201110332 六、發明說明： 【發明所屬之技術領域】 本發明有關-郷像_結構及製造像制結構的方法， 特別是有關-種能改善暗做制題的影佩測結構及製造該影像 感測結構的方法。 【先前技術】 對於數位成像裝置(例如：數位相機、數位攝影機)而言，如何 提尚影像晝質是設計上的-大重點。由傳統數位絲裝置的影像感 測器所產生的影像，其巾糾分通常較其厢部分亮，這種現象被 稱為暗影效應(lens shading effect)或周邊暗角(vignetting)現象，這是 由於當光穿透數碱像裝置的鏡職，人槪制結構的主入射角 太大，使得光感應不一致所致。發現若最大的主入射角為2〇度，則 影像周邊的亮度是影像巾㈣亮度的78%錢少。因此在習知技術 中，有各式各樣的方法來減輕暗影效應對影像的影響。 第1圖顯示一習知的CMOS影像感測器結構的剖面示意圖。 CMOS影像感測器結構10 ’其中感光二極體(ph〇t〇di〇de) 12形成於 基底(substrate) 14的表面中，三層的金屬導體層16、18、及2〇由保 護層(passivation layer) 22覆蓋’然後於各保護層22上覆蓋一介電層 24並且平坦化，再於最上層的介電層24上形成一大致上厚度均勻 且平坦的底層(under layer) 26，及於底層26上形成彩色濾光片28， 201110332 / 例如紅色濾光片30、綠色濾光片32、及藍色濾光片34。再於彩色 濾光片28上形成頂層27。於頂層(top layer) 27上形成微透鏡36。 這樣的結構’焦距的長度足以供光線聚焦於基底丨4表面中的感光二 極體12 ’因此暗影效應並不明顯。 然而，隨著數位成像裝置輕薄短小的需求，必須使焦距長度縮 短，因而主入射角(Chief-rayangle)的角度也隨之增大，於是產生暗 φ 影效應。如第2圖所示之習知CMOS影像感測器與鏡頭模組結合之 示意圖。感光二極體12位於基底14的表面15中，為了圖式簡潔易 磧，並未繪示出全部的感光二極體、各層介電層或保護層、以及内 連線。如圖所示，當焦距相對過短時，光線38自鏡頭模组4〇入射， 到達邊緣位置的光線經由邊緣的微透鏡42或43通過頂層27、彩色 遽光片28、底層26及介電層46而聚焦至八或a'點，到達中間位 置的光線經由中間的微透鏡44通過頂層27、彩色遽光片28、底層 26及介電層46而聚焦至中心的B點，可發現B點與a或A,點的位 置並非在同-平面上’ B點相對是τ凹_些深度。弧線48大致描緣 聚焦點連成的線。以-個面來看，聚焦點所形成的面是凹面，與感 光二極體12所在的基底14的水平表自15相差一個深度d，導致光 感應的不一致，使得周圍的影像較暗。 第3圖顯示-習知之改善上述暗影效應的方法，利用微透鏡位 置的往shift)及/鄉㈣光的往外遷移 (color filter shift) ’使聚焦點所形成的面與基底表面中的感光二極體 201110332 位置儘量貼合’減少位置上的差異，意即，使深度d，值(B，點)儘量為 零。但疋’隨著焦距需要更短的要求，主人射角更大，微透鏡或彩 色濾'光片遷移&距離有限’不敷墙正雜位置的差異，而仍然會有 暗影效應存在。 因此，仍需要-種新穎的影像感測器結構，能夠輕薄短小，但 不會有暗影效應。 【發明内容】 本發明之-目的是提供—種影像感測!I結構，能夠避免或減輕 暗影效應，而有一致的光感應。 依據本發明之影像感測^結構，包括—基底；—制元件陣列， 其设置於基絲面；-介電層，其覆蓋感測元件_，介電層包括 一上表面，上表面包括一凹盤結構；一底層，其填入於凹盤結構， 底層具有-折卿，折射枝於介電賴折射率；_濾、光片陣列， 设置於底層上’對麟測树陣列；及一微透鏡陣列，對應設置於 濾光片陣列上。 於本發明之另-方面，依據本發明之製造影像感測器結構的方 法’包括下列步驟。首先’提供_基底；於基底表面形成一感測元 件陣列。接著’形成一介電層覆蓋於感測元件陣列及基底上；將介 電a的上表成—凹盤結構。然後，於凹盤結構巾填人一底層， 201110332 底層具有-折射率，其大於介電層的折射率；於底層上形成一遽光 片陣列；於滤光片陣列上形成一微透鏡(micr〇lens)陣列。 依據本發明之影像感測器結構，在濾光片陣列下方設置的底層 是填充於其下方的介電層的凹盤形狀表面中，並且選擇底層材料， 使得所形成的底層除了具有黏著與提供平坦表面的功能之外，尚具 有如同凸透鏡的角色一樣，可對感測區中心與邊緣之間的焦距差異 Φ 予以補償。因此，於同一晶片中的光感應會更均勻。 【實施方式】 本發明之影像感測器結構可應用於CMOS影像感測元件(CIS) 或電荷耦合元件(CCD)。第4圖顯示一依據本發明之影像感測器結 構的具體實施例的剖面示意圖。如第4圖所示，影像感測器結構5〇 包含有一基底52、一感測元件陣列54、介電層56、一底層6〇、一 濾光片陣列62、及一微透鏡陣列64。基底52可為例如半導體基底。 ® 感測元件陣列54設置於基底52表面’是一陣列的感測元件^可依 元件性質或設計設置於基底表面中或上’並無特別限制。感測元件 可為例如光感測元件’光感測元件包括例如感光二極體。介電声56 覆蓋感測元件陣列54及基底52上。介電層56包括一上表面％， 上表面58往下凹，而形成淺凹盤狀或凹盤狀，即，為—凹盤并構。 介電層56中可進一步設置有複數層金屬内連線，可做為導電或是遮 光之用。 201110332 一底層材料填入於上表面58所形成的凹盤結構中，形成底層 60。底層60具有一折射率，此折射率可大於介電層允的折射率， 較佳為稍大於介電層56的騎率，並可進—步小於微透鏡的折射 率。例如折射率可在約1.5至ι_6之間，但不限於此，可依整體光學 性質而定。使用於本發明的底層材料，除了折射率的要求之外較 佳具有高光穿透度，進-步具有黏著層的功能以將濾光片與介電層 56黏合在一起，及進一步具有平坦化功能以提供平坦化表面供濾光 片"又置。再者，為便利製造，較佳為適合填入製程的材質。可於習 知的濾光片底層或頂層材料中挑選出，但不限於此。可舉例有例如 聚合物’其為例如壓克力聚合物，但不限於此。濾光片陣列62設置 於底層60上，位置對應於感測元件陣列54。濾光月陣列62可由複 數個濾光片排列而成，濾光片可為例如彩色濾光片或非彩色濾光 片，依產品所需而定。微透鏡陣列64則是對應設置於濾光片陣列 62上。 上述介電層56的上表面58形成的凹盤結構，其凹下的深度可 依據光學性質及所需而定，因為形成的底層即具有類似凸透鏡的性 質’因此’可搭配例如影像感測器結構的感測區域的尺寸及光學設 計、底層、微透鏡、及濾光片等的光學性質，以決定凹盤結構恰當 的深度。以數位成像裝置整體而言，使光圈數值與凹盤深度互相配 合’俾使光線均聚焦在感測元件陣列的同一平面上，可減少各感應 單元光感應的差異。 201110332 再者’依據本發明之影像感測器結構，可進一步包括一遮蔽層， 其具有遮光效果，遮蔽層係設置於介電層内，並圍繞凹盤結構。遮 蔽層可包括金屬材料，例如Ti或TiN，較介電層硬，因此在製造依 據本發明之影像感測器結構時，可利用此遮蔽層與介電層對€]^ (chemical mechanic polishing，化學機械研磨)的選擇比不同，使得介 電層在經過CMP後具有盤凹的表面。在感測區周圍部分因為有遮 蔽層的支撐，被磨除的深度較淺，在感光區中心部分因為是大片的 • 介電層中心，CMP的盤凹效應(dishingeffect)明顯，容易被磨凹，所 以被磨除的深度較深。遮蔽層可包括至少一環形結構。當遮蔽層包 括多層環形結構時，其分佈密度可由感測區外圍往中心逐漸呈梯度 狀(gradient)減少。或者，遮蔽層可為複數個不連續的區段狀其分 佈密度也可由外往中心逐漸減少。 再者’依據本發明之影像感測器結構可進—步包括—頂層的， 其設置於濾'光>1上而包魏光片，頂層可包括與底層綱的材料， 並可使濾光片陣列上方形成平坦表面。 第5圖顯示依據本發明的影像感測器結構與 士 意圖》如圖所示，光線38自鏡賴組40人射，到達邊緣== 線經由邊緣的微透鏡64a或64b通過頂層66、遽光片62、底層6〇、 及介電層56而聚焦至，點，到達中間位置的光線經由中曰間的 微透鏡咖通過頂層66 '據光片62、底層6〇、及介電層兄而聚焦 至E點，可發現E點與C或c，點即沿著感測二所在祕 201110332 底52的水平表面55。因此，對於感光二極 ，下’不同的位置可獲得相_光電轉換效率5,因Γ在圍 與中心顯現的影像亮度均句。 在〜像周圍 於本發明中’除了可將介電層形成 盤構以填入底層材料以 改交先子折射路佐之外，可再搭配習知 Α闰骷-认分祕丄々 旳調I焦距的方法。例如第 6醜不的依據本發明之影像感 光片72往外遷移設置，或是將部分微透t ^亦可進一步將部分遽 心上μ/ “ 又疋料刀微透鏡74往内遷移設置，亦即， 使獻片位在邊緣的排列節距與位在中心的節 透鏡位在細綱物咐,__。赫== 鏡陣列邊緣的微透細谢心观透鏡形狀利目同。如此，可進 -步調整焦距’而補強焦距所在平面與感測树平面的貼合。 -依據^發明之影像❹m結構可由下財法製得。如第7圖所 示首先提供基底52，然後於基；^表面形成一感測元件陣列％， 即’可於基底52表面中壯形成感測元件。織，形成一介電層 56覆蓋感測疋件陣列54及基底52。例如利用化學氣相沉積製程形 成此介電層56。或者，可於介電層兄中進-步形成複數個金屬層， 例如利用金勒連線製程來達成。可進—步沈積—由氮化碎或氧化 石夕等材料所構成的保護層0assivati〇n layer)覆蓋於金屬層。 然後’將介電層56的上表面58形成一凹盤結構，此可利用一 CMP製程研磨介電層允而達成。例如，在介電層是氧化物材料時， 201110332 鞭用研磨氧化物的CMP製程。可依所需調整⑽處方斑條件， 加研磨時間’可增加下凹程度。另可於介電層％内之圍繞預 二凹盤結構之處進—步形成一遮蔽層8G結構，於研磨時補強周圍強 缺研磨速率。第8圖顯示―平面圖，舉例說明遮蔽層8〇 ^感測請陣列54的情形，最外圍是切割道82，供影像感測器 、⑺構製做完成後切割為單獨的晶片之用。或可進—步於進行 製程之前，先於介電層56上之圍繞凹盤結構之處形成一研磨停止 (polishing卿)層，此可防止遮蔽層8〇上方的介電層被磨除而露出 遮蔽層80。若遮蔽層8〇是金屬材料而露出，易對元件造成污染。 再者’介電層的上表面形成凹盤結構，可利用CMp製程以及 蝕刻製程混合搭配而完成。詳言之，於進RCMp製程研磨介電層 56，使介電層56的上表面58形成盤凹之後，可進行蝕刻製程對介 電層56已盤凹的表面進行姓刻，由於钱刻速率與姓刻時間較容易控 制，所以可便利而準確的控制上表面所形成的凹盤結構的深度。 如第9圖所示，形成凹盤結構後，於凹盤結構中填入一底層材 料形成底層60。例如，使用一以單f基縫丙二醇乙酸酯(pGMEA，201110332 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to a structure and a method for manufacturing an image structure, and more particularly to a shadow-detecting structure capable of improving a dark-made problem and manufacturing the image. The method of measuring the structure. [Prior Art] For digital imaging devices (eg, digital cameras, digital cameras), how to improve image quality is a design-big focus. The image produced by the image sensor of the conventional digital wire device is usually brighter than the part of the car. This phenomenon is called a lens shading effect or a vignetting phenomenon. Since the main incident angle of the man-made structure is too large when the light penetrates the mirror of the alkali image device, the light induction is inconsistent. It is found that if the maximum principal incident angle is 2 degrees, the brightness around the image is less than 78% of the brightness of the image (4). Therefore, in the prior art, there are various methods to mitigate the effect of the shadow effect on the image. Figure 1 shows a cross-sectional view of a conventional CMOS image sensor structure. a CMOS image sensor structure 10' in which a photodiode 12 is formed in the surface of a substrate 14, and the three layers of metal conductor layers 16, 18, and 2 are protected by a protective layer. (passivation layer) 22 covers 'and then covers a dielectric layer 24 on each of the protective layers 22 and planarizes, and then forms a substantially uniform and flat under layer 26 on the uppermost dielectric layer 24. A color filter 28, 201110332 / for example, a red filter 30, a green filter 32, and a blue filter 34 are formed on the bottom layer 26. A top layer 27 is formed on the color filter 28. A microlens 36 is formed on the top layer 27. The length of such a structure 'focal length is sufficient for the light to be focused on the photosensitive diode 12' in the surface of the substrate 4, so the shadow effect is not significant. However, with the demand for light and thin digital imaging devices, the focal length must be shortened, and the angle of the chief incident angle (Chief-ray angle) is also increased, thus producing a dark φ shadow effect. A schematic diagram of a conventional CMOS image sensor and a lens module as shown in FIG. The photodiode 12 is located in the surface 15 of the substrate 14. For the sake of simplicity and simplicity, all of the photodiodes, the dielectric layers or protective layers, and the interconnects are not shown. As shown, when the focal length is relatively short, the light 38 is incident from the lens module 4, and the light reaching the edge position passes through the edge microlens 42 or 43 through the top layer 27, the color light sheet 28, the bottom layer 26, and the dielectric. The layer 46 is focused to the eight or a' point, and the light reaching the intermediate position is focused to the center point B through the middle microlens 44 through the top layer 27, the color light sheet 28, the bottom layer 26 and the dielectric layer 46, and B can be found. Point and a or A, the position of the point is not in the same plane - point B is opposite to τ concave_ some depth. The arc 48 is roughly drawn to the line where the focus points are connected. In a face view, the face formed by the focus point is a concave surface, and the horizontal table of the substrate 14 where the photodiode 12 is located is different from the depth d by 15 by a depth d, resulting in inconsistent light sensing, making the surrounding image dark. Figure 3 shows a conventional method for improving the above-mentioned shadow effect, using the shift of the microlens position and/or the color filter shift of the (four) light to make the surface formed by the focus point and the photosensitive surface in the surface of the substrate. Polar body 201110332 position as close as possible to reduce the difference in position, that is, to make the depth d, the value (B, point) as zero as possible. However, as the focal length requires shorter requirements, the master's angle of incidence is larger, and the microlens or color filter 'light shift & distance is limited' to the difference in the position of the wall, and there is still a shadow effect. Therefore, there is still a need for a novel image sensor structure that can be light, thin, and short, but without the shadow effect. SUMMARY OF THE INVENTION It is an object of the present invention to provide an image sensing!I structure that avoids or mitigates shadow effects with consistent light sensing. The image sensing structure according to the present invention comprises: a substrate; an array of elements disposed on the base surface; a dielectric layer covering the sensing element, the dielectric layer comprising an upper surface, the upper surface comprising a a concave disk structure; a bottom layer, which is filled in the concave disk structure, the bottom layer has a folding, the refractive index is on the dielectric refractive index; the filter, the light film array is disposed on the bottom layer, and the array of the tree is measured; The microlens array is correspondingly disposed on the filter array. In still another aspect of the invention, a method of fabricating an image sensor structure in accordance with the present invention includes the following steps. First, a substrate is provided; an array of sensing elements is formed on the surface of the substrate. Next, a dielectric layer is formed overlying the array of sensing elements and the substrate; the upper surface of dielectric a is formed into a concave disk structure. Then, the bottom surface of the concave disk structure is filled with a bottom layer having a refractive index greater than the refractive index of the dielectric layer; an array of light-emitting sheets is formed on the bottom layer; and a microlens is formed on the filter array (micr 〇lens) array. According to the image sensor structure of the present invention, the underlayer disposed under the filter array is filled in the concave shaped surface of the dielectric layer underneath, and the underlying material is selected such that the formed underlayer is adhered and provided. In addition to the function of the flat surface, it has the same function as the convex lens, which compensates for the focal length difference Φ between the center and the edge of the sensing area. Therefore, the light induction in the same wafer will be more uniform. [Embodiment] The image sensor structure of the present invention can be applied to a CMOS image sensing element (CIS) or a charge coupled device (CCD). Figure 4 is a cross-sectional view showing a specific embodiment of an image sensor structure in accordance with the present invention. As shown in FIG. 4, the image sensor structure 5A includes a substrate 52, a sensing element array 54, a dielectric layer 56, a bottom layer 6A, a filter array 62, and a microlens array 64. Substrate 52 can be, for example, a semiconductor substrate. The sensing element array 54 is disposed on the surface of the substrate 52. The sensing element is an array. The sensing element can be disposed in or on the surface of the substrate according to the nature or design of the element. There is no particular limitation. The sensing element can be, for example, a light sensing element. The light sensing element includes, for example, a photodiode. Dielectric sound 56 covers the array of sensing elements 54 and substrate 52. The dielectric layer 56 includes an upper surface %, and the upper surface 58 is recessed to form a shallow concave disk or a concave disk shape, that is, a concave disk. A plurality of metal interconnects may be further disposed in the dielectric layer 56 for conducting or shielding. 201110332 An underlying material is filled into the recessed disk structure formed by the upper surface 58 to form the bottom layer 60. The bottom layer 60 has a refractive index which is greater than the refractive index of the dielectric layer, preferably slightly greater than the ride of the dielectric layer 56, and which is further less than the refractive index of the microlens. For example, the refractive index may be between about 1.5 and ι_6, but is not limited thereto and may depend on the overall optical properties. The underlayer material used in the present invention preferably has high light transmittance in addition to the refractive index requirement, and further has an adhesive layer function to bond the filter and the dielectric layer 56 together, and further has planarization. Function to provide a flattened surface for the filter " Furthermore, in order to facilitate the manufacture, it is preferably a material suitable for filling in the process. It can be selected from conventional filter underlayer or top layer materials, but is not limited thereto. For example, a polymer 'which is, for example, an acrylic polymer, may be exemplified, but is not limited thereto. The filter array 62 is disposed on the bottom layer 60 at a position corresponding to the sensing element array 54. The filter array 62 can be arranged by a plurality of filters, such as color filters or non-color filters, depending on the desired product. The microlens array 64 is correspondingly disposed on the filter array 62. The recessed disk structure formed by the upper surface 58 of the dielectric layer 56 may have a concave depth depending on optical properties and requirements, because the formed bottom layer has a convex lens-like property. Therefore, it can be matched with, for example, an image sensor. The dimensions of the sensing region of the structure and the optical properties of the optical design, underlayer, microlenses, and filters, etc., determine the proper depth of the concave disk structure. With the digital imaging device as a whole, the aperture value and the concave disk depth are matched to each other', so that the light is focused on the same plane of the sensing element array, which can reduce the difference in light sensing of each sensing unit. 201110332 Further, the image sensor structure according to the present invention may further include a shielding layer having a light shielding effect, the shielding layer being disposed in the dielectric layer and surrounding the concave disk structure. The shielding layer may comprise a metal material, such as Ti or TiN, which is harder than the dielectric layer. Therefore, when the image sensor structure according to the present invention is fabricated, the shielding layer and the dielectric layer can be utilized. The chemical mechanical polishing has a different selection ratio such that the dielectric layer has a concave surface after passing through the CMP. The depth around the sensing area is shallow due to the support of the shielding layer. In the center of the photosensitive area, because of the large dielectric layer center, the CMP has a significant dishing effect, which is easy to be ground. So the depth of being removed is deeper. The shielding layer can include at least one annular structure. When the shielding layer comprises a multi-layered annular structure, its distribution density can be gradually reduced from the periphery of the sensing region toward the center. Alternatively, the shielding layer may be in the form of a plurality of discrete segments, and the distribution density may be gradually reduced from the outer to the center. Furthermore, the image sensor structure according to the present invention can further include a top layer, which is disposed on the filter 'light> 1 and includes a Weiguang sheet. The top layer can include the material of the bottom layer and can be filtered. A flat surface is formed over the array of light sheets. Figure 5 shows the structure of the image sensor according to the present invention. As shown in the figure, the light 38 is emitted from the mirror group 40, and the edge == line passes through the edge of the microlens 64a or 64b through the top layer 66, The light sheet 62, the bottom layer 6〇, and the dielectric layer 56 are focused to, and the light reaching the intermediate position passes through the top layer 66' of the microlens between the middle ridges, the light film 62, the bottom layer 6 〇, and the dielectric layer brother. Focusing on point E, E point and C or c can be found, and the point is along the horizontal surface 55 of the bottom 52 of the sensing second secret 201110332. Therefore, for the photosensitive diode, the phase-to-photoelectric conversion efficiency of 5 can be obtained at the lower position, because the brightness of the image appearing around the center is uniform. In the present invention, in addition to the fact that the dielectric layer can be formed into a disk structure to fill the underlying material to change the refractory road, the accompaniment can be used. I method of focal length. For example, the 6th ugly image sensor 72 of the present invention is moved outwardly, or a part of the micro-transparent t ^ can further move the part of the core/μ" and the micro-lens 74 inwardly, that is, , the arrangement of the film at the edge of the pitch and the position of the lens at the center of the lens in the fine structure, __. Her == the edge of the mirror array is slightly transparent and the shape of the lens is the same. So, you can proceed Adjusting the focal length' and reinforcing the fit of the plane of the focal length and the plane of the sensing tree. - According to the image of the invention, the structure of the structure 可由m can be obtained by the following method. As shown in Fig. 7, the substrate 52 is first provided, and then a surface is formed on the surface of the surface. The sensing element array %, ie, can form a sensing element in the surface of the substrate 52. The dielectric layer 56 is formed to cover the sensing element array 54 and the substrate 52. The dielectric is formed, for example, by a chemical vapor deposition process. Layer 56. Alternatively, a plurality of metal layers may be formed in the dielectric layer brother, for example, by using a Jinle wire process. Further deposition may be performed by a material such as nitride or oxidized stone. Protective layer 0assivati〇n layer) covered with metal Then, the upper surface 58 of the dielectric layer 56 is formed into a concave disk structure, which can be achieved by a CMP process polishing dielectric layer. For example, when the dielectric layer is an oxide material, 201110332 whip abrasive oxide The CMP process can be adjusted according to the requirements of (10) the prescription spot condition, and the grinding time can be increased to increase the degree of undercut. Alternatively, a shielding layer 8G structure can be formed in the dielectric layer % around the pre-two concave disk structure. In the case of grinding, the grinding rate is increased by the surrounding strength. Figure 8 shows the "plan view", illustrating the case where the shielding layer 8 〇 ^ senses the array 54 , and the outermost periphery is the cutting channel 82 for the image sensor, (7) After completion, the film is cut into individual wafers. Alternatively, a polishing layer can be formed on the dielectric layer 56 around the concave disk structure before the process is performed, which prevents the shielding layer 8 from being formed. The dielectric layer above the crucible is removed to expose the shielding layer 80. If the shielding layer 8 is exposed by a metal material, the component is easily contaminated. Further, the upper surface of the dielectric layer is formed into a concave disk structure, and the CMp process can be utilized. And etching process mix and match In detail, after the RCMp process is used to polish the dielectric layer 56 so that the upper surface 58 of the dielectric layer 56 is recessed, an etching process can be performed to etch the surface of the dielectric layer 56 which has been recessed. The engraving rate and the engraving time are easier to control, so the depth of the concave disc structure formed by the upper surface can be conveniently and accurately controlled. As shown in Fig. 9, after forming the concave disc structure, a bottom layer is filled in the concave disc structure. The material forms the bottom layer 60. For example, a single f-base propylene glycol acetate (pGMEA,
Propyleneglycol methyletheracetate)及乙氧基丙酸乙酯(ethyl 3-ethoxypropi〇nate，EEP)做為溶劑的聚合物材質，利用旋轉塗佈法 形成於凹盤結構中，乾燥後形成底層60，具有約95%的光穿透度。 底層填入於凹盤結構中並不侷限於要填滿或不填滿，可以實際需要 而定。圖式中的底層60是填滿凹盤結構，並溢出而覆蓋介電層56 11 201110332 的原始表面。 然後’如第10圖所示’於底層60上形成一遽光片陣列62。可 進一步於濾光片陣列62上形成一頂層66 ’其材質可與底層6〇相 同’也可利用塗佈的方法形成。然後，於濾光片陣列62上形成一微 透鏡陣列64 ’使各濾光片分別對應於一微透鏡，並且對應於一感測 tf 〇 一 早兀。 由於，於本發明之方法t，相較於習知方法，特徵是進行一 CMP 鲁 研磨以形成凹盤結構，及於凹盤結構中填入底層材料，形成底層， 底層同時具有調整焦距、平坦化、及黏著濾光片的功能，一舉數得。 底層材料可於習知用於濾光片的底層及頂層的材料中挑選(但不限 於此）’再者，習知於介電層形成時，亦是需要經過平坦化，而於本 發明中，恰可利用其CMP平坦化的製程得到所欲的凹盤結構，因 此在製程上並不會增加繁瑣的步驟或多花費的材料，相當便利。 以上所述僅為本發明之較佳實施例，凡依本發明申請專利範圍 所做之均等變化與修飾，皆應屬本發明之涵蓋範圍。 【圖式簡單說明】 第1圖顯示-習知的CMOS影像感測器結構的剖面示意圖。 第2圖顯示-習知CM0S錄感測器與鏡職組結合之示音圖。 第3圖顯示一習知的改善暗影效應的方法。 心。 12 201110332 ^圖_本發私飾❹⑻ 意圖。 再幻具體實施例的剖面示 圖第。5圖顯示依縣發明的影像感崎結構與鏡頭模組結合之示意 面不意圖。 第7至1〇圖 的不意圖。 Ϊ2，—依據本發明之影像__構之 說明依據本翻之製造影像—#構的方法的各階段Propyleneglycol methyletheracetate) and ethyl 3-ethoxypropi〇nate (EEP) as a solvent polymer material, formed by a spin coating method in a concave disk structure, and dried to form a bottom layer 60 having about 95 % light penetration. The bottom layer is filled in the concave disk structure and is not limited to being filled or not filled, and may be practically required. The bottom layer 60 in the drawing is filled with a concave disk structure and overflows to cover the original surface of the dielectric layer 56 11 201110332. Then, as shown in Fig. 10, an array of illuminating sheets 62 is formed on the underlayer 60. Further, a top layer 66' may be formed on the filter array 62, the material of which may be the same as that of the bottom layer 6'. It may also be formed by a coating method. Then, a microlens array 64' is formed on the filter array 62 such that each filter corresponds to a microlens, respectively, and corresponds to a sense tf 〇 兀 兀. Therefore, in the method t of the present invention, compared with the conventional method, a CMP is performed to form a concave disk structure, and an underlying material is filled in the concave disk structure to form a bottom layer, and the bottom layer has an adjusted focal length and a flat surface. The function of the filter and the adhesive filter is achieved in one fell swoop. The underlying material may be selected from, but not limited to, the materials used for the underlayer and the top layer of the filter. Further, in the case of the formation of the dielectric layer, it is also required to be planarized, and in the present invention The CMP flattening process can be used to obtain the desired concave disk structure, so that the process does not add cumbersome steps or costly materials, which is quite convenient. The above are only the preferred embodiments of the present invention, and all changes and modifications made to the scope of the present invention should fall within the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows a schematic cross-sectional view of a conventional CMOS image sensor structure. Figure 2 shows the sound map of the CM0S recording sensor combined with the mirror group. Figure 3 shows a conventional method of improving the shadow effect. heart. 12 201110332 ^图_本发私饰❹(8) Intention. A cross-sectional view of a specific embodiment is shown. Figure 5 shows the schematic diagram of the combination of the image sensing structure and the lens module invented by the county. The intention of Figures 7 to 1 is not intended. Ϊ2, the image according to the present invention, the description of the method according to the present invention
【主要元件符號說明】 10 CMOS影像感測器結構 12 感光二極體 14 基底 15 表面 16、18、20 金屬導體層 22 保護層 24 介電層 26 底層 27 頂層 28 彩色濾光片 30 紅色濾光片 32 綠色濾光片 34 藍色遽光片 36 微透鏡 38 光線 40 鏡頭模組 42、43、44 微透鏡 46 介電層 48 娘線 50 影像感測器結構 52 基底 54 感測元件陣列 55 水平表面 56 介電層 201110332 58 上表面 60 底層 62 濾光片陣列 64 微透鏡陣列 64a、64b、64c 微透鏡 66 頂層 70 影像感測1§結構 72 濾光片 74 微透鏡 80 遮蔽層 82 切割道[Main component symbol description] 10 CMOS image sensor structure 12 Photosensitive diode 14 Substrate 15 Surface 16, 18, 20 Metal conductor layer 22 Protective layer 24 Dielectric layer 26 Bottom layer 27 Top layer 28 Color filter 30 Red filter Sheet 32 Green filter 34 Blue enamel sheet 36 Microlens 38 Light 40 Lens module 42, 43, 44 Microlens 46 Dielectric layer 48 Niang line 50 Image sensor structure 52 Substrate 54 Sense element array 55 Level Surface 56 dielectric layer 201110332 58 upper surface 60 bottom layer 62 filter array 64 microlens array 64a, 64b, 64c microlens 66 top layer 70 image sensing 1 § structure 72 filter 74 microlens 80 shielding layer 82 cutting channel
1414