200528752 (1) 玖、發明說明 【發明所屬之技術領域】 本發明係關於空間光調變器（SLM)，特別關於具有隱 藏式鉸鏈以使像素塡充比最大、使散射及漫射最小、以及 取得高對比和高影像品質的微鏡結構。 【先前技術】 在光資訊處理、投射式顯示器、影像及圖形監視器、 電視、及電子照相列印領域中，空間光調變器具有不同的 應用。反射式S LM是以空間圖案調變入射以反射對應於 電或光輸入的影像。入射光可以在相位上、強度上、或偏 轉反向上被調變。反射式SLM典型上包含能夠反射入射 光之可尋址的畫素（像素）之區域或二維陣列。S L Μ的關鍵 參數，特別是在顯示器應用上，係光學上主動區至像素區 的部份（也以反映SLM的全部表面區之SLM表面區的一 部份作爲量測，也稱爲塡充比）。需要有高的塡充比。 習知的S L Μ具有不同的缺點。這些缺點包含但不限 於：（1 )低於最佳光學主動區，降低光學效率；（2)粗糙的 反射表面，降低鏡的反射率；（3)繞射及散射，降低顯示 器的對比；（4)所使用的材料具有長期可靠性問題；及（5) 複雜製程，增加開銷及降低裝置產能。 很多習知的裝置在它們的表面上包含實質上非反射的 區域。這提供低塡充比，以及提供低於最佳反射效率。舉 例而言’美國專利號4,2 2 9,7 3 2揭示形成於裝置表面上的 -5 - 200528752 (2) MOSFET裝置以及鏡。這些MOSFET裝置佔據表面區 低典型上是光學是主動之裝置面積的部份以及降低反 率。在裝置的表面上之 MOSFET裝置也會使入射光 ，降低顯示器的對比。此外，撞擊曝露的 MOSFET 之強光會使MOSFET裝置充電及使電路過熱，因而 裝置的適當操作。 某些S LM設計具有使入射光散射及降低反射率 糙表面。舉例而言，在某些SLM設計中，反射表面 積於LPCVD氮化矽層上的鋁膜。由於這些反射鏡的 是沈積有薄膜，所以，它們的平滑度難以控制。因此 後的產品具有粗糙表面，降低反射效率。 因某些SLM設計（特別是某些懸吊鏡設計）而 反射效率之另一問題是大的曝露鉸鏈表面積。這些曝 鉸鏈表面積會因鉸鏈結構而造成散射及漫射，相較於 參數，不利於對比。 很多傳統的S L Μ，例如美國專利號4，5 6 6 ; 9 3 5中 示之SLM具有由鋁合金製成的鉸鏈。鋁、以及其它 易於受到疲勞及塑膠變形影響，導致長期可靠度的問 而且，鋁易於受到胞「記憶」影響，其中，其餘位置 向其最經常被佔據的位置傾斜。此外，4，5 6 6 5 9 3 5專 所揭示的鏡會因移除鏡表面下方的犧牲材料而被釋放 技術通常造成精密的微鏡結構於釋放期間斷裂。其也 在鏡之間有大間隙以便蝕刻劑移除鏡下方的犧牲材料 低光學上主動的裝置區之部份。 ，降 射效 繞射 裝置 干擾 之粗 是沈 表面 ，最 降低 露的 其它 所揭 金屬 題。 開始 利中 。此 需也 ，降 -6 - 200528752 (3) 其它傳統的S LM需要多層，包含用於鏡之分離 鉸鏈、電極及/或控制電路。例如如多層S LM等的 需要使用多層薄膜堆疊以及蝕刻技術和製程。使用這 術和製程是昂貴並造成低產能。舉例而言，使用這些 通常會牽涉到大規模沈積及移除鏡板之下的犧牲材料 板表面之下的多層薄膜沈積及堆疊會造成較粗糙的鏡 ，藉因降低鏡的反射效率。此外，由於在不同層或基 具有鏡及鉸鏈，所以在鏡偏向時造成平移偏移。由於 偏移，陣列中的鏡必須相間隔以避免相鄰的鏡之間的 干擾。由於陣列中的鏡無法被設置成太接近陣列中的 鏡，所以，S L Μ會苦於比最佳光學主動區還低或較 充比。 需要具有增進的反射效率、SLM裝置長期可靠 及簡化的製程之SLM。 【發明內容】 本發明係空間光調變器（S L Μ)。在一實施例中， 具有由第一基底製成之反射選擇性可偏轉微鏡陣列， 基底接合至具有個別可尋址的電極之第二基底。第二 也具有用於微鏡陣之定址及控制電路。或者，定址及 電路的部份是在分別的基底上且連接至第二基底上的 及電極。 微鏡陣列包含設有高度反射表面以反射Α射光之 制地偏轉鏡板。此第一基底是單一材料的晶圓，在一 層、 製造 些技 技術 。鏡 表面 底中 平移 機械 其它 低塡 度、 SLM 第一 基底 控制 電路 可控 實施 -7- 200528752 (4) 例中爲單晶矽。間隔器支撐壁在鏡板及電極之間提供分離 ，該電極係與該鏡板相關連’控制鏡板的偏轉。電極設於 第二基底上，第二基底接合至微鏡陣列。 由於鉸鏈及鏡板係在相同基底中（亦即，在相同層中 )，所以，當鏡圍繞鉸鏈的縱軸旋轉時，不會有平移移動 或位移。由於沒有平移位移，所以，鏡與支撐壁之間的間 隙僅受限於製造技術及製程。鏡板的緊密間隔及鉸鏈實質 上隱藏設於反射表面下方會允許微鏡陣列有高塡充比、增 進的對比、最小化光的散射及繞射，以及實際地消除通過 微鏡陣列而撞擊第二基底上的電路之光。 此外，由於在較佳實施例中鏡板及鉸鏈是由單晶矽材 料製成，所以，所造成的鉸鏈較強固且更可靠且實際上不 會苦於記憶效應、延著晶粒邊界斷裂或疲勞。單晶矽基底 比其它材料（特別是沈積的薄膜）具有顯著較少的微缺陷 及斷裂。結果，較不易延著裝置中的晶界斷裂（或增生微 斷裂）。而且，在本發明中使用單基底會最少使用多層薄 膜堆疊以及蝕刻製程及技術。 結果，所造成的S LM能夠取得高光學效率及性能以 W »地及成本上有效地產生高品質影像。 【實施方式】 反射式空間光調變器（SLM)IOO具有可偏轉鏡202的 陣列】0 3。藉由在該鏡2 〇 2與對應的電極]2 6之間施加偏 壓’可選擇性地偏轉個別鏡2 0 2。每一鏡2 0 2的偏轉會控 200528752 (5) 制從光源反射至視頻顯示器之光。如此，控制鏡2 02的偏 轉會允許撞擊該鏡2 02之光於選擇的方向上反射，並因而 允許控制視頻顯示器上像素之出現。 空間光調變器槪述 圖1係顯示根據本發明的一實施例之SLM 100的一 般架構。所示之實施例具有三層。第一層是鏡層〗〇 3，具 有多個可偏轉的微鏡2 0 2。在一較佳實施例中，微鏡陣列 】〇 3係由例如單晶矽之單一材料的第一基底1 〇5製造。 第二層是具有多個用於控制微鏡2 02之電極126之電 極陣列104。每一電極126與微鏡202是相關連的並控制 該微鏡2 02的偏轉。尋址電路允許選取單一電極1 2 6，該 電極126係用於控制與其相關連之特定微鏡2 02。 第三層是控制電路1 06的層。此控制電路1 06具有尋 址電路，允許控制電路1 0 6控制施加至被選取的電極I 2 6 之電壓。這會允許控制電路1 06經由電極1 26來控制鏡陣 歹0】03中的鏡202之偏轉。典型上，控制電路1 06也包含 顯示控制]〇 8、線記憶體緩衝器1 1 0、脈衝寬度調變陣列 ]1 2、及用於視頻訊號1 20及繪圖訊號1 22之輸入。在某 些實施例中，微控制器1 1 4、光控制電路I 1 6、及快閃記 憶體1 1 8可以是連接至控制電路]06之外部元件、或是包 含於控制電路1 〇6中。在不同的實施例中，上述所列之控 制電路]06的某些構件可以不存在、可以是在分別的基底 上及連接至控制電路]〇 6、或是其它增加的元件可以存在 -9- 200528752 (6) 以作爲控制電路1 06的部份或連接至控制電路I 06。 在一實施例中，在單一第二基底107上，使用半導體 製造技術，製造第二層104及第三層106。亦即，第二層 I 〇4無須分開且在第三層〗06上方。然而，「層」一詞係 有助於槪念化空間光調變器1 00的不同構件。舉例而言， 在一實施例中，電極126的第二層104係製於電制電路 106的第三層的頂部上，二者均製於單一第二基底107上 。亦即，在一實施例中，電極1 26、以及顯示控制1 〇8、 線記憶體緩衝器1 1 0、和脈衝寬度調變陣列1 1 2均製於單 —基底上。相較於顯示控制1 〇 8、線記憶體緩衝器11〇、 及脈衝寬度調變陣列1 1 2製於分別的基底上之傳統的液晶 顯示裝置，控制電路1 06的數個功能元件整合於相同基底 上會提供增進的資料傳送率之優點。此外，電極陣列104 的第二層及控制電路106的第三層製於單一基底107上會 提供簡單及便宜的製造、以及輕巧的最終產品之優點。 在製造層103及107之後，它們會被接合在一起以形 成SLM 100。具有鏡陣列103之第一層會遮蓋總稱1〇7之 第二及第三層〗〇4和106。在鏡陣列103中的鏡2 02之下 的層會決定第一層103之下有多少空間用於電極]26、及 尋址和控制電路1 〇 6。在鏡陣列1 〇 3中微鏡2 0 2之下具有 有限的空間以適用於電極】2 6和適用於形成顯示控制1 〇 8 、線記憶體緩衝器Π 0、及脈衝寬度調變陣列Π 2之電子 元件。本發明允許在鏡陣列1 〇 3的微鏡之下的有限區域中 將例如顯示控制1 〇 8、線記憶體緩衝器1 1 〇、及脈衝寬度 -10- 200528752 (7) 調變陣列1 1 2等更多項目整合於與電極】2 6相同的基底上 。在與電極1 2 6相同的基底1 0 7上包含此控制電路1 0 6 ’ 會增進S L Μ 1 0 0的性能。這允許在微鏡陣列1 0 3中的微 鏡之下有限的面積中，將例如顯示控制1 〇 8、線記憶體緩 衝器1 1 〇、及脈衝寬度調變陣列1 1 2等更多項目與電極 126集成於相同基底上。在與電極126相同的基底107上 包含此控制電路1 〇 6，會增進S L Μ 1 0 0的性能。在其它實 施例中，電極1 2 6與控制電路的元件之不同組合可以製於 不同基底上及電連接。 在其它實施例中，電極1 26與控制電路的元件之不同 組合可以製於不同基底上及電連接。 鏡： 圖2是單一微鏡202的一實施例之立體視圖，圖2b 是圖2a中所示的微鏡202之角落23 6的更詳細立體視圖 。在一較佳實施例中，微鏡202包含至少一鏡板204、鉸 鏈206、連接器216及反射表面203。在另一實施例中， 微鏡202又包含間隔器支撐框210，用於支撐鏡板、鉸鏈 206、反射表面203及216。較佳地，鏡板204、鉸鏈206 、連接器2 1 6及間隔器支撐框2 ] 0係由例如單晶矽等單一 材料的晶圚所製成。如此，在此實施例之圖1中所示的第 基底105係單晶砂晶圓。從單材料晶圓製造微鏡202會 大幅地簡化鏡2 02的製造。此外，單晶矽可以被拋光以產 &平滑鏡面’此平滑鏡面之表面粗糙度在量値等級上比沈 -11 - 200528752 (8) 積膜之表面粗糙度更加平滑。由單晶矽製成的鏡2 02在機 械上是堅硬的，防止不必要的鏡表面彎曲或捲曲，以及， 由單晶矽製成的鉸鏈較強固、更可靠並實質上未遭受微鏡 陣列中所使用的很多其它材料製成的鉸鏈所共有之記憶影 響、延著晶界之斷裂等不利影響。在其它實施例中，可以 使用其它材料取代單晶矽。一種可能是使用其它型式的矽 (例如多晶矽、或非晶矽）於微鏡2 02，或是，完全從金屬 (例如鋁合金、或鎢合金）中製造鏡202。而且，在本發 明中使用單晶可以避免使用多層薄膜堆疊及蝕刻製程和技 術。 如圖2a-b、3、4a-b、7a及8所示及如上所述，微鏡 202具有鏡板204。鏡板204是微鏡202的部份，其以連 接器216耦合至鉸鏈2 06及藉由施加偏壓於鏡202與對應 的電極]2 6之間而被選擇性地偏轉。圖3中所示的實施例 中之鏡板2〇4包含三角部份2(Ma4及204b。在圖9a、9b 及1 0所示的實施例中，鏡板2 04之形狀爲實質方形，且 對於225微米平方之近似面積，幾乎爲15微米乘15微米 ’但是，其它形狀及尺寸也是可能的。鏡板2 0 4具有上表 面205及下表面201。上表面205較佳地爲高度平滑表面 ’平均粗糙度小於2埃均方根且較佳地構成微鏡204的表 面積之大邰份。在鏡板204的上表面205上及在鉸鏈206 的部份上方，沈積例如鋁或任何其它高度反射材料之反射 表面2 0 3。較佳地此反射表面2 0 3具有3 0 0人或更小的厚 度。反射表面或材料2 0 3的薄度確保其繼承上表面2 0 5之 -12 - 200528752 (9) 平滑表面。此反射表面2 03的面積大於鏡板2 04的上表面 20 5的面積，以及，以鏡板204的偏轉所決定之角度，反 射來自光源的光。注意，扭力彈簧鉸鏈2 0 6實質上形成於 鏡板204的上表面之下，且實質上由沈積於上表面205上 及鉸鏈2 0 6部份之上方的反射表面2 0 3所隱蔽。圖2 a與 3之間的差異在於圖2a顯示之鏡板204具有加至上表面 205且實質上隱蔽鉸鏈206的反射表面203，而圖3顯示 之鏡板204未具有反射表面203，因此，露出鉸鏈2 06。 由於鉸鏈206及鏡板204是在相同基底105中，且如圖 〜及7b所示，鉸鏈206的中心高度796與鏡板204的中 心高度795或7 97實質上是共平面的，當鏡2 02繞著鉸鏈 2 〇 6的縱軸旋轉時，不會有平移移動或位移。由於沒有平 移位移，所以，鏡板204與間隔器支撐框2 1 0的支撐間隔 器壁之間的間隙僅受限於製造技術及製程，典型上小於 〇·1 °鏡板204的緊密間隔與鉸鏈206實質上隱蔽在反射 表面203之下，允許微鏡陣列1 〇3有高塡充比、增加的對 比、最小的光散射及繞射、以及實質消除通過微鏡陣列 1 〇 3的光撞擊在第二基底]〇 7上的電路。 如圖 2a-b、3、4a-b、7a、8、9a、9b 及 10，鏡板 204 藉由連接器216連接至扭力彈簧鉸鏈206。扭力彈簧鉸鏈 2 0 6連接至間隔器支撐框2丨〇，支撐框2 1 0將扭力彈簧 2 〇 6、連接器2 1 6、及鏡板2 0 4固持在原位。鉸鏈2 0 6包 含第一臂206a及第二臂206b。如圖3及10所不，每一 臂2 06a及2 06b的一端連接至間隔器支撐框210 ’而另一 - 13 - 200528752 (10) 端連接至連接器2 1 6。在另一實施例中，可以在鏡板204 、鉸鏈2 0 6、及間隔器支撐框2 0 1之間使用其它彈簧、鉸 鏈及連接設計。如同圖3及4a淸楚所示，扭力鉸鏈206 較佳地相對於間隔器支撐壁2 1 0而在對角線上定向（例如 ，45度角），並將鏡板204分成二部份、或複數側邊：第 一側2 0 4 a及第二側2 0 4 b。如圖7 b所示，二電極1 2 6與 鏡202、用於第一側204a的一電極126a及用於第二側 2 0 4b之一電極126b相關連。這允許側204a或204b附著 至電極126a或126b之一之下並向下樞轉以及提供寬廣範 圍的角度運動。當藉由施加電壓於鏡202與對應的電極 1 2 6之間以將例如靜電力等力量施加至鏡板2 0 4時，扭力 彈簧206允許鏡板204繞著鉸鏈2 06的縱軸、相對於間隔 器支撐框2 1 0旋轉。此旋轉產生角度偏轉以在選取方向上 將光反射。由於鉸鏈206及鏡板204是在相同基底1 05中 ，以及，如圖7a及7b所示，鉸鏈206的中心高度796與 鏡板204的中心高度79 5或7 9 7實質上共平面，所以，鏡 2 02會繞著鉸鏈206純旋轉地移動而無平移位移。在一實 施例中，如圖7a及8所示，扭力彈簧鉸鏈206具有寬度 222，寬度222小於鉸鏈206的深度22 3 (垂直於鏡板204 之上表面205)。鉸鏈206的寬度222較佳地在約0.]2微 米至約0 · 2微米之間，且深度較佳地在約〇. 2微米與約 〇 · 3微米之間。 如圖2 a · b、3、4 a - b、6、及7 a所示，間隔器支撐框 2 1 〇將鏡板2 〇4定位於電極]2 6及尋址電路上方的預定距 -14- 200528752 (11) 離處，以致於鏡板2 04可以向下偏轉至預定角度。如圖 2a、4a、9a及10所示，間隔器支撐框210包含間隔器支 撐壁，間隔器支撐壁較佳地由相同的第一基底1 0 5形成並 較佳地正交定位。這些壁有助於界定間隔器支撐框2 1 0的 高度。根據鏡板2 04與電極1 2 6之間所需的分離、以及電 極的拓蹼設計，選擇間隔器支撐框2 1 0的高度。較大的高 度允許鏡板2 0 4更多偏轉、以及更高的偏轉角度。較大的 偏轉角度通常提供更高的對比。在一實施例中，鏡板2〇4 的偏轉角度是1 2度。在較佳實施例中，假使被供予足夠 的間隔及驅動電壓時，鏡板2 04可以旋轉多達9 0度。間 隔器支撐框2 1 0也提供支撐給鉸鏈2 0 6以及使鏡板2 0 4與 鏡陣列]03中其它的鏡板2 04相間隔。，間隔器支撐框21 0 具有間隔器壁寬度2 1 2，當加上鏡板204與支撐框2 1 0之 間的間隙時，間隔器壁寬度2 1 2實質上等於相鄰的微鏡 2 02之間的距離。在一實施例中，間隔器壁寬度2】2是I 微米或更少。在一較佳實施例中，間隔器壁寬度2 1 2是 0.5微米或更少。這會將這些鏡板204設置成緊密地在一 起以增加鏡陣列1 0 3的塡充比。 在某些實施例中，微鏡202包含元件405a或405b， 當鏡板204向下偏轉至預定角度時，元件4〇5a或4〇5b會 停止鏡板2〇4的偏轉。典型上，這些元件包含止動件 405a或405b以及著陸尖端71〇a或7]〇b。如圖4a、6、7a 、8、10及]2所示，當鏡表面2 04偏轉時，在鏡板204 上的止動件4 0 5 a或4 0 5 b會接觸著陸尖端7 ] 〇 ( 7 I 〇 3或 200528752 (12) 7 1〇b)。當此發生時，鏡板2 04不會進一步偏轉。止動件 4 0 5 a或4 0 5 b及著陸尖端710a或710b有數種可能的配置 。在圖4a、6、7a、8、10及12中所示的實施例中，止動 件是圓柱或機械止動件4 0 5 a或4 0 5 b，附著至鏡板2 0 4的 下表面201，著陸尖端710是第二基底107上對應的圓形 區。在圖7a、7b及8中所示的實施例中，著陸尖端71 0a 及7 1 Ob電連接至間隔器支撐框2 ] 0，因此，相對於止動 件405 a或405b，具有零電壓差，以防止止動件4 05 a或 4〇5b分別黏著或熔接至著陸尖端7] 0a或71 Ob。如此，當 鏡板2 04相對於間隔器支撐框2 1 0旋轉至預定角度（由機 械止動件405 a或40 5 b的長度及位置所決定）之外時，機 械止動件405a或4 05 b將與著陸尖端710a或710b分別進 入實體接觸，以及，防止鏡板2 0 4任何進一步的旋轉。 在較佳實施例中，止動件405a或405b由第一基底 105及由與鏡板2 04、鉸鏈206、連接器2] 6和間隔器支 撐框2 1 0相同的材料所製成。著陸尖端7〗〇 a或7 1 0 b也較 佳地由同於止動件4〇5a或405 b、鉸鏈206、連接器216 及間隔器支撐框2 1 0的材料所製成。在材料單晶矽之實施 例中，止動件4 〇 5 a或4 〇 5 b以及著陸尖端7 1 0 a或7 1 0 b因 而由具有長的作用壽命之硬材料製成，這允許鏡陣列1 〇 3 維持長時間。此外’由於卓晶砂是硬材料，所以，止動件 405a或405b及著陸尖端7]0a或7]0b可以由小面積製成 ，在此小面積中，止動件4 5 0 a或4 5 〇 b會分別接觸著陸尖 端7] 0a或7] 〇b，大幅降低黏著力及允許鏡板2 04自由地 -16- 200528752 (13) 偏轉。而且，此意指止動件 405 a或4 05 b及著 7 10a或710b維持在相同的電位，防止因止動件 40 5 b及著陸尖端或710b在不同電位時經由 電荷注入處理而發生之黏著。本發明不限於停止上 204的偏轉之元件或技術。可以使用此技藝中所習 何元件及技術。 圖4a是立體視圖，顯示單一微鏡202之下側 支撐壁210、鏡板204(包含側204a和204b並具有 2 0 5和下表面201)、鉸鏈2 06、連接器216和機械 405a和405b。圖4b係圖4a中所示的微鏡之角落 更詳細立體視圖。 圖5是立體視圖，顯示具有微鏡202-1至202 鏡陣列1 0 3的頂部及側邊。雖然圖5顯示具有三列 的微鏡陣列1 〇3，總共九個微鏡202，但是，其它 微鏡陣列1〇3也是可能的。典型上，每一微鏡202 視頻顯示器上的像素。如此，具有更多微鏡2 02的 列會提供具有更多像素的視頻顯示器。 如圖5所示，微鏡陣列1 03的表面具有大的塡 亦即，微鏡陣列1 〇 3的表面的大部份是由微鏡2 0 2 表面2 03製成。微鏡陣列103的表面之非常小的部 反射的。如圖5所示，微鏡陣列]03的表面之非反 是在微鏡2 02的反射表面2 0 3之間的區域。舉例而 2 0 2 -1與2 02 - 2之間的區域的寬度是由間隔器支撐 212與微鏡2 02 -1與2 02 -2的鏡板2 04與間隔器 陸尖端 4 0 5 a 或 I熔接或 .述鏡板 1知的任 丨，包含 上表面 ,止動件 2 3 7的 •9之微 和三行 尺寸的 對應於 較大陣 充比。 的反射 份是非 射部份 言，鏡 壁寬度 支撐壁 -17- 200528752 (14) 2 1 0之間的間隙之寬度總合所決定。注意，雖然單一鏡 202如圖2a、2b、3、4a及4b所不般被描述成具有其自 己的間隔器支撐框2 1 0，但是，典型上，在例如鏡2 0 2 - 1 與2 02-2等鏡與鏡之間，未具有二分開的鄰接間隔器壁 21 〇。然而，在鏡2 02-]與20-2之間典型上會有支撐框 2 I 0的一實體間隔器壁。由於在偏轉鏡板2 0 4時無平移位 移，所以，間隙與間隔器壁寬度2 1 2可以製成與製造技術 所支援的特徵尺寸一般小。因此，在一實施例中，間隙是 0.2微米，在另一實施例中，間隙是〇 · 1 3微米或更小。由 於半導體製造技術允許更小的尺寸，所以，間隔器壁2 1 0 與間隙的尺寸可以降低而允許更高的塡充比。本發明的實 施例允許高塡充比。在較佳實施例中，塡充比是96%或更 局。 圖6是立體視圖，顯示具有九個微鏡的微鏡陣列1 03 之底部及側邊。如圖6所示，微鏡2 02的間隔器支撐框 2 1 0的支撐壁界定鏡板2 04之下的穴。這些穴提供空間給 鏡板204以向下偏轉，也允許鏡板204之下的大區域用於 配置具有電極126之第二層104，以及/或用於具有控制 電路106的第三層。圖6也顯示鏡板204(包含側邊204a 和 2 04b)的下表面 20】，及間隔器支撐框21 0、扭力彈簧 錢鏈206、連接器216、及止動件405a和405b的底部。 如圖5及6所示，正交於鏡板2 04之非常少的光可以 通過微鏡陣列]03之外而到達微鏡陣列1 03下方的任何電 極或控制電路】〇 6。這是因爲間隔器支撐框2 ] 〇及鏡板 -18- 200528752 (15) 204的上表面2 0 5上和鉸鏈2 0 6的部份之上方的反射表面 203幾乎完全遮蓋微鏡陣列103之下的電路。而且，由於 間隔器支撐框2 1 0會將鏡板2 04與微鏡陣列1 03之下的電 路分開，所以，以非垂直角度行進至鏡板2 04並通至鏡板 2 04之外的光容易撞擊間隔器支撐框2 ] 0的壁以及不會到 達微鏡陣列1 03之下。由於入射於微鏡陣列1 03之少量強 光會到達電路，所以，SLM 100可以避免與強光撞擊電路 有關的問題。這些問題包含入射光將電路加熱，以及入射 光子使電路元件充電，這二者均會造成電路固障。 圖9 a係根據本發明的另一實施例之微鏡2 0 2的立體 視圖，圖9b係微鏡202的角落23 8之更詳細的立體視圖 。本實施例中的扭力鉸鏈2 0 6與間隔器支撐框2 1 0的間隔 器支撐壁平行。在鏡板2 04與對應的電極I 2 6之間施加偏 壓，會使鏡板2 04選擇性地朝向電極偏轉。圖9 a中所示 的實施例比具有對角鉸鏈2 0 6之圖2 a和2 b中所示的鏡 202提供更小的角運動的總範圍，此範圍係始於相同支撐 壁高度。然而，如同圖2 a及2 b中所示的實施例般，圖 9a和9b中所示的實施例中的鉸鏈206是在鏡板2 04的上 表面下方且由反射表面203隱蔽，造成具有高塡充比、高 光學效率、高對比、低的光繞射和散射以及可靠和成本上 有效的性能之SLM 100。圖9b是微鏡2 02的角落之更詳 細的立體視圖，並顯示鏡板2 0 4、鉸鏈2 0 6、間隔器支撐 框2 1 〇的支撐壁及反射表面2 0 3。圖1 0顯示單一微鏡2 0 2 的下側’其包含鉸鏈2 06、連接器2]6及止動件4 05 a。在 -19- 200528752 (16) 其它實施例中，鉸鏈2 0 6可以實質上平行於鏡板2〇4的一 側以及仍然設置成將鏡板204分成二部份405a和405b。 圖Π和12提供如圖9a、9b和1 0中所示的多個微鏡202 所組成之微鏡陣列的立體視圖。 圖1 3係形成於第二基底1 〇 7上的電極1 2 6之一實施 例的立體視圖。在本實施例中，每一微鏡2 02具有對應的 電極1 2 6。在此所示的實施例中，電極1 2 6係被製成高於 第二基底上的電路的其它部份。在較佳實施例中，電極 1 2 6設於與第二基底上的電路之其它部份相同水平。在另 一實施例中，電極1 2 6延伸至電路上方。在本發明的一實 施例中，電極1 26係配接於微鏡板之下的個別鋁墊。電極 的形狀係取決於微鏡202的實施例。舉例而言，在圖2a 、2 b及3所示的實施例中，較佳地有二電極1 2 6在鏡2 0 2 之下，每一電極126具有如圖7b所示之三角形。在圖9a 、9b及1 0中所示的實施例中，較佳地有單一的、方形的 電極126在鏡2 02之下。這些電極126係製於第二基底 1 〇 7的表面上。在本實施例中電極1 2 6的大表面積會造成 下拉鏡板2 04至機械止動所需之相當低的尋址電壓，因而 造成微板2 04之全預角偏轉。 選項： 在操作上，個別反射式微鏡202會被選擇性地偏轉並 用以在空間上調變入射至鏡2 02及由其反射之光。 圖7 a及8係顯示延著圖2 a中的虛線2 5 0所示之微鏡 -20- 200528752 (17) 2 02的剖面視圖。注意，此剖面視圖係偏移微 心對角線，藉以顯示鉸鏈2 0 6的輪廓。圖7 c 圖2a中的虛線25 0所示的微鏡2 02之不同舀 意，此剖面視圖是延著中心對角線，垂直於較 7a、7c及8是顯示電極126上方的微鏡20 2。 將電壓施加至微鏡202的一側上之電極126 1 2 6上方的鏡板2 0 4之對應部份的偏轉（圖8 ή 。如圖8所示，當電壓施加至電極1 2 6時，g 一半會附著至電極126，而鏡板204b的另一 2 04的結構及剛性而被移離電極1 2 6及第二基 會造成鏡板204圍繞扭力彈簧鉸鏈206旋轉。 電極126時，如圖7a所示，錢鏈2 06會造成 回至其未經偏移的位置。或者，在具有如圖2 所示的對角鉸鏈206之實施例中，電壓可以 2 04的另一側上的電極126，以使鏡202在相 轉。如此，撞擊鏡2 02之光會在藉由施加電壓 而受控之方向上反射。 一實施例如下述般操作。起先，鏡202如 般未經偏轉。在此未偏移的狀態下，自光源歪 SLM 100之入射光會由平面鏡202反射。外離 射的光會由例如光泵所接收。從未經偏轉的鏡 光不會被反射至視頻顯示器。 當電壓偏壓施加於鏡板2 04 a的半部與其 ]26之間時，鏡202會因靜電吸引而偏轉。在 鏡2 0 2的中 係顯示延著 面視圖。注 鏈 206 。圖 在操作上， 以控制電極 3的側2 0 4 a) |板204a的 半會因鏡板 底107 。這 當電壓移離 鏡板204彈 a、2 b 及 3 施加至鏡板 反方向上偏 至電極126 圖7 a及7 c 斜地入射至 的、經過反 2 0 2反射的 下方的電極 一實施例中 -21 - 200528752 (18) ’當鏡板2 0 4 a如圖8所示般向下偏轉時，v e!較佳地爲 1 2伏特，V b爲-1 〇伏特，及v e2爲〇伏特。同樣地（或相 反地），當微板2 〇 4 b向下偏轉時，v e】較佳地爲〇伏特， Vb爲-10伏特，及Vd爲12伏特。由於鉸鏈2〇6的設計 ’鏡板2 04a或204b的一側（亦即，位於具有偏壓的電極 1 2 6的上方的側）會向下偏轉（朝向第二基底丨〇 7 )，而鏡板 2 0 4 b或2 0 4 a的另一側會移離第二基底7。注意，在一 較佳實施例中，實質上所有的彎曲發生於鉸鏈2 〇 6中的而 非鏡板2 04。在一實施例中，藉由使鉸鏈寬度222薄，以 及連接鉸鏈2 0 6至僅位於二端上的支撐柱，而達成此點。 如上所述般，鏡板2 0 4的偏轉受限於止動件4 〇 5 &或4 0 5 b .。鏡板2 0 4的全部偏轉會使外離的反射光偏轉至成像光件 及視頻顯示器。 當鏡板2 0 4偏轉通過「快動」或「下拉」電壓時（在 一實施例中幾乎爲1 2伏特或更低），鉸鏈2 0 6之恢復的機 械力或扭力無法再平衡靜電力或扭力，且在其下具有靜電 力之鏡板204的一半204a或204b會快速朝向其下的電極 1 2 6以取得完全偏轉，於所需時僅受限於止動件4 〇 5 a或 4 0 5 b。在如圖9 a、9 b及1 0所示之鉸鏈2 0 6平行於間隔器 支撐框2 1 0的支撐壁之實施例中，爲了將鏡板204從其完 全偏轉的位置釋放，電壓必須被關閉。在如圖2a、2b及 3所示之鉸鏈爲對角線的實施例中，爲了將鏡板2 04從其 完全偏轉的位置釋放，當其它電極正被致能時電壓必須被 關閉，且鏡2 02附著至另一側。 -22- 200528752 (19) 微鏡2 0 2是機電雙穩態裝置。在釋放電壓與快動 之間給予特定電壓，則取決於鏡2 0 2偏轉的歷史’ 2 0 4會有二種可能的偏轉角度。因此，鏡2 0 2偏轉表 同佇鎖。由於鏡202的偏轉所需之機械力相對於偏轉 大致上爲線性的，所以’這些雙穩定性及彳宁鎖特性會 ，而相反的靜電力是與鏡板2 0 4與電極1 2 6之間的距 反比。 由於鏡板2 0 4與電極1 2 6之間的靜電力取決於 2 04與電極1 2 6之間的總電壓差，所以’施加至鏡板 的負電壓會降低施加至電極1 2 6所需之正電壓而取得 的偏轉量。如此，施加電壓至鏡陣列1 〇 3可以降低 1 26的電壓量値需求。此點是有的，舉例而言，在某 用中，因爲5V的切換能力在半導體工業中是更加通 成本上更有效，所以，需要使必須施加至電極]2 6的 電壓保持在12V以下。 由於鏡202的最大偏轉是固定的，所以，假使 1 〇〇以超過快動電壓之電壓操作，則其可以以數位方 作。在如圖2 a、2 b及3所示之鉸鏈平行於間隔器支 210的支撐壁之實施例中，由於鏡板2 04會因電壓施 相關連的電極1 2 6而完全向下偏轉，或是無電壓施加 關連的電極1 2 6時，允許鏡板2 0 4向上彈，所以，操 本上是數位的。在具有如圖1 2 a、] 2 b及1 3所示的 2 0 6對角線之實施例中，當使鏡板2 0 4的另一側上的 電極1 2 6致能時，鏡板2 0 4會因電壓施加至鏡板2 0 4 電壓 鏡板 現如 角度 存在 離成 鏡板 204 給定 電極 些應 用且 最大 SLM 式操 撐框 加至 至相 作基 鉸鏈 其它 的一 -23- 200528752 (20) 側上之相關連的電極1 26而完全向下偏轉至鏡板204的另 一側。造成鏡板2 04完全向下偏轉直到由停止鏡板2 0 4的 偏轉之實體元件停止爲止的電壓係稱爲「快動」或「下拉 」電壓。如此，爲了使鏡板2 0 4完全向下偏轉，將等於或 大於快動電壓之電壓施加至對應的電極I 2 6。在視頻顯示 應用中，當鏡板2 0 4完全向下偏轉時，入射於鏡板2 〇 4上 的入射光會被反射至視頻顯示螢幕上對應的像素，且像素 會呈現明亮的。當鏡板2 04被允許向上彈時，光會以不會 撞擊視頻顯不螢幕之方式被偏轉，且像素呈現暗的。 在此數位操作期間，在相關連的鏡板204被完全偏轉 之後’無須在電極1 26上保持完全快動電壓。在「尋址階 段」期間，用在對應於應被完全偏轉的鏡板2 〇 4之被選取 電極1 2 6的電壓會被設定於偏轉鏡板2 〇 4所需的位準。在 所討論之鏡板2 0 4因電極〗2 6上的電壓而被偏轉時，用以 將鏡板2 0 4固持於偏轉位置所需之電壓會小於真正偏轉所 需的電壓。這是因爲被偏轉的鏡板204與尋、址電極126之 間的間隙比鏡板204在被偏轉的過程中時還小。因此，在 尋址階段之後的.「固持階段」中，施加至所選取的電極 1 2 6之壓會從其原先所需的位準縮減，卻不會實質地影 響鏡板2 (Η的偏轉狀態。具有較低固持階段之一優點係附 近的未被偏轉之鏡板2 〇4會遭受較小的靜電吸力，且它們 因而保持較接近零偏轉位置。這會改進偏轉鏡板2〇4與未 偏轉的鏡板2 0 4之間的光學對比。 藉由適當選取尺寸（在一實施例中，鏡板2〇4與電極 - 24 - 200528752 (21) 1 2 6之間的支撐框2 1 0分離取決於鏡結構及偏轉角度需求 而爲1至5微米，且鉸鏈206厚度爲〇.〇5至0.45微米）及 材料（例如單晶矽（100))，可以將反射式SLM 100製成操 作電壓僅爲數伏特。由單晶矽製成的扭力彈簧2 0 6的剪力 模數可以爲5 X 101()牛頓/半徑平方米。將鏡板204維持 在適當電壓（負偏壓）而非接地，可以使電極126操作以完 全偏轉相關連的鏡板2 0 4之電壓更低。對於施加至電極 1 2 6之給定電壓，這會造成更大的偏轉角度。最大的負偏 壓是釋放電壓，所以，當尋址電壓降至零時，鏡板204可 以快動回至未偏轉的位置。 也能夠以更「類比」的方式，控制鏡板2 04的偏轉。 施加小於「快動電壓」之電壓以將鏡板2 0 4偏轉以及控制 入射光被反射的方向。 其它應用 除了視頻顯示器之外，空間光調變器1 〇〇在其它應用 中也是有用的。一種此應用是無掩罩微影術，其中，空間 光調變器]00會導引光以使所沈積的光阻顯影。這將不需 掩罩而能以所需圖案使光阻正確地顯影。 雖然已參考多個實施例’特別地顯示及說明本發明， 但是，習於相關技藝者應瞭解’在不悖離本發明的精神及 範圍之下，可以在形式上及細節上作不同的改變。舉例而 言，鏡板2 04可以藉由靜電吸引以外的其它方法而偏轉。 替代地，可以使用磁、熱或壓電致動以偏轉鏡板2 0 4。 -25- 200528752 (22) 【圖式簡單說明】 圖1係說明根據本發明的一實施例之空間光調變器的 一般架構。 圖2 a係本發明的一實施例中單一微鏡之立體視圖。 圖2 b係圖2 a的微鏡之角落的立體視圖。 圖3係無反射表面之單一微鏡的立體視圖，顯示本發 明一實施例中的微鏡陣列的鏡板之頂部及側邊。 圖4 a係本發明的一實施例中單一微鏡的底部及側邊 〇 圖4b係圖4a的微鏡之角落的立體視圖。 圖5係立體視圖，顯示本發明的一實施例中微鏡的頂 部及側邊.。 圖6係立體視圖，顯示本發明的一實施例中微鏡陣列 的底部及側邊。 圖7 a係延著偏移對角剖面之圖2 a中所示的未經偏轉 的微鏡之剖面視圖。 圖7 b係在本發明的一實施例中形成於第二基底中的 鏡板下方的電極及著陸尖梢。 圖7 c係延著中心對角線剖面之圖2 a中所示的未經偏 轉的微鏡之剖面視圖。 圖8係顯示於圖2 a中的偏轉的微鏡之剖面視圖。 圖9 a係微鏡的另一實施例中的頂部及側邊的立體視 圖。 圖9b係圖9a的微鏡之角落之立體視圖。 -26- 200528752 (23) 圖]〇係立體視圖，顯示微鏡的另一實施例的底部及 側邊。 圖1 1係立體視圖，顯示微鏡陣列的另一實施例之頂 部及側邊。 圖1 2係立體視圖，顯示微鏡陣列的另一實施例之底 部及側邊。 圖1 3係立體視圖，顯示形成於第二基底上的電極之 一實施 主要元 件對 昭 表 100 空 間 光 調 變 器 103 可 偏 轉 的 鏡 陣 列 1 04 電 極 陣 列 105 第 一 基 底 106 控 制 電 路 107 第 二 基 底 108 顯 示 控 制 110 線 記 憶 體 緩 衝 器 112 脈 衝 調 變 陣 列 1 14 微 控 制 器 116 光 控 制 電 路 1 1 8 快 閃 記 億 體 1 20 視 頻 訊 號 1 22 繪 圖 訊 號 -27- 200528752 (24) 126 電極 126a 電極 126b 電極 20 1 下表面 202 微鏡 202-1 ^ -202-9 微鏡 203 反射表面 204 鏡板 204a 第一側 204b 第二側 205 上表面 206 鉸鏈 2 0 6 a 第一臂 ' 20 6b 第二臂 2 10 間隔器支撐框 2 1 2 間隔器壁寬度 2 16 連接器 222 寬度 223 深度 23 6 角落 23 7 角落 23 8 角落 4 0 5 a 止動件 4 0 5 b 止動件 200528752 (25) 710a 著陸尖端 710b 著陸尖端 >29-200528752 (1) 发明 Description of the invention [Technical field to which the invention belongs] The present invention relates to a spatial light modulator (SLM), and more particularly, to having a hidden hinge to maximize the pixel filling ratio, to minimize scattering and diffusion, and Micromirror structure for high contrast and high image quality. [Prior Art] In the fields of optical information processing, projection displays, image and graphic monitors, televisions, and electrophotographic printing, spatial light modulators have different applications. Reflective S LM is a spatial pattern that modulates incident light to reflect an image corresponding to electrical or light input. Incident light can be modulated in phase, intensity, or inversion of polarization. Reflective SLM typically contains an area or two-dimensional array of addressable pixels (pixels) capable of reflecting incident light. The key parameters of SL Μ, especially in display applications, are the optically active area to the pixel area (also a part of the SLM surface area that reflects the entire surface area of the SLM as a measurement, also known as battery charging) ratio). Requires a high charge-to-charge ratio. The conventional SLM has different disadvantages. These disadvantages include, but are not limited to: (1) lower than the optimal optical active area, reducing optical efficiency; (2) rough reflective surface, reducing the reflectivity of the mirror; (3) diffraction and scattering, reducing the contrast of the display; ( 4) the materials used have long-term reliability issues; and (5) complex processes, which increase overhead and reduce device capacity. Many conventional devices include substantially non-reflective areas on their surface. This provides a low fill ratio and provides sub-optimal reflection efficiency. For example, 'US Patent No. 4,2 2 9,7 3 2 discloses -5-200528752 (2) MOSFET device and mirror formed on the surface of the device. These MOSFET devices occupy a low surface area, typically a portion of the device area where optics are active and reduce reflectance. A MOSFET device on the surface of the device will also cause incident light and reduce the contrast of the display. In addition, the strong light that strikes the exposed MOSFET can cause the MOSFET device to charge and overheat the circuit, thus proper operation of the device. Some S LM designs have a rough surface that scatters incident light and reduces reflectivity. For example, in some SLM designs, an aluminum film with a reflective surface on the LPCVD silicon nitride layer. Since these mirrors are deposited with a thin film, their smoothness is difficult to control. The resulting product therefore has a rough surface and reduces the reflection efficiency. Another issue with reflection efficiency due to some SLM designs (especially some pendant mirror designs) is the large exposed hinge surface area. These exposed hinge surface areas will cause scattering and diffusion due to the hinge structure, which is not good for comparison compared to the parameters. Many conventional SLMs, such as the SLM shown in U.S. Patent Nos. 4,5 6; 9 35, have hinges made of aluminum alloy. Aluminium and others are susceptible to fatigue and plastic deformation, leading to long-term reliability issues. Moreover, aluminum is susceptible to cellular "memory", of which the remaining positions are tilted towards the positions they are most often occupied. In addition, the mirrors disclosed in 4, 5 6 6 5 9 3 5 are released by removing sacrificial material beneath the mirror surface. The technology usually causes the delicate micromirror structure to break during release. It also has a large gap between the mirrors so that the etchant removes the sacrificial material beneath the mirrors and parts of the low optically active device area. The effect of the interference reduction of the diffraction device is the rough surface, which minimizes the exposure of other exposed metal problems. Start to profit. (3) Other traditional S LMs require multiple layers, including separate hinges, electrodes, and / or control circuits for the mirror. For example, the need for multilayer S LM, etc. requires the use of multiple thin film stacks as well as etching techniques and processes. Using this technique and process is expensive and results in low productivity. For example, the use of these usually involves large-scale deposition and removal of sacrificial material beneath the mirror plate. The deposition and stacking of multiple layers of thin film beneath the surface of the plate can result in rougher mirrors by reducing the reflection efficiency of the mirrors. In addition, since there are mirrors and hinges on different layers or bases, translational shifts are caused when the mirrors are deflected. Due to the offset, the mirrors in the array must be spaced to avoid interference between adjacent mirrors. Since the mirrors in the array cannot be set too close to the mirrors in the array, SLM suffers from being lower or more adequate than the optimal optical active area. There is a need for an SLM with improved reflection efficiency, long-term reliability of the SLM device, and a simplified process. SUMMARY OF THE INVENTION The present invention is a spatial light modulator (SLM). In one embodiment, there is a reflective selectively deflectable micromirror array made of a first substrate, and the substrate is bonded to a second substrate having individually addressable electrodes. The second also has an addressing and control circuit for the micromirror array. Alternatively, the addressing and circuit parts are on separate substrates and connected to the and electrodes on the second substrate. The micromirror array includes a deflection mirror plate provided with a highly reflective surface to reflect A-ray light. This first substrate is a single-material wafer, with a few layers of manufacturing technology. Mirror surface, bottom-to-bottom translation mechanism, other low-degree, SLM first substrate control circuit, controllable implementation -7- 200528752 (4) In the example, it is monocrystalline silicon. The spacer support wall provides separation between the mirror plate and an electrode, which is associated with the mirror plate 'to control the deflection of the mirror plate. An electrode is disposed on the second substrate, and the second substrate is bonded to the micromirror array. Since the hinge and the mirror plate are in the same base (ie, in the same layer), there is no translational movement or displacement when the mirror rotates about the longitudinal axis of the hinge. Because there is no translational displacement, the gap between the mirror and the support wall is limited only by manufacturing techniques and processes. The tight spacing of the mirror plates and the hinges that are substantially hidden under the reflective surface will allow the micromirror array to have a high fill ratio, enhanced contrast, minimize light scattering and diffraction, and practically eliminate the impact of passing through the micromirror array to the second Light from the circuit on the substrate. In addition, since the mirror plate and the hinge are made of single crystal silicon material in the preferred embodiment, the resulting hinge is stronger and more reliable, and does not actually suffer from memory effects, fractures or fatigue along the grain boundary. Single crystal silicon substrates have significantly fewer micro-defects and fractures than other materials, especially deposited films. As a result, it is less likely to propagate grain boundary fractures (or proliferative microfractures) in the device. Moreover, the use of a single substrate in the present invention will minimize the use of multilayer film stacking and etching processes and techniques. As a result, the resulting S LM can achieve high optical efficiency and performance to efficiently produce high-quality images W and ground. [Embodiment] A reflective spatial light modulator (SLM) 100 has an array of deflectable mirrors 202]. Individual mirrors 202 can be selectively deflected by applying a biasing voltage 'between the mirror 202 and the corresponding electrode] 26. The deflection of each mirror 202 will control the light reflected from the light source to the video display. In this way, the deflection of the control mirror 202 will allow the light hitting the mirror 202 to be reflected in the selected direction, and thus to control the appearance of pixels on the video display. Description of Spatial Light Modulator FIG. 1 shows a general architecture of an SLM 100 according to an embodiment of the present invention. The illustrated embodiment has three layers. The first layer is a mirror layer [03], which has a plurality of deflectable micromirrors 202. In a preferred embodiment, the micromirror array is fabricated from a first substrate 105 of a single material such as single crystal silicon. The second layer is an electrode array 104 having a plurality of electrodes 126 for controlling the micromirror 202. Each electrode 126 is associated with the micromirror 202 and controls the deflection of the micromirror 202. The addressing circuit allows the selection of a single electrode 126, which is used to control a particular micromirror 202 associated with it. The third layer is the layer of the control circuit 106. This control circuit 106 has an addressing circuit which allows the control circuit 106 to control the voltage applied to the selected electrode I 2 6. This would allow the control circuit 106 to control the deflection of the mirror 202 in the mirror array 歹 0] 03 via the electrodes 126. Typically, the control circuit 106 also includes display control] 8. Line memory buffer 110, pulse width modulation array 1.12, and inputs for video signal 120 and drawing signal 122. In some embodiments, the microcontroller 1 1 4, the light control circuit I 1 6, and the flash memory 1 1 8 may be external components connected to the control circuit] 06 or included in the control circuit 1 〇 6 in. In different embodiments, certain components of the control circuit listed above may be absent, may be on separate substrates and connected to the control circuit], or other additional components may be present. 200528752 (6) as part of control circuit 1 06 or connected to control circuit I 06. In one embodiment, the second layer 104 and the third layer 106 are fabricated on a single second substrate 107 using semiconductor manufacturing technology. That is, the second layer 104 does not need to be separated and is above the third layer 06. However, the term "layer" helps to memorize the different components of the spatial light modulator 100. For example, in one embodiment, the second layer 104 of the electrode 126 is fabricated on top of the third layer of the electrical circuit 106, and both are fabricated on a single second substrate 107. That is, in one embodiment, the electrodes 126, the display control 108, the line memory buffer 110, and the pulse width modulation array 1 12 are all fabricated on a single substrate. Compared with the conventional liquid crystal display device which is composed of display control 108, line memory buffer 11 and pulse width modulation array 1 12 on separate substrates, several functional elements of control circuit 106 are integrated in The same substrate offers the advantage of increased data transfer rates. In addition, the second layer of the electrode array 104 and the third layer of the control circuit 106 are fabricated on a single substrate 107 to provide the advantages of simple and inexpensive manufacturing, and a lightweight final product. After manufacturing layers 103 and 107, they are joined together to form SLM 100. The first layer with the mirror array 103 will cover the second and third layers collectively referred to as 107 and 04 and 106. The layer below the mirror 202 in the mirror array 103 determines how much space is used for the electrodes under the first layer 103], and the addressing and control circuit 106. There is limited space under the micromirror 2 0 2 in the mirror array 1 0 3 to be suitable for the electrode] 2 6 and suitable for forming the display control 10 8, the line memory buffer Π 0, and the pulse width modulation array Π 2 of the electronic components. The present invention allows, for example, display control 1 08, line memory buffer 1 1 0, and pulse width -10- 200528752 (7) modulation array 1 1 in a limited area under the micromirror of the mirror array 1 03. 2 and more items are integrated on the same substrate as the electrode] 2 6. The inclusion of this control circuit 10 6 ′ on the same substrate 10 7 as the electrode 1 2 6 will improve the performance of the SL M 100. This allows for more items such as display control 108, line memory buffer 1 1 0, and pulse width modulation array 1 1 2 in the limited area under the micromirror in the micro mirror array 103. Integrated with the electrode 126 on the same substrate. The inclusion of this control circuit 106 on the same substrate 107 as the electrode 126 improves the performance of the SL 100. In other embodiments, different combinations of the electrodes 126 and the components of the control circuit can be made on different substrates and electrically connected. In other embodiments, different combinations of the electrodes 126 and the components of the control circuit can be made on different substrates and electrically connected. Mirror: FIG. 2 is a perspective view of an embodiment of a single micromirror 202, and FIG. 2b is a more detailed perspective view of the corner 23 6 of the micromirror 202 shown in FIG. 2a. In a preferred embodiment, the micromirror 202 includes at least one mirror plate 204, a hinge 206, a connector 216, and a reflective surface 203. In another embodiment, the micromirror 202 further includes a spacer support frame 210 for supporting the mirror plate, the hinge 206, and the reflective surfaces 203 and 216. Preferably, the mirror plate 204, the hinge 206, the connector 2 16 and the spacer support frame 2] 0 are made of a single material such as monocrystalline silicon. As such, the first substrate 105 shown in FIG. 1 of this embodiment is a single crystal sand wafer. Manufacturing the micromirror 202 from a single material wafer will greatly simplify the fabrication of the mirror 202. In addition, single crystal silicon can be polished to produce & smooth mirror surface. The surface roughness of the smooth mirror surface is smoother than that of Shen -11-200528752 (8). The mirror 2 02 made of single crystal silicon is mechanically rigid to prevent unnecessary bending or curling of the mirror surface, and the hinge made of single crystal silicon is stronger, more reliable, and substantially free of micromirror arrays Many other materials used in the hinges have common memory effects, fractures extending along grain boundaries, and other adverse effects. In other embodiments, other materials may be used instead of single crystal silicon. One possibility is to use other types of silicon (such as polycrystalline silicon or amorphous silicon) for the micromirror 202, or to make the mirror 202 entirely from a metal (such as an aluminum alloy or a tungsten alloy). Moreover, the use of single crystals in the present invention can avoid the use of multiple thin film stacking and etching processes and techniques. As shown in FIGS. 2a-b, 3, 4a-b, 7a, and 8 and as described above, the micromirror 202 has a mirror plate 204. The mirror plate 204 is part of the micromirror 202, which is coupled to the hinge 206 with a connector 216 and is selectively deflected by applying a bias voltage between the mirror 202 and the corresponding electrode] 26. The mirror plate 204 in the embodiment shown in FIG. 3 includes triangular portions 2 (Ma4 and 204b. In the embodiment shown in FIGS. 9a, 9b, and 10, the shape of the mirror plate 204 is substantially square, and for the The approximate area of 225 microns square is almost 15 microns by 15 microns '. However, other shapes and sizes are possible. The mirror plate 204 has an upper surface 205 and a lower surface 201. The upper surface 205 is preferably a highly smooth surface' average The roughness is less than 2 angstroms and preferably constitutes a large portion of the surface area of the micromirror 204. On the upper surface 205 of the mirror plate 204 and above the portion of the hinge 206, for example, aluminum or any other highly reflective material is deposited. Reflective surface 2 0. Preferably this reflective surface 2 0 3 has a thickness of 3 0 0 people or less. The thinness of the reflective surface or material 2 3 ensures that it inherits the upper surface 2 5-5-200528752 ( 9) Smooth surface. The area of this reflective surface 203 is larger than the area of the upper surface 20 5 of the mirror plate 204 and reflects the light from the light source at an angle determined by the deflection of the mirror plate 204. Note that the torsion spring hinge 2 0 6 Substantially formed under the upper surface of the mirror plate 204, And is substantially hidden by the reflective surface 2 0 3 deposited on the upper surface 205 and above the hinge portion 206. The difference between Figs. 2a and 3 is that the mirror plate 204 shown in Fig. 2a has an upper surface 205 and The reflective surface 203 of the hinge 206 is substantially hidden, while the mirror plate 204 shown in FIG. 3 does not have the reflective surface 203, so the hinge 206 is exposed. Since the hinge 206 and the mirror plate 204 are in the same base 105, as shown in Figures 7 and 7b It is shown that the center height 796 of the hinge 206 and the center height 795 or 7 97 of the mirror plate 204 are substantially coplanar. When the mirror 202 is rotated about the longitudinal axis of the hinge 206, there will be no translational movement or displacement. There is no translational displacement, so the gap between the mirror plate 204 and the spacer wall of the spacer support frame 2 10 is limited only by the manufacturing technology and process. Typically, the tight distance between the mirror plate 204 and the hinge 206 is substantially less than 0.1 ° The upper cover is hidden under the reflective surface 203, allowing the micromirror array 103 to have a high fill ratio, increased contrast, minimal light scattering and diffraction, and substantially eliminating the impact of light passing through the micromirror array 103 to the second Substrate]. As shown in Figures 2a-b, 3, 4a-b, 7a, 8, 9a, 9b, and 10, the mirror plate 204 is connected to the torsion spring hinge 206 by a connector 216. The torsion spring hinge 2 0 6 is connected to the spacer support frame 2 〇, the support frame 2 10 holds the torsion spring 2 06, the connector 2 16 and the mirror plate 2 0 4 in place. The hinge 2 6 includes the first arm 206a and the second arm 206b. See FIGS. 3 and 10 No, one end of each of the arms 2 06a and 2 06b is connected to the spacer support frame 210 ′ and the other-13-200528752 (10) end is connected to the connector 2 1 6. In another embodiment, other springs, hinges, and connection designs may be used between the mirror plate 204, the hinge 206, and the spacer support frame 201. As shown in FIGS. 3 and 4a, the torsional hinge 206 is preferably oriented diagonally (eg, at a 45 degree angle) with respect to the spacer support wall 2 1 0, and divides the mirror plate 204 into two parts, or plural Sides: the first side 2 0 4 a and the second side 2 4 4 b. As shown in FIG. 7b, the two electrodes 126 are associated with the mirror 202, an electrode 126a for the first side 204a, and an electrode 126b for the second side 204b. This allows the side 204a or 204b to attach under one of the electrodes 126a or 126b and pivot downwards as well as provide a wide range of angular motion. When a voltage is applied between the mirror 202 and the corresponding electrode 1 2 6 to apply a force such as an electrostatic force to the mirror plate 204, the torsion spring 206 allows the mirror plate 204 to surround the longitudinal axis of the hinge 2006 with respect to the interval The support frame 2 1 0 rotates. This rotation creates an angular deflection to reflect light in the chosen direction. Since the hinge 206 and the mirror plate 204 are in the same base 105, and as shown in FIGS. 7a and 7b, the center height 796 of the hinge 206 and the center height 79 5 or 7 9 7 of the mirror plate 204 are substantially coplanar, so the mirror 2 02 will move purely around the hinge 206 without translational displacement. In one embodiment, as shown in Figs. 7a and 8, the torsion spring hinge 206 has a width 222 which is smaller than the depth 22 3 of the hinge 206 (perpendicular to the upper surface 205 of the mirror plate 204). The width 222 of the hinge 206 is preferably at about 0. ] 2 micrometers to about 0.2 micrometers, and the depth is preferably about 0. Between 2 microns and about 0.3 microns. As shown in Figures 2 a · b, 3, 4 a-b, 6, and 7 a, the spacer support frame 2 1 〇 positioning the mirror plate 2 〇4 on the electrode] 2 6 and the predetermined distance -14 above the addressing circuit -200528752 (11) so that the mirror plate 04 can be deflected downward to a predetermined angle. As shown in Figures 2a, 4a, 9a, and 10, the spacer support frame 210 includes a spacer support wall. The spacer support wall is preferably formed from the same first substrate 105 and is preferably positioned orthogonally. These walls help define the height of the spacer support frame 210. According to the required separation between the mirror plate 20 04 and the electrode 1 2 6 and the design of the electrode extension, the height of the spacer support frame 2 10 is selected. The larger height allows more deflection of the mirror plate 204 and a higher deflection angle. Larger deflection angles usually provide higher contrast. In one embodiment, the deflection angle of the mirror plate 204 is 12 degrees. In a preferred embodiment, the mirror plate 204 can be rotated up to 90 degrees if it is supplied with sufficient intervals and driving voltages. The spacer support frame 2 10 also provides support for the hinge 206 and spaced the mirror plate 204 from the other mirror plates 204 in the mirror array 03. The spacer support frame 21 0 has a spacer wall width 2 1 2. When the gap between the mirror plate 204 and the support frame 2 1 0 is added, the spacer wall width 2 1 2 is substantially equal to the adjacent micromirror 2 02 the distance between. In one embodiment, the spacer wall width 2] 2 is 1 micron or less. In a preferred embodiment, the spacer wall width 2 1 2 is 0. 5 microns or less. This places the mirror plates 204 closely together to increase the fill ratio of the mirror array 103. In some embodiments, the micromirror 202 includes an element 405a or 405b. When the mirror plate 204 is deflected downward to a predetermined angle, the element 405a or 405b stops the deflection of the mirror plate 204. Typically, these elements include a stop 405a or 405b and a landing tip 71oa or 7] ob. As shown in Figures 4a, 6, 7a, 8, 10, and 2), when the mirror surface 204 is deflected, the stopper 4 5 a or 40 5 b on the mirror plate 204 will contact the landing tip 7] 〇 ( 7 I 03 or 200528752 (12) 7 10b). When this happens, the mirror plate 204 will not be further deflected. There are several possible configurations of the stopper 4 0 5 a or 4 5 b and the landing tip 710a or 710b. In the embodiments shown in Figs. 4a, 6, 7a, 8, 10 and 12, the stopper is a cylindrical or mechanical stopper 4 5 a or 4 0 5 b, attached to the lower surface of the mirror plate 2 0 4 201. The landing tip 710 is a corresponding circular area on the second base 107. In the embodiment shown in Figs. 7a, 7b, and 8, the landing tips 71 0a and 7 1 Ob are electrically connected to the spacer support frame 2] 0, and therefore have a zero voltage difference with respect to the stopper 405 a or 405b. To prevent the stopper 4 05 a or 405b from sticking or welding to the landing tip 7] 0a or 71 Ob, respectively. In this way, when the mirror plate 2 04 is rotated relative to the spacer support frame 2 10 beyond a predetermined angle (determined by the length and position of the mechanical stopper 405 a or 40 5 b), the mechanical stopper 405a or 4 05 b will come into physical contact with the landing tip 710a or 710b, respectively, and prevent any further rotation of the mirror plate 204. In the preferred embodiment, the stopper 405a or 405b is made of the first base 105 and the same material as the mirror plate 204, the hinge 206, the connector 2] 6 and the spacer support frame 2 10. The landing tip 7a or 7a 0b is also preferably made of the same material as the stopper 40a or 405b, the hinge 206, the connector 216, and the spacer support frame 2 10. In the embodiment of the material monocrystalline silicon, the stopper 4 0 5 a or 4 0 5 b and the landed tip 7 1 0 a or 7 1 0 b are thus made of a hard material with a long working life, which allows the mirror Array 1 03 is maintained for a long time. In addition, because the chert sand is a hard material, the stopper 405a or 405b and the landing land 7] 0a or 7] 0b can be made of a small area. In this small area, the stopper 4 5 0 a or 4 5 〇b will touch the landing tip 7] 0a or 7] 〇b, respectively, greatly reducing the adhesion and allowing the mirror plate 2 04 to freely -16- 200528752 (13) deflection. Moreover, this means that the stopper 405 a or 4 05 b and the book 7 10a or 710b are maintained at the same potential, preventing the stopper 40 5 b and the land contact tip or 710b from occurring through charge injection processing at different potentials. Sticky. The invention is not limited to elements or techniques that stop the deflection of the upper 204. You can use the components and techniques used in this technique. Fig. 4a is a perspective view showing the lower side supporting wall 210, the mirror plate 204 (including the sides 204a and 204b and having a 205 and the lower surface 201), the hinge 206, the connector 216, and the machinery 405a and 405b under the single micromirror 202. Fig. 4b is a more detailed perspective view of the corner of the micromirror shown in Fig. 4a. FIG. 5 is a perspective view showing the top and sides with the micromirrors 202-1 to 202 mirror array 103. Although Fig. 5 shows a micromirror array 103 having three columns for a total of nine micromirrors 202, other micromirror arrays 103 are also possible. Typically, each micromirror 202 is a pixel on a video display. As such, a column with more micromirrors 202 will provide a video display with more pixels. As shown in FIG. 5, the surface of the micromirror array 103 has a large 塡. That is, most of the surface of the micromirror array 103 is made of the micromirror 2 02 surface 202. A very small portion of the surface of the micromirror array 103 is reflective. As shown in FIG. 5, the non-reflection of the surface of the micromirror array] 03 is a region between the reflective surfaces 2 0 3 of the micromirror 202. For example, the width of the area between 2 0 2 -1 and 2 02-2 is supported by the spacer 212 and the micromirror 2 02 -1 and 2 02 -2. The mirror plate 2 04 and the spacer land tip 4 0 5 a or I splice or. Any of the known mirror plates 1 includes the upper surface, a stopper 2 3 7 of • 9, and three rows of dimensions corresponding to a larger array charge ratio. The reflection part is the non-radiation part. The width of the mirror wall is determined by the total width of the gap between the support wall -17- 200528752 (14) 2 1 0. Note that although the single mirror 202 is depicted as having its own spacer support frame 2 1 0 as shown in FIGS. 2a, 2b, 3, 4a, and 4b, typically, for example, the mirrors 2 0 2-1 and 2 Between 02-2 and other mirrors, there is no two separate adjacent spacer walls 21 〇. However, there will typically be a solid spacer wall supporting the frame 2 I 0 between the mirrors 2 02-] and 20-2. Since there is no translational displacement during the deflection mirror plate 204, the gap and the width of the spacer wall 2 1 2 can be made and the feature size supported by the manufacturing technology is generally small. Therefore, in one embodiment, the gap is 0. 2 microns, in another embodiment, the gap is 0.13 microns or less. Since the semiconductor manufacturing technology allows a smaller size, the size of the spacer wall 2 10 and the gap can be reduced to allow a higher charge ratio. Embodiments of the present invention allow high charge ratios. In the preferred embodiment, the charge ratio is 96% or more. FIG. 6 is a perspective view showing the bottom and sides of the micromirror array 103 with nine micromirrors. As shown in FIG. 6, the support wall of the spacer support frame 2 10 of the micromirror 202 defines a cavity under the mirror plate 204. These holes provide space for the mirror plate 204 to deflect downwards, and also allow a large area under the mirror plate 204 to be used for the second layer 104 having electrodes 126 and / or for the third layer having control circuits 106. Fig. 6 also shows the lower surface 20 of the mirror plate 204 (including the sides 204a and 204b), and the bottom of the spacer support frame 210, the torsion spring money chain 206, the connector 216, and the stops 405a and 405b. As shown in Figs. 5 and 6, very little light orthogonal to the mirror plate 204 can pass through the micromirror array] 03 and reach any electrode or control circuit below the micromirror array 103]. This is because the spacer support frame 2] 〇 and the mirror plate -18- 200528752 (15) 204 on the upper surface 2 0 5 and the reflection surface 203 above the hinge 2 0 6 almost completely cover the micromirror array 103 Circuit. Moreover, since the spacer support frame 2 10 separates the mirror plate 04 from the circuit under the micromirror array 103, the light that travels to the mirror plate 2 04 at a non-vertical angle and passes outside the mirror plate 20 04 is easy to hit The walls of the spacer support frame 2] 0 and will not reach below the micromirror array 103. Since a small amount of strong light incident on the micromirror array 103 can reach the circuit, the SLM 100 can avoid problems related to strong light hitting the circuit. These problems include heating the circuit with incident light and charging circuit elements with incident photons, both of which can cause circuit failure. Fig. 9a is a perspective view of a micromirror 202 according to another embodiment of the present invention, and Fig. 9b is a more detailed perspective view of a corner 23 8 of the micromirror 202. The torsion hinge 206 in this embodiment is parallel to the spacer support wall of the spacer support frame 210. Applying a bias between the mirror plate 20 04 and the corresponding electrode I 2 6 will cause the mirror plate 04 to be selectively deflected toward the electrode. The embodiment shown in Fig. 9a provides a smaller overall range of angular motion than the mirror 202 shown in Figs. 2a and 2b with diagonal hinges 206, which starts at the same support wall height. However, like the embodiment shown in Figs. 2a and 2b, the hinge 206 in the embodiment shown in Figs. 9a and 9b is below the upper surface of the mirror plate 204 and hidden by the reflective surface 203, resulting in a high SLM 100 with high charge ratio, high optical efficiency, high contrast, low light diffraction and scattering, and reliable and cost-effective performance. Figure 9b is a more detailed perspective view of the corner of the micromirror 202, and shows the mirror plate 204, the hinge 206, the support wall of the spacer support frame 21, and the reflective surface 203. FIG. 10 shows the lower side of a single micromirror 20 2 ', which includes a hinge 20 06, a connector 2] 6 and a stopper 4 05 a. In other embodiments of -19-200528752 (16), the hinge 206 may be substantially parallel to one side of the mirror plate 204 and still be arranged to divide the mirror plate 204 into two parts 405a and 405b. Figures Π and 12 provide perspective views of a micromirror array composed of a plurality of micromirrors 202 as shown in Figs. 9a, 9b, and 10. FIG. 13 is a perspective view of an embodiment of an electrode 1 2 6 formed on a second substrate 107. In this embodiment, each micromirror 202 has a corresponding electrode 1 2 6. In the embodiment shown here, the electrodes 1 2 6 are made higher than other parts of the circuit on the second substrate. In a preferred embodiment, the electrodes 1 2 6 are disposed at the same level as the other parts of the circuit on the second substrate. In another embodiment, the electrodes 1 2 6 extend above the circuit. In one embodiment of the present invention, the electrodes 126 are attached to individual aluminum pads under the micromirror plate. The shape of the electrodes depends on the embodiment of the micromirror 202. For example, in the embodiment shown in Figs. 2a, 2b, and 3, it is preferable that two electrodes 1 2 6 are under the mirror 2 02, and each electrode 126 has a triangle as shown in Fig. 7b. In the embodiments shown in Figs. 9a, 9b, and 10, preferably, a single, square electrode 126 is below the mirror 202. These electrodes 126 are formed on the surface of the second substrate 107. In this embodiment, the large surface area of the electrodes 1 2 6 will cause a relatively low addressing voltage required to pull down the mirror plate 20 04 to the mechanical stop, thereby causing the full pre-angle deflection of the micro plate 20 04. Options: In operation, the individual reflective micromirrors 202 are selectively deflected and used to spatially modulate the light incident on and reflected by the mirror 202. Figures 7a and 8 show cross-sectional views of the micromirror -20-200528752 (17) 2 02 shown along the dotted line 2 50 in Figure 2a. Note that this section view is offset diagonally from the center of the center to show the outline of the hinge 206. Figure 7c Figure 2a shows the difference of the micromirror 202 shown by the dashed line 25 0 in this figure. This cross-sectional view is along the center diagonal and is perpendicular to the micromirror 20 above the display electrode 126 than 7a, 7c and 8 2. A voltage is applied to the deflection of the corresponding portion of the mirror plate 2 0 4 above the electrode 126 1 2 6 on one side of the micromirror 202 (FIG. 8). As shown in FIG. 8, when a voltage is applied to the electrode 1 2 6, Half of g will be attached to the electrode 126, while the structure and rigidity of the other 204 of the mirror plate 204b will be removed from the electrode 1 2 6 and the second base will cause the mirror plate 204 to rotate around the torsion spring hinge 206. For the electrode 126, as shown in Figure 7a As shown, the money chain 206 will cause it to return to its unbiased position. Alternatively, in an embodiment having a diagonal hinge 206 as shown in FIG. 2, the voltage may be an electrode 126 on the other side of the 206 In order to make the mirror 202 in phase rotation. In this way, the light striking the mirror 202 will be reflected in a direction controlled by the application of a voltage. An embodiment operates as follows. At first, the mirror 202 is not deflected as usual. In this unshifted state, the incident light from the light source SLM 100 will be reflected by the plane mirror 202. The externally deflected light will be received by, for example, an optical pump. The light from the undeflected mirror will not be reflected to the video display. When a voltage bias is applied between the half of the mirror plate 2 04 a and its] 26, the mirror 202 will be attracted by static electricity. In the deflected mirror system 202 to display the delay chain 206 in FIG face view in the injection operation, in order to control side electrode 3 2 0 4 a) |... A half plate 204a due to the bottom 107 of the mirror plate. In this embodiment, when the voltage is removed from the mirror plate 204 and a, 2 b, and 3 are applied to the mirror plate, they are biased to the electrode 126 in the opposite direction. Figure 7 a and 7 c. 21-200528752 (18) 'When the mirror plate 2 0 4 a is deflected downward as shown in FIG. 8, ve! Is preferably 12 volts, V b is −10 volts, and v e2 is 0 volts. Similarly (or vice versa), when the microplate 2 0 4 b is deflected downward, v e] is preferably 0 volts, Vb is -10 volts, and Vd is 12 volts. Due to the design of the hinge 20, one side of the mirror plate 04a or 204b (that is, the side above the biased electrode 12 26) is deflected downward (toward the second base 丨 07), while the mirror plate The other side of 2 0 4 b or 2 0 4 a moves away from the second substrate 7. Note that in a preferred embodiment, substantially all of the bending occurs in the hinge 206 rather than the mirror plate 204. In one embodiment, this is achieved by making the hinge width 222 thin and connecting the hinge 206 to a support post located only on the two ends. As described above, the deflection of the mirror plate 2 0 4 is limited by the stopper 4 0 5 & 4 5 b. . The total deflection of the mirror plate 2 0 4 deflects the external reflected light to the imaging light element and the video display. When the mirror plate 2 0 4 is deflected through the "fast-moving" or "pull-down" voltage (almost 12 volts or lower in one embodiment), the restored mechanical force or torque of the hinge 2 06 can no longer balance the electrostatic force or Torque, and half of the mirror plate 204 with electrostatic force 204a or 204b under it will quickly move towards the electrode 1 2 6 below to obtain complete deflection, limited only by the stopper 405 a or 40 when required. 5 b. In the embodiment shown in Figs. 9a, 9b and 10 where the hinge 2 06 is parallel to the support wall of the spacer support frame 2 10, in order to release the mirror plate 204 from its fully deflected position, the voltage must be shut down. In the embodiment where the hinges shown in Figures 2a, 2b, and 3 are diagonal, in order to release the mirror plate 204 from its fully deflected position, the voltage must be turned off while other electrodes are being enabled, and the mirror 2 02 attaches to the other side. -22- 200528752 (19) The micromirror 2 0 2 is an electromechanical bistable device. Given a specific voltage between the release voltage and the snap action, there are two possible deflection angles depending on the history of the mirror's deflection of 202. Therefore, the mirror 2 0 2 deflection table is the same with the shackles. Since the mechanical force required for the deflection of the mirror 202 is approximately linear with respect to the deflection, these 'bi-stable and stern lock characteristics will occur, while the opposite electrostatic force is between the mirror plate 2 0 4 and the electrode 1 2 6 The distance is inversely proportional. Since the electrostatic force between the mirror plate 2 0 4 and the electrode 1 2 6 depends on the total voltage difference between 20 04 and the electrode 1 2 6, the negative voltage applied to the mirror plate will reduce the amount required to apply to the electrode 1 2 6 The amount of deflection obtained with a positive voltage. As such, applying a voltage to the mirror array 103 can reduce the amount of voltage 126 required. This is true, for example, in some applications, because the 5V switching capability is more versatile and cost effective in the semiconductor industry, it is necessary to keep the voltage that must be applied to the electrode] 2 6 below 12V. Since the maximum deflection of the mirror 202 is fixed, if it is operated at a voltage exceeding the snap-action voltage, it can be operated digitally. In the embodiment shown in Figs. 2a, 2b and 3 where the hinge is parallel to the support wall of the spacer support 210, the mirror plate 2 04 will be completely deflected downward due to the voltage applied to the electrode 1 2 6, or When the electrodes 1 2 6 connected with no voltage are applied, the mirror plate 2 0 4 is allowed to spring upward, so the operation is digital. In an embodiment having a diagonal of 206 as shown in Figs. 12a, 2b, and 13, when the electrode 1 2 6 on the other side of the mirror plate 2 0 4 is enabled, the mirror plate 2 0 4 will be applied to the mirror plate due to the voltage 2 0 4 The voltage mirror plate is now separated from the mirror plate 204 if the angle is present Some applications and the largest SLM-type operating frame is added to the other one of the base hinge -23- 200528752 (20) The associated electrodes 126 on one side are fully deflected down to the other side of the mirror plate 204. The voltage that causes the mirror plate 2 04 to deflect completely downwards until it is stopped by the solid element that stops the deflection of the mirror plate 2 04 is referred to as "quick action" or "pull-down" voltage. In this way, in order to fully deflect the mirror plate 204, a voltage equal to or greater than the snap-action voltage is applied to the corresponding electrode I 2 6. In video display applications, when the mirror plate 204 is fully deflected downward, the incident light incident on the mirror plate 204 will be reflected to the corresponding pixels on the video display screen, and the pixels will appear bright. When the mirror plate 20 is allowed to pop up, the light will be deflected in such a way that it will not hit the video display screen, and the pixels will appear dark. During this digital operation, after the associated mirror plate 204 is fully deflected ', it is not necessary to maintain a full snap-action voltage on the electrode 126. During the "addressing phase", the voltage applied to the selected electrode 1 2 4 corresponding to the mirror plate 2 0 which should be fully deflected is set to the level required for the deflection mirror 2 0 4. When the mirror plate 204 in question is deflected due to the voltage on the electrode 26, the voltage required to hold the mirror plate 204 in the deflected position will be less than the voltage required for true deflection. This is because the gap between the deflected mirror plate 204 and the address and address electrodes 126 is smaller than when the mirror plate 204 is deflected. So after the addressing phase. In the "holding stage", the pressure applied to the selected electrode 1 2 6 will be reduced from its original required level, but it will not substantially affect the deflection state of the mirror plate 2 (Η. It has one of the advantages of a lower holding stage The undeflected mirror plates 204 near the system are subject to less electrostatic attraction, and they thus remain closer to the zero deflection position. This improves the optical contrast between the deflected mirror plates 204 and the undeflected mirror plates 204. By appropriately selecting the size (in one embodiment, the support frame 2 between the mirror plate 204 and the electrode-24-200528752 (21) 1 2 6 is separated from 1 to 5 depending on the mirror structure and deflection angle requirements Microns, and the thickness of the hinge 206 is 0. 〇5 to 0. 45 microns) and materials (such as monocrystalline silicon (100)), the reflective SLM 100 can be made to operate at voltages of only a few volts. The shear modulus of the torsion spring 206 made of single crystal silicon can be 5 X 101 () Newton / radius square meter. Maintaining the mirror plate 204 at an appropriate voltage (negative bias) instead of ground allows the electrode 126 to operate to fully deflect the voltage of the associated mirror plate 204. For a given voltage applied to the electrodes 1 2 6 this results in a greater deflection angle. The largest negative bias voltage is the release voltage, so when the address voltage drops to zero, the mirror plate 204 can quickly move back to the undeflected position. It is also possible to control the deflection of the mirror plate 204 in a more "analogous" manner. A voltage smaller than the "quick-action voltage" is applied to deflect the mirror plate 204 and control the direction in which incident light is reflected. Other applications In addition to video displays, the spatial light modulator 100 is also useful in other applications. One such application is maskless lithography, where a spatial light modulator directs light to develop the deposited photoresist. This will enable the photoresist to be developed correctly in a desired pattern without a mask. Although the present invention has been particularly shown and described with reference to various embodiments, those skilled in the relevant arts should understand that, without departing from the spirit and scope of the present invention, various changes in form and details can be made. . For example, the mirror plate 204 can be deflected by means other than electrostatic attraction. Alternatively, magnetic, thermal, or piezoelectric actuation may be used to deflect the mirror plate 204. -25- 200528752 (22) [Brief description of the drawings] FIG. 1 illustrates a general architecture of a spatial light modulator according to an embodiment of the present invention. FIG. 2 a is a perspective view of a single micromirror according to an embodiment of the present invention. Fig. 2b is a perspective view of a corner of the micromirror of Fig. 2a. Fig. 3 is a perspective view of a single micromirror without a reflection surface, showing the top and sides of a mirror plate of a micromirror array in an embodiment of the present invention. Fig. 4a is a bottom view and a side view of a single micromirror according to an embodiment of the present invention. Fig. 4b is a perspective view of a corner of the micromirror of Fig. 4a. Figure 5 is a perspective view showing the top and sides of a micromirror in an embodiment of the present invention. . Fig. 6 is a perspective view showing the bottom and sides of a micromirror array according to an embodiment of the present invention. Fig. 7a is a cross-sectional view of an undeflected micromirror shown in Fig. 2a along an offset diagonal section. Fig. 7b shows an electrode and a land landing tip formed under the mirror plate in the second substrate according to an embodiment of the present invention. Fig. 7c is a cross-sectional view of the undeflected micromirror shown in Fig. 2a along the central diagonal section. Fig. 8 is a sectional view of the deflected micromirror shown in Fig. 2a. Fig. 9 is a perspective view of the top and sides of another embodiment of a micromirror. FIG. 9b is a perspective view of a corner of the micromirror of FIG. 9a. -26- 200528752 (23) Figure] is a stereoscopic view showing the bottom and sides of another embodiment of the micromirror. Fig. 11 is a perspective view showing the top and sides of another embodiment of the micromirror array. Fig. 12 is a perspective view showing the bottom and sides of another embodiment of the micromirror array. FIG. 1 is a series of three-dimensional views showing one of the electrodes formed on the second substrate. The main components of the display table 100 are the spatial light modulator 103. The deflectable mirror array 1 04 the electrode array 105 the first substrate 106 the control circuit 107 the second Substrate 108 display control 110 line memory buffer 112 pulse modulation array 1 14 microcontroller 116 light control circuit 1 1 8 flash memory 1 20 video signal 1 22 drawing signal -27- 200528752 (24) 126 electrode 126a electrode 126b electrode 20 1 lower surface 202 micromirror 202-1 ^ -202-9 micromirror 203 reflective surface 204 mirror plate 204a first side 204b second side 205 upper surface 206 hinge 2 0 6 a first arm '20 6b second arm 2 10 Spacer support frame 2 1 2 Spacer wall width 2 16 Connector 222 Width 223 Depth 23 6 Corner 23 7 Corner 23 8 Corner 4 0 5 a Stopper 4 0 5 b Stopper 200528752 (25) 710a Landing Tip 710b landing tip > 29-