的第1折 之構成例的圖示。此處，圖7係將第1折 在ΧΖ面作切斷後的剖面圖，圖8係將第1 Α在ΥΖ面作切斷後的剖面圖。 第1折射光學單元A，係含有被配置在 折射光學元件A2 1、和被配置在像面I側的 A22。折射光學元件A2 1，係作爲像面I側 面而具有凸的柱形面CA11，作爲物體面0 2面而具有曲面（球面或非球面）S11，全 的功率。折射光學元件A22，係在物體面〇 學系PO的第 去除後之狀態 :形面CB11、 在光軸AX方 、例如掃描方 就可個別地進 圖案來個別地 率，可使第二 處沒有特別言 射光學單元A 射光學單元A 折射光學單元 物體面0側的 折射光學元件 的面亦即第1 側的面亦即第 體而言具有凸 側具有凹的柱 -13- 201003119 形面CA12，全體是具有凹的功率。第1折射光學 ，係可具有更多的柱形面。亦即’第1折射光學單 係可具有2個以上的柱形面。第1折射光學單兀 個以上的柱形面C A 1 1、C A 1 2的母線係平行於’（ 彎折反射鏡M D時則是將其去除後之狀態下的）投 系ΡΟ之光軸ΑΧ所正交的第3方向（Υ方向）。 正投影光學系ΡΟ的第1方向（當具有彎折反射鏡 則是將其去除後之狀態下的X方向）上的投影倍率 以上的柱形面C A 1 1、C A 1 2當中的至少1個’是被 動器AA在光軸AX方向上進行驅動。 圖9、圖10係第2折射光學單元B之構成例 。此處，圖9係將第2折射光學單元B在XZ面作 的剖面圖，圖1 〇係將第2折射光學單元B在Y Z 斷後的剖面圖。第2折射光學單元B，係含有被配 體面〇側的折射光學元件B2 1、和被配置在像面I 射光學元件B22。折射光學元件B21，係在物體面 有凹的柱形面CB11，全體是具有凹的功率。折射 件B 2 2，係作爲物體面Ο側的面亦即第1面而具有 形面C B 1 2，作爲像面I側的面亦即第2面而具有 球面或非球面）S22，全體而言具有凸的功率。第 光學單元B，係可具有更多的柱形面。亦即，第2 學單元B，係可具有2個以上的柱形面。第2折射 元B的2個以上的柱形面CB11、CB12的母線係平 (當具有彎折反射鏡MD時則是將其去除後之狀態 單元A 元A， A的2 當具有 影光學 爲了補 MD時 ’ 2個 第1致 的圖示 切斷後 面作切 置在物 側的折 I側具 光學元 凸的柱 曲面（ 2折射 折射光 光學單 行於， 下的） -14 - 201003119 投影光學系PO之光軸ΑΧ所正交的第4方向（X方向） 。爲了補正投影光學系ΡΟ的第2方向（當具有彎折反射 鏡MD時則是將其去除後之狀態下的Υ方向）上的投影倍 率，2個以上的柱形面CB11、CB12當中的至少1個’是 被第2致動器ΑΒ在光軸ΑΧ方向上進行驅動。 上記例子中’第1折射光學單元Α及第2折射光學 單元B之雙方係在柱形面的背面具有曲面’但亦可爲’第 1折射光學單元A及第2折射光學單元B之任一方是在柱 形面的背面具有曲面。 若依據此實施形態，則藉由將柱形面的背面設計成曲 面（球面或非球面），就可構成小型且高性能的投影光學 系。尤其是，藉由將該當曲面設計成非球面形狀，就可保 持軸外光學性能（非點像散差（astigmatic difference )、 像面彎曲）之性能不變，而擴大帶狀（例如圓弧狀）的良 像區域。此外，若除了第1、第2折射光學單元以外’配 置球面或非球面透鏡以實現所望之軸外光學性能（非點像 散差（astigmatic difference)、像面彎曲），則投影光學 系中的折射光學元件會增加。因此，若爲了提升曝光的效 率而將曝光波長設計成廣頻帶化，則投影光學系中的折射 光學元件之增加所帶來的色像差之增加，就會變成無法忽 視。於是如上記，在倍率補正用的柱形面的背面設置曲面 較爲理想，藉此就可構成小型且高性能的投影光學系。 [其他] -15- 201003119 第1、第2實施形態中，令投影光學系P〇的全體之 成像倍率爲1倍（等倍）時，第1凹反射面Ml與第2凹 反射面M3係被構成爲具有同一曲率半徑。若投影光學系 PO的成像倍率是設成1倍以外時，則第1凹反射面Ml 的曲率半徑與第2凹反射面M3的曲率半徑，係隨著成像 倍率而決定。 [元件製造方法] 本發明的理想實施形態之元件製造方法，係例如適合 於液晶元件、半導體元件之製造，其係含有：在塗佈了感 光劑之基板的該感光劑，使用上記曝光裝置而將原版的圖 案進行轉印之工程，和使該感光劑顯影之工程。進而，經 過其他周知的工程（蝕刻、光阻剝離、硏磨、打線、封裝 等），就可製造出元件。 以上雖然參照實施例來說明本發明，但這僅是爲了便 於理解本案發明所作的例示。在界定本案發明的範圍時， 應參酌申請專利範圍中所記載之事項。當業者在不脫離本 發明範圍內，當然可做各種變更，而這些等價的結構或功 能，當然也都被本發明之範圍所包含。 【圖式簡單說明】 [圖1 ]本發明的理想實施形態之曝光裝置的槪略構成 之圖示。 [圖2]圖1所示之曝光裝置的投影光學系之構成的模 -16- 201003119 式性圖示。 [圖3]第1實施形態的第 之剖面圖。 [圖4]第1實施形態的第 之剖面圖。 [圖5]第1實施形態的第 之剖面圖。 [圖6]第1實施形態的第 之剖面圖。 [圖7]第2實施形態的第 之剖面圖。 [圖8]第2實施形態的第 之剖面圖。 [圖9]第2實施形態的第 之剖面圖。 [圖10]第2實施形態的第 之剖面圖。 【主要元件符號說明】 A :第1折射光學單元 A 1 1 :折射光學元件 A 1 2 :折射光學元件 A21 :折射光學元件 A22 :折射光學元件 1折射光學單元的XZ平面 1折射光學單元的YZ平面 2折射光學單元的XZ平面 2折射光學單元的YZ平面 1折射光學單元的XZ平面 1折射光學單元的YZ平面 2折射光學單元的XZ平面 2折射光學單元的YZ平面 -17- 201003119 AA :第1致動器 AB :第2致動器 AX :光軸 B :第2折射光學單元 B 1 1 :折射光學元件 B 1 2 :折射光學元件 B 2 1 :折射光學元件 B22 :折射光學元件 B A :第2致動器 CA1 1 :柱形面 CA12 :柱形面 CB1 1 :柱形面 C B 1 2 :柱形面 EX :曝光裝置 IL :照明光學系 I :像面 Μ :原版 Μ1 :第1凹反射面 M2 :凸反射面 M3 :第2凹反射面 ΜΑ :反射面 MB :反射面 MD :彎折反射鏡 M S T .原版平台 201003119 〇 :物體面 ρ :基板 ΡΟ :投影光學系 P S Τ :基板平台 S 1 1 :曲面（球面或非球面） S22 :曲面（球面或非球面） -19-[Technical Field] The present invention relates to a projection optical system, an exposure device including the projection optical system, and a device manufacturing method for manufacturing an element using the exposure device. [Prior Art] A flat panel display (hereinafter referred to as FPD) is manufactured by a photolithography project. In the photolithography project, a substrate coated with a photosensitive agent (generally a glass plate in the production of FPD) is projected by a projection optical system to expose the substrate. In an exposure apparatus for exposing a substrate, it is important to reduce the coincidence error. The coincidence error is caused, for example, by positioning error, aberration (distortion), and magnification error. Among these, the positioning error can be reduced by adjusting the relative position of the original plate and the substrate with high precision, but the aberration and the magnification error cannot be completely corrected by the positioning. The magnification error is caused by the expansion and contraction of a substrate such as a glass plate due to factors such as heat generated in the production process. Due to the shrinkage of the substrate, the coincidence accuracy below the second process is deteriorated. Also, the scanning direction in the scanning exposure is different from the vertical direction. In order to reduce the error caused by the expansion and contraction of the substrate, a method of performing exposure by giving a speed difference between the original plate and the substrate in consideration of the expansion and contraction in the scanning direction is considered. Further, in the vertical direction of the scanning direction, it is conceivable to arrange the parallel plate glass in the vicinity of the original plate or the substrate, and to bend the parallel plate glass to bend the light to finely adjust the magnification. A method of correcting magnification is disclosed in Japanese Laid-Open No. 62-3 5620 and Japanese Patent No. 3 3 4 1 269. Japanese Laid-Open No. 62-3 5 620 discloses that in the projection optical system, two lenses that are rotationally symmetrical with respect to the optical axis are arranged close to each other, and one of them is driven in the optical axis direction. Japanese Patent No. 3 3 4 1 No. 3 discloses that at least one of at least two annular optical members that are non-rotationally symmetrical with respect to the optical axis of the projection optical system is rotated in the optical axis or in the optical axis direction. Move on. However, in the method of correcting the magnification in the scanning direction of the substrate and providing a speed difference between the original plate and the substrate, if the good image region is arcuate, a special asymmetric distortion aberration occurs. In order to correct this, it is necessary to drive additional optical components for aberration correction. As a result, the mechanical mechanism is complicated and the drive aberration remains. Further, in the method of correcting the magnification in the direction perpendicular to the scanning direction of the substrate and bending the parallel plate glass, it is difficult to bend the parallel plate glass into a target shape with high precision, which causes a nonlinear image to be generated. Difference error component. Further, in order to increase the size of the substrate in recent years, the parallel plate glass must also be enlarged, which also causes errors. In the method described in Japanese Laid-Open No. Hei. No. 334-26920, Japanese Patent No. 334 1 269, the magnification correction in the scanning direction and the magnification correction in the vertical direction are not individually performed. SUMMARY OF THE INVENTION -6 - 201003119 In the present invention, the magnification correction in the orthogonal direction is expressed in view of the above background, and one side of the present invention can be individually arranged, and the first concave is sequentially arranged in the relevant optical path. The reflection surface and the projection surface are projection optical systems for the object to be placed on the surface of the object, and the projection optical system unit and the second refractive optical unit are arranged to drive the first refraction. The actuator is configured to drive a cylindrical surface of the second refractive optical unit and the second refractive optical unit, and the projection magnification unit of the front projection optical system in the direction of the first refractive optical unit is configured to be The projection magnification of the pre-recorded projection optical system orthogonal to the pre-recording is adjusted; the projection magnification in the first direction of the two or more pre-recording first refracting optical units is the first second actuator, and the second refraction is changed before the second refraction. The interval between the light-columnar surfaces is such that the target projection magnification in the second direction is described. Further points of the present invention will be understood with reference to the embodiments. [Embodiment] The present invention will be described below with reference to the accompanying drawings, and it is intended that the invention can be carried out in two places. The image projected from the object surface to the image plane convex reflection surface and the second concave surface is projected to the front image surface: the first refractive optical is in the front optical path; and the optical unit; and the second unit; the first refractive system There are two or more first actuators arranged in the second direction in which the first and second refracting optical directions are adjusted, and the interval between the cylindrical surfaces is changed so as to achieve the target projection magnification; The projection magnification of one or more is the second embodiment and the preferred embodiment described below. [First Embodiment] Fig. 1 is a schematic view showing a schematic configuration of an exposure apparatus EX according to a preferred embodiment of the present invention. Fig. 2 is a schematic illustration showing the configuration of the projection optical system PO of the exposure apparatus EX shown in Fig. 1. Further, in Fig. 2, a bending mirror (mirror of zero power) MD required for bending the optical axis AX of the projection optical system PO is omitted. It is also conceivable that the bending mirror MD bends the optical path. The bending mirror MD is configured to, for example, bend the optical axis AX by 90 degrees between the object surface 配置 disposed on the original Μ and the first concave reflecting surface M1 by the reflecting surface MA. Between the concave reflecting surface Μ 3 and the image plane I on which the substrate P is placed, the optical axis is bent at 90 degrees by the reflecting surface MB. In the actual configuration, it is advantageous to provide the bending mirror MD in order to reduce the size of the projection optical system PO, but the bending mirror MD is not essential. In the present specification, in order to explain the spatial direction, the XYZ coordinate system defining the Z direction parallel to the optical axis AX is defined when the state of the bending mirror MD is removed. The exposure apparatus EX is preferably used for manufacturing a large-area element such as a flat panel display such as a liquid crystal display, but can be applied to a semiconductor element such as an LSI or other elements. The exposure device EX includes an illumination optical system IL for illuminating an original plate (also referred to as a mask or a mask), an original plate MST for holding the original plate, a projection optical system, and a substrate for holding the substrate (for example, a glass plate). The tablet platform PST of the flat plate). The original lanthanum is disposed on the object plane 0 of the projection optical system 201003119 PO, and the substrate P is disposed on the image plane I of the projection optical system PO. The exposure apparatus EX is typically configured as a scanning exposure apparatus. In the exposure apparatus EX, the original printing plate is formed into a strip shape (for example, an arc shape or a rectangular stripe shape) by the illumination optical system IL, and the original plate and the substrate are scanned while the original plate is scanned. The platform MST and the substrate platform PST are scanned and driven. Thereby, the pattern of the original enamel is transferred to the sensitizer on the substrate 。. The projection optical system Ρ0 has a good image area at a position shifted from the optical axis ,, and the good image area is used to project the pattern of the original enamel onto the substrate Ρ. The projection optical system Ρ0 is arranged in the optical path from the object plane to the image plane I in order: the first refracting optical unit A, the first concave reflecting surface M1, the convex reflecting surface M2, the second concave reflecting surface M3, and the first 2 Refractive optical unit B. A reflection surface 弯 of the bending mirror MD is disposed between the first refracting optical unit A and the first concave reflecting surface Μ1. A reflection surface MB of the bending mirror MD is disposed between the second concave reflecting surface M3 and the second refractive optical unit 。. The first refracting optical unit A and the second refracting optical unit B are not limited to the above-described positions, and may be disposed in the optical path from the object plane 0 to the image plane I. The first concave reflecting surface M1 and the second concave reflecting surface M3 are typically reflecting surfaces having the same center of curvature. The first concave reflecting surface M1 has a positive power, the convex reflecting surface M2 has a negative power, and the second concave reflecting surface has a positive power. The first refracting optical unit A and the second refracting optical unit B are optical units for correcting the projection magnification of the projection optical system p -9-201003119. More specifically, the first refracting optical unit A is disposed so as to be able to adjust the projection optical system PO in the first direction (the X direction in a state where the bending mirror MD is removed) Projection magnification. Further, the second refracting optical unit A is disposed so as to be able to adjust the projection in the second direction orthogonal to the first direction (the Y direction in the state where the bending mirror MD is removed) Projection magnification of the optical system PO. Each of the first refracting optical unit A and the second refracting optical unit B has two or more cylindrical surfaces. The first refractive optical unit A is driven by the first actuator AA so that the projection magnification in the first direction (the X direction in the state where the bending mirror MD is removed) is removed. It becomes the first target projection magnification. Specifically, the first actuator AA changes the interval between the two or more cylindrical surfaces of the first refracting optical unit A so that the projection magnification in the first direction becomes the first target projection magnification. The second refractive optical unit B is driven by the second actuator BA so that the projection magnification in the second direction (the Y direction in the state where the bending mirror MD is removed) is removed. It becomes the second target projection magnification. Specifically, the second actuator BA changes the interval between the two or more cylindrical surfaces of the second refractive optical unit B so that the projection magnification in the second direction becomes the second target projection magnification. Here, the first target projection magnification and the second target projection magnification are determined by, for example, a measurement unit such as a locator (not shown) that detects the position of the positioning mark formed on the substrate P. It. The first refractive optical unit A typically has two cylindrical faces -10- 201003119 . As a combination of the characteristics (power) of the two cylindrical faces, convex and convex, concave and concave, convex and concave can be considered. Here, in order to make the change in the projection magnification with respect to the change in the focus position, it is preferable to design the projection optical system P 〇 to be a telecentric type. Therefore, the combination of the characteristics (power) of the two cylindrical surfaces is preferably a combination of convex and concave '. The bus bars ' of the cylindrical faces included in the first refractive optical unit A are typically parallel to each other. The second refractive optical unit B is also typically composed of two cylindrical surfaces, and the combination of characteristics (power) is preferably a combination of convex and concave. Further, the bus bars ' of the cylindrical faces included in the second refractive optical unit B are typically parallel to each other. The bus bars of the two or more cylindrical surfaces included in the first refracting optical unit A are parallel to each other (in the state where the bending mirror MD is removed, the optical axis of the projection optical system PO) The third direction (Y direction) orthogonal to AX. The bus bars of the two or more cylindrical faces included in the second refractive optical unit B are parallel to each other (in the state in which the bending mirror MD is removed), the optical axis of the projection optical system PO AX and the fourth direction (X direction) orthogonal to the third direction. 3 and 4 are views showing a configuration example of the first refracting optical unit A. Here, Fig. 3 is a cross-sectional view showing the first refracting optical unit A cut along the XZ plane, and Fig. 4 is a cross-sectional view showing the first refracting optical unit A cut along the Y Z plane. The first refracting optical unit A' includes a refracting optical element A 1 1 disposed on the object plane side and a refracting optical element A 1 2 disposed on the image plane I side. The refracting optical element A 1 1 has a convex cylindrical surface CA11 on the image plane I side, and has a convex power as a whole. The refractive optical element A12 -11 - 201003119 ' has a concave cylindrical surface CA12 on the side of the object surface, and has a concave power as a whole. The first refractive optical unit A may have more cylindrical faces. In other words, the first refractive optical unit A may have two or more cylindrical surfaces. The bus bars of the two or more cylindrical surfaces CA11 and CA12 of the first refractive optical unit A are parallel to each other (when the bending mirror MD is removed, the optical axis of the projection optical system PO is removed) The third direction (Y direction) orthogonal to AX. In order to correct the projection magnification in the first direction of the projection optical system PO (in the X direction in the state where the bending mirror MD is removed), two or more cylindrical surfaces CA 1 1 and CA 1 At least one of the two is driven by the first actuator AA in the optical axis ΑΧ direction. 5 and 6 are diagrams showing a configuration example of the second refractive optical unit Β. Here, Fig. 5 is a cross-sectional view showing the second refracting optical unit Β in the ΧΖ plane, and Fig. 6 is a cross-sectional view showing the second refracting optical unit Β in the ΥΖ plane. The second refractive optical unit Β ' includes a refracting optical element Β 1 1 disposed on the object plane side and a refracting optical element Β 1 2 disposed on the image plane I side. The refracting optical element Β 1 1 ' has a concave cylindrical surface CB11 on the object surface I side, and has a concave power as a whole. The refracting optical element Β 1 2 has a convex cylindrical surface CB 1 2 on the side of the object surface, and the whole has a convex power. The second refractive optical unit can have more cylindrical faces. That is, the second refractive optical unit Β' may have two or more cylindrical surfaces. The bus bars of the two or more cylindrical surfaces CB11 and CΒ 1 2 of the second refractive optical unit 平行 are parallel to 'the projection optical system P0 in the state where the bending mirror MD is removed. The optical axis AX is the -12th - 201003119 intersection of the 4th direction (X direction). In order to correct the projection magnification in the direction of the projection light 2 (the Υ direction in which the bending mirror MD is provided), at least one of the two or more cedar CB12 is the second actuator ΑΒ upward Drive. According to this embodiment, the correction of the projection magnifications in the two orthogonal directions and in the direction orthogonal thereto is performed. Thereby, for example, the superimposition accuracy of the image after the projection magnification which has been formed in the complementary scanning direction of the substrate and the direction orthogonal thereto can be improved. [Second embodiment] Hereinafter, a second embodiment of the present invention will be described. These matters are the same as in the first embodiment. Fig. 7 and Fig. 8 are views showing a configuration example of the first fold in the second embodiment. Here, Fig. 7 is a cross-sectional view in which the first fold is cut off from the kneading surface, and Fig. 8 is a cross-sectional view in which the first turn is cut off from the kneading surface. The first refracting optical unit A includes A22 disposed on the refracting optical element A2 1 and disposed on the image plane I side. The refracting optical element A2 1 has a convex cylindrical surface CA11 as the image plane I side surface, and has a curved surface (spherical surface or aspherical surface) S11 as the object plane 0 2 surface, and has full power. The refracting optical element A22 is in a state after the removal of the object surface learning system PO: the shape surface CB11, on the optical axis AX side, for example, the scanning side can individually enter the pattern to individually rate, so that the second portion can be eliminated. In particular, the surface of the refracting optical element on the object plane 0 side of the refracting optical unit, that is, the surface of the first side, that is, the surface of the first side, which has a convex side and a concave side, has a concave column-13-201003119. The whole is concave power. The first refractive optics can have more cylindrical faces. That is, the 'first refractive optical unit' may have two or more cylindrical surfaces. The bus bar of the first refracting optical unit or more of the plurality of cylindrical surfaces CA 1 1 and CA 1 2 is parallel to the optical axis of the ' (the state in which the mirror MD is removed) The third direction orthogonal to the direction (Υ direction). At least one of the cylindrical surfaces CA 1 1 and CA 1 2 having a projection magnification equal to or greater than the projection magnification in the first direction of the orthographic projection optical system (the X direction in the state in which the bending mirror is removed) 'It is driven by the actuator AA in the optical axis AX direction. Fig. 9 and Fig. 10 show an example of the configuration of the second refracting optical unit B. Here, Fig. 9 is a cross-sectional view showing the second refracting optical unit B on the XZ plane, and Fig. 1 is a cross-sectional view showing the second refracting optical unit B after Y Z is broken. The second refracting optical unit B includes a refracting optical element B2 1 on the side of the surface of the ligand, and is disposed on the image surface I optical element B22. The refracting optical element B21 has a concave cylindrical surface CB11 on the object surface, and has a concave power as a whole. The refracting member B 2 2 has a surface CB 1 2 as a surface on the side of the object surface, and has a spherical surface or an aspheric surface as a surface on the image surface I side, that is, a spherical surface or an aspheric surface. It has a convex power. The first optical unit B can have more cylindrical faces. That is, the second unit B may have two or more cylindrical faces. The busbars of the two or more cylindrical surfaces CB11 and CB12 of the second refracting element B are flat (when the bending mirror MD is provided, the state unit A is removed, and the second of the A is A When the MD is complemented, the two first diagrams cut off the surface of the cylinder that is cut on the object side and has the optical element convex (2 refracted refracting optical single line, below) -14 - 201003119 Projection optics The fourth direction (X direction) orthogonal to the optical axis PO of the PO. In order to correct the second direction of the projection optical system ( (when the bending mirror MD is provided, the Υ direction is removed) In the upper projection magnification, at least one of the two or more cylindrical surfaces CB11 and CB12 is driven by the second actuator ΑΒ in the optical axis 。 direction. In the above example, the first refractive optical unit Α (2) Both of the refractive optical units B have a curved surface on the back surface of the cylindrical surface. However, either one of the first refractive optical unit A and the second refractive optical unit B may have a curved surface on the back surface of the cylindrical surface. In this embodiment, the back surface of the cylindrical surface is designed as a curved surface ( A faceted or aspherical surface can form a small and high-performance projection optical system. In particular, by designing the curved surface into an aspherical shape, off-axis optical properties (astigmatic difference, The performance of the image plane is constant, and the band-like (for example, arc-shaped) image area is enlarged. In addition, if a spherical or aspheric lens is disposed in addition to the first and second refractive optical units to achieve the desired axis, Optical performance (astigmatic difference, field curvature) increases the number of refractive optical elements in the projection optical system. Therefore, if the exposure wavelength is designed to be wide-band in order to improve the efficiency of exposure, the projection The increase in chromatic aberration caused by the increase in the refractive optical element in the optical system is not negligible. Therefore, it is preferable to provide a curved surface on the back surface of the cylindrical surface for magnification correction as described above. Small and high-performance projection optical system. [Others] -15- 201003119 In the first and second embodiments, the imaging magnification of the entire projection optical system P〇 is 1 time (equal times) The first concave reflecting surface M1 and the second concave reflecting surface M3 are configured to have the same radius of curvature. When the imaging magnification of the projection optical system PO is set to be one time, the radius of curvature of the first concave reflecting surface M1 is obtained. The radius of curvature of the second concave reflecting surface M3 is determined by the imaging magnification. [Element Manufacturing Method] The device manufacturing method according to a preferred embodiment of the present invention is suitable for, for example, production of a liquid crystal element or a semiconductor element. : the sensitizer on the substrate coated with the sensitizer, the transfer of the original pattern using the exposure device, and the development of the sensitizer. Further, other well-known engineering (etching, photoresist) Components can be fabricated by stripping, honing, wire bonding, packaging, etc.). The present invention has been described above with reference to the embodiments, but this is only for the purpose of understanding the invention. In defining the scope of the invention, the matters recited in the scope of the patent application shall be considered. Various changes may be made by those skilled in the art without departing from the scope of the invention, and such equivalent structures or functions are of course included in the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic view showing a schematic configuration of an exposure apparatus according to a preferred embodiment of the present invention. Fig. 2 is a view showing the configuration of the projection optical system of the exposure apparatus shown in Fig. 1 -16-201003119. Fig. 3 is a cross-sectional view showing the first embodiment. Fig. 4 is a cross-sectional view showing the first embodiment. Fig. 5 is a cross-sectional view showing the first embodiment. Fig. 6 is a cross-sectional view showing the first embodiment. Fig. 7 is a cross-sectional view showing a second embodiment. Fig. 8 is a cross-sectional view showing a second embodiment. Fig. 9 is a cross-sectional view showing a second embodiment. Fig. 10 is a cross-sectional view showing a second embodiment. [Description of main component symbols] A: First refractive optical unit A 1 1 : Refractive optical element A 1 2 : Refractive optical element A21 : Refractive optical element A22 : Refractive optical element 1 XZ plane of refractive optical unit 1 YZ of refractive optical unit XZ plane of the plane 2 refractive optical unit YZ plane of the refractive optical unit 1 XZ plane of the refractive optical unit 1 YZ plane of the refractive optical unit 2 XZ plane of the refractive optical unit 2 YZ plane of the refractive optical unit -17- 201003119 AA : 1 actuator AB: second actuator AX: optical axis B: second refractive optical unit B 1 1 : refractive optical element B 1 2 : refractive optical element B 2 1 : refractive optical element B22: refractive optical element BA: Second actuator CA1 1 : cylindrical surface CA12 : cylindrical surface CB1 1 : cylindrical surface CB 1 2 : cylindrical surface EX : exposure device IL : illumination optical system I : image plane 原 : original Μ 1 : first concave Reflecting surface M2: convex reflecting surface M3: second concave reflecting surface ΜΑ: reflecting surface MB: reflecting surface MD: bending mirror MST. Original platform 201003119 〇: object surface ρ: substrate ΡΟ: projection optical system PS Τ : substrate platform S 1 1 : surface (spherical or aspherical) S22 : curved surface Spherical or aspherical) -19-