509975 7527twf.doc/〇〇6509975 7527twf.doc / 〇〇6
發明說明（ 經濟部智慧財產局員工消費合作社印製 '又月疋：有關方< 〜種微影製程（photolithography Process)，且特別是有關於一種預測曝光能量 Energy)的方法。 隨者積體電路之積集度的提高，整個積體電路之元件 尺、于也必須隨之縮小。而在半導體製程中最舉足輕重的可 說是微影製程，凡是與金氧半導體（Metal-Oxide-Semkonductor ; M0S)元件結構相關的，例如··各層薄膜 的圖案（Pattern)，及摻有雜質（Dopants)的區域，都是由微 影适個步驟來決定的。 習知對於微影製程之關鍵尺寸（Critical Dimension，CD) 之控制’係於進行曝光製程之後，依照其最後所量測之關 鍵尺寸的結果，來對曝光能量進行補償，以使下一批次晶 圓之關鍵尺寸値能更接近目標値。 然而’習知之方法係於得到曝光圖案之關鍵尺寸量測 結果之後，才對下一批次之晶圓之曝光能量進行補償，如 此將犧牲該批次之晶圓。此外，習知之方法對於一新產品 之曝光條件，僅能比照相同或類似製程產品，或由經驗判 斷以給定其曝光能量値，而此方式並未將光罩關鍵尺寸之 差異與線上產品之關鍵尺寸偏差趨勢（CD Baseline)考慮進 去，因此常常造成最後曝光圖案之關鍵尺寸無法達到目標 値，甚至超過R]忍受之偏差値，而必須重新修改該產品之 曝光條件。再者，習知之方法對於不同批次之材料（例如 光阻材料），並無法預測其對關鍵尺寸之影響，而事先瞭 解並加以補償。 -----------AW ^ · 11-----訂--------- (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準（CNS)/V1規烙（210x297公釐） 509975 7527twf.doc/006 A7 B7 五 發明說明（Z) 因此，本發明的目的就是在提供一種頊_曝 光能 的 行曝 方法，利用事先預測補償的方式，以避免_ 光之晶圓。 本發明的另一目的是提供一種預測曝光能數 利用事先預測補償之方式，以減少最後曝光、方法， 寸之變異，並使曝光圖案之關鍵尺寸更接近自擦 本發明的再一目的是提供一種預測曝光能鐵 以改善習知方法無法對新的產品預測其曝光能鐵。'方法， 本發明提出一種預測曝光能量的方法，此方法 依照一欲曝光層之一光罩關鍵尺寸計算一欲 後由此欲曝光層之一'前層薄膜厚度’以得知\ _ % 之 寸差異値，並由第一關鍵尺寸差異値計算出 償値。接著查詢一光阻靈敏度之前後批次笺幾値_ ^*補 1%。倘若光阻靈敏度之前後批次差異値小於1%&杏小於 光能量與第一能量補償値之加總即爲此欲曝光騰則欲曝 能量。否則依照光阻靈敏度之前後批次差異偷 曝光 ，再由第二關鍵尺寸差異値 保首先 裊 經濟部智慧財產局員工消費合作社印製 能量補償値，而欲曝光能量、第一能量補償値與第 補償値之加總係爲欲曝光層之曝光能量。 本發明利用預先補償之方式，可減少曝光圖案之關鍵 尺寸之變異，並使曝光圖案之關鍵尺寸較接近目標値。 本發明利用事先預測曝光能量之方法，可避免習知方 法中僅能對下一批次之晶圓進行補償，而必須犧牲該批次 之晶圓。 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公爱） ------------t--------^-------------.---*_____ (請先閱讀背面之注咅？事項再填寫本頁) 509975 經濟部智慧財產局員工消費合作社印製 7 527twf. doc/00 6 五、發明說明（> ) 本發明可較精確的預測新產品之曝光能量，因此可改 善較習知之方法中無法對新產品預測其曝光能量。 爲讓本發明之上述和其他目的、特徵、和優點能更明 顯易懂，下文特舉一較佳實施例，並配合所附圖式，作詳 細說明如下： 圖式之簡單說明： 第1圖爲依照本發明一較佳實施例之預測曝光能量的 方法之流程圖； 第2圖是光罩關鍵尺寸與曝光能量之線性回歸關係 圖； 第3圖是薄膜沈積厚度與關鍵尺寸之關係曲線圖； 第4圖是光阻靈敏度與關鍵尺寸之線性回歸關係圖； 以及 第5圖關鍵尺寸與曝光能量之線性回歸關係圖。 圖式之標示說明： 100至112 :步驟 實施例 第1圖所示，其繪示爲依照本發明一較佳實施例之預 測曝光能量的方法之流程圖；第2圖所示，其繪示爲光罩 關鍵尺寸與曝光能量之線性回歸關係圖。 請先參照第1圖，本實施例用以預測曝光能量之方法 係首先依照一欲曝光層之一光罩關鍵尺寸，計算出一欲曝 光能量（步驟1 〇 〇)。其中計算一欲曝光能量之方法請爹照 第2圖，第2圖係將光罩之關鍵尺寸與曝光能量之歷史資 -----------裝--------訂---------^9. (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準（CNS)A4規格（210 x 297公f ) 509975 經濟部智慧財產局員工消費合作社印製 7527twf.doc/006 A7 五、發明說明（f ) 料，製作成一線性回歸關係圖，其中，橫座標係爲光罩關 鍵尺寸之大小，而縱座標係爲曝光能量値。因此，將欲曝 光層之光罩關鍵尺寸代入第2圖中之線性回歸方程式，可 得到一曝光能量値，此曝光能量値即爲對欲曝光層進行曝 光步驟的一欲曝光能量。 接著，請繼續參照第1圖，在計算出欲曝光能量之後， 進行步驟102,即由欲曝光層之前層薄膜沈積厚度推算一 關鍵尺寸差異値，並藉由此關鍵尺寸差異値計算出一第一 能量補償値。意即利用此欲曝光層之前層薄膜沈積厚度與 關鍵尺寸規格之目標値，來推算一關鍵尺寸差異値，再由 所推算出之關鍵尺寸差異値求出第一能量補償値。 第3圖所示，其繪示爲薄膜沈積厚度與關鍵尺寸之關 係曲線圖。 由前層薄膜厚度推算一關鍵尺寸差異値之方法請參照 第3圖。第3圖係將薄膜沈積厚度與關鍵尺寸之歷史資料 製作成一曲線圖，其中橫座標係爲薄膜之沈積厚度，而縱 座標係爲關鍵尺寸。利用第3圖可由前層之薄膜沈積厚度 得到前層薄膜之一關鍵尺寸値，將前層薄膜之關鍵尺寸値 減去前層薄膜關鍵尺寸規格之目標値之中心値，即爲關鍵 尺寸差異値。例如，前層薄膜沈積厚度之規格目標値爲8000 ± 800埃，其中心値爲8000埃，在第3圖中所對應之關鍵 尺寸値爲0.46微米，意即前層薄膜之關鍵尺寸規格目I標値 係爲0.46微米。而實際欲曝光層之前層沈積厚度爲8300 埃，在第3圖屮所對應之關鍵尺寸値係爲0.48埃。由此可 -ϋ ϋ n ϋ ϋ ϋ ϋ I ^1 · ϋ ϋ n n n ϋ n 一 · ·ϋ ·ϋ n 1 n ·ϋ n I #, (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐） 509975 7527twf.doc/006 A7 ms B7 五、發明說明（孓） 知’可得知關鍵尺寸差異値係爲〇 〇2微米（0.48微米-0.46 微米=0.02微米）。 第5圖所$，其繪示爲關鍵尺寸與曝光能量之線性回 歸關係圖。 之後’由第3圖所求出之關鍵尺寸差異値計算出第一 能量補償値’其方法係利用第5圖之關鍵尺寸與曝光能量 之線性回歸關係方程式以求出第一能量補償値，其中第5 圖中之橫座標係爲關鍵尺寸値，而縱座標係爲曝光能量 値。利用第5圖求得第一能量補償値之方法係將於第3圖 中所求出之關鍵尺寸差異値代入第5圖之線性回歸方程式 之中，即可求得第一能量補償値。例如，將上述第3圖所 求出之關鍵尺寸_値G.G2微米以第5圖中之線性回歸 (Υ-144〇·7Χ+15_〇12)，可求出第—能量補償値爲43似 之二第1圖，查詢光阻靈敏度前後批 技異疋《灿1%(頻刚）。倘若步驟ig 每^，批次之差異是小於1%，_行步驟m，= ;:Γ一膜口光能量與步驟102中之第-能量補償値 帕疋-曝如⑼，葛即將步驟1GQ ^ 加上步驟1。2之所求出之第—能量補償値即== 之後，進仃抓m ’即量測曝光圖 = 認曝光圖案之_尺寸之偏差値是否小於;^穴以 値。 夂又乙 ϋ ϋ n im ϋ ϋ ϋ ϋ n I n · n alai I n n n Hi 一一口V I I n an mM— §MB (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 509975 7 52 7twf·doc/006 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明（彡） 倘若步驟104中之光阻靈敏度前後批次之差異大於 1 %，則進行步驟1 〇 6，即依照光組靈敏度之前後批次差異 値計算出一第二關鍵尺寸差異値。 第4圖所示，其繪示爲光阻靈敏度與關鍵尺寸之線性 回歸關係圖。 計算第二關鍵尺寸差異値之方法，係利用第4圖之光 阻靈敏度與關鍵尺寸線性回歸方程式，以求出第二關鍵尺 寸差異値。第4圖中之橫座標係爲光阻靈敏度，縱座標係 爲關鍵尺寸，將上一批次光阻材料之光阻靈敏度代入第4 圖之線性回歸方程式，即可計算出使用上批次光阻材料之 一關鍵尺寸値。將該批次光阻材料之光阻靈敏度代入第4 圖之線性回歸方程式可得到該批次光阻材料之一關鍵尺寸 値，然後將該批次光阻材料之關鍵尺寸値減去上一批次之 光阻材料之關鍵尺寸値即爲一第二關鍵尺寸差異値。例 如，上一批次光阻材料之光阻靈敏度爲97%，其所計算出 之關鍵尺寸値爲0.492微米，而該批次光阻材料之光阻靈 敏度爲99%，其所計算出之關鍵尺寸値爲0.477微米，因 此，可得知第二關鍵尺寸差異値爲-0.015微米（0.477微米-0.492 微米二-0.015 微米）。 之後，請參照第1圖，由第二關鍵尺寸差異値以計算 出一第二能量補償値（步驟108)，其中由第二關鍵尺寸差 異値以計算出一第二能量補償値，係將第二關鍵尺寸差異 値代入第5圖之線性回歸方程式，以求出第二能量補償値。 例如，將上述步驟106所求出第二關鍵尺寸差異値-0.015 8 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公爱） (請先閱讀背面之注意事項再填寫本頁) · ϋ n ϋ n n n an 一：0V n n I ί I ϋ I I · ^U9975 A7 B7 ^527twf.doc/006 五、發明說明（q) 微米代入第5圖之線性回歸關係（Υ=1440·7Χ+15.012)中可 得知第二能量補償値爲-6.5985 ms。 接著，進行步驟11〇，即由步驟1〇〇中之所求出之欲 曝光能量、步驟102中所求出之第一能量補償値以及步驟 108中所求出之第二能量補償値給定一曝光能量。意即欲 曝光能量、第一能量補償値與第二能量補償値之加總，即 爲欲曝光層進行曝光製程所需之曝光能量條件。 最後，於進行曝光製程之後，量測曝光圖案之關鍵尺 寸（步驟112)，以確認曝光圖案之關鍵尺寸之偏差値是否 小於可接受之偏差値。 由最後之曝光圖案之關鍵尺寸之量測結果可發現，本 發明所預測出之曝光能量條件，可使曝光圖案之關鍵尺寸 較接近目標値，並且可使曝光圖案之關鍵尺寸皆在可接受 之偏差値內，因此，可避免習知無法事先預測該次曝光製 程所應設定之曝光能量’而僅能於下一批晶圓進行補償， 必須犧牲上一批次之晶圓。此外，本發明對於一新產品可 以事先較精準的預測其曝光能量，因此可避免習知無法對 新的試產產品進f了曝光能量預測。 綜合以上所述，本發明具有下列優點： 1. 本發明利用預先補償之方式，可減少曝光圖案之關 鍵尺寸之變異，且司使曝光圖案之關鍵尺寸較接近目標 値。 2. 本發明利用事先預測曝光能量之方法，可避免習知 Λ法中僅能封卜-批次之晶圓進行補償，而必須犧牲該批 -----------裝--------訂--------- (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 尺 張 紙 規 4 A S) N (C 準 標 97 509975 7527twf.doc/006 _B7 五、發明說明（？） 、 次之晶圓。 3.本發明可較精確的預測新產品之曝光能量，因此可 改善較習知之方法中無法對新產品預測其曝光能量。 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 神和範圍內，當可作些許之更動與潤飾，因此本發明之保 護範圍當視後附之申請專利範圍所界定者爲準。 ------------·裝 (請先閱讀背面之注意事項再填寫本頁) 訂--------- 經濟部智慧財產局員工消費合作社印製 10 本紙張尺度適用中國國家標準（CNS)A4規格（Z10 X 297公釐）Description of the invention (printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, "Yueyue 疋: interested parties < ~ a kind of photolithography process, and in particular, a method for predicting exposure energy Energy). With the increase of the integration degree of the integrated circuit, the component size of the entire integrated circuit must also be reduced accordingly. The most important one in the semiconductor process is the lithography process, which is related to the metal-Oxide-Semkonductor (MOS) device structure, such as the pattern of each layer of thin film (Pattern), and doped with impurities ( The area of Dopants) is determined by a suitable step of lithography. The control of the critical dimension (CD) of the lithography process is known after the exposure process is performed, and the exposure energy is compensated according to the result of the last measured critical dimension to make the next batch The critical size of the wafer cannot be closer to the target. However, the conventional method is to compensate the exposure energy of the next batch of wafers only after obtaining the critical dimension measurement results of the exposure pattern. This will sacrifice the wafers of this batch. In addition, the known method can only compare the exposure conditions of a new product with the same or similar process products, or by experience to determine the exposure energy, and this method does not make the difference between the key size of the mask and the online product. The critical dimension deviation trend (CD Baseline) is taken into consideration, so the critical dimension of the final exposure pattern often fails to reach the target 甚至, and even exceeds the tolerance tolerance of R], and the exposure conditions of the product must be modified again. In addition, the conventional method cannot predict the impact on critical dimensions of different batches of materials (such as photoresist materials), and understand and compensate in advance. ----------- AW ^ · 11 ----- Order --------- (Please read the notes on the back before filling this page) This paper size applies to Chinese national standards (CNS) / V1 gauge (210x297 mm) 509975 7527twf.doc / 006 A7 B7 Five invention descriptions (Z) Therefore, the purpose of the present invention is to provide a line exposure method of 顼 _exposure energy. Way to avoid _ light of the wafer. Another object of the present invention is to provide a method for predicting the exposure energy using pre-prediction compensation to reduce the variation of the final exposure, method, and size, and make the key size of the exposure pattern closer to self-wiping. Another object of the present invention is to provide A method of predicting exposure energy to improve the conventional method cannot predict the exposure energy of a new product. 'Method, The present invention proposes a method for predicting exposure energy. This method calculates a' front layer film thickness' of one of the layers to be exposed according to the key dimensions of a mask of one of the layers to be exposed to obtain the \ _% of Inch difference 値, and the compensation is calculated from the first key size difference 尺寸. Then check the photoresistance sensitivity before and after the batch. ^ ^ ^ * 1%. If the difference between the photoresist sensitivity before and after the batch 値 is less than 1% & apricot is less than the sum of the light energy and the first energy compensation, the energy is to be exposed for this purpose. Otherwise, the exposure will be stolen according to the difference between the previous and subsequent batches of the photoresist sensitivity, and then the second key size difference guarantee will be printed first by the consumer property cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, and the energy compensation will be printed. The sum of the compensations is the exposure energy of the layer to be exposed. The present invention utilizes a pre-compensation method to reduce the variation of the key size of the exposure pattern and make the key size of the exposure pattern closer to the target frame. The present invention uses a method of predicting the exposure energy in advance, which can avoid that in the conventional method, only the next batch of wafers can be compensated, and the wafers of the batch must be sacrificed. This paper size applies to China National Standard (CNS) A4 specification (210 X 297 public love) ------------ t -------- ^ --------- ----.--- * _____ (Please read the note on the back? Matters before filling out this page) 509975 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 7 527twf. Doc / 00 6 V. Description of the Invention (> ) The present invention can more accurately predict the exposure energy of a new product, and therefore can improve the inability to predict the exposure energy of a new product in a more conventional method. In order to make the above and other objects, features, and advantages of the present invention more comprehensible, a preferred embodiment is given below in conjunction with the accompanying drawings for detailed description as follows: Brief description of the drawings: FIG. 1 FIG. 2 is a flowchart of a method for predicting exposure energy according to a preferred embodiment of the present invention; FIG. 2 is a linear regression relationship between the key size of the photomask and the exposure energy; FIG. 3 is a relationship curve between the film deposition thickness and the key size Figure 4 is a linear regression diagram of photoresist sensitivity and key size; and Figure 5 is a linear regression diagram of key size and exposure energy. Description of the diagrams: 100 to 112: The steps are shown in the first figure, which is a flowchart of the method for predicting the exposure energy according to a preferred embodiment of the present invention; It is a linear regression diagram of the key size of the mask and the exposure energy. Please refer to FIG. 1 first. The method for predicting the exposure energy in this embodiment is to first calculate an exposure energy according to the key size of a mask of a layer to be exposed (step 100). Among them, please calculate the method for calculating the exposure energy. Please refer to the second picture. The second picture is the historical information of the key size of the photomask and the exposure energy. -Order --------- ^ 9. (Please read the precautions on the back before filling this page) This paper size is applicable to China National Standard (CNS) A4 (210 x 297 male f) 509975 Ministry of Economic Affairs Wisdom Printed by the Consumer Cooperative of the Property Bureau, 7527twf.doc / 006 A7 V. Explanation of the material (f), a linear regression diagram is made, where the horizontal coordinate system is the size of the key mask size, and the vertical coordinate system is the exposure energy. . Therefore, substituting the key size of the mask of the layer to be exposed into the linear regression equation in Fig. 2 can obtain an exposure energy 値, which is the desired exposure energy for the exposure step of the layer to be exposed. Next, please continue to refer to FIG. 1. After calculating the energy to be exposed, proceed to step 102, in which a key size difference 値 is calculated from the film thickness of the layer before the layer to be exposed, and a first An energy compensation 値. That is to use the target thickness of the film thickness of the layer to be exposed before and the target size of the key size specification to calculate a key size difference, and then calculate the first energy compensation from the calculated key size difference. Figure 3 shows the relationship between the film thickness and the critical dimensions. Refer to Figure 3 for a method to calculate a key size difference from the thickness of the previous layer. Figure 3 is a graph of the historical data of film deposition thickness and key dimensions, where the horizontal coordinate is the film thickness and the vertical coordinate is the key dimension. Using Figure 3, one of the key dimensions of the front film can be obtained from the thickness of the front film deposition. The key dimension of the front film 値 is subtracted from the center of the target dimension specification of the front film 値, which is the key dimension difference. . For example, the target thickness of the front film deposition thickness is 8000 ± 800 Angstroms, and its center 値 is 8000 Angstroms. The corresponding key size in Figure 3 is 0.46 microns, which means the key size specification of the front film. The standard series is 0.46 microns. The thickness of the layer before the actual exposure layer is 8300 angstroms, and the key dimension corresponding to Fig. 3 is 0.48 angstroms. From this-ϋ ϋ n ϋ ϋ ϋ ϋ I ^ 1 · ϋ ϋ nnn ϋ n a · · ϋ · ϋ n 1 n · ϋ n I #, (Please read the notes on the back before filling this page) This paper The scale is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 509975 7527twf.doc / 006 A7 ms B7 V. Description of the invention (孓) Knowing that the key size difference is 〇2 μm (0.48 Micron-0.46 Micron = 0.02 Micron). Figure 5 shows the linear regression relationship between key dimensions and exposure energy. Afterwards, 'the first energy compensation is calculated from the key size difference obtained in Fig. 3'. The method is to use the linear regression equation of the key size and exposure energy in Fig. 5 to obtain the first energy compensation. The horizontal coordinate system in Figure 5 is the key dimension 値, and the vertical coordinate system is the exposure energy 値. The method of using Fig. 5 to obtain the first energy compensation 系 is to substitute the key size difference obtained in Fig. 3 into the linear regression equation in Fig. 5 to obtain the first energy compensation 値. For example, the key dimension _ 値 G.G2 microns obtained in the above figure 3 is linearly regressed (Υ-144〇 · 7 × + 15_〇12) in figure 5 to obtain the first energy compensation 値 as 43 Like the second picture 1, query the photoresistance sensitivity before and after the batch "Chang 1% (frequency rigid)". If the difference between batches in step ig is less than 1%, _ go to step m, =;: Γ a film mouth light energy and the first-energy compensation 値 値-exposure as in step 102, Ge is about to step 1GQ ^ Add the first — energy compensation 步骤 obtained in step 1.2, that is, after ==, enter 仃 to measure the exposure image = check whether the deviation of the _ size of the exposure pattern 小于 is less than; ^ points are 値.夂 又 乙 ϋ ϋ n im ϋ ϋ ϋ ϋ n I n · n alai I nnn Hi Yikou VII n an mM— §MB (Please read the precautions on the back before filling this page) Employees of the Intellectual Property Bureau of the Ministry of Economic Affairs Printed by the cooperative 509975 7 52 7twf · doc / 006 A7 B7 Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (彡) If the difference between the photoresist sensitivity before and after the batch in step 104 is greater than 1%, proceed to step 10, that is, a second critical size difference is calculated according to the difference between the previous and subsequent batches of the light group sensitivity. Figure 4 shows the linear regression relationship between photoresist sensitivity and critical dimensions. The method of calculating the second critical size difference ， is to use the linear regression equation of the photoresist sensitivity and the critical size in Fig. 4 to obtain the second critical size difference 値. The horizontal coordinate system in Figure 4 is the photoresistance sensitivity, and the vertical coordinate system is the key dimension. Substituting the photoresist sensitivity of the previous batch of photoresist materials into the linear regression equation in Figure 4, you can calculate the photoresistance of the previous batch. One of the key dimensions of resistance materials is 値. Substituting the photoresistance sensitivity of the batch of photoresistive materials into the linear regression equation in Fig. 4 can obtain one of the critical size of the batch of photoresistive materials, and then subtracting the key batch size of the batch of photoresistive materials from the previous batch. The second critical dimension of the photoresist material is a second critical dimension difference. For example, the photoresist sensitivity of the previous batch of photoresist materials was 97%, and its calculated critical dimension was 0.492 microns, while the photoresist sensitivity of this batch of photoresist materials was 99%, and its calculated key The size 値 is 0.477 microns, so the second critical dimensional difference 知 is -0.015 microns (0.477 microns-0.492 microns and -0.015 microns). After that, please refer to FIG. 1 to calculate a second energy compensation 由 from the second key size difference 値 (step 108), wherein the second key size difference 値 to calculate a second energy compensation 値 is the first The two key size differences 値 are substituted into the linear regression equation in Figure 5 to find the second energy compensation 値. For example, the second key size difference obtained in the above step 106 is -0.015 8 This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 public love) (Please read the precautions on the back before filling this page) · Ϋ n ϋ nnn an 1: 0V nn I ί I ϋ II · ^ U9975 A7 B7 ^ 527twf.doc / 006 V. Description of the invention (q) The linear regression relationship of micrometers in Fig. 5 (Υ = 1440 · 7 × + 15.012 ) Shows that the second energy compensation 値 is -6.5985 ms. Next, step 11 is performed, which is given by the desired exposure energy obtained in step 100, the first energy compensation 求 obtained in step 102, and the second energy compensation 求 obtained in step 108. One exposure energy. It means that the sum of the exposure energy, the first energy compensation 値 and the second energy compensation ， is the exposure energy condition required for the exposure process of the layer to be exposed. Finally, after the exposure process is performed, the key size of the exposure pattern is measured (step 112) to confirm whether the deviation 关键 of the key size of the exposure pattern is smaller than the acceptable deviation 値. From the measurement results of the critical dimensions of the final exposure pattern, it can be found that the exposure energy conditions predicted by the present invention can make the critical dimensions of the exposure pattern closer to the target, and can make the critical dimensions of the exposure pattern acceptable. The deviation is within the range. Therefore, it is avoided that the conventional method cannot predict the exposure energy that should be set for the exposure process in advance, and can only be compensated in the next batch of wafers. The wafers in the previous batch must be sacrificed. In addition, the present invention can predict the exposure energy of a new product more accurately in advance, so it can be avoided that it is impossible to predict the exposure energy of a new trial product. In summary, the present invention has the following advantages: 1. The present invention utilizes a pre-compensation method, which can reduce the variation of the key size of the exposure pattern, and make the key size of the exposure pattern closer to the target 値. 2. The present invention utilizes a method of predicting the exposure energy in advance, which can avoid the conventional Λ method that can only be used to seal the wafers of a batch for compensation, and the batch must be sacrificed. ------- Order --------- (Please read the notes on the back before filling out this page) Printed Rule Sheets for Employee Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 4 AS) N (C Standard 97 509975 7527twf.doc / 006 _B7 V. Description of the invention (?), Followed by wafer. 3. The present invention can more accurately predict the exposure energy of new products, so it can improve that new products cannot be used in more conventional methods. Predict its exposure energy. Although the present invention has been disclosed as above with a preferred embodiment, it is not intended to limit the present invention. Anyone skilled in the art can make some changes without departing from the spirit and scope of the present invention. And retouching, so the protection scope of the present invention shall be determined by the scope of the attached patent application. ------------ · Installation (please read the precautions on the back before filling this page) Order --------- Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. NS) A4 size (Z10 X 297 mm)