1274733 玖、發明說明： 【發明所屬之技術領域】 本發明係有關微小流路之形成方法。特別是有關在微化 學之領域所使用之在微晶片形成微小流路之方法。 【先前技術】 近年，利用微機械技術，將執行溶液的混合、反應、分離、 以及檢測等的系統積體化在數公分方材的玻璃基板上之所 謂的實驗室晶片（Laboratory on a Chip)的裝置技術係 熱烈地被硏究。實驗室晶片係因應積體化的系統，也稱爲# -TAS(微總分析系統）、微反應器等等。 通常，實驗室晶片係具備有形成在厚度1 mm程度的基板 上之溝寬爲數十〜數百// m的微小流路，在微小流路中執行 溶液的混合等作業。在微小流路中由於比界面積變大，因尺 寸效應而難以反應者係會反應、難以混合者係會混合，可有 效率地執行溶液的混合及反應。藉由把微小流路的溝寬設 定爲1 0 // m〜5 0 // m，可使流路阻力比較小，可獲得良好的尺 寸效應。又，微小流路的形狀係對流體的送液特性有大的影 響，所以微小流路係具備平滑的壁面且以高精度地製成爲 佳。 以往，實驗室晶片的微小流路係利用以阻體膜被覆基板 表面，且依利用紫外線或電子線的微影成像術將阻體膜圖案 化之後，再以此作爲遮罩將基板蝕刻之半導體加工技術所形 成。微影成像術係利用在半導體製程所使用的密接曝光裝 置而被執行。其曝光方式係使用有遮罩校準的類比曝光方 -6- 1274733 式，例如1平方米的大面積係難以高速曝光。 【發明所欲解決之課題】 然而，以往的微小流路之形成方法中，係利用遮罩曝光以 執行圖案化，所以光阻膜的厚度被限制，具有難以高精度形 成微小流路之問題。亦即，當光阻膜薄時，則在蝕刻基板之 際，側面蝕刻變容易，溝寬的製作精度降低的同時，變成不能 達成足夠的溝深。 又，在遮罩曝光中，對各圖案之高精度的玻璃遮罩等係成 爲必要，所以具有成本變高，造成大面積化困難的同時也不 適合少量多樣的生產之問題。 一方面，也考量到將微影成像術工程以數位曝光方式來執 行，但是使用紫外線的以往之數位曝光裝置係以單一光束的 掃描曝光，曝光時間係花費過多。特別是，光束直徑爲丨〇 # m 以下且定址能力爲1 V m程度的高精細曝光時，具有曝光時 間過長的問題。 本發明係爲解決上述問題而成者，本發明之目的係提供一 可高速且高精度形成微小流路之微小流路之形成方法。又， 本發明之其他目的係提供一種可低成本形成任意圖案的微 小流路之微小流路的形成方法。 【發明内容】 【解決課題之手段】 爲達成上述目的，本發明之微小流路之形成方法的特徵爲 具備有：曝光工程，以對應微小流路之形成圖案資料而在空 間作調變之波長3 5 0nm〜450nm的雷射光，曝光被形成在基 1274733 板上之阻體膜；圖案化工程，對應曝光圖案部分地去除該阻 體膜以形成指定圖案之阻體膜；蝕刻工程，使用該指定圖案 之阻體膜，由表面蝕刻該基板且去除以形成微小流路。 在此微小流路的形成方法中，因爲使用波長3 5 0 n m〜4 5 0 n m 的雷射光，所以沒有必要使用如準分子雷射之紫外線對應的 特殊材料的光學系統，與可視域之雷射曝光裝置同樣地，係 可使用DMD等之空間光調變元件。藉此，因應微小流路之形 成圖案資料，能以空間調變之雷射光將阻體膜曝光。亦即， 可將在任意圖案之阻體膜予以高速且高精細地作數位曝 光。 如此，在曝光工程中，因爲可高速且高精細地曝光在任意 圖案之阻體膜，經過下一個圖案化工程及蝕刻工程之後，可 高速且高精度地形成任意圖案之微小流路。又，因爲是數位 曝光、所以不要各圖案之遮罩，可低成本地形成微小流路。 在上述的曝光工程係可使用具備有照射雷射光之雷射光 源、和具有因應各自控制信號而光調變狀態會變化之多數 個畫素部以矩形狀配置在基板上且用以把由該雷射裝置所 照射的雷射光予以調變的空間光調變元件、以及使在各畫 素部被調變的雷射光成像在曝光面上之光學系統的曝光 頭。又，使此曝光頭相對於阻體膜的曝光面，在與指定方向 交叉的方向相對移動，可將形成在基板上的阻體膜作掃描曝 光。 爲了將阻體膜更高精細地曝光，空間光調變元件係使其各 畫素部的配列方向爲與垂直於副掃描方向之方向成指定角 1274733 度~般地稍傾斜配置而作多重曝光爲較佳。藉此，可以光束 直徑1 0 # m以1 // m之定址能力高精細地曝光。傾斜角度0 爲1〜5的範圍爲較佳。。 又，在空間調變元件的出射側，更好爲配置有對應空間調 變元件之各畫素部而設置且具備在各畫素集光雷射光之微 透鏡的微透鏡陣列。在配置有微透鏡陣列之場合，在空間調 變元件之各畫素部所調變之雷射光係因應微透鏡陣列之各 微透鏡而對應各畫素而集光，所以即使在被曝光面中之曝光 區域被放大時，也可縮小各光束光點的尺寸，可執行高精細 的曝光。依使用此縮小光學系統，係能以1 // m的光束直徑， 以0 . 1 // m的定址能力超高精細地曝光。 如此將阻體膜作高精細的曝光，可形成非常平順的微小流 路之壁面，可減低流路阻力以獲得良好的尺寸效應。 爲高精度形成微小流路，阻體膜之厚度係厚者爲較佳。 在形成溝寬1 Ο μ m〜5 0 // m之微小流路的場合，阻體膜之厚 度係l〇/zm〜50//m爲較佳，10//m〜100/zm爲更好。特別是， 使阻體膜成爲2層及3層般地疊層複數層而執行曝光者爲 更好。因爲係將阻體膜數位曝光，所以利用數位定比機能可 筒精度地執行曝光時及顯影後％之延伸等的補正，第1層之 曝光位置和第2層等等之複數層之曝光位置的定位係可高 精度地實現。其結果爲以往之2倍厚度的阻體膜、高精度 且高長寬比的圖案化係成爲可能，可用蝕刻形成高精度且深 的微小流路。此外，所謂的長寬比係阻體膜所形成之溝的 溝寬a對溝深b之比率a / b。 -9- 1274733 在上述形成方法之曝光工程中，藉由使用高亮度光源以 深的焦點深度作曝光，可以更局精度將阻體膜曝光。以局売 度光源而言，係將複數之雷射光合波使入射至各自光纖之合 波雷射光源爲合適。又，厚膜化之阻體膜的曝光係需要高輸 出之雷射光源。振盪波長3 50〜45 Oiim之半導體雷射要以單 一元件之高輸出化雖然難，但藉由合波可圖謀高輸出化。 合波雷射光源係可爲例如：（1 )包含有複數個半導體雷 射、和1條光纖、以及將該複數個半導體雷射各自所出射 的雷射光予以集光，使集光束結合至該光纖的入射端之集光 光學系統之構成；（2)包含有具備複數個發光點之多腔雷 射、和1條光纖、以及將該複數個發光點各自所出射的雷 射光予以集光，使集光束結合至該光纖的入射端之集光光學 系統之構成；或者爲（3 )包含有複數個多腔雷射、和丨條 光纖、以及將該複數個多腔雷射之該複數個發光點各自所 出射之雷射光予以集光，使集光束結合至該光纖之入射端的 集光光學系統之構成。 可以把在上述之合波雷射光源之光纖的出射端中之發光 點各自予以陣列狀配列作成光纖陣列光源、把發光點各自 予以束狀配列作成光纖束光源。藉由束化或陣列化，可更加 圖謀高輸出化。又，由圖謀高亮度化之觀點來說，較佳爲使 用核心直徑係均一且出射端的包層直徑比入射端的包層直 徑還小的光纖。 從使發光點的直徑成爲小的觀點，光纖出射端之包層直徑 係比1 2 5 // m小者較好，8 0 // m以下更好，6 〇 #仍以下係特別 1274733 好。、核心直徑爲均一且出射端之包層直徑爲比入射端之包 層直徑更小的光纖係，例如可以將核心直徑相同且包層直徑 · 爲不同之複數個光纖予以結合而構成。依此，可使陣列化之 際的發光區域更小而可高亮度化。又，藉由構成爲將複數的 光纖連接成在連接器可裝卸，使得當光源模組有部分破損時， 其交換係變容易。 特別是，在將如上述之空間光調變元件傾斜配置且使用縮 小光學系統或等倍光學系統以執行超高精細曝光之場合中， 藉由使用前述之高亮度光纖陣列光源或光纖束光源，因爲可 鲁 使空間調變元件之照明NA設小，所以通過空間調變元件後 之成像光束的焦點深度可取深，可獲得深的焦點深度，在阻 體表面及阻體內光束不會太寬，而成爲可更高精度且高長寬 比之圖案。又，在形成壁面爲傾斜的傾斜流路之場合，也可 獲得平順的圖案。 上述的曝光工程中，雷射光係例如，照射於基板上配列有 因應各個控制信號而光調變狀態會變化之多數個畫素部空 間光調變元件，而在該空間光調變元件之各畫素部被調變。 魯 以空間調變元件而言，可使用在基板上（例如，矽基板） 以2維狀配列有因應各假控制信號可變更反射面角度之多 數個微鏡所構成之微鏡裝置（DMD ;數位微鏡裝置）。又、 空間調變元件也可構成爲，把具備帶狀反射面且因應控制信 號而可移動的可動格子和具備帶狀反射面的固定格子予以 交互地多數個並列配置所構成之1維的光柵燈泡（GLV )。 又，也可以使用在基板上以2維狀配列有可因應各個控制信 - 1 1 - 1274733 號以遮斷透過光之多數個液晶胞所構成之液晶遮板陣列。 在此等空間調變元件之出射側，較佳爲配置有對應空間調 變元件之各畫素部而設置且具備在各畫素將雷射光集光之 微透鏡的微透鏡陣列。在配置有微透鏡陣列之場合，在空間 調1變元件之各畫素部被調變的雷射光係利用微透鏡陣列之 各微透鏡、以對應各畫素而被集光，所以在被曝光面中之曝 光區域被放大之場合，也可縮小各光束光點的尺寸，即使在 大面積化的場合時也可執行高精細的曝光。 【實施方式】 以下，參照圖面以詳細說明使用本發明的微小流路之形成 方法來製造合成反應用微晶片之實施形態。 〔其他空間調變元件〕 上述的實施形態中，雖然已針對作爲空間調變元件之具 備有DMD的曝光頭加以說明，但是例如可使用GLV等之MEMS (微機電系統）型之空間調變元件（SLM )、依電氣光學效 果而調變透過光之光學元件（PLZT元件）及液晶光遮板 (FLC )等之MEMS型以外之空間調變元件。此外，所謂的MEMS 係以I C製程爲基礎的微機械技術所成之微尺寸的感測器、 致動器，然後把控制電路予以積體化的微系統之總稱，所謂 的MEMS型之空間調變元件係意味著利用靜電力之電氣機械 動作所驅動之空間調變元件。 〔合成反應用微晶片〕 合成反應用微晶片係如第1圖所示，係在玻璃等所形成的 平板狀基板150上重疊保護基板202而構成。基板150低 1274733 厚度通常爲0.5mm〜2.0mm程度，保護基板202之厚度通常 爲0 · 1關〜2 · Omni程度。在各保護基板20 2上係貫通設置有 用以注入試藥之注入口 204 a、204b ,以及將試藥反應後得到 之反應液予以排出的排出口 2 0 6。在基板1 5 0係設置有使試 藥或反應液流通之微小流路2 0 8。微小流路2 0 8係配設成在 由注入口 2 0 4 a、2 0 4 b被各自注入的試藥在合流點2 1 0合流 之後，朝排出口 2 0 6被排出。微小流路之溝寬係數^--數百 //m，而10/zm〜50/zm係特別好。溝寬爲lOem〜50//m的 微小流路係流路阻力比較小，所以可獲得良好的尺寸效應。 對此反應用微晶片的注入口 204a、204b各自注入試藥， 當由排出口 2 0 6側吸引時，試藥係在微小流路2 〇 8流通且在 合流點2 1 0被混合而反應。依此可合成所期望的物質。獲 得之反應液係在微小流路208流通且自排出口 206被排出。 依自此排出口 206獲得之反應液的分析，可相同於在通常定 比之反應，執行反應生成物之識別或定量。 〔微晶片之製造方法〕 接著參照第2圖以針對上述合成反應用微晶片的製造方 法加以說明。此製造方法係由把光阻膜曝光之曝光工程、 將光阻膜部分除去以圖案化之圖案化工程、將基板融刻以 形成微小流路的蝕刻工程、以及將形成有微小流路的基板 和保護基板予以接合的接合工程等所構成。以下茲說明各 工程。 如第2 ( A )圖所示，在基板1 5 0上以旋轉塗布法等方式 形成光阻膜2 1 2之後，如第2 ( B )圖所示，依微小流路2 0 8 1274733 的圖案將光阻膜2 1 2曝光，如第2 ( C )圖所示，使曝光部分 2 1 4溶解於顯影液且加以除去。在此，藉由將光阻膜2 1 2以 高的位置精度予以圖案化，可高精度地形成微小流路20 8。 此外，有關光阻膜2 1 2的曝光工程係在後面敘明。 接著，如第2 ( D )圖所示，使用被圖案化的光阻膜2 1 2 ,由 表面蝕刻基板150以形成微小流路208，如第2( E )圖所示， 除去剩餘的光阻膜2 1 2。基板1 5 0的鈾刻係可執行乾式鈾刻 及濕式蝕刻中任一者，但是由於是微細加工，所以高速原子 線（F A B )鈾刻等之乾式蝕刻係適合的。 接著，如第 2 ( F )圖所示，利用超音波加工等方式在保護 基板202形成成爲注入口 204a、204b及排出口 206的貫通 孔。然後如第 2 ( G)圖所示，使保護基板202和基板150 之形成有微小流路208的面成爲對向般地將兩基板重疊密 接而加以固定。例如可使用UV接著劑來固定。在保護基板 2 〇 2之形成有微小流路2 0 8的面以旋轉塗布法等方式塗布UV 接著劑，在使基板1 5 0和保護基板2 0 2密接之後，再照射紫 外線而接著。 此外，基板150和保護基板202爲以玻璃形成之場合，以 氫氟酸將兩基板的表面溶解再接合也可以。 〔光阻膜之曝光〕 以下，針對光阻膜之曝光工程加以詳細說明。在此曝光工 程中，使用空間光調變兀件，把波長3 5 0 n m〜4 5 0 n m的雷射光; 因應微小流路的形成圖案資料作調變，再以調變後的雷射光 將光阻膜212作數位曝光。爲了以更高精度來執行曝光，以 1274733 由高亮度光源所出射之深的焦點深度之雷射光來曝光係較 佳。 〔光阻膜〕 光阻膜 212係可使用在印刷配線基板（PWB ; Printed Wirmg Board)的製造工程所使用之乾式阻體膜（DFR ; Dry Fi lm Res i s t )或電極沈積阻體。此等DFR或電極沈積阻體， 與在半導體製程所使用的阻體相較下係可厚膜化，可形成厚 度10//ΙΏ〜40//m的膜。 又，藉由將光阻膜積層複數層而可圖謀更厚膜化。此時， 如第3 ( A)圖所示，形成第1光阻膜21 2a，將指定區域214a 曝光後，如第3 ( B )圖所示，在第1光阻膜2 1 2 a上形成第2 光阻膜2 1 2b，使用數位曝光之定比機能，將對應指定區域 214a的區域214b作曝光。如第3 ( C )圖所示，當除去被曝 光之區域2 1 4 a及區域2 1 4b時，係形成依阻體的深溝。此外， 在此例中，係針對把阻體膜作2層積層的例子加以說明，但 是把阻體膜作3層、4層的積疊、利用數位曝光之定比機能 將相同位置曝光係可形成更深的溝。此外，在此，曝光係在 不透過顯影工程之下，重疊2層以上作說明，但是把第1層 曝光，其後顯影.，顯影後的基板之延伸或阻體之膨潤等係利 用數位定比作補正，把第2層曝光般地在顯影後作曝光且第 3層、第4層也同樣作曝光也可以。依此、顯影時之圖案的 位置偏差也可高精度地補正。 又，如此藉由將光阻膜2 1 2厚膜化，可形成依阻體的深溝， 可藉蝕刻而在基板202形成精度佳的深溝（微小流路）。 1274733 例如，由第 4 ( A )及 4 ( B )圖可知，在利用FAB蝕刻來形 成相同溝寬之微小流路時，當光阻膜2 1 2爲薄時，依斜向光 基板1 5 0可易於側面蝕刻，光阻膜2 1 2爲厚時，因受阻使得 斜向光係難以入射，基板1 5 0係變難以被側面蝕刻。藉此， 可在基板1 50形成精度佳的深溝。 在形成溝寬1 〇 # m〜5 〇 // m之微小流路的場合，光阻膜2 1 2 之厚度係10//m〜50/zm較好，10//m〜100//m更好。 X ,在藉由利用蝕刻溶液之濕式蝕刻來形成微小流路之場 合，如第5圖所示，在光阻膜2 1 2也可把推拔狀擴開的開口 2 1 6形成圖案。因爲開口 2 1 6係推拔狀擴開，所以容易使鈾 刻溶液浸入。 〔曝光裝置〕 在光阻膜的曝光工程所使用的曝光裝置方面、例如可使 用第6圖所示之裝置，此裝置具備有將形成有阻體膜之基板 1 5 0吸附在表面而予以保持之平板狀的載物台1 5 2。而在由 4根腳部1 5 4所支持之厚板狀之設置台1 5 6的上面，係設置 有沿著載物台移動方向延伸之2根導引部1 5 8。載物台1 5 2 係使其長度方向朝載物台移動方向而配置，同時依導引部 1 5 8以可往復移動地被支持著。此外，在此曝光裝置設置有 用以使載物台152沿著導引部158驅動之未圖示的驅動裝 置。 設置台156的中央部係設置有跨越載物台152的移動路 徑般之〕字狀閘門160。〕字狀之閘門160的端部係各自固 定在設置台1 5 6之兩側面。挾住此閘門1 6 0而在一側係設 - 1 6 - 1274733 置有掃描器162,他側係設置有用以檢測感光材料15〇之前 端及後端的複數（例如2個）個檢測感測器丨6 4。掃描器1 6 2 及檢測感測器1 64係各自被安裝在閘門丨60且固定配置在 載物台1 52之移動路徑的上方。此外，掃描器丨62及檢測感 測器1 6 4係連接在未圖示之用以控制此等之控制器上。 f币描窃162如弟7圖及弟8(B)圖所不，係具備有ill行η 列（例如，3行5列）之略矩陣狀配列的複數個（例如，！ 4 個）曝光頭166。在此例中，因爲與感光材料150之寬度的 關係，在第3行配置了 4個曝光頭1 6 6。此外，在表示配列在 第 m行的第η列之各個曝光頭之場合時，係表示成曝光頭 1 6 6mn 〇 依曝光頭166的曝光區域168係以副掃描方向爲短邊之 矩形狀。因此，隨著載物台1 52之移動，形成於基板1 50之 阻體膜係形成有各曝光頭166之帶狀的已曝光區域170。此 外，在表示第m行之第η列所配列之各個曝光頭的曝光區域 之場合，係表示爲曝光區域168mn。 又，如第8(A)圖及8(B)圖所示，帶狀之已曝光區域170 係無間隙地在與副掃描方向正交的方向排列，線狀配列之各 行的曝光頭各自係在配列方向以指定間隔（曝光區域之長 邊的自然數倍，本實施形態中爲2倍）偏移而配置著。因此， 在第1行的曝光區域1 6 8: i和曝光區域1 6 8 i 2之間之不能曝 光的部分係可依第2行之曝光區域1 6 8 21和第3行的曝光區 域1 6 8 3 i而曝光。 曝光頭166M〜166mn係各自如第9，10(A)及10(B)圖 1274733 所示，具備數位微鏡裝置（DMD ) 5 0以作爲因應畫像資料把 入射光束因應畫像資料而對各畫素作調變之空間光調變元 件。此DMD5 0係連接在未圖示之具有資料處理部和鏡驅動 控制部之控制器上。此控制器之資料處理部係依據輸入的 畫像資料，生成用以對各曝光頭166之DMD50之應控制區 域內的各微鏡驅動控制之控制信號。又，鏡驅動控制部係依 據在畫像資料處理部生成的控制信號，控制各曝光頭1 66 之DMD50之各微鏡的反射面之角度。此外有關反射面之角 度控制係在後面加以敘述。 在DMD50的光入射側係以如下之順序配置即：備有光纖的 出射端部（發光點）沿著與曝光區域1 68之長邊方向對應 之方向成一列配列的雷射出射部之光纖陣列光源6 6 ;把由光 纖陣列光源66所出射之雷射光作補正且使集光於DMD上之 透鏡系67 ;以及將透射透鏡系67的雷射光朝DMD50反射之 鏡69。 透鏡系67，係由使光纖陣列光源66所出射的雷射光平行光 化之1對組合透鏡7 1、使被平行光化的雷射光之光量分布成 爲均一*般而加以補正之1對組合透鏡73、以及把光量分布被 補正的雷射光集光於DMD上之集光透鏡7 5所構成。組合透鏡 7 3係具備有，對雷射出射端之配列方向，接近透鏡的光軸之 部分爲擴大光束且離開光軸的部分係光束縮減，且在與此配 列方向正交的方向使光照其原樣通過之機能，使光量分布成 爲均一般地補正雷射光。 又，在DMD5 0的光反射側配置有使在DMD50反射的雷射光成 1274733 像於感光材料150的掃描面（被曝光面）56上之透鏡系54、 58。透鏡系54及58係配置成使DMD50和被曝光面56成爲共軛 的關係。 DMD50係如第11圖所示，在SRAM胞（記憶體胞）60上，微 小鏡（微鏡）62係由支柱所支持而配置者，係使構成畫素 (PIXEL)之多數個（例如，600個X 800個）微小鏡以格子 狀配列所構成之鏡裝置。各畫素之最上部係設置有由支柱 所支持的微鏡62 ,微鏡62的表面係蒸鍍有鋁等之反射率高的 材料。此外，微鏡62的反射率係90%以上。且在微鏡62的正 下係透過包含有鉸鏈及軛架的支柱配置有在通常的半導體 記憶體之生產線所製造之矽閘門的CMOS之SRAM胞60，全體係 構成爲整塊（一體型）。 當DMD50的SRAM胞60被寫入數位信號時，則由支柱所 支撐的微鏡62係以對角線爲中心，被以相對於配置有DMD50 的基板側，以α .度（例如控1 0度）的範圍傾斜。第1 2 ( A ) 圖係表示微鏡62在開啓狀態之傾斜在+ α度的狀態，第1 2 (Β )圖係微鏡62在關閉狀態之傾斜在-α度的狀態。因此， 因應畫像信號，藉由把在DMD50之各畫素的微鏡62之傾斜 控制成如第6圖，則入射至DMD50的光係朝各自的微鏡62 之傾斜方向反射。 又，第1 1圖係放大DMD50之一部分，表示微鏡62係被控 制+ α度或一 α度之一狀態例。各自的微鏡6 2之開啓、關 閉控制係由連接在DMD5 0之未圖示的控制器所執行。此外， 在依關閉狀態的微鏡62、光束會被反射之方向上係配置有 -19- 1274733 光吸收體（未圖示）。 又，D M D 5 0係配置成其短邊與副掃描方向成指定角度0 (例 如，1 °〜5 ° )般地稍微傾斜者爲較佳。第〗3 ( a )圖係表 示不使DMD50傾斜時之依各微鏡的反射光像（曝光束）53 之掃描軌跡，第8 ( B)圖係使DMD50傾斜時之曝光束53的 掃描軌跡。 在DMD5 0中，於長度方向配置有多數個微鏡（例如，800 個）之微鏡列係在寬度方向配置有多數組（例如，6 0 0組）， 如第13 ( B)圖所示，藉由傾斜DMD50，使得依各微鏡的曝 光束5 3之掃描軌跡（掃描線）的間距P1係變得比不傾斜 DMD 5 0時之掃描線的間距P2還狹小，可使解像度大幅地提 升。一方面，因爲DMD50之傾斜角微小之故，所以使DMD50 傾斜時之掃描寬度W2和使DMD50不傾斜時之掃描寬度W1 係略相同。 又，依不同的微鏡列、相同掃描線上係成爲重疊被曝光（多 重曝光）。如此，藉由被多重曝光，而可控制曝光位置的微 少量，可實現高精細的曝光。又，藉由微少量的曝光位置控 制等之數位畫像處理，可無段差地把配列在主掃描方向之 複數個曝光頭間之連接處予以連繫。 此外，取代DMD50之傾斜，而改以使各微鏡列在與副掃描 方向正交的方向，以指定間隔偏移作棋盤狀配置，也可獲得 同樣的效果。 光纖陣列光源6 6，係如第1 4 ( A )圖所示，具備複數（例 如，6個）個雷射模組64，各雷射模組64係結合在多模光纖 -20- 1274733 30之一端。多模光纖30之他端係結合有核心直徑爲與多模 光纖3 0相同且包層直徑較多模光纖3 0小的光纖3 1 ,如第1 4 (C )圖所示，光纖3 1的出射端部（發光點）係沿著與副掃 描方向正交的主掃描方向配置1列而構成雷射出射部6 8。 此外，如第1 4 ( D )圖所示，也可把發光點沿著主掃描方向成 2列地配列。 光纖3 1之出射端部係如第1 4 ( B )圖所示，表面係被平坦 的2片支持板6 5挾住而固定著。又，光纖3 1之光出射側係 配置有玻璃等之透明的保護板6 3，以保護光纖3 1之端面。 保護板6 3也可與光纖3 1的端面密接配置，也可使光纖3 1 之端面被密封般地配置。光纖31之出射端部雖然光密度且 容易集塵而劣化，但是藉由配置保護板6 3，不但可防止塵埃 對端面之附著同時可延緩劣化。 在本例中，爲了將包層直徑小的光纖3 1之出射端無間隙 地配列成1列，在以包層直徑大的部分鄰接的2條多模光纖 3 0之間將多模光纖3 0聚集，而被聚集的多模光纖3 0所結合 之光纖3 1的出射端，係配列成被挾於以包層直徑爲大的部 分鄰接之2條多模光纖3 0所結合的光纖3 1之2個出射端 之間。 這樣的光纖，例如第1 5圖所示，係藉由在包層直徑爲大的 多模光纖30之雷射光出射側的前端部分，將長度1〜30cm 之包層直徑爲小的光纖31予以同軸地結合而可獲得。2條 的光纖係光纖31之入射端面在多模光纖3 0之出射端面以 兩光纖的中心軸呈一致般地熔接而被結合著。如同上述，光 -21- 1274733 纖31之核心3 1 a的直徑係與多模光纖3 0之核心3 0 a的直 徑相同大小。 又，也可以使長度爲短包層直徑爲大的光纖中熔接有包層 直控爲小的光纖之短尺寸光纖，經由一套圈或光連接器等而 結合至多模光纖3 0之出射端。藉由利用連接器等以可裝卸 地結合，以在包層直徑爲小的光纖破損時等場合，使前端部 分的交換變成容易，可減低曝光頭的維修所要之成本。此外， 以下有時把光纖3 1稱爲多模光纖30之出射端部。 以多模光纖30及光纖31而言，也可以是STEP INDEX型 光纖、GRATED INDEX型光纖、及複合型光纖之中任一。例 如，可使用由三菱電線工業株式會社所製造的STEP INDEX 型光纖。在本實施形態中，多模光纖30及光纖31係STEP INDEX型光纖，多模光纖30係包層直徑=125 // m、核心直徑 = 25//m、NA二0.2、入射端面塗層的透過率= 99· 5%以上， 光纖31係包層直徑= 60//m、核心直徑= 25//m、ΝΑ=0·2。 一般，以紅外線區域的雷射光而言，若光纖的包層直徑設 定小則傳送損失會增加。因此，係因應雷射光之波長帶域以 決定合適的包層直徑。然而，波長越短傳送損失係變少，以 由GaN系半導體雷射所出射的波長405nm之雷射光而言，即 使包層的厚度{(包層直徑-核心直徑）/ 2 }爲傳送8 0 0 n m 之波長帶域的紅外光時之1 / 2左右、或爲傳送通信用之1 . 5 // m之波長頻帶的紅外光時之約1 / 4，傳送損失也幾乎不會 增加。因此，可把包層直徑設小成爲6 0 # m。藉由使用G a N 系的LD而可容易獲得光密度高之光束。 - 22 - 1274733 但是，光纖31的包層直徑不限定爲60/zm。以往在光纖光 源所使用之光纖的包層直徑爲1 2 5 // m，但是包層直徑越小則 焦點深度係變越深，所以多模光纖的包層直徑係8 0 // m以下 較好，6 0 // m以下更好，4 0 // m以下更佳。一方面，核心直徑 有必要至少爲3〜4 // ιώ，所以光纖3 1的包層直徑係1 〇 // m以 上較佳。 雷射模組6 4係由第1 6圖所示之合波雷射光源（光纖光 源）所構成。此合波雷射光源係由如下所構成：即，配列固 定在熱塊1 0上之複數（例如7個）個晶片狀之橫多模或單 模之 GaN 系半導體雷射 LD1、LD2、LD3、LD4、LD5、LD6、 及LD7;對應GaN系半導體雷射LD1〜LD7各自而設置之准 直透鏡11、12、13、14、15、16、及17; 1個集光透鏡20; 1條多模光纖3 0。此外，半導體雷射之個數不受限爲7個。 例如，包層直徑=6 0 // m、核心直徑=5 0 // m、N A = 0 · 2的多 模光纖係可入射2 0多個半導體雷射光，實現曝光頭5之必 要光量，且可將光纖條數減爲更少。1274733 玖, invention description: [Technical Field to Which the Invention Is Ascribed] The present invention relates to a method of forming a minute flow path. In particular, it relates to a method of forming a micro flow path in a microchip used in the field of microchemistry. [Prior Art] In recent years, a so-called laboratory on a chip in which a system for performing mixing, reaction, separation, and detection of a solution is integrated on a glass substrate of a square centimeter square by using a micromechanical technique is used. The device technology is enthusiastically being researched. Laboratory wafers are systems that are integrated into the body, also known as #-TAS (Micro Total Analysis System), microreactors, and so on. In general, the laboratory wafer is provided with a micro flow path having a groove width of several tens to several hundreds / m on a substrate having a thickness of about 1 mm, and mixing of the solution is performed in the micro flow path. In the micro flow path, since the specific area is larger, it is difficult for the responder to react due to the size effect, and it is difficult to mix the mixture, and the mixing and reaction of the solution can be efficiently performed. By setting the groove width of the minute flow path to 1 0 // m to 5 0 // m, the flow path resistance can be made small, and a good size effect can be obtained. Further, since the shape of the minute flow path greatly affects the liquid feeding characteristics of the fluid, it is preferable that the minute flow path has a smooth wall surface and is formed with high precision. In the past, the micro flow path of the laboratory wafer was coated with a resist film to cover the surface of the substrate, and the resist film was patterned by lithography using ultraviolet rays or electron lines, and then used as a semiconductor for etching the substrate. Processing technology is formed. Photolithography is performed using a close exposure device used in semiconductor fabrication. The exposure method is based on the analog exposure of the mask calibration -6-1274733 type, for example, a large area of 1 square meter is difficult to high-speed exposure. [Problems to be Solved by the Invention] However, in the conventional method of forming a minute flow path, the mask is exposed to perform patterning. Therefore, the thickness of the photoresist film is limited, and it is difficult to form a minute flow path with high precision. In other words, when the photoresist film is thin, the side etching is facilitated when the substrate is etched, and the groove width is less likely to be formed, and a sufficient groove depth cannot be obtained. Further, in the mask exposure, it is necessary to provide a high-precision glass mask or the like for each pattern. Therefore, the cost is increased, and it is difficult to increase the area, and it is not suitable for a small amount of production. On the one hand, it is also considered that the lithography process is performed by digital exposure, but the conventional digital exposure device using ultraviolet rays is scanned with a single beam, and the exposure time is excessive. In particular, when the beam diameter is 丨〇 # m or less and the address is capable of a high-definition exposure of about 1 V m, there is a problem that the exposure time is too long. The present invention has been made to solve the above problems, and an object of the present invention is to provide a method of forming a minute flow path capable of forming a minute flow path at high speed and high precision. Further, another object of the present invention is to provide a method of forming a micro flow path in which a micro flow path of an arbitrary pattern can be formed at low cost. [Means for Solving the Problems] In order to achieve the above object, the method for forming a micro flow path according to the present invention is characterized in that it includes an exposure process and a wavelength which is modulated in space in accordance with formation of pattern data of a minute flow path. 3 5 0 nm ~ 450 nm of laser light, exposure is formed on the base 1247333 on the resist film; patterned engineering, corresponding to the exposure pattern partially remove the resist film to form a resistive film of the specified pattern; etching engineering, using the The resist film of the pattern is specified, and the substrate is etched by the surface and removed to form a minute flow path. In the method of forming the minute flow path, since the laser light having a wavelength of 550 nm to 4500 nm is used, it is not necessary to use an optical system of a special material corresponding to the ultraviolet light of the excimer laser, and the mine of the visible region Similarly to the exposure apparatus, a spatial light modulation element such as DMD can be used. Thereby, the resist film can be exposed by the spatially modulated laser light in response to the formation of the pattern data by the minute flow path. That is, the resist film of any pattern can be subjected to high-speed and high-definition digital exposure. As described above, in the exposure process, since the resist film of any pattern can be exposed at high speed and high definition, the micro flow path of an arbitrary pattern can be formed at high speed and high precision after the next patterning process and etching process. Further, since it is digitally exposed, the mask of each pattern is not required, and the minute flow path can be formed at low cost. In the exposure engineering described above, a laser light source having irradiated laser light and a plurality of pixel portions having a light modulation state change in response to respective control signals may be disposed on the substrate in a rectangular shape for use in the exposure. A spatial light modulation element that modulates the laser light irradiated by the laser device, and an exposure head that images the laser light modulated in each pixel portion on the exposure surface. Further, the exposure head is relatively moved in a direction intersecting the designated direction with respect to the exposure surface of the resist film, and the resist film formed on the substrate can be subjected to scanning exposure. In order to expose the resist film to a higher precision, the spatial light modulation element is arranged such that the arrangement direction of each pixel portion is slightly inclined with respect to the direction perpendicular to the sub-scanning direction at a specified angle of 1274735 degrees. It is better. Thereby, the beam diameter 10 0 m can be exposed with high precision in a positional capacity of 1 // m. A range in which the inclination angle 0 is 1 to 5 is preferable. . Further, on the exit side of the spatial modulation element, it is more preferable to provide a microlens array in which microlenses of the laser light are collected in each pixel in accordance with the respective pixel portions of the spatial modulation element. When the microlens array is disposed, the laser light modulated by each pixel portion of the spatial modulation element is collected by the respective microlenses of the microlens array, so that even in the exposed surface When the exposure area is enlarged, the size of each beam spot can also be reduced, and high-definition exposure can be performed. According to the use of this reduction optical system, it is capable of ultra-high-definition exposure with a beam diameter of 1 // m at an addressability of 0.1 μm. Thus, the barrier film is subjected to high-definition exposure to form a wall surface of a very smooth micro flow path, which can reduce the flow path resistance to obtain a good size effect. It is preferable to form a minute flow path for high precision, and the thickness of the resist film is thick. In the case of forming a minute flow path having a groove width of 1 Ο μ m to 5 0 // m, the thickness of the resist film is preferably l〇/zm 5050/m, and 10//m to 100/zm is more. it is good. In particular, it is more preferable to laminate a plurality of layers by forming a plurality of layers of a resist film as two or three layers. Since the resistive film is digitally exposed, the correction of the extension of the exposure and the % after development can be performed accurately by the digital scaling function, the exposure position of the first layer, and the exposure position of the plurality of layers of the second layer or the like. The positioning system can be realized with high precision. As a result, it is possible to form a resist film having twice the thickness of the conventional one, and a patterning system with high precision and a high aspect ratio, and it is possible to form a fine flow path with high precision and deepness by etching. Further, the aspect ratio is a ratio a / b of the groove width a to the groove depth b of the groove formed by the film. -9- 1274733 In the exposure process of the above-described formation method, by using a high-intensity light source to expose with a deep depth of focus, the resist film can be exposed with greater precision. In the case of a localized light source, a plurality of laser light beams are combined to make a combined laser light source incident on the respective fibers suitable. Moreover, the exposure system of the thick film barrier film requires a high output laser source. Oscillation wavelength 3 50 to 45 Oiim's semiconductor laser is difficult to output with a single component, but it can be plotted and combined with high output. The multiplexed laser light source may be, for example, (1) including a plurality of semiconductor lasers, and one optical fiber, and collecting laser light emitted from each of the plurality of semiconductor lasers to integrate the concentrated light beam a light collecting optical system of an incident end of the optical fiber; (2) comprising a multi-cavity laser having a plurality of light-emitting points, and one optical fiber, and concentrating the laser light emitted from the plurality of light-emitting points, a constituent optical optical system that combines the collected beam to the incident end of the optical fiber; or (3) includes a plurality of multi-cavity lasers, and a beam of optical fibers, and the plurality of multi-chamber lasers The laser light emitted from each of the light-emitting points is collected to combine the collected light beam with the light collecting optical system of the incident end of the optical fiber. The light-emitting points in the exit end of the optical fiber of the above-described combined laser light source may be arranged in an array to form an optical fiber array light source, and the light-emitting points are respectively bundled to form a fiber bundle light source. By beam or array, it is possible to achieve higher output. Further, from the viewpoint of increasing the luminance, it is preferable to use an optical fiber having a uniform core diameter and a smaller cladding diameter than the incident end. From the viewpoint of making the diameter of the light-emitting point small, the cladding diameter of the exit end of the optical fiber is preferably smaller than 1 2 5 // m, more preferably 80 0 / m or less, and 6 〇 # is still the following special 1274733. The optical fiber system in which the core diameter is uniform and the cladding diameter of the exit end is smaller than the diameter of the cladding layer at the incident end, for example, a plurality of optical fibers having the same core diameter and different cladding diameters may be combined. Accordingly, the light-emitting area at the time of array formation can be made smaller and brighter. Further, by arranging a plurality of optical fibers to be detachably attached to the connector, when the light source module is partially broken, the exchange system becomes easy. In particular, in the case where the spatial light modulation element as described above is obliquely arranged and a reduction optical system or an equal magnification optical system is used to perform ultra-high-definition exposure, by using the aforementioned high-intensity optical fiber array light source or fiber bundle light source, Because the illumination NA of the spatial modulation component is set to be small, the depth of focus of the imaging beam after passing through the spatial modulation component can be deep, and a deep depth of focus can be obtained, and the beam is not too wide on the surface of the resistor body and the resistor body. It becomes a pattern with higher precision and a high aspect ratio. Further, in the case where an inclined flow path having a slanted wall surface is formed, a smooth pattern can be obtained. In the above-described exposure process, for example, a plurality of pixel spatial light modulation elements in which a light modulation state changes depending on respective control signals are arranged on a substrate, and each of the spatial light modulation elements is arranged. The picture department was changed. In the case of a spatially modulated component, a micromirror device (DMD) composed of a plurality of micromirrors that can change the angle of the reflecting surface in response to each false control signal can be arranged in a two-dimensional manner on a substrate (for example, a germanium substrate). Digital micromirror device). Further, the spatial modulation element may be configured as a one-dimensional grating in which a movable lattice having a strip-shaped reflecting surface and movable in response to a control signal and a fixed lattice having a strip-shaped reflecting surface are alternately arranged in a plurality of rows. Light bulb (GLV). Further, it is also possible to use a liquid crystal shutter array in which a plurality of liquid crystal cells which can block the transmitted light in response to the respective control signals - 1 1 - 1274733 are arranged in two dimensions on the substrate. Preferably, on the emission side of the spatial modulation element, a microlens array provided with a microlens corresponding to the spatial modulation element and provided with a microlens for collecting the laser light in each pixel is provided. When a microlens array is arranged, the laser light modulated by each pixel portion of the spatial modulation variable element is collected by the respective microlenses of the microlens array, and is collected by corresponding pixels. When the exposure area in the surface is enlarged, the size of each beam spot can be reduced, and high-definition exposure can be performed even in the case of a large area. [Embodiment] Hereinafter, an embodiment in which a microchannel for a synthetic reaction is produced by using the method for forming a microchannel of the present invention will be described in detail with reference to the drawings. [Other Spatial Modulation Element] In the above-described embodiment, an exposure head including a DMD as a spatial modulation element has been described. For example, a MEMS (Micro Electro Mechanical System) type spatial modulation element such as GLV can be used. (SLM) A spatial modulation element other than the MEMS type such as an optical element (PLZT element) that transmits light and a liquid crystal shutter (FLC) is modulated by an electro-optical effect. In addition, the so-called MEMS is a general term for a micro-system of micro-mechanical sensors and actuators based on IC process-based micro-mechanical technology, and then a control system is integrated. The so-called MEMS type space modulation A variable component means a spatially modulated component driven by an electromechanical action of electrostatic force. [Microchip for Synthesis Reaction] As shown in Fig. 1, the microchip for synthesis reaction is formed by superposing a protective substrate 202 on a flat substrate 150 formed of glass or the like. The thickness of the substrate 150 is 1274733, and the thickness is usually about 0.5 mm to 2.0 mm. The thickness of the protective substrate 202 is usually 0 · 1 off ~ 2 · Omni. On each of the protective substrates 20 2 , injection ports 204 a and 204 b for injecting the reagents and a discharge port 2 0 6 for discharging the reaction liquid obtained after the reagent reaction are disposed. The substrate 150 is provided with a minute flow path 208 for circulating a reagent or a reaction solution. The microchannels 2 0 8 are arranged such that the reagents injected from the injection ports 2 0 4 a and 2 0 4 b are merged at the junction point 2 1 0 and then discharged toward the discharge port 2 06. The groove width coefficient of the micro flow path is - hundreds of meters, and 10/zm~50/zm is particularly good. The flow path resistance of the micro flow path having a groove width of lOem to 50//m is relatively small, so that a good size effect can be obtained. In this reaction, each of the injection ports 204a and 204b of the microchip is injected with a reagent, and when sucked by the discharge port 206, the reagent flows through the microchannels 2 to 8 and is mixed at the junction 2 1 0 to react. . According to this, the desired substance can be synthesized. The obtained reaction liquid flows through the minute flow path 208 and is discharged from the discharge port 206. The analysis of the reaction liquid obtained from the discharge port 206 can be carried out to identify or quantify the reaction product in the same manner as in the usual ratio. [Method for Producing Microchip] Next, a method for producing the above-described microchip for synthesis reaction will be described with reference to Fig. 2 . The manufacturing method is an exposure process of exposing a photoresist film, a patterning process of partially removing a photoresist film, a etching process of dicing a substrate to form a minute flow path, and a substrate on which a micro flow path is to be formed. It is composed of a bonding process or the like for bonding a protective substrate. The following is a description of each project. As shown in the second (A) diagram, after the photoresist film 2 1 2 is formed on the substrate 150 by a spin coating method or the like, as shown in the second (B) diagram, the micro flow channel 2 0 8 1274733 The pattern exposes the photoresist film 2 12 , and as shown in the second (C) diagram, the exposed portion 2 14 is dissolved in the developer and removed. Here, by patterning the photoresist film 2 1 2 with high positional accuracy, the minute flow path 208 can be formed with high precision. Further, the exposure engineering relating to the photoresist film 2 1 2 will be described later. Next, as shown in the second (D) diagram, the patterned photoresist film 2 1 2 is used to etch the substrate 150 from the surface to form the minute flow path 208, as shown in the second (E) diagram, and the remaining light is removed. Barrier film 2 1 2 . The uranium engraving of the substrate 150 can perform either dry uranium etching or wet etching, but since it is microfabrication, dry etching such as high-speed atomic (F A B ) uranium etching is suitable. Next, as shown in Fig. 2(F), through holes for the injection ports 204a and 204b and the discharge port 206 are formed in the protective substrate 202 by ultrasonic machining or the like. Then, as shown in Fig. 2(G), the surfaces of the protective substrate 202 and the substrate 150 on which the minute flow paths 208 are formed are fixed and fixed by overlapping the substrates in the opposite direction. For example, a UV adhesive can be used for fixing. The surface of the protective substrate 2 〇 2 on which the minute flow path 208 is formed is applied by a spin coating method or the like, and after the substrate 150 and the protective substrate 208 are adhered to each other, the ultraviolet ray is further irradiated. Further, when the substrate 150 and the protective substrate 202 are formed of glass, the surfaces of the two substrates may be dissolved and rejoined by hydrofluoric acid. [Exposure of Photoresist Film] Hereinafter, the exposure process of the photoresist film will be described in detail. In this exposure project, the spatial light modulation element is used to convert the laser light with a wavelength of 550 nm to 4500 nm; the modulated light is modulated according to the pattern of the micro flow path, and then the modulated laser light will be used. The photoresist film 212 is digitally exposed. In order to perform exposure with higher precision, it is better to expose the laser light with a deep depth of focus of 1274733 emitted by a high-intensity light source. [Photoresist Film] The photoresist film 212 can be a dry resist film (DFR; Dry Film Res i s t) or an electrode deposition resist used in a manufacturing process of a printed wiring board (PWB; Printed Wirmg Board). These DFR or electrodeposition resistors can be thickened in comparison with the resistors used in the semiconductor process to form a film having a thickness of 10//40/m. Further, by stacking a plurality of layers of the photoresist film, it is possible to draw a thicker film. At this time, as shown in the third (A) diagram, the first photoresist film 21 2a is formed, and after the designated region 214a is exposed, as shown in the third (B) diagram, on the first photoresist film 2 1 2 a The second photoresist film 2 1 2b is formed, and the region 214b corresponding to the designated region 214a is exposed by using a predetermined function of digital exposure. As shown in Fig. 3(C), when the exposed area 2 1 4 a and the area 2 1 4b are removed, a deep groove depending on the resist is formed. In addition, in this example, an example in which the resistive film is laminated in two layers is described, but the barrier film is used as a stack of three layers and four layers, and the same position is exposed by a digital exposure. Form a deeper ditch. In addition, here, the exposure system is overlapped by two or more layers under the non-transmission development process, but the first layer is exposed, and then developed. The extension of the substrate after development or the swelling of the resist body is determined by the number position. For comparison, the second layer may be exposed to light after exposure, and the third layer and the fourth layer may be exposed as well. Accordingly, the positional deviation of the pattern at the time of development can be corrected with high precision. Further, by thickening the photoresist film 2 1 2, a deep groove depending on the resist can be formed, and a deep groove (small flow path) having excellent precision can be formed on the substrate 202 by etching. 1274733 For example, as shown in the fourth (A) and 4 (B) diagrams, when the microchannels having the same groove width are formed by FAB etching, when the photoresist film 2 12 is thin, the light substrate 15 is obliquely oriented. 0 can be easily etched on the side, and when the photoresist film 2 1 2 is thick, the oblique light is hard to be incident due to the hindrance, and the substrate 150 becomes difficult to be etched by the side. Thereby, a deep groove having excellent precision can be formed on the substrate 150. In the case of forming a minute flow path having a groove width of 1 〜#m~5 〇//m, the thickness of the photoresist film 2 1 2 is preferably 10//m to 50/zm, and 10//m to 100//m. better. X, in the case where a minute flow path is formed by wet etching using an etching solution, as shown in Fig. 5, the opening 2166 of the lift-off pattern may be patterned in the resist film 2 1 2 . Since the opening 2 16 is pushed out, it is easy to immerse the uranium engraving solution. [Exposure Device] For the exposure device used in the exposure process of the photoresist film, for example, a device as shown in Fig. 6 can be used. The device is provided with a substrate 150 on which a resist film is formed, and is held on the surface to be held. The flat-shaped stage 1 5 2 . Further, on the upper surface of the thick plate-shaped mounting table 156 supported by the four leg portions 154, two guide portions 158 extending in the moving direction of the stage are provided. The stage 1 5 2 is disposed such that its longitudinal direction is moved toward the stage of the stage, and is supported by the guide portion 158 so as to be reciprocally movable. Further, the exposure apparatus is provided with a driving device (not shown) for driving the stage 152 along the guiding portion 158. The central portion of the installation table 156 is provided with a font gate 160 that extends across the movement path of the stage 152. The ends of the gates 160 are fixed to the side faces of the setting table 156. Hold the gate 160 and tie it on one side - 1 6 - 1274733. There is a scanner 162. The side is provided with a plurality of (for example, 2) detection sensors for detecting the front end and the rear end of the photosensitive material 15〇.丨 6 4. The scanner 1 6 2 and the detecting sensor 1 64 are each mounted on the shutter 丨 60 and fixedly disposed above the moving path of the stage 152. Further, the scanner 丨 62 and the detection sensor 164 are connected to a controller (not shown) for controlling these. f coin piracy 162, such as the brother 7 map and the younger 8 (B) map, is a plurality of (for example, ! 4) exposures with a slightly matrix-like arrangement of ill rows η columns (for example, 3 rows and 5 columns) Head 166. In this example, four exposure heads 166 are arranged in the third row because of the relationship with the width of the photosensitive material 150. Further, in the case of indicating the respective exposure heads arranged in the nth column of the mth row, the exposure region 168 which is the exposure head 1 6 6 mn is formed in a rectangular shape in which the sub-scanning direction is the short side. Therefore, as the stage 1 52 moves, the resist film formed on the substrate 150 is formed with the strip-shaped exposed regions 170 of the respective exposure heads 166. Further, in the case of indicating the exposure areas of the respective exposure heads arranged in the nth column of the mth row, it is shown as the exposure region 168mn. Further, as shown in Figs. 8(A) and 8(B), the strip-shaped exposed regions 170 are arranged in a direction orthogonal to the sub-scanning direction without a gap, and the exposure heads of the respective rows in the line arrangement are each The arrangement direction is arranged at a predetermined interval (a natural multiple of the long side of the exposure region, twice in the present embodiment). Therefore, the unexposed portion between the exposure region 1 6 8: i and the exposure region 1 6 8 i 2 in the 1st row can be in the exposure region 1 6 8 21 of the 2nd row and the exposure region 1 in the 3rd row. 6 8 3 i and exposure. The exposure heads 166M to 166mn are each shown in Figures 9, 10(A) and 10(B), and 1274473, and are provided with a digital micromirror device (DMD) 50 as a corresponding image data for the incident light beam to correspond to the image data. It is used as a spatial light modulation component for modulation. The DMD 105 is connected to a controller having a data processing unit and a mirror drive control unit (not shown). The data processing unit of the controller generates control signals for driving control of the respective micromirrors in the control region of the DMD 50 of each exposure head 166 based on the input image data. Further, the mirror drive control unit controls the angle of the reflection surface of each of the micromirrors of the DMDs 50 of the respective exposure heads 1 66 in accordance with a control signal generated by the image data processing unit. Further, the angle control of the reflecting surface will be described later. The light incident side of the DMD 50 is arranged in the following order: an optical fiber array in which the exit end portion (light-emitting point) of the optical fiber is arranged in a row along the longitudinal direction of the exposure region 168 The light source 66; the lens system 67 that corrects the laser light emitted from the fiber array light source 66 and condenses the light on the DMD; and the mirror 69 that reflects the laser light from the transmission lens system 67 toward the DMD 50. The lens system 67 is a pair of combined lenses in which a pair of combined lenses 71 that collimate the laser light emitted from the optical fiber array light source 66 and a light quantity distribution of the laser light that is parallelized to be uniform are corrected. 73. A light collecting lens 75 that collects the laser light whose light quantity distribution is corrected is collected on the DMD. The combined lens 713 is provided with a direction in which the laser emitting end is arranged, and a portion close to the optical axis of the lens is a portion that expands the light beam and is separated from the optical axis, and the light beam is reduced in a direction orthogonal to the arrangement direction. The function of passing through as it is, the light quantity distribution is generally corrected for the laser light. Further, on the light reflection side of the DMD 50, lens lines 54, 58 which reflect the laser light reflected by the DMD 50 on the scanning surface (exposed surface) 56 of the photosensitive material 150 are disposed. The lens systems 54 and 58 are arranged such that the DMD 50 and the exposed surface 56 are in a conjugate relationship. As shown in Fig. 11, the DMD 50 is a small mirror (microscope) 62 supported by a pillar on a SRAM cell (memory cell) 60, and is configured to constitute a plurality of pixels (PIXEL) (for example, 600 X 800) micro mirrors are arranged in a lattice arrangement. The uppermost part of each pixel is provided with a micromirror 62 supported by a pillar, and the surface of the micromirror 62 is vapor-deposited with a material having a high reflectance such as aluminum. Further, the reflectance of the micromirror 62 is 90% or more. In addition, the CMOS SRAM cell 60 of the gate of the normal semiconductor memory production line is disposed in the front of the micromirror 62 through the pillar including the hinge and the yoke, and the whole system is configured as a monolith (integrated type). . When the SRAM cell 60 of the DMD 50 is written with a digital signal, the micromirror 62 supported by the post is centered on the diagonal and is at a degree relative to the side of the substrate on which the DMD 50 is disposed (for example, control 10) The range of degrees) is inclined. The 1 2 (A) diagram shows a state in which the tilt of the micromirror 62 in the open state is +α degrees, and the 1 2 (Β) graph is in a state where the tilt of the micromirror 62 in the closed state is -α degrees. Therefore, in response to the image signal, by controlling the tilt of the micromirrors 62 of the respective pixels of the DMD 50 as shown in Fig. 6, the light beams incident on the DMD 50 are reflected toward the oblique directions of the respective micromirrors 62. Further, Fig. 1 is an example of amplifying a portion of the DMD 50, and indicates that the micromirror 62 is controlled to have one state of + degree or one degree of degree. The opening and closing control of the respective micromirrors 6 2 is performed by a controller (not shown) connected to the DMD 105. Further, a light absorber (not shown) of -19-1274733 is disposed in the direction in which the micromirror 62 in the closed state and the light beam are reflected. Further, it is preferable that the D M D 5 0 is arranged such that the short side thereof is slightly inclined at a predetermined angle 0 (for example, 1 ° to 5 °) in the sub-scanning direction. The third figure (a) shows the scanning trajectory of the reflected light image (exposure beam) 53 of each micromirror when the DMD 50 is not tilted, and the scanning trajectory of the exposure beam 53 when the DMD 50 is tilted by the eighth (B) image. . In the DMD50, a micromirror array in which a plurality of micromirrors (for example, 800) are arranged in the longitudinal direction is arranged in a plurality of arrays in the width direction (for example, a group of 600), as shown in Fig. 13(B). By tilting the DMD 50, the pitch P1 of the scanning track (scanning line) of the exposure beam 5 3 of each micromirror becomes narrower than the pitch P2 of the scanning line when the DMD 50 is not tilted, so that the resolution can be greatly improved. Upgrade. On the one hand, since the tilt angle of the DMD 50 is small, the scan width W2 when the DMD 50 is tilted is slightly the same as the scan width W1 when the DMD 50 is not tilted. Further, depending on the different micromirror columns and the same scanning line, the overlap is exposed (multiple exposure). Thus, by being subjected to multiple exposures, a small amount of exposure position can be controlled, and high-definition exposure can be realized. Further, by the digital image processing such as a small amount of exposure position control, the connection between the plurality of exposure heads arranged in the main scanning direction can be connected without any difference. Further, in place of the tilt of the DMD 50, the same effect can be obtained by arranging the respective micromirrors in a direction orthogonal to the sub-scanning direction and shifting at a predetermined interval in a checkerboard shape. The fiber array light source 66 is as shown in Figure 14 (A), and has a plurality of (for example, six) laser modules 64, each of which is combined with a multimode fiber -20-1274733. One end. The other end of the multimode optical fiber 30 is combined with an optical fiber 3 1 having a core diameter which is the same as that of the multimode optical fiber 30 and having a larger cladding diameter and a larger mode fiber 30, as shown in the first 4 (C) diagram, the optical fiber 3 1 The emission end portions (light-emitting points) are arranged in one line along the main scanning direction orthogonal to the sub-scanning direction to constitute the laser emission portion 68. Further, as shown in Fig. 14(D), the light-emitting points may be arranged in two rows along the main scanning direction. The exit end of the optical fiber 3 1 is fixed as shown in Fig. 1 (B), and the surface is held by the flat two support plates 65. Further, the light-emitting side of the optical fiber 31 is provided with a transparent protective plate 63 such as glass to protect the end faces of the optical fibers 31. The protective plate 63 may be disposed in close contact with the end surface of the optical fiber 31, and the end surface of the optical fiber 3 1 may be disposed in a sealed manner. The exit end of the optical fiber 31 is deteriorated due to the optical density and easy to collect dust. However, by arranging the protective plate 63, it is possible to prevent the adhesion of dust to the end surface and to delay deterioration. In this example, in order to arrange the exit ends of the optical fibers 31 having a small cladding diameter in a row without gaps, the multimode optical fibers 3 are interposed between two multimode optical fibers 30 adjacent to each other with a large cladding diameter. 0 is gathered, and the exit end of the optical fiber 3 1 combined with the aggregated multimode optical fiber 30 is arranged to be bundled with the optical fiber 3 combined with the two multimode optical fibers 30 adjacent to each other with a large cladding diameter. Between 1 and 2 exits. Such an optical fiber, for example, as shown in Fig. 15, is an optical fiber 31 having a small cladding diameter of 1 to 30 cm in length, by a front end portion on the exit side of the laser light of the multimode optical fiber 30 having a large cladding diameter. Available in combination with coaxial. The incident end faces of the two fiber-optic fibers 31 are joined at the exit end faces of the multimode fibers 30 in such a manner that the central axes of the two fibers are uniformly welded. As described above, the diameter of the core 3 1 a of the light -21-1274733 fiber 31 is the same as the diameter of the core 30 a of the multimode fiber 30. Further, a short-length optical fiber in which a short-clad diameter is large and a small-capacity optical fiber having a small cladding diameter is welded to the short-length optical fiber, which is directly controlled to be small, is coupled to the outgoing end of the multimode optical fiber 30 via a ring or an optical connector. . By detachably joining by a connector or the like, it is easy to exchange the tip end portion when the optical fiber having a small cladding diameter is broken, and the cost of maintenance of the exposure head can be reduced. Further, the fiber 31 is sometimes referred to as an exit end of the multimode fiber 30 hereinafter. The multimode fiber 30 and the optical fiber 31 may be any of a STEP INDEX type fiber, a GRATED INDEX type fiber, and a composite type fiber. For example, a STEP INDEX type optical fiber manufactured by Mitsubishi Electric Industries, Ltd. can be used. In the present embodiment, the multimode fiber 30 and the optical fiber 31 are STEP INDEX type fibers, and the multimode fiber 30 is a cladding diameter = 125 // m, a core diameter = 25//m, a NA of 0.2, and an incident end face coating. Transmittance = 99·5% or more, fiber 31-series cladding diameter = 60//m, core diameter = 25//m, ΝΑ = 0.22. Generally, in the case of laser light in the infrared region, if the cladding diameter of the optical fiber is set small, the transmission loss increases. Therefore, the appropriate cladding diameter is determined by the wavelength band of the laser light. However, the shorter the wavelength, the less the transmission loss is. In the case of laser light having a wavelength of 405 nm emitted by a GaN-based semiconductor laser, even if the thickness of the cladding {(cladding diameter - core diameter) / 2 } is transmitted 80 Approximately 1 / 2 of the infrared light in the wavelength range of 0 nm or about 1 / 4 of the infrared light in the wavelength band of 1.5 / m, the transmission loss hardly increases. Therefore, the cladding diameter can be made small to 60 # m. A light beam having a high optical density can be easily obtained by using an LD of the G a N system. - 22 - 1274733 However, the cladding diameter of the optical fiber 31 is not limited to 60/zm. In the past, the fiber diameter of the fiber used in the fiber source is 1 2 5 // m, but the smaller the cladding diameter is, the deeper the depth of focus is. Therefore, the cladding diameter of the multimode fiber is less than 80 // m. Good, 6 0 // m is better, and 4 0 // m is better. On the one hand, the core diameter needs to be at least 3 to 4 // ι , so the cladding diameter of the fiber 3 1 is preferably 1 〇 // m or more. The laser module 64 is composed of a combined laser light source (fiber source) as shown in Fig. 16. The multiplexed laser light source is composed of a plurality of (for example, seven) wafer-shaped transverse multimode or single mode GaN-based semiconductor lasers LD1, LD2, LD3 fixed on the thermal block 10. LD4, LD5, LD6, and LD7; collimating lenses 11, 12, 13, 14, 15, 16, and 17 respectively corresponding to the GaN-based semiconductor lasers LD1 to LD7; one collecting lens 20; Multimode fiber 30. In addition, the number of semiconductor lasers is not limited to seven. For example, a multimode fiber with a cladding diameter = 6 0 // m, a core diameter = 5 0 // m, and NA = 0 · 2 can inject more than 20 semiconductor lasers to achieve the necessary amount of light for the exposure head 5, and The number of fiber strips can be reduced to less.
GaN系半導體雷射LD1〜LD7係振盪波長全部共通（例 如，405nm)，最大輸出也全部共通（例如，多模雷射爲i00mw、 單模雷射爲30mW )。此外，以GaN系半導體雷射LD1〜LD7 而言，在350nm〜450nm的波長範圍，也可使用具備有上述之 40 5 nm以外的振盪波長之雷射。 上述之合波雷射光源係如第1 7及1 8圖所示，連同其他光 學要素一起被收納在上方有開口之箱狀的封裝4 0內。封裝 4 0係具備有關閉其開口般所作成之封裝蓋4 1，在脫氣處理 -23- 1274733 後導入封止氣體，藉由把封裝4 0之開口以封裝蓋4 1閉合， 而在由封裝40和封裝蓋41所形成之閉空間（封止空間） 內，氣密封止上述合波雷射光源。 在封裝40的底面係固定有基板42 ,此基板42的上面係安 裝有：該熱塊1 0 ;保持集光透鏡20的集光透鏡保持器；以 及用以保持多模光纖30的入射端部之光纖保持器46。多模 光纖3 0的出射端部係由形成於封裝40之壁面的開口被引 出至封裝外。 又，在熱塊10的側面係安裝有准直透鏡保持器44,准直透 鏡1 1〜1 7係被保持著。在封裝40之橫壁面形成有開口，通 過此開口，用以對GaN系半導體雷射LD1〜LD7供給驅動電 流的配線47係被引出至封裝外。 此外，在第1 8圖中，爲避免圖面之煩雜化，僅由複數個GaN 系半導體雷射之中、對GaN系半導體雷射LD7附加編號，複 數個准直透鏡之中僅對賦予准直透鏡1 7附加編號。 第19圖係表示上述准直透鏡11〜17之安裝部分的正面 形狀。准直透鏡1 1〜1 7係各自形成爲以平行的平面，細長 地切取包含有具備非球面的圓形透鏡之光軸的區域。此細 長形狀的准直透鏡，例如係可藉由將樹.脂或光學玻璃予以模 製成形而形成。准直透鏡11〜17係，長度方向爲與GaN系 半導體雷射LD1〜LD7之發光點的配列方向（第19圖之左 右方向）成正交般地被密接配置在上述發光點之配列方向° 一方面，以GaN系半導體雷射LD1〜LD7而言，係使用具備 發光寬度爲2 // m的活性層，與活性層平行的方向、直角的 - 24 - 1274733 方向之視角各自爲例如1 Ο ° 、3 0 °的狀態之發射各個雷射 光束 B1〜B7之雷射。此等GaN系半導體雷射LD1〜LD7係 _ 在與活性層平行的方向上發光點成1列排列地配設著。 因此，由各發光點所發出之雷射光束B 1〜B7係如上述般、 對細長形狀之各准直透鏡1 1〜1 7，係成爲以視角角度爲大的 方向與長度方向一致，視角角度爲小的方向係與寬度方向 (與長度方向正交之方向）一致的狀態入射。亦即，各准直 透鏡11〜17之寬度爲1.1mm、長度爲4.6 mm,入射至此等之 雷射光束B1〜B7的水平方向、垂直方向的光束直徑係各自 鲁 爲0 . 9mm、2 . 6mm。又，准直透鏡1 1〜1 7係各自爲焦點距離 fl = 3mm、NA 二 0.6、透鏡配置間距=1.25mm。 集光透鏡20，係以平行的平面，細長地切取包含有具備非 球面之圓形透鏡的光軸之區域，准直透鏡1 1〜1 7的配列方 向，亦即形成爲在水平方向爲長、且在與其垂直的方向爲 短的形狀。此集光透鏡2 0係焦點距離f 2 = 2 3 mm、NA = 0 . 2。 此集光透鏡20也係藉由例如將樹脂或光學玻璃予以模製成 形而形成。 ♦ 以下，針對上述曝光裝置的動作加以說明。 在掃描器162之各曝光頭166,由構成光纖陣列光源66 之合波雷射光源的GaN系半導體雷射LD1〜LD7各自以發散 光狀態所出射之雷射光束Bl、B2、B3、B4、B5、B6、及B7 各自係由對應的准直透鏡1 1〜1 7而被平行光化。被平行光 化之雷射光束B1〜B7係由集光透鏡20所集光而收束至多 模光纖3 0之核心3 0 a的入射端面。 -25 - 1274733 本例中，由准直透鏡11〜17及集光透鏡20構成了集光光 學系統，由其集光光學系統和多模光纖3 〇而構成合波光學 系統。亦即，利用集光透鏡20、如同上述之被集光之雷射光 束B1〜B7係入射至此多模光纖30之核心30a以在光纖內 傳送，而被合波成1條雷射光束B再由結合至多模光纖3 0 之出射端部的光纖3 1出射。 於各雷射模組中，雷射光束B 1〜B7對多模光纖3 0之結合 效率係0.85、且GaN系半導體雷射LD1〜LD7之各輸出爲30mW 時，被陣列狀配列的各光纖3 1係可獲得輸出約18〇11^(= 3 OMw X 0 · 8 5 X 7 )之合波雷射光束B。因此，以陣列配列有6 條光纖31的雷射出射部68之輸出約爲1W ( = 180mWX 6 )。 光纖陣列光源6 6之雷射出射部6 8上係沿著主掃描方向 呈一列地配列有此種高亮度之發光點。由於把來自單一半 導體雷射之雷射光結合至1條光纖之以往的光纖光源係低 輸出，所以若未配列多數列則不能獲得所期望的輸出，但在 本實施形態所使用之合波雷射光源係高輸出，所以少數列， 例如即使1列也可獲得所期望的輸出。 例如，在將半導體雷射和光纖以1對1結合之以往的光纖 光源中，通常，以半導體雷射而言，係使用輸出爲30mW (毫 瓦）程度之雷射，以光纖而言，因爲係使用核心直徑5 〇 m、 包層直徑1 2 5 μ m、ΝΑ (開口數）〇 . 2之多模光纖，所以若欲 獲得約1 W (瓦）的輸出，則多模光纖必需把4 8條（8 X 6 ) 成一束，發光區域之面積爲〇.62mm2 ( 0.675mmX0.925mm)， 所以在雷射出射部68之亮度爲1.6 X106 (W/m2)，每1條 1274733 光纖之亮度爲3.2xi06(W/m2)。 相對地，在本實施形態中，如同上述，以多模光纖6條約可 獲得1 1 V的輸出，在雷射出射部6 8之發光區域的面積爲 0.0081_2(0.325mmX0.025mm)，所以雷射出射部 68 之亮 度成爲123 XI 06(W/m2)，相較於以往約可圖謀80倍的高 亮度化。又，每1條光纖之亮度爲90 X 1 06 ( W/m2 )，相較於 以往約可圖謀28倍的高亮度化。 在此，參照第20 ( A )及20 ( B )圖，針對以往的曝光頭和 本實施形態的曝光頭之焦點深度的差異加以說明。以往的 曝光頭之束狀光纖光源的發光區域之副掃描方向的直徑爲 0 · 6 7 5 mm，本實施形態之曝光頭的光纖陣列光源之發光區域 的副掃描方向的直徑爲0.025mm。如第20(A)圖所示，在 以往的曝光頭中，光源（束狀光纖光源）1之發光區域大，所 以對DMD3入射的光束之角度變大，其結果，對掃描面5入 射的光束之角度變大。爲此，相對於集光方向（焦點方向之 偏差），光束直徑係易過寬。 —方面，如第2 0 ( B )圖所示，在本實施形態的曝光頭中， 光纖陣列光源6 6之發光區域的副掃描方向之直徑小，所以 通過透鏡系67對DMD50入射的光束之角度變小，其結果， 對掃描面5 6入射的光束之角度變小。亦即，焦點深度變深。 在本例中，發光區域之副掃描方向的徑係約爲以往的3 0倍， 可獲得與略繞折界限相當的焦點深度。因此適於微小光點 之曝光。對此焦點深度之效果係在曝光頭的必要光量越大 越顯著且有效。在此例中，被投影在曝光面之1畫素尺寸係 -27- 1274733 1 〇 M m X 1 〇 # m。此外，DMD係反射型的空間調變元件，如第20 (A)及2 0 ( B )圖係用以說明光學方面之關係的展開圖。 因應曝光圖案的畫像資料係被輸入到連接至DMD50之未 圖示的控制器，而暫時記憶在控制器內之圖框記憶體。此畫 像資料係把構成畫像之各畫素的濃度以2進制（點記錄之 有無）所表示的資料。 吸附有形成光阻膜的基板1 5 0之載物台1 5 2，係依未圖示 的驅動裝置，沿著導引部1 5 8由閘門1 60之上游側往下游側 被以一定速度移動。載物台152係在通過閘門160下方之 際，當安裝在閘門1 60的檢測感測器1 64檢測到基板1 50的 前端時，則被記憶在圖框記憶體的畫像資料係依序被讀出複 數線，依據在資料處理部讀出的畫像資料以生成對各曝光頭 1 66之控制信號。然後，利用鏡驅動控制部，依據所生成之控 制信號，各曝光頭1 6 6之DMD5 0的微鏡各自被控制開啓、 關閉。亦即，於DMD50，在主掃描方向配列有800個微鏡的微 鏡列係在副掃描方向配列有6 0 0組，其全部要使用。 當由光纖陣列光源66對DMD50照射雷射光時，在DMD50 的微鏡爲開啓狀態時所反射的雷射光係利用透鏡系54、58 而成像於形成在基板1 5 0上之光阻膜的被曝光面5 6上。如 此一來，由光纖陣列光源6 6所出射的雷射光係在各畫素被 開啓、關閉，光阻膜係在與DMD 50之使用畫素數略同數量的 畫素單位（曝光區域168)被曝光。又，藉由基板15〇連 同載物台152 —起被以一定速度移動，使得形成在基板15〇 上之光阻膜經由掃描器1 62而在與載物台移動方向相反之 -28- 1274733 方向上被執行副掃描，形成各曝光頭丨6 6帶狀的已曝光區 域 1 7 0。 依掃描器1 62的光阻膜之副掃描終了而以檢測感測器1 64 檢測基板1 5 0的後端時，載物台丨5 2係依未圖示的驅動裝置， 沿著導引部158返回位在閘門16〇之最上游側的原點，再次 沿著導引邰1 5 8自閘門1 6 0的上游側至下游側以一定速度 移動。 如同以上說明，在本實施形態中，因爲在光阻膜的曝光工 程中使用DMD等之空間光調變元件，所以因應微小流路的形 成圖案而可將雷射光在各畫素作調變，可以被調變的雷射光 將光阻膜予以局速且高精細地曝光。如此，在曝光工程中， 因爲可把在任思圖案之光阻fl吴予以局速且局精細地曝光，經 由次一圖案化工程及蝕刻工程，可高速且高精度地形成任意 圖案之微小流路。 如同上述，因爲在任意圖案之曝光爲可能，所以可容易地 形成複雜圖案之微小流路。又，因爲可高速曝光，所以可於 大面積的玻璃基板上短時間形成微小流路。再者，因爲係數 位曝光、所以係不要各圖案之遮罩，可低成本地形成微小流 路。 又，由於在光阻膜使用DFR或電極沈積阻體，所以與在半 導體製程所使用的阻體相較下係可厚膜化，可形成厚度1 〇 # m〜4 0 // m的光阻膜。如此，藉由將先阻膜厚吴化，可依倉虫 刻形成精度佳之深溝的微小流路。 又，可將光阻膜積層複數層以圖謀厚膜化。此場合係使用 -29- 1274733 數位曝光之定比機能，可將複數積層之光阻膜的相同位置曝 光。 又，本實施形態中，於曝光裝置，使用合波雷射光源構成 光纖陣列光源，同時將光纖之出射端包層直徑設爲比入射端 包層直徑還小，所以發光部直徑變更小，被圖謀光纖陣列光 源之高亮度化。藉此，可以深的焦點深度之雷射光將光阻膜 更高精細地曝光。例如，可爲光束直徑1 # m以下、解像度〇 . i # Π]以下之超高解像度的曝光，在精度佳地形成溝寬i 〇 # m 〜5 0 // πι的微小流路上係很充分。 以下針對本實施形態之變形例作說明。 〔高速驅動方法〕 通常，於DMD，在主掃描方向配列有8 0 0微鏡的微鏡列係 在副掃描方向配列有6 0 0組，但是依控制器僅控制一部分的 微鏡列（例如，800個X 1 0列）被驅動也可以。DMD之資料 處理速度有其限度，由於與使用的畫素數成比例且每1線 之調變速度被決定，所以僅使用一部分的微鏡列，每1線的 調變速度係變快。藉此可縮短曝光時間。一方面，在連續地 使照射頭對曝光面相對移動之掃描方式的場合，副掃描方向 之畫素沒有必要全部使用。 例如，600組的微鏡列之中，在僅使用3 00組之場合，與600 組全部使用之場合相比較下，係可將每1線調變快2倍。 又，600組的微鏡列之中，在僅使用200組之場合，與600組 全部使用之場合相比較下，係可將每1線調變快3倍。亦 即，可在副掃描方向將500mm的區域以17秒曝光。再者，在 1274733 僅使用1 00組之場合時，係可將每1線調變快6倍。亦即， 可在副掃描方向將5 0 0 m m的區域以9秒曝光。 欲使用之微鏡列的數目，亦即，配置在副掃描方向之微鏡 的個數係10以上且200以下較好，10以上且100以下更好。 由於相當於1畫素之每1個微鏡的面積爲15//mXl5/zm，所 以若換算爲DMD50的使用區域，則12mmX 150 // m以上且12mm X3mm以下的區域較好，12_Xl50//m以上且12mmXl.5_ 以下的區域更好。 欲使用之微鏡列的數目若在上述範圍，則如第1 0圖所示， 使由光纖陣列光源66所出射的雷射光在透鏡系67施以略 平行光化而可對DMD50照射。由DMD50照射雷射光的照射 區域與DMD50之使用區域係一致者爲較佳。照射區域若較 使用區域還寬則雷射光之利用效率降低。 一方面，因應透鏡系67之在副掃描方向配列之微鏡的個 數，雖然有必要將集光於DMD50上之光束的副掃描方向之 直徑設定小，但是當使用之微鏡列的數目未滿1 0時，則入射 於DMD50之光束的角度係變大，在掃描面56中之光束的焦 點深度變淺，所以並不佳。又，以調變速度的觀點來說，使 用之微鏡列數爲200以下係較佳。. 〔微晶片之其他製造方法〕 上述的實施形態中，雖然以在構成微晶片之基板上直接 形成微小流路的例子加以說明，但是在模製作用的基板上形 成微小流路以製作模型，再藉由使用有此模型的模衝鍛或玻 璃模製，也可能製造具備微小流路之微晶片。 -31- 1274733 〔具備有微小流路之微晶片〕 上述的實施形態中，雖然以製造合成反應用微晶片爲例 加以說明，但是在本發明之微小流路的形成方法，係也適用 在製造具備有微小流路之其他種類的微晶片之場合。 以其他種類之微晶片而言，係可舉例有癌症診斷晶片、細 胞生化學晶片、環境計測晶片、層析晶片、電泳晶片、蛋 白質晶片、以及免疫分析晶片等等。此等晶片雖然因應各 晶片的機能而形成不同圖案之微小流路，但依本發明之微小 流路之形成方法，微小流路之形成圖案所因應的數位曝光可 形成蝕刻遮罩，所以可容易對應多種類的生產。又，也可容 易形成具備有複數個機能之微小流路。 又，本發明之微小流路之形成方法並不局限在實驗室晶片 之微小流路，可作爲在基板上形成微細溝的方法而廣範地使 用。 〔其他空間調變元件〕 上述的實施形態中，雖然已針對作爲空間調變元件之具 備有DMD的曝光頭加以說明，但是例如可使用GLV等之MEMS (微機電系統）型之空間調變元件（SLM )、依電氣光學效 果而調變透過光之光學元件（PLZT元件）及液晶光遮板 (FLC )等之 MEMS型以外之空間調變元件。 〔其他雷射 裝置（光源）〕 上述的實施形態中，係針對使用具備有複數個合波雷射光 源的光纖陣列光源之例子加以說明，但是雷射裝置並不局限 在把合波雷射光源予以陣列化的光纖陣列光源。例如，可使 -32 - 1274733 用把具備1條用以出射由具有1個發光點的單一半導體雷 射所入射之雷射光之光纖的光纖光源被陣列化的光纖陣列 光線。 〔雷射陣列〕 又，以具備有複數個發光點之光源而言，例如，如第2 1圖 所示，可使用在熱塊1 0 0上配列有複數個（例如7個）晶片 狀之半導體雷射LD1〜LD7的雷射陣列。 〔多腔雷射〕 又，如第22 ( A )圖所示，在指定方向配列有複數（例如，5 個）個發光點1 1 0 a之晶片狀的多腔雷射1 1 0係爲人所知悉。 多腔雷射1 1 0與配列晶片狀的半導體雷射相較下，係可高精 度地配列發光點，可容易地把各發光點所出射的雷射光束予 以合波。但是，發光點變多則於雷射製造時在多腔雷射1 1 〇 變得容易產生變形，所以發光點1 1 0 a之個數係設定爲5個 以下較佳。 本發明之曝光頭中，可將此多腔雷射110或如第22(B) 圖所示，在熱塊1 0 0上與各晶片之發光點1 1 0 a之配列方向 相同方向上配列有複數個多腔雷射1 1 0之多腔雷射陣列作 爲雷射裝置（光源）來使用。 〔使用多腔雷射之合波雷射光源〕 又，合波雷射光源並不被限定於用以把由複數個晶片狀之 半導體雷射所出射的雷射光予以合波者。例如，如第23圖 所示，可使用具備有複數（例如，3個）個發光點1 1 〇 a之晶 片狀的多腔雷射1 1 0之合波雷射光源。此合波雷射光源係 - 33 - 1274733 構成爲具備有多腔雷射1 1 〇、1條多模光纖1 3 Ο、以及集光 透鏡120。多腔雷射1 10係例如可以振盪波長爲405 nm的GaN 系雷射二極體來構成。 上述的構成中，由多腔雷射1 10之複數個發光點1 1 0 a所 出射的雷射光束B係各自由集光透鏡1 2 0所集光而入射於 多模光纖1 3 0的核心1 3 0 a。入射到核心1 3 0 a的雷射光係在 光纖內傳送且合波爲1條而出射。 在與上述多模光纖130之核心直徑略等寬度內並設多腔 雷射110之複數個發光點110a，同時作爲集光透鏡120,係 使用與多模光纖1 3 0之核心直徑略等焦點距離之凸透鏡或 來自多腔雷射110之出射光束僅在垂直其活性層之面內准 直的杆式透鏡，藉此可提升雷射光束B對多模光纖1 3 0的結 合效率。 〔使用多腔雷射陣列之合波雷射光源〕 又，如第24圖所示，可使用具備有複數（例如，3個）個發 光點之多腔雷射1 1 0、在熱塊1 1 1上具備有以等間隔配列複 數（例如，9個）個多腔雷射1 1 0之雷射陣列1 4 0的合波雷 射光源。複數個多腔雷射1 1 0係配列在與各晶片之發光點 1 1 0 a的配列.方向相同方向而固定。 此合波雷射光源係具備有：雷射陣列1 4 0 ;對應各多腔雷 射1 1 0而配置之複數個透鏡陣列1 1 4 ;配置在雷射陣列1 40 與複數個透鏡陣列1 1 4之間的1條杆式透境1 1 3 ; 1條多模 光纖1 3 0 ;以及集光透鏡1 2 0。透鏡陣列1 1 4係具備有對應 多腔雷射1 1 0之發光點的複數個微透鏡。 - 34- 1274733 上述的構成中，複數多腔雷射110之複數個發光點10a之 各自出射的雷射光束B ,係各自依杆式透境1 1 3而被集光在 指定方向之後，藉透鏡陣列1 1 4之各微透鏡而平行光化。 被平行光化的雷射光束L係由集光透鏡1 20集光而入射至 多模光纖1 3 0的核心1 3 0 a。入射至核心1 3 0 a的雷射光係在 光纖內傳送、合波成1條而出射。 〔多段構成之合波雷射光源〕 接著要介紹其他合波雷射光源的例子。此合波雷射光源 係如第2 5 ( A )及2 5 ( B )圖所示，在略矩形狀之熱塊1 8 0 上搭載有光軸方向的斷面爲L字狀的熱塊1 8 2，在2個熱塊 間形成有收納空間。在L字狀的熱塊1 82上面，以陣列狀 配列有複數個發光點（例如，5個）的複數（例如，2個）多 腔雷射110係在與各晶片之發光點110a的配列方向相同方 向以等間隔配列而固定。 略矩形狀的熱塊180形成有凹部，在熱塊180的空間側上 面，以陣列狀配列有複數個發光點（例如，5個）複數（例 如，2個）之多腔雷射1 1 0，係其發光點被配置成位在與配置 在熱塊1 8 2之上面的雷射晶片之發光點相同的鉛直面上。 多腔雷射1 1 0之雷射光出射側係配置有，因應各晶片的發 光點110a而配列有准直透鏡之准直透鏡陣列184。准直透 鏡陣列1 84 ,係各准直透鏡之長度方向和和雷射光束之視角 爲大的方向（速軸方向）一致，而各准直透鏡之寬度方向和 視角爲小的方向（遲軸方向）一致般地配置。如此，藉由 將准直透鏡陣列化而成一體化，雷射光之空間利用效率係提 - 35 - 1274733 升而可謀求合波雷射光源之高輸出化，同時可使零件數減少 且低成本化。 又，准直透鏡陣列1 8 4之雷射光出射側係配置有，1條多模 光纖1 3 0、以及把雷射光束集光至此多模光纖1 3 0的入射端 且結合的集光透鏡1 2 0。 上述的構成中，配置在雷射塊180、182上之複數多腔雷 射1 1 0之複數個發光點1 0 a所各自出射的雷射光束B係各 自被准直透鏡陣列184所平行光化，依集光透鏡120而被集 光以入射至多模光纖1 3 0之核心1 3 0 a。入射至核心1 3 0 a之 · 雷射光係在光纖內傳送且被合波成1條而出射。 此合波雷射光源係如同上述，藉由多腔雷射之多段配置 和准直透鏡之陣列化，特別可圖謀高輸出化。藉由使用此合 波雷射光源，因爲可構成高亮度之光纖陣列光源或束光纖光 源，所以特別適合作爲構成本發明之曝光裝置的雷射光源之 光纖光源。 此外，把上述之各合波雷射光源收納至罩內，可構成把多 模光纖1 3 0之出射端部由其罩引出的雷射模組。 鲁 又，在上述實施形態中，已說明了在合波雷射光源之多模 光纖的.出射端，與核心直徑爲與多模光纖相同且包層直徑爲 較多模光纖還小之其他光纖結合，以圖謀光纖陣列光源之高 亮度化的例子，例如第29圖所示，把包層直徑爲1 25 # m、8〇 A m、60// m等之多模光纖30在出射端不結合其他光纖之下 來使用也可以。 〔其他的成像光學系統〕 ~36- 1274733 上述的實施形態中，雖然在曝光頭所使用之DMD的光反 射側設置了作爲成像光學系統之2組透鏡，但也可配置將雷 射光放大而成像之成像光學系統。藉由放大由DMD所反射 之光束線的斷面積，可將在被曝光面中之曝光區域面積（畫 像區域）放大成所期望之大小。 例如，曝光頭可由如第26( A)圖所示構成：對DMD5 0，DMD5 0 照射雷射光之照明裝置144;把在DMD50反射之雷射光予以 放大而成像之透鏡系45 4,4 5 8;對應DMD50之各畫素而配置 有多數微透鏡474之微透鏡陣列47 2 ;對應微透鏡陣列472 之各微透鏡而配置有多數光圏478之光圈陣列476;以及使 通過光圈之雷射光成像於被曝光面56之透鏡系480,482。 以此曝光頭而言，由照明裝置144照射雷射光時，由DMD50 在開啓方向所反射之光束線的斷面積係經由透鏡系454、458 而被放大數倍（例如，2倍）。被放大的雷射光係由微透鏡 陣列472的各微透鏡而對應DMD50之各畫素被集光，通過光 圈陣列47 6之對應的光圈。通過光圈之雷射光係經由透鏡 系480、482而成像於被曝光面56上。 在此成像光學系統中，由DMD50所反射之雷射光係經由放 大透鏡4 5 4、4 5 8被放大數倍而投影至被曝光面56,所以全 體的畫像區域變廣。此時，若未配置有微透鏡陣列4 7 2及光 圈陣列4 7 6 ,則如第2 6 ( B )圖所示，投影至被曝光面5 6之 各光束光點BS之1畫素尺寸（光點尺寸）係因應曝光區域 468的尺寸而成爲大者，表不曝光區域468之鮮銳度的 MTF (光學傳遞函數）特性會降低。 1274733 一方面，在配置有微透鏡陣列472及光圈陣列47 6之場合， 由DMD50所反射之雷射光係依微透鏡陣列47 2的各微透鏡， 對應DMD50之各畫素而被集光。藉此，如第26 ( C)圖所示， 即使是在曝光區域被放大的場合，也可把各光束光點BS的 光點尺寸縮小成所期望之大小（例如，1 0 // m 1 0 // m )，可 防止MTF特性之降低以執行高精細的曝光。此外，曝光區域 468之所以傾斜係，爲了使畫素間沒有間隙而將DMD50傾斜 地配置所致。 又，即使依微透鏡之像差的光束爲寬，也可利用光圈使被 鲁 曝光面5 6上之光點尺寸成爲一定大小般地將光束整形，同 時藉由使其通過對應各畫素所設置的光圈，可防止在鄰接之 畫素間的串音。 再者，藉由在照明裝置1 44上使用與上述實施形態同樣 的高亮度光源，因爲由透鏡45 8入射至微透鏡陣列472的各 微透鏡之光束角度變小，所以可防止鄰接的畫素之光束的一 部分之入射。亦即，可實現高消光比。 〔發明之效果〕 鲁 依本發明的微小流路之形成方法，係能獲得可高速且高精 密形成微小流路、且能以低成本形成任意圖案之微小流路 的效果。 【圖式簡單說明】 第1圖係表示合成反應用微晶片的構成之斜視圖。 第2(A)〜2(G)圖係表示第1圖所示之合成反應用微晶片 之製造工程的順序之斷面圖。 -38- 1274733 第3 ( A )〜3 ( C )圖係阻體膜之厚膜化之例子的斷面圖。 第4 ( A )〜4 ( B )圖係用以說明伴隨著阻體膜之厚膜化，倉虫 刻精度會提升之說明圖。 第5圖係表示推拔狀圖案化的阻體膜之斷面圖。 第6圖係表示在本發明實施形態之曝光裝置的外觀斜視 圖。 第7圖係表示第6圖所示之曝光裝置的掃描器之構成斜 視圖。 第8(A)圖係表示形成在光阻膜之已曝光的區域之平面圖， 第8(B)圖係表示依各曝光頭的曝光區域之配列圖。 第9圖係表示第6圖所示之曝光裝置的曝光頭之槪略構 成斜視圖。 第10(A)係表示第9圖所示之曝光頭的構成之沿著光軸的 副掃描方向的斷面圖，第1 0 ( B )圖係1 0 ( A )圖之側面圖。 第11圖係表示數位微鏡裝置（DMD)的構成之部分放大 圖。 第12(A)及12(B)圖係用以說明DMD的動作之說明圖。 第1 3 ( A )、1 3 ( B )圖係表示DMD在未傾斜配置和傾斜配置 時之曝光束的配置及掃描線比較平面圖 。 第1 4 ( A )圖係表示光纖陣列光源的構成之斜視圖，第1 4 ( B ) 圖係第1 4 ( A )圖所示之光纖陣列光源的部分放大圖，第 1 4 ( C )、1 4 ( D )圖係表示在雷射出射部中之發光點的配列平 面圖 。 第15圖係表示多模光纖的構成圖。 -39- 1274733 第1 6圖係表示合波雷射光源的構成之平面圖。 第1 7圖係表示雷射模組的構成之平面圖。 第1 8圖係表示第1 7圖所示之雷射模組的構成之側面圖。 第1 9圖係表示第1 7圖所示之雷射模組的構成之部分側 面圖。 第20(A)、（B)圖係表示以往的曝光裝置中之焦點深度與 本實施形態之曝光裝置中之焦點深度的差異之沿著光軸的 斷面圖。 第2 1圖係表示雷射陣列的構成之斜視圖。 馨 第22(A)圖係表示多腔雷射的構成之斜視圖，第22(B)圖係 將第22 ( A )圖所示之多腔雷射予以陣列配列的多腔雷射陣列 之斜視圖。 第2 3圖係表示合波雷射光源之其他構成的平面圖。 第24圖係表示合波雷射光源之其他構成的平面圖。 第25(A)圖係表示合波雷射光源之其他構成之平面圖，第 2 5 ( B )圖係沿著第2 5 ( A )圖之光軸的斷面圖。 第26 ( A )圖係表示沿著結合光學系統之其他不同的曝光頭 ® 的構成之光軸的斷面圖，第26(B)圖係表示在不使用微透鏡 陣列等之場合時、投影至被曝光面之光像的平面圖。第26 ( C ) 圖係表示在使用有微透鏡陣列等之場合時、投影至被曝光 面之光像的平面圖。 【主要元件符號說明】 LD1〜LD7· · .GaN系半導體雷射 10......熱塊 -40 - 1274733 1 1〜1 7 · · · ·准直透鏡 2 0......集光透鏡 30......多模光纖 5 3......反射光像（曝光束） 5 4、5 8 · · · ·透鏡系 56......掃描面（被曝光面） 6 6......光纖陣列光源 73......組合透鏡The GaN-based semiconductor lasers LD1 to LD7 are all common in oscillation wavelengths (for example, 405 nm), and the maximum output is also common (for example, multi-mode laser is i00mw and single-mode laser is 30mW). Further, in the GaN-based semiconductor lasers LD1 to LD7, a laser having an oscillation wavelength other than the above-described 40 5 nm can be used in the wavelength range of 350 nm to 450 nm. The above-described combined laser light source is housed in a box-like package 40 having an opening above, together with other optical elements, as shown in Figs. The package 40 is provided with a package cover 4 1 made by closing the opening thereof. After the degassing process -23-1274733, the sealing gas is introduced, and the opening of the package 40 is closed by the package cover 41. The sealed laser light source is hermetically sealed in a closed space (sealed space) formed by the package 40 and the package cover 41. A substrate 42 is fixed on the bottom surface of the package 40. The upper surface of the substrate 42 is mounted with the thermal block 10; a collecting lens holder for holding the collecting lens 20; and an incident end for holding the multimode optical fiber 30. Fiber holder 46. The exit end of the multimode fiber 30 is led out of the package by an opening formed in the wall surface of the package 40. Further, a collimator lens holder 44 is attached to the side surface of the thermal block 10, and the collimator lenses 1 1 to 17 are held. An opening is formed in the lateral wall surface of the package 40, and the wiring 47 for supplying the driving current to the GaN-based semiconductor lasers LD1 to LD7 is led out of the package. Further, in Fig. 18, in order to avoid cumbersome drawing, only the plurality of GaN-based semiconductor lasers are numbered with respect to the GaN-based semiconductor laser LD 7, and only a plurality of collimating lenses are given. Straight lens 1 7 is numbered. Fig. 19 is a view showing the front shape of the mounting portions of the above-described collimating lenses 11 to 17. Each of the collimator lenses 1 1 to 17 is formed so as to be elongated in a plane parallel to the optical axis including the aspherical circular lens. The elongated collimating lens can be formed, for example, by molding a resin or optical glass. The collimator lenses 11 to 17 are arranged such that the longitudinal direction thereof is closely arranged in the direction in which the light-emitting points of the GaN-based semiconductor lasers LD1 to LD7 are aligned (the left-right direction in FIG. 19) in the arrangement direction of the light-emitting points. On the other hand, in the GaN-based semiconductor lasers LD1 to LD7, an active layer having a light-emitting width of 2 // m is used, and a direction parallel to the active layer and a right-angle angle of view of -24 - 1274733 are each, for example, 1 Ο. The laser of each of the laser beams B1 to B7 is emitted in a state of ° and 30 °. These GaN-based semiconductor lasers LD1 to LD7 are arranged in a line in a direction parallel to the active layer. Therefore, the laser beams B 1 to B7 emitted from the respective light-emitting points are aligned with the longitudinal direction of the collimating lenses 1 1 to 17 7 having an elongated shape as described above. The direction in which the angle is small is incident in a state in which the width direction (the direction orthogonal to the longitudinal direction) coincides. That is, each of the collimating lenses 11 to 17 has a width of 1.1 mm and a length of 4.6 mm, and the beam diameters in the horizontal direction and the vertical direction of the laser beams B1 to B7 incident thereon are each 0. 9 mm, 2 . 6mm. Further, each of the collimator lenses 1 1 to 17 has a focal length fl = 3 mm, a NA of 0.6, and a lens arrangement pitch of 1.25 mm. The collecting lens 20 is formed by extending a region including an optical axis of a circular lens having an aspherical surface in a parallel plane, and the alignment direction of the collimator lenses 1 1 to 17 is formed to be long in the horizontal direction. And a shape that is short in a direction perpendicular thereto. The collecting lens 20 has a focal length f 2 = 2 3 mm and NA = 0.2. The collecting lens 20 is also formed by, for example, molding a resin or an optical glass. ♦ The operation of the above exposure apparatus will be described below. In each of the exposure heads 166 of the scanner 162, the GaN-based semiconductor lasers LD1 to LD7 constituting the multiplexed laser light source of the optical fiber array light source 66 are each emitted in a divergent light state, and the laser beams B1, B2, B3, and B4 are emitted. B5, B6, and B7 are each independently actinic by the corresponding collimating lenses 1 1 to 17 . The laser beams B1 to B7 which are parallelized are collected by the collecting lens 20 and contracted to the incident end faces of the core 30 a of the multimode fiber 30. -25 - 1274733 In this example, the collimating lenses 11 to 17 and the collecting lens 20 constitute a collecting optical system, and the collecting optical system and the multimode optical fiber 3 constitute a combined optical system. That is, the laser beam 20 is collected by the collecting lens 20, and the laser beam B1 to B7 collected as described above is incident on the core 30a of the multimode fiber 30 to be transmitted in the optical fiber, and is combined into one laser beam B. The fiber 31 is coupled to the exit end of the multimode fiber 30. In each of the laser modules, when the bonding efficiency of the laser beams B 1 to B7 to the multimode fiber 30 is 0.85 and the outputs of the GaN semiconductor lasers LD1 to LD7 are 30 mW, the fibers arranged in an array are arranged. The 3 1 series can obtain a combined laser beam B of about 18〇11^(= 3 OMw X 0 · 8 5 X 7 ). Therefore, the output of the laser output portion 68 in which six fibers 31 are arranged in an array is about 1 W (= 180 mWX 6 ). The laser emitting portion 6 8 of the optical fiber array light source 66 has such high-luminance light-emitting points arranged in a row along the main scanning direction. Since a conventional optical fiber light source that combines laser light from a single semiconductor laser to one optical fiber has a low output, if a plurality of columns are not arranged, a desired output cannot be obtained, but the combined laser used in the present embodiment is used. The light source is high output, so a few columns, for example even one column, can obtain the desired output. For example, in a conventional optical fiber source in which a semiconductor laser and an optical fiber are combined in a one-to-one manner, generally, in the case of a semiconductor laser, a laser having an output of 30 mW (milliwatt) is used, in terms of an optical fiber, because A multimode fiber with a core diameter of 5 〇m, a cladding diameter of 1 2 5 μm, and a ΝΑ (number of openings) 〇. 2 is used, so if an output of about 1 W (Watt) is to be obtained, the multimode fiber must be 4 8 strips (8 X 6 ) are bundled, and the area of the light-emitting area is 〇.62mm2 (0.675mmX0.925mm), so the brightness of the laser exit portion 68 is 1.6 X106 (W/m2), and the brightness of each 1127735 optical fiber It is 3.2xi06(W/m2). On the other hand, in the present embodiment, as described above, the output of the multimode optical fiber 6 can be obtained as a 1 1 V output, and the area of the light emitting region of the laser emitting portion 68 is 0.0081_2 (0.325 mm×0.025 mm), so that Ray The brightness of the emission and emission unit 68 is 123 XI 06 (W/m 2 ), which is 80 times higher than that of the conventional image. Further, the brightness of each of the optical fibers is 90 X 1 06 (W/m2), which is about 28 times higher than that of the conventional one. Here, the difference between the focus depths of the conventional exposure head and the exposure head of the present embodiment will be described with reference to Figs. 20(A) and 20(B). The diameter of the light-emitting region of the bundled fiber-optic light source of the conventional exposure head in the sub-scanning direction is 0 · 6 7 5 mm, and the diameter of the light-emitting region of the fiber array light source of the exposure head in the sub-scanning direction is 0.025 mm. As shown in Fig. 20(A), in the conventional exposure head, since the light-emitting area of the light source (bundle fiber light source) 1 is large, the angle of the light beam incident on the DMD 3 is increased, and as a result, the scanning surface 5 is incident. The angle of the beam becomes larger. For this reason, the beam diameter is easily too wide with respect to the collecting direction (deviation of the focus direction). On the other hand, as shown in the 20th (B) diagram, in the exposure head of the present embodiment, since the diameter of the light-emitting region of the optical fiber array light source 6 6 is small in the sub-scanning direction, the light beam incident on the DMD 50 by the lens system 67 is As the angle becomes smaller, as a result, the angle of the light beam incident on the scanning surface 56 becomes smaller. That is, the depth of focus becomes deeper. In this example, the diameter of the light-emitting region in the sub-scanning direction is about 30 times that of the conventional one, and the depth of focus corresponding to the slightly-wound limit can be obtained. Therefore, it is suitable for exposure of minute spots. The effect of this depth of focus is that the greater the amount of light necessary for the exposure head, the more significant and effective. In this example, the 1 pixel size projected on the exposure surface is -27-1274733 1 〇 M m X 1 〇 # m. In addition, the DMD-based reflective spatial modulation elements, such as the 20th (A) and 20 (B) diagrams, are used to illustrate the development of the optical relationship. The image data corresponding to the exposure pattern is input to a controller (not shown) connected to the DMD 50, and the frame memory in the controller is temporarily memorized. This image data is data represented by the density of each pixel constituting the image in binary (the presence or absence of dot recording). The stage 150 of the substrate 150 on which the photoresist film is formed is driven by a driving device (not shown) at a constant speed from the upstream side of the shutter 1 60 to the downstream side along the guiding portion 158. mobile. When the stage 152 is below the gate 160, when the detecting sensor 1 64 attached to the shutter 160 detects the leading end of the substrate 150, the image data stored in the frame memory is sequentially The complex line is read, and a control signal for each exposure head 166 is generated based on the image data read by the data processing unit. Then, using the mirror drive control unit, the micromirrors of the DMDs 50 of the respective exposure heads 166 are controlled to be turned on and off in accordance with the generated control signals. That is, in the DMD 50, a micromirror array having 800 micromirrors arranged in the main scanning direction is arranged in the sub-scanning direction with 600 sets, all of which are used. When the DMD 50 is irradiated with the laser light by the optical fiber array light source 66, the laser light reflected when the micromirror of the DMD 50 is turned on is imaged by the lens system 54 and 58 on the photoresist film formed on the substrate 150. Exposure surface 5 6 on. In this way, the laser light emitted by the fiber array light source 66 is turned on and off in each pixel, and the photoresist film is in the same number of pixel units as the DMD 50 (exposure area 168). Was exposed. Further, the substrate 15 is moved at a constant speed together with the stage 152, so that the photoresist film formed on the substrate 15 is rotated by the scanner 1 62 in the opposite direction to the stage -28-1274733 A sub-scan is performed in the direction to form an exposed region 170 of each exposure head 丨6 6 . When the sub-scan of the photoresist film of the scanner 1 62 is terminated and the rear end of the substrate 150 is detected by the detecting sensor 1 64, the stage 丨 52 is guided by a driving device (not shown). The portion 158 returns to the origin on the most upstream side of the gate 16〇, and moves again at a constant speed from the upstream side to the downstream side of the gate 160 from the guide 邰1 58. As described above, in the present embodiment, since the spatial light modulation element such as DMD is used in the exposure process of the photoresist film, the laser light can be modulated in each pixel in response to the formation pattern of the minute flow path. The laser light that can be modulated changes the photoresist film to a high speed and high precision. In this way, in the exposure engineering, since the photoresist of the Rensi pattern can be exposed to the local speed and finely, the micro flow path of any pattern can be formed at high speed and high precision through the second patterning process and the etching process. . As described above, since exposure in an arbitrary pattern is possible, a minute flow path of a complicated pattern can be easily formed. Further, since the film can be exposed at a high speed, a minute flow path can be formed on a large-area glass substrate for a short period of time. Further, since the coefficient bits are exposed, the mask of each pattern is not required, and the minute flow path can be formed at low cost. Moreover, since the DFR or the electrode deposition resist is used in the photoresist film, the film can be thickened in comparison with the resistor used in the semiconductor process, and a photoresist having a thickness of 1 〇#m~4 0 // m can be formed. membrane. In this way, by thickening the thickness of the first resist film, it is possible to form a minute flow path of the deep groove with high precision by the worm. Further, a plurality of layers of the photoresist film can be laminated to form a thick film. In this case, the same function as the -29-1274733 digital exposure can be used to expose the same position of the multiple laminated photoresist films. Further, in the present embodiment, the optical fiber array light source is formed by using the multiplexed laser light source in the exposure apparatus, and the diameter of the exit end cladding of the optical fiber is made smaller than the diameter of the incident end cladding layer, so that the diameter of the light-emitting portion is small and is changed. The high brightness of the fiber array light source is plotted. Thereby, the laser light can be exposed to a deeper depth of focus to expose the photoresist film to a higher precision. For example, the exposure of the ultra-high resolution of the beam diameter of 1 # m or less and the resolution 〇. i # Π] is sufficient to form the groove width i 〇# m ～5 0 // πι on the micro-flow path with excellent precision. . Modifications of the embodiment will be described below. [High-speed driving method] Generally, in the DMD, a micromirror array in which 800 micromirrors are arranged in the main scanning direction is arranged in the sub-scanning direction with 60 sets, but only a part of the micromirror columns are controlled by the controller (for example) , 800 X 1 0 columns) can also be driven. The data processing speed of DMD has a limit. Since the modulation speed per line is determined in proportion to the number of pixels used, only a part of the micro mirror array is used, and the modulation speed per line is fast. This can shorten the exposure time. On the other hand, in the case of a scanning method in which the irradiation head is relatively moved to the exposure surface continuously, it is not necessary to use all of the pixels in the sub-scanning direction. For example, among the 600 sets of micromirror columns, when only 300 sets are used, compared with the case where all 600 sets are used, the line can be adjusted twice as fast. Further, among the 600 sets of micromirror rows, when only 200 sets are used, compared with the case where all 600 sets are used, it is possible to change the number of lines per line three times faster. That is, an area of 500 mm can be exposed in the sub-scanning direction for 17 seconds. Furthermore, when 1274733 is used only for 100 sets, it is possible to change the frequency of each line by 6 times. That is, an area of 500 m in the sub-scanning direction can be exposed for 9 seconds. The number of micromirror columns to be used, that is, the number of micromirrors arranged in the sub-scanning direction is preferably 10 or more and 200 or less, more preferably 10 or more and 100 or less. Since the area of each micromirror corresponding to one pixel is 15//m×l5/zm, if converted to the use area of DMD50, a region of 12 mm×150 // m or more and 12 mm×3 mm or less is preferable, 12_Xl50// The area above m and below 12mmXl.5_ is better. If the number of micromirror rows to be used is within the above range, as shown in Fig. 10, the laser light emitted from the optical fiber array light source 66 can be irradiated to the DMD 50 by slightly parallelizing the lens system 67. It is preferable that the irradiation area in which the laser light is irradiated by the DMD 50 coincides with the use area of the DMD 50. If the irradiation area is wider than the use area, the utilization efficiency of the laser light is lowered. On the one hand, in view of the number of micromirrors arranged in the sub-scanning direction of the lens system 67, although it is necessary to set the diameter of the sub-scanning direction of the light beam collected on the DMD 50 to be small, the number of micromirror columns used is not When the time is 10, the angle of the light beam incident on the DMD 50 becomes large, and the depth of focus of the light beam on the scanning surface 56 becomes shallow, which is not preferable. Further, from the viewpoint of the modulation speed, it is preferable that the number of micromirrors used is 200 or less. [Other manufacturing method of microchip] In the above-described embodiment, an example in which a minute flow path is directly formed on a substrate constituting a microchip is described. However, a micro flow path is formed on a substrate for molding to form a mold. It is also possible to manufacture a microchip having a minute flow path by using die punching or glass molding having this model. -31- 1274733 [Microchip having micro flow path] In the above embodiment, a microchip for synthesizing a reaction is described as an example. However, the method of forming the microchannel in the present invention is also applicable to manufacturing. There are other types of microchips with tiny flow paths. For other types of microchips, for example, a cancer diagnostic wafer, a cell biochemical wafer, an environmental measurement wafer, a chromatography wafer, an electrophoresis wafer, a protein wafer, and an immunoassay wafer can be exemplified. Although these wafers form minute flow paths of different patterns in accordance with the functions of the respective wafers, according to the method of forming the micro flow paths of the present invention, the digital exposure corresponding to the formation pattern of the micro flow paths can form an etching mask, so that it can be easily Corresponding to a variety of types of production. Further, it is also easy to form a minute flow path having a plurality of functions. Further, the method of forming the minute flow path of the present invention is not limited to the minute flow path of the laboratory wafer, and can be widely used as a method of forming fine grooves on the substrate. [Other Spatial Modulation Element] In the above-described embodiment, an exposure head including a DMD as a spatial modulation element has been described. For example, a MEMS (Micro Electro Mechanical System) type spatial modulation element such as GLV can be used. (SLM) A spatial modulation element other than the MEMS type such as an optical element (PLZT element) that transmits light and a liquid crystal shutter (FLC) is modulated by an electro-optical effect. [Other laser device (light source)] In the above embodiment, an example of using a fiber array light source including a plurality of multiplexed laser light sources is described, but the laser device is not limited to the multiplexed laser light source. Arrayed fiber array light sources. For example, -32 - 1274733 can be used to array optical fiber arrays having a fiber source for emitting laser light incident from a single semiconductor laser having one light-emitting point. [Laser Array] Further, as a light source including a plurality of light-emitting points, for example, as shown in FIG. 2, a plurality of (for example, seven) wafers may be arranged in the thermal block 100. Laser array of semiconductor lasers LD1 to LD7. [Multi-cavity laser] Also, as shown in Fig. 22 (A), a wafer-shaped multi-cavity laser 1 1 0 system in which a plurality of (for example, 5) light-emitting points 1 1 0 a are arranged in a specified direction is People know. Compared with a semiconductor laser equipped with a wafer, the multi-cavity laser 110 can arrange the light-emitting points with high precision, and can easily combine the laser beams emitted from the respective light-emitting points. However, when the number of light-emitting points is increased, the multi-cavity laser 1 1 〇 is easily deformed during laser manufacturing, so that the number of light-emitting points 1 1 0 a is preferably 5 or less. In the exposure head of the present invention, the multi-cavity laser 110 can be arranged in the same direction as the arrangement direction of the light-emitting points 1 1 0 a of the respective wafers on the thermal block 100 as shown in the 22nd (B) diagram. There are a plurality of multi-cavity laser 110-cavity laser arrays for use as laser devices (light sources). [Combined laser light source using multi-chamber laser] Further, the multiplexed laser light source is not limited to those for multiplexing laser light emitted from a plurality of wafer-shaped semiconductor lasers. For example, as shown in Fig. 23, a multi-chamber laser 110 laser light source having a plurality of (for example, three) light-emitting points 1 1 〇 a can be used. The combined laser light source system - 33 - 1274733 is configured to have a multi-cavity laser 1 1 〇, a multimode fiber 13 Ο, and a collecting lens 120. The multi-cavity laser 1 10 system can be constructed, for example, by oscillating a GaN-based laser diode having a wavelength of 405 nm. In the above configuration, the laser beam B emitted from the plurality of light-emitting points 1 1 0 a of the multi-cavity laser 1 10 is collected by the collecting lens 1 2 0 and is incident on the multimode fiber 1 130. Core 1 3 0 a. The laser light incident on the core 130 h is transmitted in the optical fiber and is split into one and is emitted. A plurality of light-emitting points 110a of the multi-cavity laser 110 are disposed in a slightly equal width to the core diameter of the multimode fiber 130, and at the same time, as the collecting lens 120, a focus is slightly equal to the core diameter of the multimode fiber 130. The convex lens from the convex lens or the exiting beam from the multi-chamber laser 110 is only aligned in a plane perpendicular to the plane of its active layer, thereby enhancing the bonding efficiency of the laser beam B to the multimode fiber 130. [Using a multi-cavity laser array multiplexed laser light source] Further, as shown in Fig. 24, a multi-cavity laser having a plurality of (for example, three) light-emitting points can be used. The 1 1 has a combined laser light source with a plurality of (for example, 9) multi-chamber laser 1 1 1 laser arrays 1 40 at equal intervals. A plurality of multi-cavity lasers 1 1 0 are arranged in the same direction as the arrangement of the light-emitting points of each wafer, 1 10 a. The multiplexed laser light source is provided with: a laser array 140; a plurality of lens arrays 1 1 4 configured corresponding to each multi-cavity laser 110; and a laser array 1 40 and a plurality of lens arrays 1 1 bar type 1 1 3 between 1 4; 1 multimode fiber 1 3 0 ; and collecting lens 1 2 0. The lens array 1 14 is provided with a plurality of microlenses corresponding to the light-emitting points of the multi-cavity laser 110. - 34- 1274733 In the above configuration, the laser beam B emitted from each of the plurality of light-emitting points 10a of the plurality of multi-cavity lasers 110 is collected in a specified direction according to the rod-type permeability 1 1 3 Each of the microlenses of the lens array 112 is parallelized to be photochemically. The laser beam L that is collimated in parallel is collected by the collecting lens 120 and incident on the core 1300 of the multimode fiber 130. The laser light incident on the core 130 h is transmitted through the optical fiber and merged into one to be emitted. [Multi-section compositing laser source] Next, an example of other multiplexed laser sources will be described. The multiplexed laser light source is provided with a heat block having an L-shaped cross section in the optical axis direction on the slightly rectangular heat block 180 as shown in Figs. 2 5 (A) and 2 5 (B). 1 8 2, a storage space is formed between the two thermal blocks. On the upper surface of the L-shaped thermal block 1 82, a plurality of (for example, two) multi-cavity lasers 110 arranged in an array of a plurality of light-emitting points (for example, five) are arranged in association with the light-emitting points 110a of the respective wafers. The directions are fixed in the same direction at equal intervals. The slightly rectangular heat block 180 is formed with a concave portion, and a plurality of light-emitting points (for example, five) of a plurality of (for example, two) multi-cavity lasers 1 1 0 are arranged in an array on the space side of the heat block 180. The light-emitting point is arranged to be positioned on the same vertical plane as the light-emitting point of the laser chip disposed above the thermal block 128. The multi-chamber laser 110 light exiting side is provided with a collimating lens array 184 in which a collimating lens is arranged in response to the light emitting point 110a of each wafer. The collimating lens array 1 84 has a length direction of each collimating lens and a direction in which the viewing angle of the laser beam is large (the direction of the speed axis), and the width direction and the viewing angle of each collimating lens are small (latent axis) The direction is consistently configured. In this way, by integrating the collimating lens into an array, the space utilization efficiency of the laser light is 35 - 1274733 liters, and the output of the multiplexed laser light source can be increased, and the number of parts can be reduced and the cost can be reduced. Chemical. Further, the laser light exiting side of the collimator lens array 184 is provided with one multimode fiber 130, and a collecting lens that combines the laser beam to the incident end of the multimode fiber 130. 1 2 0. In the above configuration, the laser beams B respectively emitted by the plurality of light-emitting points 10 a of the plurality of multi-cavity lasers 110 arranged on the laser blocks 180 and 182 are respectively collimated by the collimating lens array 184. The light is collected by the collecting lens 120 to be incident on the core 130 0 a of the multimode fiber 130. The laser light incident on the core 1130a is transmitted in the optical fiber and is combined into one to be emitted. The multiplexed laser light source is as described above, and is multi-chambered in a multi-segment laser array and an array of collimating lenses, and is particularly capable of high output. By using this combined laser light source, it is particularly suitable as a fiber source for a laser light source constituting the exposure apparatus of the present invention because it can constitute a high-intensity fiber array light source or a bundle fiber light source. Further, by accommodating the above-described respective combined laser light sources in the cover, a laser module in which the exit end portion of the multimode optical fiber 130 is led out from the cover can be constructed. Lu, in the above embodiment, the output end of the multimode fiber in the multiplexed laser source has been described, and the other fiber having the same core diameter as the multimode fiber and having a larger cladding fiber diameter than the other mode fiber In combination, in order to exemplify the high brightness of the optical fiber array light source, for example, as shown in Fig. 29, the multimode optical fiber 30 having a cladding diameter of 1 25 # m, 8 〇A m, 60//m or the like is not at the exit end. It can also be used in combination with other optical fibers. [Other imaging optical system] ~36- 1274733 In the above embodiment, although two sets of lenses as imaging optical systems are provided on the light reflection side of the DMD used for the exposure head, it is also possible to arrange the laser light to be magnified and imaged. Imaging optical system. The area of the exposed area (image area) in the exposed surface can be enlarged to a desired size by enlarging the sectional area of the beam line reflected by the DMD. For example, the exposure head can be constructed as shown in Fig. 26(A): an illumination device 144 that irradiates the DMD 50, the DMD 50 with laser light, and a lens system that magnifies the laser light reflected by the DMD 50 and images 45 4, 4 5 8 a microlens array 47 2 configured with a plurality of microlenses 474 corresponding to respective pixels of the DMD 50; an aperture array 476 of a plurality of apertures 478 disposed corresponding to each microlens of the microlens array 472; and imaging of laser light passing through the aperture The lens system 480, 482 on the exposed surface 56. With this exposure head, when the laser beam is irradiated by the illumination device 144, the area of the beam line reflected by the DMD 50 in the opening direction is amplified several times (for example, twice) via the lens systems 454 and 458. The amplified laser light is collected by the respective microlenses of the microlens array 472 corresponding to the respective pixels of the DMD 50, and passes through the corresponding apertures of the aperture array 476. The laser light passing through the aperture is imaged on the exposed surface 56 via the lens systems 480, 482. In this imaging optical system, the laser light reflected by the DMD 50 is magnified several times by the magnification lenses 4 5 4 and 4 5 8 and projected onto the exposure surface 56, so that the entire image area is widened. At this time, if the microlens array 4 7 2 and the aperture array 4 7 6 are not disposed, as shown in the second 6 (B) diagram, the 1 pixel size of each of the beam spots BS projected onto the exposed surface 56 The (spot size) is larger depending on the size of the exposure region 468, and the MTF (optical transfer function) characteristic of the sharpness of the non-exposed region 468 is lowered. 1274733 On the one hand, in the case where the microlens array 472 and the aperture array 47 6 are disposed, the laser light reflected by the DMD 50 is collected by the respective microlenses of the microlens array 47 2 corresponding to the respective pixels of the DMD 50. Thereby, as shown in Fig. 26(C), even when the exposure area is enlarged, the spot size of each beam spot BS can be reduced to a desired size (for example, 1 0 // m 1 0 // m ) prevents the MTF feature from being lowered to perform high-definition exposure. Further, the exposure region 468 is inclined so that the DMD 50 is obliquely arranged in order to prevent a gap between the pixels. Further, even if the beam of the aberration of the microlens is wide, the beam can be shaped by the aperture to make the size of the spot on the exposed surface 56 a certain size, and by passing the corresponding pixels. The aperture is set to prevent crosstalk between adjacent pixels. Further, by using the high-intensity light source similar to that of the above-described embodiment in the illumination device 1 44, since the beam angle of each microlens incident on the microlens array 472 by the lens 45 8 becomes small, adjacent pixels can be prevented. The incident of a portion of the beam. That is, a high extinction ratio can be achieved. [Effects of the Invention] According to the method for forming a minute flow path of the present invention, it is possible to obtain a micro flow path capable of forming a minute flow path at a high speed and with high precision, and capable of forming an arbitrary pattern at a low cost. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a perspective view showing the configuration of a microchip for synthesis reaction. The second (A) to (G) drawings are cross-sectional views showing the order of the manufacturing process of the microchip for synthesis reaction shown in Fig. 1. -38- 1274733 Section 3 (A) to 3 (C) A cross-sectional view of an example of thick film formation of a resist film. The 4th (A) to 4th (B) diagrams are used to explain the increase in the precision of the worms with the thick film formation of the barrier film. Fig. 5 is a cross-sectional view showing a resistive patterned resist film. Fig. 6 is a perspective view showing the appearance of an exposure apparatus according to an embodiment of the present invention. Fig. 7 is a perspective view showing the configuration of a scanner of the exposure apparatus shown in Fig. 6. Fig. 8(A) is a plan view showing an exposed region of the photoresist film, and Fig. 8(B) is a view showing an arrangement of exposure regions of the respective exposure heads. Fig. 9 is a perspective view showing a schematic configuration of an exposure head of the exposure apparatus shown in Fig. 6. Fig. 10(A) is a cross-sectional view showing the configuration of the exposure head shown in Fig. 9 along the optical axis in the sub-scanning direction, and the 10th (B) diagram is a side view of the 10 (A) diagram. Fig. 11 is a partially enlarged view showing the configuration of a digital micromirror device (DMD). The 12th (A) and 12(B) drawings are explanatory diagrams for explaining the operation of the DMD. The 1 3 (A) and 1 3 (B) diagrams show the arrangement of the exposure beam and the comparison of the scanning lines of the DMD in the untilted configuration and the oblique configuration. The first 4 (A) diagram shows a perspective view of the configuration of the optical fiber array light source, and the first 4 (B) diagram is a partially enlarged view of the optical fiber array light source shown in the first 4 (A) diagram, and the first 4 (C) The 1 4 (D) diagram shows a plan view of the arrangement of the light-emitting points in the laser exit portion. Fig. 15 is a view showing the configuration of a multimode fiber. -39- 1274733 Figure 16 is a plan view showing the construction of a combined laser light source. Fig. 17 is a plan view showing the configuration of the laser module. Fig. 18 is a side view showing the configuration of the laser module shown in Fig. 17. Fig. 19 is a side elevational view showing the configuration of the laser module shown in Fig. 17. 20(A) and (B) are cross-sectional views along the optical axis showing the difference between the depth of focus in the conventional exposure apparatus and the depth of focus in the exposure apparatus of the embodiment. Fig. 2 is a perspective view showing the configuration of a laser array. Xin 22(A) is a perspective view showing the configuration of a multi-chamber laser, and 22 (B) is a multi-chamber laser array in which the multi-chamber laser shown in FIG. 22(A) is arrayed. Oblique view. Fig. 2 is a plan view showing another configuration of the multiplexed laser light source. Fig. 24 is a plan view showing another configuration of the multiplexed laser light source. Fig. 25(A) is a plan view showing another configuration of the multiplexed laser light source, and Fig. 25(B) is a sectional view taken along the optical axis of Fig. 25(A). Figure 26 (A) shows a cross-sectional view of the optical axis along the other different exposure heads of the optical system, and Figure 26 (B) shows the projection when the microlens array or the like is not used. A plan view of the light image to the exposed surface. The 26th (C) diagram is a plan view showing an optical image projected onto the exposure surface when a microlens array or the like is used. [Description of main component symbols] LD1 to LD7·. .GaN semiconductor laser 10... Thermal block-40 - 1274733 1 1~1 7 · · · · Collimating lens 2 0... Collecting lens 30...multimode fiber 5 3...reflected light image (exposure beam) 5 4,5 8 · · · ·Lens system 56...scanning surface Exposure surface) 6 6...fiber array light source 73...combined lens
15 0.....感光材料 152......載物台 1 6 2......掃描器 166......曝光頭 168......曝光區域 170......已曝光區域 202 ......搬運滾筒 204a、204b· •注入口15 0..... photosensitive material 152... stage 1 6 2 ... scanner 166 ... exposure head 168 ... exposure area 170. ..... exposed area 202 ... carrying roller 204a, 204b · • injection port
206 ......排出口 210......合流點 212......光阻膜 214......曝光部分 -41 -206 ... discharge port 210 ... confluence point 212 ... photoresist film 214 ... exposure section -41 -