經濟部中央標率局貝工消费合作社印策 A7 __B7 五、發明説明（1 ) 本發明是有關於一種新穎的半導體微透鏡製程，且特 別疋有關於一種利用複晶梦氧化物製程(p〇iy 〇xide process) 來製作微透鏡的製程。 近來，藉著半導體積體電路科技所製造之固態影像感 測器，已廣泛地使用在影像攝取裝置之光電轉換元件中， 以便將物件影像轉換成電子訊號。而微透鏡(micro lens)則 是經常以On-chip的方式製作在影像感測器的上方，以便 將入射光聚合導仑感測器的光接受元件，藉此吾人得以製 作出高感光度的影像感測器。 第1圖所示為習知中之一種電荷耦合裝置（Charge CoupledDevice ; CCD)影像感測器的剖面結構，其主要是 由光二極體10、垂直電荷耦合元件(V-CCD)14、及通道阻 絕區12所構成。在V-CCD上的轉移閘極20通常是由複晶 矽所組成’而在轉移閘極20的上方則有一金屬材質的遮光 層22 (light shield)，用以避免光線進入V-CCD區14，造 成拖影現象（smear phenomenon) ’致使畫面模糊。一般而 言，遮光層22的材質通常是鋁或矽化鎢（WSix)。而在遮光 層之上，還包括硼磷矽玻璃層24、内透銳層26、據光層 28、以及微透鏡30。其中硼磷矽玻璃層24經過再熱流 (reflow)後，會在光二極體的上方形成凹槽，此時再填入高 透光性的有機樹脂，可形成有聚焦作用的内透鏡層26，以 進一步提高其感光性及避免拖影現象。 如第1圖所示，微透鏡30通常製作在每個光二極艘1〇 的上方，使虚線箭頭所代表的入射光，經過微透鏡30及内 3 本紙張尺度適用中國国家梯準（CNS ) Α4規格（210Χ297公釐） (請先閲讀背面之注$項再填寫本頁) α. 訂 A7 A7Imprint A7, __B7, Shellfish Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs 5. Description of the Invention (1) The present invention relates to a novel semiconductor microlens process, and particularly to a process using a polycrystalline dream oxide process (p. 0). iy 〇xide process). Recently, solid-state image sensors manufactured by semiconductor integrated circuit technology have been widely used in photoelectric conversion elements of image pickup devices to convert object images into electronic signals. The micro lens is often fabricated on the image sensor in an on-chip manner, so that the incident light can be used to aggregate the light-receiving element of the sensor, thereby enabling us to produce high-sensitivity Image sensor. Figure 1 shows the cross-sectional structure of a conventional Charge Coupled Device (CCD) image sensor, which is mainly composed of a photodiode 10, a vertical charge-coupled device (V-CCD) 14, and a channel. The exclusion zone 12 is formed. The transfer gate 20 on the V-CCD is usually composed of polycrystalline silicon, and a metal light shield 22 (light shield) is provided above the transfer gate 20 to prevent light from entering the V-CCD area 14 , Causing smear phenomenon ', resulting in blurred picture. Generally, the material of the light shielding layer 22 is usually aluminum or tungsten silicide (WSix). The light-shielding layer further includes a borophosphosilicate glass layer 24, an inner transparent layer 26, a light-receiving layer 28, and a microlens 30. After the borophosphosilicate glass layer 24 is reflowed, a groove is formed above the photodiode. At this time, a highly translucent organic resin is filled to form a focusing inner lens layer 26. In order to further improve its sensitivity and avoid smear. As shown in Figure 1, the microlenses 30 are usually made above each photodiode 10, so that the incident light represented by the dotted arrow passes through the microlenses 30 and 3 ) Α4 size (210 × 297mm) (Please read the note on the back before filling in this page) α. Order A7 A7
發明説明（2) 透鏡層26的聚焦，得以進入光二極體ι〇 讀取光二極體10所釋出的電荷。 再由 V-CCD 14Description of the invention (2) The focusing of the lens layer 26 allows the photodiode 10 to enter the photodiode 10 to read the electric charge released by the photodiode 10. V-CCD 14
在習知技術中，晶片上的微透鏡3Q 區域具有相當穿透率的光阻騎料加㈣作並經過洪烤 的步驟後，形成如第！囷所示之圓滑曲面。以下將配合第 2A〜2D圖來說明習知製作微透鏡的流程，為方面起見，其 中與第1圖相同的元件將沿用相同的標號。 首先請參照第2A圖，在基底上已形成有影像感測器 元件’包括光二極體10、V-CCD Μ、轉移閘極π、以 及滅光層28。首先，為了製作出間隔一致的微透鏡，必須 先對遽光層28進行平坦化’例如可利用壓克力樹脂 (acrylate resin)旋塗在濾光層28上，再配合熱墊板的烘烤， 以得到一平坦化層 40 (planarization layer)。 接著’請參照第2B圏，在平坦化層40上塗佈適當的 光阻材料42，例如深紫外光光阻、g-線光阻等，然後利用 光罩100配合適當的步進機（stepper)進行微影製程。經過 顯影後，可在晶片上形成等間隔排列的光阻區塊42a，如 第2C圖所示。這些整齊排列的光阻區塊經過適當的烘烤 加以圓化(rounding off)後，就成了如第2D圖所示的圓頂形 透镜（dome-lens) 42b 〇 雖然目前的微透銑製程都是以光阻严當作透鏡的材 料，但是利用光阻材所得的微透鏡’卻不可避免地會有以 下的缺點：In the conventional technology, the photoresist material with a considerable transmittance in the 3Q area of the microlens on the wafer is processed and flooded to form the first! The smooth surface shown by 囷. The process of making microlenses according to the prior art will be described with reference to Figures 2A to 2D. For the sake of brevity, the same components as those in Figure 1 will use the same reference numerals. First, referring to FIG. 2A, an image sensor element 'including a photodiode 10, a V-CCD M, a transfer gate π, and a light-extinguishing layer 28 has been formed on a substrate. First, in order to produce uniformly spaced microlenses, the calender layer 28 must be first planarized. For example, an acrylic resin may be spin-coated on the filter layer 28, and then the baking of the hot pad is performed. To obtain a planarization layer 40 (planarization layer). Next, please refer to Section 2B 圏, apply a suitable photoresist material 42 on the planarization layer 40, such as deep ultraviolet photoresist, g-line photoresist, etc., and then use the photomask 100 with an appropriate stepper (stepper ) Perform a lithography process. After development, photoresist blocks 42a can be formed on the wafer at equal intervals, as shown in FIG. 2C. After the neatly arranged photoresist blocks are rounded off after proper baking, they become dome-lens 42b as shown in Figure 2D. 〇 Although the current micro-through milling process The photoresist is strictly used as the lens material, but the microlenses obtained using the photoresist will inevitably have the following disadvantages:
1.首先，由於光阻本身材質的限制’使得現今的CCD 本紙張尺度適用中國國家樣準（CNS ) A4規格（210XM7公* ) - λι, (請先聞讀背面之注f項再填寫本頁) 訂 經滴部中央標率局負工消费合作社印掣 A71. First of all, due to the limitation of the material of the photoresist itself, the current CCD paper size is applicable to the Chinese National Standard (CNS) A4 specification (210XM7 male *)-λι, (please read the note f on the back before filling in this (Page) Order Book A7 of the Central Standards Bureau of the Ministry of Work
經濟部中央標率局貝工消费合作社印¾ 五、發明説明（3) 影像感測器無法承受高溫的環境。 2.其次，光阻的穿透性會隨著時間而衰減，使得影像 感測器的光感度降低。 / 3·此外，由光阻所形成的透鏡輪廓也很難在製程令作 精確的控制》 有鑑於此’本發明的主要目的就是提供一種新賴的微 透鏡製程，係利用複晶碎氧化物(P〇ly 〇Xide)來作為微透鏡 的材料’由於氧体矽的耐溫性與穿透性都比光阻來得好， 因此所形成的微透鏡不但品質較佳而且可耐高溫。 本發明的另一個目的是提供一種微透鏡製程，藉由控 制其製程條件，吾人可調整所形成的透鏡輪廓，而得到任 何所需要的聚焦深度。 本發明的再一個目的是提供一種微透鏡製程，係利用 複晶矽氧化物製程(P〇ly oxide process)來製作氧化矽微透 鏡’所形成之微透鏡具有足夠的聚焦能力，因此毋需製作 額外的光阻透鏡，而且BPSG層也不需經過再熱流的步 根據上述目的，本發明提供一種新穎的微透鏡製程， 包括下列步驟：（a)提供一半導體基底，其上形成有一光接 收元件，以及一覆蓋住此元件之平坦層；（b)在該平坦層上 依序形成一透鏡承載層、一複晶矽層、以及一氮化矽層； (c)定義上述氮化矽層，形成一開口而露出此光接收元件上 方之複晶矽層；（d)以濕式氡化法將開口處之複晶矽層氧 化，形成一具有雙凸面之複晶矽氧化物，並使此氧化物之 〆 本纸張尺度適用中國國家橾準（CNS ) A4規格（2丨OX 297公羞） (請先W讀背面之注意事項再填寫本頁) •訂 經濟部中央標隼局貝工消費合作社印家 A7 B7 五、發明说明（4 ) .底部與上述透鏡承載層連接；（e)去除氮化矽層；以及⑴ 去除複晶矽層，得到一氧化矽材質之雙凸面微透鏡（Biconvex)。 而 在步驟（0 之後 ，可利用旋塗玻璃製程將此複晶 矽氧化物的下半部覆蓋住，即可形成一單凸面微透鏡 (Planar convex)° 根據本發明的微透鏡製程，所形成的透鏡輪廓可藉由 步驟（b)中複晶矽層的厚度，以及步驟（c)中開口的寬度加以 調整，因此吾人便可藉此得到各種不同的聚焦深度。而其 中步驟（d)則是以習知製作場氧化層的區域氧化法(LOCOS 法）來形成一具有雙凸面的氧化矽微透鏡《由於在後序的製 程中，將會陸續去除氮化矽層與複晶矽層，因此所形成的 複晶矽氧化物必須與透鏡承載層連接，以作為固定之用。 根據本發明的較佳實施例，步驟(b)所述之透鏡承載層 可為厚度約0.1〜1 μπι的氧化層；步驟(e)通常是以濕蝕刻的 方式剝除此氮化矽層；而其中步驟（f)可包括二階段的去除 步驟：例如先以複晶矽氧化物為罩幕，等向性地蝕刻此複 晶矽層後，再以濕浸泡的方式去除複晶矽氧化物底下殘餘 之複晶矽層。 為讓本發明之上述和其他目的、特徵、和優點能更明 顯易僅，下文特舉一較佳實施例，並配合所附圖式，作詳 細說明如下： 圖式之簡辈銳明 、 第1圖為習知一種CCD影像感測器的剖面示意圈。 「'第2A〜2D圖為一系列剖面沒，用以U習知—製作微透 ( CNS ) ( 210X297^t ) (請先閲讀背面之注意事項再填寫本頁)Printed by the Shell Standard Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs Ⅴ. Description of the invention (3) The image sensor cannot withstand high temperature environment. 2. Secondly, the penetrability of the photoresist will decrease over time, which will reduce the light sensitivity of the image sensor. / 3 · In addition, the lens profile formed by the photoresist is difficult to accurately control in the manufacturing process. "In view of this, the main purpose of the present invention is to provide a novel microlens manufacturing process, which uses multicrystalline broken oxide (Poly 〇Xide) as the material of the micro lens' Since oxygen silicon has better temperature resistance and penetration than photoresist, the micro lens formed is not only of good quality but also resistant to high temperature. Another object of the present invention is to provide a micro-lens manufacturing process. By controlling the process conditions, we can adjust the lens profile formed to obtain any desired depth of focus. Yet another object of the present invention is to provide a microlens process, which uses a polycrystalline silicon oxide process to make silicon oxide microlenses. The microlenses formed have sufficient focusing ability, so there is no need to make An additional photoresistive lens, and the BPSG layer does not need to be reheated. According to the above purpose, the present invention provides a novel microlens process, including the following steps: (a) providing a semiconductor substrate on which a light receiving element is formed; And a flat layer covering the element; (b) sequentially forming a lens bearing layer, a polycrystalline silicon layer, and a silicon nitride layer on the flat layer; (c) defining the above silicon nitride layer, Forming an opening to expose the polycrystalline silicon layer above the light-receiving element; (d) oxidizing the polycrystalline silicon layer at the opening by a wet process to form a polycrystalline silicon oxide having a double convex surface, and The size of the oxide paper is applicable to the Chinese National Standard (CNS) A4 specification (2 丨 OX 297). (Please read the precautions on the back before filling out this page.) Consumer cooperatives A7 B7 family V. invention is described in (4) above the bottom of the lens carrier layer is connected;. (E) removing the silicon nitride layer; ⑴ removing the polycrystalline silicon layer, a silicon oxide material to obtain the double convex microlens (Biconvex). After step (0), a spin-on-glass process can be used to cover the lower half of the polycrystalline silicon oxide to form a single convex microlens (Planar convex). According to the microlens process of the present invention, The lens profile can be adjusted by the thickness of the polycrystalline silicon layer in step (b) and the width of the opening in step (c), so we can use this to get a variety of different depths of focus. Among them, step (d) is The area oxidation method (LOCOS method) for forming a field oxide layer is used to form a silicon oxide microlens with a double convex surface. "Since the subsequent process, the silicon nitride layer and the polycrystalline silicon layer will be successively removed. Therefore, the formed polycrystalline silicon oxide must be connected to the lens bearing layer for fixing. According to a preferred embodiment of the present invention, the lens bearing layer described in step (b) may have a thickness of about 0.1 to 1 μm. Oxide layer; step (e) usually strips the silicon nitride layer by wet etching; and step (f) may include a two-stage removal step: for example, using a polycrystalline silicon oxide as a mask, isotropic After the polycrystalline silicon layer is etched The remaining polycrystalline silicon layer under the polycrystalline silicon oxide is removed by wet immersion. In order to make the above and other objects, features, and advantages of the present invention more obvious and easy, only a preferred embodiment is given below, and A detailed description is given in conjunction with the attached drawings: The simple and sharp generation of the drawings, Fig. 1 is a schematic cross-section of a conventional CCD image sensor. "The 2A ~ 2D pictures are a series of cross-sections. Learning by U—Creating Micro Penetration (CNS) (210X297 ^ t) (Please read the precautions on the back before filling this page)
C 丁 ΑΊ Β7 經濟部中央橾準局貝工消费合作社印製 五、發明説明（5 ) 鏡的流程。 第3A~3E圖為一系，列剖电圖，用以說明本發風二較袭_ 實施例製作微透鏡的流程。 符號說明 10〜光二極體；12〜通道阻絕區；14〜V-CCD ; 20〜轉 移閘極；22〜遮光層；24~硼磷矽玻璃層；26〜内透鏡層； 28〜濾光層；30、42b〜微透鏡；42〜光阻層；100〜光罩； 50〜透鏡承載層；、52〜複晶矽層；54〜氮化矽層；55〜開 口； 56~複晶矽氧化物；58〜SOG層？ 實施例 在此實施例中，是將本發明之微透鏡製程應用在一 CCD影像感測器上，以下將配合第3A圖至第3D圖詳細說 明之，為方面起見，其中與第1圖相同的元件將沿用相同 的標號。 請參照第3A圖，首先提供一半導體基底2，例如是n-底材上的p-井區，其上已形成有影像感測器元件，包括光 二極體10、V-CCD 14、轉移閘極20、遮光層22、以及 將前述元件覆蓋的一層内層介電層24，其材質為硼磷矽玻 璃。其中光二極體10與V-CCD 14通常以陣列（matrix)的方 式形成在基底中，而轉移閘極20係經由一絕緣層形成在 V-CCD 14的上方。此外，在轉移閘極20上方則形成有一 矽化鎢材質的遮光層22。 請繼續參照第3A圖，在上述的内層介電層24上，依 序沈積一透鏡承載層50、複晶矽層52、及氮化矽層54。 (請先閲讀背面之注意事項再填寫本頁) -X. 訂. 本紙張尺度適用中國國家標準（CNS ) A4規格（210X 297公釐） 經滴部中央標準局貝工消费合作社印簟 A7 B7 五、發明説明（6) 其中透鏡承載層50為厚度約0.1〜1 μιη的氧化層，而複晶 矽52的厚度則視微透鏡的厚度而定。 接下來，以類似習知中形成場氧化層的方式來製作一 具有雙凸面的氧化矽微透鏡。請參照第3Β圖，以傳統的 微影與蝕刻製程定義氮化矽層54，以在光二極體10的上 方形成一開口 55而露出部份的複晶矽層。之後，以濕式氧 化法將開口處之複晶矽層氧化，形成一具有雙凸面之複晶 石夕氧化物（P〇ly〇xide)56 ;由於在後序的製程中，將會陸續 去除氮化矽層54a與複晶矽層52，因此所形成的複晶矽氧 化物56必須與透鏡承載層50相連接，以作為固定之用。 由第3B圖可看出，複晶矽層52的厚度大小將會決定 微透鏡的厚度，而開口 55的大小則決定了微透鏡的徑寬。 根據這兩項參數，吾人便可控制微透鏡所形成的輪廓，進 而調整所需要的聚焦深度。 完成氧化矽微透鏡的製作後，先以濕蝕刻的方式將氮 化矽層54a剝除，然後再將複晶矽層52去除。複晶矽層52 的去除可分兩階段進行：首先，以複晶矽氧化物56為罩幕 進行等向性#刻，如第3C圖所示；之後，再以済、浸泡（wet dip)的方式將複晶矽氧化物56底下殘餘的複晶矽52a去 除，即可得到如第3D圖所示之微透鏡。 至此，吾人已經完成一雙凸面微透鏡(Bi-Convex)的製 作，若要進一步形成如第3E圖所示的單凸面微透鏡(Planar convex)，只要利用旋塗玻璃層58將複晶石夕氧化物56的下 半部覆蓋住即可。 8 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公羞) (請先閲讀背面之注意事項再填寫本頁) Λ. 訂 經濟部中央標準局貝工消费合作社印聚 A7 B7 五、發明説明（7) 為了進一步了解上述兩種微透鏡的聚焦情形，我們進 行了以下的模擬試驗。 1.雙凸面微透鏡 此處是根據第3D圖所示之雙凸面微透鏡進行模擬， 在固定厚度D下調整不同的徑寬A，以造成不同的曲率半 徑R!、R2，再根據此曲率半徑計算其聚焦深度f，其計算 公式如下： 1^=(1.46-1)(1^^1/^) 式（1) 其中心代表透鏡上曲面的曲率半徑；R2代表下曲面的曲 率半徑；1.46=氧化矽的折射率；1 =空氣的折射率。 所得之模擬結果如下：（單位：μιη) ΓΤ- Ri r2 f 0.1 0.3 0.112803 0.104537 -3.10126 [0.2 0.3 0.203712 0.215648 8.001013 丨丨 0.3 0.3 0.355227 0.400833 6.787186 0.4 0.3 0.567348 0.660093 8.778304 0.5 0.3 0.840076 0.993426 11.83074 0.6 0.3 1.173409 1.400833 15.71236 0.7 0.3 1.567348 1.882315 20.36272 I 0.8 0.3 2.021894 2.437870 25.75980 0.9 0.3 2.537045 3.067500 31.89384 1.0 0.3 3.112803 3.771204 38.75998 由式（1)可知，若要使入射光在基底上聚焦，必須使 ReRz。而根據實驗顯示，在上述製程中經由熱氧化法所 形成的微透鏡會有45%的氧化矽長在複晶矽中，而有55% 是長在空氣中（請參照第3B圖），因此其曲率半徑心勢必 本紙張尺度適用中國困家橾準（CNS ) A4規格（210X297公釐） (請先閲讀背面之注意事項再填寫本頁)C 丁 ΑΊ Β7 Printed by the Shellfish Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs. 5. Description of invention (5) Mirror process. Figures 3A to 3E are a series of cross-section electrical diagrams, which are used to explain the process of making microlenses in this second example. Explanation of symbols 10 ~ photodiode; 12 ~ channel barrier; 14 ~ V-CCD; 20 ~ transfer gate; 22 ~ light-shielding layer; 24 ~ borophosphosilicate glass layer; 26 ~ inner lens layer; 28 ~ filter layer 30, 42b ~ microlens; 42 ~ photoresist layer; 100 ~ photomask; 50 ~ lens bearing layer; 52 ~ polycrystalline silicon layer; 54 ~ silicon nitride layer; 55 ~ opening; 56 ~ polycrystalline silicon oxide Material; 58 ~ SOG layer? EXAMPLES In this example, the microlens manufacturing process of the present invention is applied to a CCD image sensor, which will be described in detail below with reference to Figures 3A to 3D. Identical components will use the same reference numerals. Referring to FIG. 3A, a semiconductor substrate 2 is first provided, for example, a p-well region on an n-substrate, on which an image sensor element has been formed, including a photodiode 10, a V-CCD 14, and a transfer gate. The electrodes 20, the light-shielding layer 22, and an inner dielectric layer 24 covering the aforementioned elements are made of borophosphosilicate glass. The photodiode 10 and the V-CCD 14 are generally formed in a matrix in a substrate, and the transfer gate 20 is formed above the V-CCD 14 through an insulating layer. In addition, a light-shielding layer 22 made of tungsten silicide is formed above the transfer gate 20. Please continue to refer to FIG. 3A. On the inner dielectric layer 24, a lens bearing layer 50, a polycrystalline silicon layer 52, and a silicon nitride layer 54 are sequentially deposited. (Please read the notes on the back before filling this page) -X. Order. This paper size is applicable to China National Standard (CNS) A4 (210X 297 mm) V. Description of the invention (6) The lens bearing layer 50 is an oxide layer having a thickness of about 0.1 to 1 μm, and the thickness of the polycrystalline silicon 52 depends on the thickness of the microlens. Next, a silicon oxide microlens with a biconvex surface is fabricated in a manner similar to the conventional field oxide layer formation. Referring to FIG. 3B, a conventional silicon lithography and etching process is used to define the silicon nitride layer 54 to form an opening 55 above the photodiode 10 to expose a portion of the polycrystalline silicon layer. After that, the polycrystalline silicon layer at the opening is oxidized by a wet oxidation method to form a biconvex polycrystalline oxide (Polyoxide) 56; it will be successively removed in the subsequent process The silicon nitride layer 54a and the polycrystalline silicon layer 52, therefore, the formed polycrystalline silicon oxide 56 must be connected to the lens bearing layer 50 for fixing. It can be seen from FIG. 3B that the thickness of the polycrystalline silicon layer 52 will determine the thickness of the microlenses, and the size of the opening 55 determines the diameter and width of the microlenses. Based on these two parameters, we can control the contour formed by the microlenses and adjust the required depth of focus. After the fabrication of the silicon oxide microlenses, the silicon nitride layer 54a is first stripped by wet etching, and then the polycrystalline silicon layer 52 is removed. Removal of the polycrystalline silicon layer 52 can be performed in two stages: first, the isotropic #etching is performed with the polycrystalline silicon oxide 56 as a mask, as shown in FIG. 3C; and then, wetting and immersion (wet dip) The residual polycrystalline silicon 52a under the polycrystalline silicon oxide 56 is removed in a manner such that the microlens shown in FIG. 3D is obtained. At this point, I have completed the production of a bi-convex micro-lens (Bi-Convex). To further form a single-convex micro-lens (Planar convex) as shown in Figure 3E, as long as the polycrystalline stone is spin-coated with a glass layer 58 The lower half of the oxide 56 may be covered. 8 This paper size applies to Chinese National Standard (CNS) A4 specifications (210X297). (Please read the notes on the back before filling this page) Λ. Order A7 B7 printed by the Bayong Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs Explanation (7) In order to further understand the focusing conditions of the above two types of microlenses, we performed the following simulation tests. 1. Double-convex microlenses are simulated here based on the double-convex microlenses shown in Figure 3D. Different diameter widths A are adjusted under a fixed thickness D to cause different curvature radii R !, R2, and then according to this curvature The radius is used to calculate the focal depth f, and the formula is as follows: 1 ^ = (1.46-1) (1 ^^ 1 / ^) Formula (1) The center represents the curvature radius of the upper surface of the lens; R2 represents the curvature radius of the lower surface; 1.46 = refractive index of silicon oxide; 1 = refractive index of air. The simulation results obtained are as follows: (unit: μιη) ΓΤ- Ri r2 f 0.1 0.3 0.112803 0.104537 -3.10126 [0.2 0.3 0.203712 0.215648 8.001013 丨 丨 0.3 0.3 0.32727 0.400833 6.787186 0.4 0.3 0.567348 0.660093 8.778304 0.5 0.3 0.840076 0.993426 11.83074 0.6 0.3 1.173409 1.400833 15.71236 0.7 0.3 1.567348 1.882315 20.36272 I 0.8 0.3 2.021894 2.437870 25.75980 0.9 0.3 2.537045 3.067500 31.89384 1.0 0.3 3.112803 3.771204 38.75998 It can be known from equation (1) that if the incident light is to be focused on the substrate, ReRz must be made. According to experiments, during the above process, microlenses formed by the thermal oxidation method will have 45% silicon oxide growing in polycrystalline silicon and 55% growing in air (please refer to Figure 3B), so The radius of curvature is bound to the size of this paper. Applicable to China Standards (CNS) A4 (210X297 mm) (Please read the precautions on the back before filling this page)
- yA-C 訂 經濟部中央標率局員工消费合作社印袈 A7 B7 五、發明説明（8) 小於r2，f>o，表示入射光將可通過微透鏡往基底方向聚 焦。 2、單凸面微透鏡： 此處是根據第3E圖所示之單凸面微透鏡進行模擬， 其中由於透鏡的下半部已經被SOG所覆蓋，因此R2為無 限大，因此新的計算公式為： l/f=(1.46-l)(l/R〇 式（2) 所得之模擬結果如下：（單位：μιη) A D Ri f 0.1 0.3 0.112803 0.245224 0.2 0.3 0.203712 0.442852 0.3 0.3 0.355227 0.772233 0.4 0.3 0.567348 1.233366 0.5 0.3 0.840076 1.826252 0.6 0.3 1.173409 2.550889 0.7 0.3 1.567348 3.407279 0.8 0.3 2.021894 4.395422 0.9 0.3 2.537045 5.515316 1.0 0.3 3.112803 6.766963 由以上可知，在本發明的方法中，微透鏡的曲率半徑 可藉由複晶矽層52的厚度，以及開口 55的寬度加以控制， 進而得到各種所需要的聚焦深度，此為習知的光阻式微透 鏡所沒有的優點。此外，由於氧化矽本身的折射率較光阻 材料穩定，再加上透光性與耐溫性質又較光阻為佳，因此 依本法所製得之氧化矽微透鏡更適合用在各種的影像感測 器上。 10 ϋ張尺度適用中國國家揉準（CNS ) Α4規格（210X297公着1 (請先閲讀背面之注意事項再填寫本頁) JA. A7 B7 五、發明説明（9 ) 雖然上述實施例是以CDD影像感測器為例，然其並非 用以限定本發明，根據本發明之微透鏡製程，亦可應用在 其他的光電元件或其他的光學用途上，特別是上述中具備 雙凸狀的氧化矽微透鏡，亦可應用在光纖的連結上。 雖然本發明已以一較佳實施例揭露如上，然其並非用 以限定本發明，任何熟習此技藝者，在不脫離本發明之精 神和範圍内，當可作各種之更動與潤飾，因此本發明之保 護範圍當視後附芩申請專利範圍所界定者為準。 (請先閱讀背面之注意事項再填寫本頁) 訂 Ψ 經滴部中央揉準局貝工消費合作社印製 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）-yA-C Order A7 B7 Employee Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention (8) Less than r2, f > o, which means that the incident light will be focused through the micro lens toward the substrate. 2. Single-convex micro-lens: Here is a simulation based on the single-convex micro-lens shown in Figure 3E, where the lower half of the lens has been covered by SOG, so R2 is infinite, so the new calculation formula is: l / f = (1.46-l) (1 / R0 Formula (2) The simulation results obtained are as follows: (Unit: μιη) AD Ri f 0.1 0.3 0.112803 0.245224 0.2 0.3 0.203712 0.442852 0.3 0.3 0.355227 0.772233 0.4 0.3 0.567348 1.233366 0.5 0.3 0.840076 1.826252 0.6 0.3 1.173409 2.550889 0.7 0.3 1.567348 3.407279 0.8 0.3 2.021894 4.395422 0.9 0.3 2.537045 5.515316 1.0 0.3 3.112803 6.766963 From the above, the radius of curvature of the microlens can be determined by the thickness of the polycrystalline silicon layer 52, and The width of the opening 55 is controlled to obtain various required focusing depths, which is an advantage that conventional photoresistive microlenses do not have. In addition, since the refractive index of silicon oxide itself is more stable than that of photoresistive materials, plus light transmission The performance and temperature resistance are better than the photoresist. Therefore, the silicon oxide microlens produced according to this method is more suitable for use in a variety of films. On the sensor, 10 scales are applicable to China National Standards (CNS) Α4 specifications (210X297 publication 1 (please read the precautions on the back before filling out this page) JA. A7 B7 5. Description of the invention (9) Although the above The embodiment takes the CDD image sensor as an example, but it is not intended to limit the present invention. The microlens process according to the present invention can also be applied to other optoelectronic elements or other optical applications, especially the above-mentioned dual-lens The convex silicon oxide microlens can also be applied to the connection of optical fibers. Although the present invention has been disclosed above with a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art will not depart from the present invention. Within the spirit and scope, various modifications and retouching can be made, so the scope of protection of the present invention shall be determined by the scope of the attached patent application. (Please read the precautions on the back before filling out this page) Printed on the paper printed by the Central Bureau of the Central Government of Zhuhai Bureau, Cooperate Cooperative, the size of the paper is applicable to the Chinese National Standard (CNS) A4 (210X297 mm)