I3022lffPwf*doc/g 九、發明說明： 【發明所屬之技術領域】 本發明是有關於一種探針之針壓的檢測方法與調整 方法，且特別是有關於可以將探針之針壓量化的一種探針 之針壓的檢測方法與調整方法。 【先前技術】 一般來說，半導體的封裝測試可區分為兩大部份，分 別是在晶圓加工完成後的晶圓測試（wafer pr〇be and sort)，以及封裝完成後的成品測試(fmaltest)。晶圓測試是 利用晶圓針測機(prober)上探針卡(pr〇be card)的探針與受 測晶圓的切割道(scribe line)上所配置的測試鍵(test_ke力連 接來進行測試，而此測試鍵與晶粒上之某一元件結構相 同。然後，將測得的資料送往測試機⑽㈣作分析與判斷， =正理出各晶粒的可修補資料，經由雷射修補機依據此資 =將不良的元件替換掉’之後再經晶圓針測通過後，即告 亓，成 〇 f而’捸針與測試鍵連接時因為探針之針壓大小不 ^墊(Pad)造成大小不—的刻痕。也就是 而冬二广’探針會在薛墊上造成大面積的刻痕， 針=探針會在銲墊上造成小面積的刻痕。探 結果。；例來::彺$響電性的量測，而造成不同的測試 性測試的結果n。 過大會使得接觸電阻過大而影響電 針對上述問題，目前多使用人工利用顯微鏡觀測的方 I30220Twfdoc/g 式來觀測刻痕的面積以判定料段 、， 高度來調整針壓。然而，使用古亚精由調整探針的 的時間，且由於無法將目式往往會浪費過多 而導致判定針壓大小的精準度I較低:之面積量化，進 【發明内容】 本發明的目的就是在提供一藉 法，可以將針壓量化以精確__。之針壓的檢測方 本發明的另一目的是括版 ^ 法，可以增加電性量測的可H舆m之針壓的調整方 本發明提出-種探針之針壓的檢 提供一晶圓，此晶圓之切割道 去此方法疋先 域，其中測試區域分為多個或、，有區域與測試區 每-個次區域之面積相等°°5 ^區域之面積與 件，而每-個魏域中具有具有—第一元 觸，並施加電壓’以使探針與參ΐΐ域 “域，亚施加，以使探針與至少一個： 接著，將測試電流值與參考電流2 試電流值。 弈描供一 a if] . , I „ 正力/:£： 此方法 二f Γ割道上具有參考區糊 〃域刀為多個次區域’而參考區域之面積與 13 022(^l°twf doc/g 每-個次區域之面積相等，且參考區域中具有—第_元 件，而每-個次區域中具有一第二元件。然後，將探針接 齡考區域，並施加電壓，以使探針與參考區域中之第— 兀件包性連接’並對應得到一參考電流值，以及將探針接 觸至少-，次區域，並施加電壓，以使探針與至少一個次 區域中之第一元件電性連接，並對應得到—則試電流值。 接著，將測忒電流值與參考電流值進行數據處理步驟，以 付到至少-個次區域之總面積。繼之，將總面積與標準值 0 随比對’以得到探針之針壓與標準針壓之間的相對關 係。之後，依照探針之針壓與標準針壓之間的相對關係對 探針進行校正步驟。 本發明因採用探針與測試區域及參考區域電性連接 時所測得的測試電流值與參考電流值來進行數據處理，以 得到具有導通的二極體的次區域之面積，並將此面積與標 準值做比較，關定探針之針壓是否過大或過小。如此一 來，便=將探針之針壓進行量化，並藉由量化的結果對 _ 探針進行彳X準避免了以人卫方式利用顯微鏡觀測刻痕面 積所產生的疾差，且縮短了檢測探針之針壓的時間。 “為讓本發明之上述和其他目的、特徵和優點能更明顯 易懂，下文特舉實施例，並配合所附圖式，作詳細說明如 下。 【實施方式】 方林發明實施例崎示的探収針壓的檢測 方法之步驟流程圖。請參照圖卜首先，在步驟削中， 130220^¾001^°10^ 提供一晶圓，此晶圓之切割道上具有參考區域與測試區 域。測試區域分為多個次區域。這些次區域例如是以矩陣 方式排列。此外，參考區域之面積與每一個次區域之面積 相等。參考區域中具有一第一元件，而每一個次區域中具 ^一第二元件。第一元件與第二元件例如為二極體。在本 貫鈀例中，第一元件的形成方法，例如是先在位於測試區 域的基底中形成例如是淺溝渠隔離結構的元件隔離結構。 淺溝渠隔離結構將測試區域的基底分為多個以矩陣方式排 列的次區域，而這些次區域也就是所謂的主動區（active area)。然後，進行離子植入製程，將與基底具有不同導電 型態的摻質植入次區域的基底中，以在每一個次區域中形 成一個二極體。之後，於每一個二極體上進行一般熟知的 内連線製程，並於最上層的金屬上形成銲墊。因此，若以 上視的角度來看，在測試區域可看到多個以矩陣方式排列 的銲墊。特別一提的是，上述的内連線製程是與晶粒上的 内連線製程同時進行。 此外，在進行測試區域的製程時，一般會同時於參考 區域的基底中形成與第一元件相同的第二元件。參考區域 與測試區域的差異在於··測試區域中具有多個以矩陣方式 排列的一極體，而參考區域中僅具有一個二極體，也就是 說，在參考區域中，淺溝渠隔離結構僅定義出一個主動區^ 同樣地，在參考區域的二極體形成之後，與晶粒同時進行 内連線製程，並於最上層的金屬上形成銲墊。特別一提的 是，在參考區域中，由於僅具有一個二極體，為了避=因 130220totwfd〇c/g 田然，在測試區域的旁邊必須形 試區域中的二極體下方的端點接出， ^件來將測 施加電料形成一個完整的迴路。上^十接觸銲墊並 本領域中具有通常知識者所熟知，於此:再=作方法為 然後，在步驟102中，將探針 、= 觸至少一個次區域，並施加電璧，以使探^域^同時接 之第一元件以及與至少—個奸域域中 接，即導通探針所接觸到的二極體，並對庫得:/ 連 以及至少-個次區域，並施加電壓/ =觸茶考區域 區域中之第-元件以及至少一個次區二十:別考 連接接ΠΓ得到一參考電流值與-測以^ 行數據處理步與參考電流值進 據處理步驟例如是將測試電流值除積以 婁值即為測試區域中經探針導通心 $:後’將茶考區域之面積乘以數值，即可得到 區域中具有導通之二極體的次區域之總面積。 ㈣之後二在步驟106中’將上述的總面積與標準值進行 之針壓與標準針壓之間的娜 以，視產叩的需要而有所不同，其可以是一個數值或 -個粑圍。在-實施例中，當總面積大於標準值時，探針 130220^0-4 之針壓與標準針壓之間的相對關係被判定為該探針之針壓 大於標準針塵。在另一實施例中’當總面積小於標準值時， 探針之針壓與標準針壓之間的相對關係被判定為該探針之 針壓小於標準針壓。如此一來，便可以將每一次以探針進 行電性量測時的針壓量化，以方便檢測探針的針壓是否過 大或過小。此外，上述方法還可以避免以人工方式利用顯 微鏡觀測刻痕面積所產生的誤差，以及縮短了檢測的時間。 除此之外，在以上述的檢測方法判定探針的針壓是否 過大或過小之後’遷必須將採針加以調整，以使電性量測 的可靠廑輿籍破疮接4。ra ，太！驗1 λ，、_I3022lffPwf*doc/g IX. Description of the Invention: [Technical Field] The present invention relates to a method and a method for detecting a needle pressure of a probe, and more particularly to a method for quantifying a needle of a probe The detection method and adjustment method of the needle pressure of the probe. [Prior Art] In general, semiconductor package testing can be divided into two parts, namely wafer pr〇be and sort after wafer processing, and finished product testing after packaging (fmaltest) ). The wafer test is performed by using a probe of a probe card on a probe and a test button (test_ke force connection) arranged on a scribe line of the test wafer. Test, and the test key has the same structure as a component on the die. Then, the measured data is sent to the test machine (10) (4) for analysis and judgment, = the repairable data of each die is corrected, and the laser repair machine is passed through the laser repair machine. According to this resource = replace the defective component, and then pass the wafer needle to pass, then the warning is made, and the needle is connected with the test button because the needle pressure of the probe is not the pad (Pad). Causes the size of the notch - that is, the winter two wide 'probe will cause a large area of the mark on the Xue mat, the needle = probe will cause a small area of the mark on the pad. : 彺 $ 电 电 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Adjust the needle pressure by determining the section and height However, the use of the ancient fine is adjusted by the time of the probe, and since the eye can not be wasted too much, the accuracy I for determining the needle pressure is low: the area is quantified, and the present invention is The purpose is to provide a borrowing method, which can quantify the needle pressure to accurately detect the acupressure of the needle. Another object of the invention is to include a plate method, which can increase the needle pressure of the electrical measurement. The invention proposes that the detection of the needle pressure of the probe provides a wafer, and the scribe line of the wafer goes to the method first, wherein the test area is divided into multiple or, and the area and the test area are each - The area of the sub-regions is equal to the area of the region of the region, and each region has a -first-element touch, and a voltage is applied to make the probe and the para-domain "domain, sub-applied, Make the probe with at least one: Next, test the current value and the reference current 2 to test the current value. Game for a if]. , I „ positive force /: £: This method has a reference area paste on the f cutting path The domain knife is a plurality of sub-regions' and the area of the reference region is 13 022 (^l°twf doc/g per - the area of each sub-area is equal, and the reference area has - the - element, and each sub-area has a second element. Then, the probe is connected to the age test area, and a voltage is applied to make the probe a reference element in the reference region - and correspondingly obtaining a reference current value, and contacting the probe with at least - a sub-region and applying a voltage to electrically connect the probe to the first component of the at least one sub-region Sexual connection, and correspondingly get - the test current value. Next, the measured current value and the reference current value are subjected to a data processing step to pay at least the total area of the sub-region. Then, the total area and the standard value are 0. As a result of the comparison, the relative relationship between the needle pressure of the probe and the standard acupressure is obtained. Thereafter, the probe is subjected to a calibration step in accordance with the relative relationship between the needle pressure of the probe and the standard acupressure. The invention performs data processing by using the test current value and the reference current value measured when the probe is electrically connected to the test area and the reference area to obtain the area of the sub-region having the turned-on diode, and the area is Compare with the standard value to determine if the needle pressure of the probe is too large or too small. In this way, the needle pressure of the probe is quantified, and the _ probe is quasi-quantified by the result of the quantification, thereby avoiding the difference caused by observing the scoring area by the microscope in a human-assisted manner, and shortening The time of the needle pressure of the probe is detected. The above and other objects, features, and advantages of the present invention will become more apparent and understood. Flow chart of the steps of detecting the needle pressure detection method. Please refer to the drawing. First, in the step of cutting, 130220^3⁄4001^°10^ provides a wafer having a reference area and a test area on the cutting track. Divided into a plurality of sub-regions. The sub-regions are arranged, for example, in a matrix. In addition, the area of the reference region is equal to the area of each sub-region. The reference region has a first component, and each sub-region has a The second component. The first component and the second component are, for example, diodes. In the present palladium example, the first component is formed by, for example, forming a component such as a shallow trench isolation structure in a substrate located in the test area. Isolation structure The shallow trench isolation structure divides the base of the test area into a plurality of sub-areas arranged in a matrix, and these sub-areas are also called active areas. Then, an ion implantation process is performed, and dopants having different conductivity types from the substrate are implanted into the substrate of the sub-region to form a diode in each sub-region. Thereafter, on each of the diodes Generally known as the interconnect process, the pads are formed on the uppermost metal. Therefore, if viewed from the above perspective, a plurality of pads arranged in a matrix can be seen in the test area. The above-described interconnect process is performed simultaneously with the interconnect process on the die. Further, in the process of performing the test region, the same second component as the first component is generally formed in the substrate of the reference region. The difference between the reference area and the test area is that there are a plurality of matrix bodies arranged in a matrix in the test area, and the reference area has only one diode, that is, in the reference area, the shallow trench isolation structure is only Defining an active region ^ Similarly, after the formation of the diode of the reference region, the interconnect process is performed simultaneously with the die, and a pad is formed on the uppermost metal. In the reference area, since there is only one diode, in order to avoid = 130220totwfd〇c/g, the endpoint below the diode in the test area must be connected next to the test area, ^ To form a complete loop of the applied electrical material. The top ten contact pads are well known to those of ordinary skill in the art, where: the method is then, in step 102, the probe, = Touching at least one sub-region and applying an electric sputum so that the probe element is connected to the first component and to at least one of the trait domains, that is, the diode contacted by the probe is contacted, and : / Connect and at least - sub-area, and apply voltage / = touch the first component in the tea test area and at least one sub-zone twenty: Do not test the connection to get a reference current value and - test data The processing step and the reference current value processing step are, for example, dividing the test current value by the 娄 value, that is, the probe conducting the center of the test area in the test area: after the 'the area of the tea test area is multiplied by the value, the area is obtained. The total area of the sub-region with the turned-on diode. (4) After the second step in step 106, 'the ratio between the needle pressure and the standard needle pressure of the total area and the standard value mentioned above is different depending on the need of calving, which may be a numerical value or a range. . In the embodiment, when the total area is larger than the standard value, the relative relationship between the needle pressure of the probe 130220^0-4 and the standard needle pressure is determined to be that the needle pressure of the probe is larger than the standard needle dust. In another embodiment, when the total area is less than the standard value, the relative relationship between the needle pressure of the probe and the standard needle pressure is determined to be that the needle pressure of the probe is less than the standard needle pressure. In this way, the needle pressure can be quantified each time the probe is electrically measured to facilitate detection of whether the needle pressure of the probe is too large or too small. In addition, the above method can also avoid the error caused by manually observing the scoring area by the microscope and shortening the detection time. In addition, after determining whether the needle pressure of the probe is too large or too small by the above-mentioned detection method, the needle must be adjusted to make the electrical measurement reliable. Ra, too! Test 1 λ,, _
度，以使探針遠離晶圓。或者，當探針之針 壓時，降低探針的高度，以使探針接近晶圓。如此一來， 在進行電性量_便可以精確地控他針的遞，你借見Degree to keep the probe away from the wafer. Alternatively, when the probe is needled, the height of the probe is lowered to bring the probe close to the wafer. In this way, in the electrical quantity _ can accurately control the hand of the needle, you borrow
I3022&t0twfdoc/^ 有導ΐ:出:M區域中導通的二極體的數目，進而得到且 比較，以判定探針之是否:大==標= 除了可以將探針之針磨進行量化之外，還可以藉由X 結果對探針進行校準，避免了以人工方式利用^二、、、 刻痕面積所產生的誤差，以及縮短了檢測的時間Γ域測 此外，利用本發明之方法還可以在機台運作 檢測晶圓吸座的平坦度，而不需將機台停止，可以、辟= :影響製程產量的問題，並可以隨時檢測晶圓吸二 雖然本發明已以實施例揭露如上，然其 :明丄=熟習此技藝者，在不脫離本發明之精神： ，内’虽可作些許之更動與潤飾，因此本發明之保 虽視後附之申請專利範圍所界定者為準。 &圍 【圖式簡單說明】 方法=為驟^圖發明實施例所繪示的探計之針壓的檢測 【主要元件符號說明】 100〜106 :步驟I3022&t0twfdoc/^ Guided: Out: The number of conducting diodes in the M region, and then obtained and compared to determine whether the probe is: large == mark = in addition to the needle grinding of the probe can be quantified In addition, the probe can be calibrated by the X result, thereby avoiding the manual use of the error caused by the area of the scoring, and shortening the time of the detection. Further, the method of the present invention is further utilized. It is possible to operate the machine to detect the flatness of the wafer holder without stopping the machine. It can solve the problem of process throughput and can detect the wafer suction at any time. Although the present invention has been disclosed by way of example However, it is clear that: if you are familiar with the art, you can make some changes and refinements within the spirit of the present invention. Therefore, the warranty of the present invention is subject to the definition of the patent application scope attached. . & [Simplified description of the method] Method = detection of the needle pressure of the probe shown in the embodiment of the invention [Description of main component symbols] 100~106: Step