200914385 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種藉由對脆性材料基板照射雷射光束 進行局部加熱，繼而冷卻加熱部位，以利用生成於基板之 熱應力（拉伸應力）形成裂痕之脆性材料基板之加工方法。 此處所谓脆性材料基板係指玻璃基板、燒結材料之陶 瓦單日日矽、半導體晶圓、藍寶石基板、陶瓷基板等。 f \, 又 裂痕」在其深度方向之前端未到達基板之背面 刖為』線’到達基板之背面時則為分割線（全分割線，_ cut lme )。就劃線而言，藉由進行沿著劃線施加彎矩之分 割處理而將其分割’或施加使裂痕進—步深人基板之更深 處之後處理將其分割。 另外，於以下說明中，為了便於說明，所謂「裂痕之 =入]係指裂痕沿基板之深度方肖（厚度方向）前進之狀 ’先、以4痕基板之表面方向前進之狀態（稱為「裂痕 之延伸」）區別使用。 又，於以下說明中「較漤裂疳 及此_ — . τ仪久衮痕」係指深度方向之前端 未到達基板背面之裂痕（即形成查丨蟪、 ^取里丨j踝）之中，欲直接分割 必須沿著裂痕施加較大之彎矩，田士 V & ^ ^ 因此必須使裂痕沿厚度方 向進入後再分割之裂痕。 【先前技術】 玻璃基板等脆性材料基板，於各 Α合禋產品中加工成適當 之大小或形狀後得以使用。例如， 液顯不面板通常係經 過製成貼合2 >!玻璃基板之大面穑之签把 八卸槓之母板，將液晶注入該 等2片玻璃基板之間隙後，切成 战各早位顯不基板之步驟而 200914385 製成。 於分割玻璃基板之步驟中，有例如專利文獻^ 利用如下方法，即，一面對基板壓接刀輪一面移動’ 措此於基板上刻出劃線，再沿著該劃線於基板之厚-’ 上施加彎矩以折斷。 又方向 在使用刀輪形成劃線（裂痕）_，除沿厚度 之垂直裂痕以外，亦可能於劃線附近產生玻璃肩。於對： 反’即大面積之玻璃基板縱橫地形成 出較小單位基板之情形下，由刀輪所形成之劃線之；= ^變得極長。伴隨該累計長度變長，產^機= 會變高，又，玻制之產生量亦會增加。—旦產生^:率 更=頻繁清掃劃線形成裝置（裂痕形成裝置），或飛散 /口著劃線對基板施加彎矩一 屑。 π研教罝之+台上的玻璃 相對於此’作為可抑制玻璃屑產生量之加工方 =二之雷射劃線法已被實用（參照專利文’ 並以藉由沿著預定劃線： = : =形成光朿點， 以下進行局部加熱所板之軟化溫度 喷射>桩、隹~、人，、诹，、，口者先束點所通過之軌跡局部 貝射々媒進订“之步驟，使基板 該熱應力形成劃線。 …、應力並利用 形成=，自母板切出較小單位基板時，需要實施縱橫地 ί線之Γ劃線（咖如be)，而當利用雷射劃線 產叉劃線時，於劃線之交又點附近有時會產生導致 不良之較大裂痕（不良裂痕）。為了防止該情形，有 200914385 揭示控制雷射劃線之條件以 淺於第-方向之垂直裂痕之深度痕之深度 體而言’係揭示與第一方向相比；、’文.3)。具 線時之雷射輸出，或者使第二=成第二方向上之劃 φί ^ * 向之掃描速度快於第一方 内之拎描速度，從而控制垂直裂痕之深度。 ί200914385 IX. Description of the Invention: [Technical Field] The present invention relates to a method of heating a portion of a brittle material substrate by irradiating a laser beam, and then cooling the heating portion to utilize thermal stress (tensile stress) generated on the substrate A method of processing a cracked brittle material substrate. Here, the brittle material substrate means a glass substrate, a ceramic tile of a sintered material, a semiconductor wafer, a sapphire substrate, a ceramic substrate, or the like. f \, and the crack" does not reach the back surface of the substrate at the front end in the depth direction. When the line "reach" reaches the back surface of the substrate, it is a dividing line (full dividing line, _ cut lme ). In the case of scribing, it is divided by performing a dividing process by applying a bending moment along a scribe line, or applying a crack to the deeper depth of the human substrate to divide it. In the following description, for the sake of convenience of explanation, the term "crack = in" refers to a state in which the crack advances in the depth direction (thickness direction) of the substrate, and is advanced in the surface direction of the four-mark substrate (referred to as "Extension of the crack" is used differently. In addition, in the following description, "the 漤 疳 疳 此 τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ In order to directly divide, it is necessary to apply a large bending moment along the crack, and the field V & ^ ^ must therefore cause the crack to enter the crack in the thickness direction and then divide the crack. [Prior Art] A substrate of a brittle material such as a glass substrate can be used after being processed into an appropriate size or shape in each of the composite products. For example, the liquid display panel is usually formed by bonding a double-sided slab of a double-sided squeegee of a glass substrate, and the liquid crystal is injected into the gap between the two glass substrates, and then cut into wars. The step of displaying the substrate in the early position is made in 200914385. In the step of dividing the glass substrate, for example, the patent document uses a method in which a surface of the substrate is crimped to the cutter wheel, and the substrate is scribed with a scribe line, and then the scribe line is thick along the substrate. -' Apply a bending moment to break. Further direction In the use of the cutter wheel to form a scribe line (crack) _, in addition to the vertical crack along the thickness, it is also possible to produce a glass shoulder near the scribe line. In the case where the large-area glass substrate is formed vertically and horizontally to form a small unit substrate, the scribe line formed by the cutter wheel becomes extremely long. As the cumulative length becomes longer, the production machine = becomes higher, and the amount of glass produced increases. Once the ^: rate is more = the scribing forming device (crack forming device) is frequently cleaned, or the flying/stitching line is applied to the substrate to apply a bending moment. π 研 罝 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + = : =The formation of the pupil point, the softening temperature spray of the local heating plate below >Pile, 隹~, person, 诹,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the step, the thermal stress of the substrate is formed into a scribe line, the stress is used to form a =, and when the small unit substrate is cut out from the mother board, the 纵 line of the vertical and horizontal lines is required to be performed, and when the mine is utilized, When the cross-hatching is produced, the large cracks (bad cracks) that cause defects are sometimes generated near the intersection of the scribing lines. In order to prevent this, there are 200914385 revealing that the conditions for controlling the laser scribing are shallower than The depth of the depth crack of the vertical crack in the first direction is 'relatively compared with the first direction; 'Text. 3'. The laser output with the line, or the second = second direction Φί ^ * scan speed faster than the scan speed in the first square, thus controlling The depth of the rift straight. Ί
:當形成第一方向之垂直裂痕時，必須使 =二文獻所記載，藉由提高雷射照射之輸出 :广入熱罝，或者降低雷射之掃描速度以增加每一單位 ，度之加熱時間，藉此增加入熱量。基板溫度會上升入敎 里增加之量，其後冷卻時溫度差增大，藉此可產生較大之 熱應力（拉伸應力），從而可使裂痕深入。 然而，必須增加對基板之入熱量以提高基板溫度作 隨產品不同可能希望避免基板溫度上升…若降低掃描 速度會對生產率造成影響，@此較佳為儘可能高速地形成 裂痕。 相對於此，使裂痕深入之其他方法（參照專利文獻4: 亦已有揭示。首先，藉由熱能強度於長軸方向之各端部為 取大之光束點，以2階段對母玻璃基板之表面進行加熱。 即，藉由較大熱能強度進行加熱後，在藉由較小熱能強度 進行加熱期間，使最初施加之較大熱量確實傳導至内部。 此時’並未以較大熱能強度持續加熱基板表面，故可防止 基板表面軟化。於進行此種加熱之情形下，由於處於熱量 傳遞至基板内部之狀態’故僅藉由形成冷卻點（冷卻p〇int) 之冷媒進行冷卻’並無法於與冷卻點之間獲得充分之熱應 力梯度，從而無法形成較深的垂直裂痕。 200914385 因此’為了高效率且確實形成劃線，於光束點之後方， 車父冷部點（主冷卻點）更靠近雷射點側，且沿著劃線之區 域（例如於專利文獻4之實施例i中距離光束點後端55瓜爪 之後方位置附近），形成預先噴射冷媒進行冷卻之輔助冷 钾點，又，將輔助冷卻點之冷媒溫度設定為高於主冷卻點 之冷媒溫度，且掃描輔助冷卻點以及主冷卻點。: When forming the vertical crack in the first direction, it must be recorded in the second document, by increasing the output of the laser irradiation: widening the heat, or reducing the scanning speed of the laser to increase the heating time of each unit. To increase the heat input. The substrate temperature rises to an increase in 敎, and the temperature difference increases after cooling, whereby a large thermal stress (tensile stress) is generated, so that the crack can be deepened. However, it is necessary to increase the heat input to the substrate to increase the substrate temperature. Depending on the product, it may be desirable to avoid the substrate temperature rise. If the scanning speed is lowered to affect the productivity, it is preferable to form a crack as high as possible. On the other hand, another method of making the crack deeper is also disclosed (refer to Patent Document 4: First, the thermal energy intensity is a large beam spot at each end portion in the long axis direction, and the mother glass substrate is applied in two stages. The surface is heated. That is, after heating by a large thermal energy intensity, the first applied large amount of heat is actually conducted to the inside during heating by the smaller thermal energy intensity. At this time, 'there is not continued with a large thermal energy intensity. By heating the surface of the substrate, it is possible to prevent the surface of the substrate from softening. In the case of such heating, since the heat is transferred to the inside of the substrate, it is only cooled by the refrigerant forming the cooling point (cooling). A sufficient thermal stress gradient is obtained between the cooling point and the cooling point, so that deep vertical cracks cannot be formed. 200914385 Therefore 'for high efficiency and indeed forming a scribe line, behind the beam point, the cold part of the car (main cooling point) Closer to the side of the laser spot, and along the area of the scribe line (for example, in the embodiment i of Patent Document 4, near the rear end of the beam point of the beam point 55) Pre-injection auxiliary refrigerant cooled cold spot of potassium, and the auxiliary cooling point temperature of the refrigerant is set higher than the temperature of the primary coolant of the cooling point and the scanning point, and a main cooling auxiliary cooling point.
凄該輔助冷卻點係緩和多餘之熱衝擊，使原本會因熱衝 擊：損失之能量得用於裂痕之形成。BI 6係說明沿著分割 預疋線P猎由光束點Bs進行基板加熱，其後依次進行輔 助々部、主冷卻時各點上之寬度方向（短轴方向）之溫度 :布變化：熱應力變化的示意圖。圖6U)係表示以光束 = 進仃加熱以及以輔助冷卻點as、主冷卻點则進行 V冲之區域之位置關係的俯視圖。 % α加熱後，在即將以 p 2冷卻點AS進行冷卻之位置正交於分割預定線 交於:為Α·Α’剖面，以輔助冷卻點AS進行冷卻時之正 預定線P的剖面為心剖面，以主冷卻點MS進 日正又於分割預定線p的剖面為c-c·剖面。 圖6 ( b )係表示於A-A'剖面上夕A把主二 部（厚度方向之中間點）之”八：基板表面以及基板内 溫度分布a 布圖，以及基板内部之 布與熱應力的示意圖。以 將以後方之Μ ^ 元果點BS加熱後，在即 方之辅助冷部點AS進行冷卻 表面之溫度分布呈現且有以〜 R A A °J面上，基板 之溫度尖峰之分布；有：=定線p為中心向上凸出 分布以及埶應力 。彳面上之基板内部之溫度 叹熟愿力而言，於弁击& ^ 、先束點BS通過後，熱源之中心 200914385 於分割預定線p上朝向基板内部前進，並且自熱源中心放 射狀傳遞熱量’其結果’形成橢圓狀之溫度分布，且自基 板表面附近至内部受到壓縮應力。 圖ό (c)係表示於Β-Βι剖面上之基板表面以及基板内 部之溫度分布圖，以及基板内部之溫度分布與熱應力的示 意圖。於以輔助冷卻點AS進行冷卻之B_B，剖面上，由於 係使在基板表面上受輔助冷卻點AS冷卻之寬度與由光束 點BS所加熱之寬度相同，故變為如下狀態，即，基板表 面附近於整個寬度方向（短軸方向）上得以緩緩冷卻，使 得溫度尖峰整體降低（圖中，虛線部分為因冷卻而產生之 變化）。此時基板内部因自表面傳遞至基板内部之熱量而 歹欠存有向上凸出之尖峰。以輔助冷卻點AS進行冷卻雖 會於基板表面上產生較弱拉伸應力，，旦此時產生之拉伸應 力（由於應力集中不充分）仍無法形成裂痕。 圖6 ( d )係表示c_Ci剖面上之基板表面以及基板内部 之孤度/刀布圖，以及基板内部之溫度分布與熱應力的示意 圖。於以主冷卻點MS進行冷卻< c_c’剖面上，藉由基板 表面上之主冷卻點Ms @力冷卻整個基板表面附近而於整 個寬度方肖（短轴方向）上產生較大拉伸應力。此時基板 内部因自表面傳遞至基板内部之熱量而殘存有向上凸出之 大峰’故產生壓縮應力。其結果，於產生由輔助冷卻點Μ 引起之拉伸應力之狀態下，進而藉由形成主冷卻點MS之 冷媒進仃冷卻’有較強之拉伸應力將對基板表面產生作 用。此時便會形成深入之裂痕。 卜亦已揭不實行此種輔助冷卻，與不實行辅助冷 200914385 卻之情形相比較，裂痕之深度增加1〇%左右。 【專利文獻1】日本特開平！卜〗1626〇號公報 【專利文獻2】曰本特表平8_5〇9947號公報 【專利文獻3】日本特開2〇〇1_58281號公報 【專利文獻4】W02004/014625號公報 【發明内容】 藉由專利文獻4所記载之方法使裂痕深入之情形，盥 丨使輔助冷卻點發揮作用之情形相比，雖可散地形成裂 ί 痕並使裂痕進入，然而對於光束點之寬度方向之裂痕位置辅助 The auxiliary cooling point mitigates the extra thermal shock, which would otherwise be due to thermal shock: the lost energy is used for the formation of cracks. The BI 6 system explains that the substrate is heated by the beam spot Bs along the split pre-twist line P, and then the temperature in the width direction (short axis direction) at each point of the auxiliary crotch portion and the main cooling is sequentially performed: cloth change: thermal stress Schematic diagram of the change. Fig. 6U) is a plan view showing the positional relationship of the area where the V-rush is performed by the beam = heating and the auxiliary cooling point as and the main cooling point. After the % α is heated, the position at which the cooling is performed by the p 2 cooling point AS is orthogonal to the division planned line to the section of the 预定·Α' section, and the section of the positive predetermined line P when the auxiliary cooling point AS is cooled is the heart. In the cross section, the section where the main cooling point MS is in the day and the dividing line p is the cc section. Figure 6 (b) shows the "A" of the main part (the middle point of the thickness direction) on the A-A' section, the surface area of the substrate and the temperature distribution in the substrate, and the cloth and thermal stress inside the substrate. Schematic diagram of the temperature distribution of the cooling surface of the substrate after the heating of the 果 ^ 果 点 BS , 在 辅助 辅助 辅助 辅助 辅助 辅助 辅助 辅助 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 〜 :=The alignment line p is the upward convex distribution and the 埶 stress. The temperature inside the substrate on the 彳 surface is sighed by the force, and after the sniper & ^, the first beam point BS passes, the center of the heat source is 200914385 The predetermined line p advances toward the inside of the substrate, and the heat transfer from the center of the heat source 'the result' forms an elliptical temperature distribution and is subjected to compressive stress from the vicinity of the substrate surface to the inside. Fig. ό (c) is expressed in Β-Βι A schematic diagram of the surface of the substrate and the temperature profile inside the substrate, and the temperature distribution and thermal stress inside the substrate. The B_B is cooled by the auxiliary cooling point AS, and the profile is on the surface of the substrate. The width of the cooling by the auxiliary cooling point AS is the same as the width heated by the beam spot BS, so that the vicinity of the surface of the substrate is gradually cooled in the entire width direction (the short-axis direction), so that the temperature peak is lowered as a whole. (In the figure, the broken line portion is a change due to cooling.) At this time, the inside of the substrate is immersed in the upward bulging peak due to the heat transferred from the surface to the inside of the substrate. The cooling is performed on the auxiliary cooling point AS. A weak tensile stress is generated on the surface, and the tensile stress generated at this time (due to insufficient stress concentration) cannot form a crack. Fig. 6 (d) shows the surface of the substrate on the c_Ci profile and the degree of resolution inside the substrate/ a knife layout, and a schematic diagram of the temperature distribution and thermal stress inside the substrate. On the cooling < c_c' section at the main cooling point MS, the entire substrate surface is cooled by the main cooling point Ms @force on the surface of the substrate. A large tensile stress is generated in the entire width square (short axis direction). At this time, the inside of the substrate remains upward due to the heat transferred from the surface to the inside of the substrate. The large peaks thus generate compressive stress. As a result, in the state where the tensile stress caused by the auxiliary cooling point 产生 is generated, the refrigerant is cooled by the formation of the main cooling point MS, and the tensile stress is strong. The surface has an effect. At this time, a deep crack is formed. Bu has not revealed such auxiliary cooling, and the depth of the crack is increased by about 1% compared with the case where the auxiliary cold 200914385 is not applied. [Patent Document 1] Japanese Patent Laid-Open Publication No. Hei. No. Hei. No. Hei. No. Hei. No. Hei. No. Hei. DISCLOSURE OF THE INVENTION In the case where the crack is deepened by the method described in Patent Document 4, the crack is formed and the crack is entered as compared with the case where the auxiliary cooling point is acted upon, but for the beam spot Crack position in the width direction
的控制性並不充分，難以使裂痕位置準確地對準分割預定 線。其原因在於’辅助冷卻點雖如上所述緩和多餘熱衝擊， 使原本因熱衝擊而失去之能量得用於裂痕形成，以使較弱 之拉伸應力產生於基板表面，但亦造成裂痕係於其次之主 冷卻點通過後才形成。因此，輔助冷卻點與主冷卻點此2 個冷卻點均會影響裂痕形成位置之定位，因此，因2個冷 卻點之形狀或冷媒喷射壓力、2個冷卻點間之距離、甚I I 2 ^冷卻點之掃描速度等減，皆會使裂痕產生之位置微 妙變動。 因此，本發明之目的在於提供一種以比已往更高之穩 定性形成裂痕，而且可深入，並且可使所形成之裂痕沿： 告'J預定線準確定位之加工方法。 為解決上述課題，本發明之加工方法係利用於脆性材 料基板上δ又疋分割預定線，使雷射光束之光束點沿著上述 分割預t線相對㈣而以低於軟化溫度之溫度加熱上述基 板之加熱步驟、與使藉由冷媒喷射所形成之冷卻點沿著掃 200914385 描上述光束點之執跡相對移動以冷卻之冷卻步驟，沿著上 述分割預定線形成裂痕之脆性材料基板之加工方法，上 :卻步驟係連續實施（a)第一冷卻步驟與（b)第二冷： 4 ’上述U)第-冷卻步驟係使將上述 :至小於上述光束點之寬度而形成之第一冷卻點緊= 束點作相對㈣，藉此使較淺之裂痕W申，上述（b 二冷：步驟係使將上述冷卻點之寬度擴大至大於上述光束 點之寬度而形成之第二冷卻點沿著掃描第一冷卻點之執跡 作相對移動，藉此使先前所形成之較淺裂痕沿 方向進入。 千良 根據本發明，沿著設定於基板上之分割預定線相對移 動先束點（即相對於基板移動光束點，或相對於光束點移 動基板），藉此局部加熱基板。於加熱後，以緊跟隨光束 點之方式，相對移動縮小至寬度小於光束點寬度之第一冷 卻點實施驟冷處理H點剛移動後之 通 過之表面料受❹熱之狀態。因此，第—冷卻 點之後方近處實施驟冷處理，僅會於分割預定線之上方近 處且表面附近局部產生較大之熱應力（拉伸應力），因此 將沿著分割預定線準確地形成裂痕，x，確實形成裂痕。 繼而γ藉由使冷卻點之寬度擴大至大於光束點之寬度而形 成的第二·冷卻‘點，整體性冷卻光束點溫度上升之區 域。此時之冷卻面積變大，較大之拉伸應力施加於受第二 冷卻點冷卻之部分整體。於第一冷卻步驟已形成有較淺裂 痕之區域中，進而施加較大之拉伸應力，其結果，較淺之 裂痕將沿深度方向深入。 200914385 即以第一冷卻步驟形成較淺之裂濟 割預定線於準確之彳 裂痕，以確實沿著分 千崎之位置形成裂痕 以先前所形成之較淺裂痕為起點使之深入冷部步驟中 根據本發明，益丄 成較淺裂痕丄不同：件之冷卻點連續進行形 裂痕深入。皁確之位置上穩定地形《痕，而且可使 =述發明中’較佳為使光束點之寬 度於光束點中央達到最大。 …、此強 藉此，光束點中央受到最大加熱，進行第一冷卻 於光束點中央溫度差最大，因此可使熱應力.( 伸=)之集中部分為光束點中央，即分割預定線上，故 可&著刀割預疋線準確地形成較淺之裂痕。 於上述發明中，第一冷卻步驟較佳為第一冷卻點剛通 過之垂直於分割預定線之面内溫度分布夾著分割預定線於 左右兩側形成-對高溫區域’並進行冷卻使於分割預定線 上幵y成恤度低於上述尚溫區域之溫度極小區域，並藉由因 上述一對咼溫區域與上述溫度極小區域之溫度差而產生之 熱應力來形成較淺之裂痕。 藉此，垂直於分割預定線之面内之溫度分布成為溫度 極小區域夾於2個溫度尖峰（高溫區域）之間之形狀，因 此熱應力（拉伸應力）集中於溫度極小區域，可於溫度極 小區域準確地形成裂痕。 於上述發明令，亦可使冷媒中包含水分，第一冷卻點 藉由噴附將水分霧化之冷媒的噴霧器（mist jet )形成，第 12 200914385 二冷卻點藉由大範圍地喷附將水分氣化後自我冷卻而溫度 低於上述噴霧器之冷媒的氣化冷卻形成。 此處所謂「霧化之冷媒」係指朝向較小區域集中喷射 包含水分之冷媒’並於難以氣化之狀態下自喷嘴進行喷 射’使水分氣化引起之自我冷卻現象不會發生而被喷射之 狀態之冷媒。使用霧化之冷媒時，由於不會產生自我冷卻， 故冷媒溫度不會降低，可集中冷卻較小之區域。 所谓「氣化冷卻」係指以使水分擴散之方式噴射包含 f 水分之冷媒，並於易於氣化之狀態下自喷嘴嘴射利用冷 媒中水刀氣化所引起之自我冷卻現象，使冷媒溫度降低， 並以低溫化之冷媒進行冷卻。利用氣化冷卻，雖難以集中 冷卻較小之區域，但可以低溫冷卻廣泛之區域。 ^ 利用本發明，可易於藉由經霧化之冷媒進行局部之冷 部，另一方面，藉由經氣化冷卻之冷媒強力冷卻較廣之範 圍。 & 於上述發明中，較佳為使光束點、第一冷卻點、第二 I 冷卻點之相對移動速度為l〇〇 mm/秒〜72〇 mm/秒。 +根據本發明，雖因板厚、材質、加工方法而有差異， 但藉由連續進行形成較淺裂痕之步驟與形成深入之裂痕之 步驟，即使進行ΐθθ爪―秒〜72〇瓜…秒之高速化，亦可沿 著分割預定線確實形成裂痕，因此可以高速形成裂痕。 【實施方式] 以下，根據圖式說明本發明之實施形態。圖丨係本發 明一實施形痗之基板加工裝置LSI之概略構成圖。此處以 加工玻璃基板之情形為例進行說明。 13 200914385 就基板加工裝置Ls〗 敕 著於水平架A 正體構成進行說明。沿 σ 上平行配置之一對導勤3 . 紙面前後方向(以下稱為Υ方向)二動4設有於圖1 於兩導軌3、4之門、κ — 彺谈私動之滑動台2。 螺桿5上㈣配置有導螺桿5,於該導 以馬達(:a Γ上述滑動…固定件6,並藉由 著導執3 4=复移動導螺桿5正、反轉動’使滑動台2沿 4在设移動於γ方向上。 右方動台2上配置有沿著導執8往復移動於圖1之左 Vi之支:下稱”方向)之水平台座7。於固定於台ί 萨由導蟫:、MU貝通螺合有以馬達9轉動之導螺桿10， 精由¥螺桿1〇正、反轅 搞△产 動於X方向。反轉動，使台座7沿著導軌8往復移 曰广台& 7上設置有以旋轉機構11轉動之旋轉台12， 麵基板A等脆性材料基板在水平之狀g下安裝於該旋 ° 2上。該玻璃⑽反A係'例如用於切出較小單 1母板。旋轉機構U係使旋轉台12繞垂直之軸旋轉^ $成為可以相對於基準位置成為任意旋轉角度之方式進行 旋轉。又，玻璃基板AH由吸引夾頭固定於旋轉台12上。 於旋轉台12之上方，雷射敦置13與光學保持器14保 持於安裝框架15上。 ’、 田射裝置13 ’作為脆性材料基板之加卫用途，使用通 常之雷射裝置即可’具體而言，使用準分子雷射、YAG雷 射—氧化石厌雷射或一氧化碳雷射等。於玻璃基板A之加 工中，較佳為使用可振盪出破璃材料之能量吸收效率較大 之波長之光的二氧化碳雷射。 14 200914385 自雷射裝置13射出之雷射光束，其預先設定之形狀之 光束點藉由組裝有用於調整光束形狀之透鏡光學系統的光 學保持器14照射至玻璃基板a上。關於光束點之形狀， 雖具有長軸之形狀（橢圓形、長圓形等）可沿著分割預定 線高效率地進行加熱這一方面較為優異，但只要可在低於 軟化溫度之溫度下進行加熱之形狀，光束點之形狀並無特 別限定。 於本實施形態中，為形成橢圓形狀之光束點。圖2係 表示所照射之橢圓形狀光束點之俯視圖（a)以及寬度方向 (短轴方向）之熱能強度分布（b)。光束點bS之寬度方 向之熱能強度分布，例如高斯分布般，分布成左右對稱且 光束點中央為最大之熱能，該光束點Bs於基板上進行移 動時，光束點BS之中央部分所通過之執跡被加熱為最高 度（但為軟化溫度以下）。關於光束點B s之長軸方向 之熱能強度分布，其既可為高斯分布，亦可為提高加熱二 率而採用不同分布形狀之熱能強度分布。 於安裝框架15,接近光學固持器14設置有第一冷卻 喷嘴16a以及第二冷卻喷嘴16b。冷媒由該等冷卻喷嘴i^、 ⑹進行喷射。冷媒可使用冷卻水、壓縮空氣、氦氣、二 氧化破等，於本實施形態中係喷射冷卻水與壓縮空氣之混 合流體。 & 自第-冷卻嘴嘴16a噴射之冷媒朝向自光學保持器Μ 照射至玻璃基板A之光束點BS之長邊方向之後端部，於 玻璃基板A之表面形成第一冷卻點⑶。自第一冷卻喷嘴 16a中所噴射之冷媒以成為霧狀之方式由噴嘴喷射。 15 200914385 自第二冷卻喷嘴丨6b喷射之冷媒朝向第一冷卻點cs^ 之後端部，或者與第一冷卻點部分重疊之位置，或者略微 離開第-冷卻點CS1後端之位置，於玻璃基板八之表面形 成第二冷卻點CS2。自第二冷卻噴嘴16b噴射之冷媒以大 範圍擴散之方式由喷嘴進行噴射，故可使水分氣=有 行冷卻。 1^疋 圖3係表示於玻璃基板A上，沿著分割預定線p所形 成之光束點BS、第-冷卻，點CS1及第二冷卻點CS2之位 置關係以及尺寸關係之俯視圖。 使光束點BS、第一冷卻點CS1及第二冷卻點cs2皆 為具有長軸之橢圓形狀或長圓’且長軸方向分別與分割預 定線P為同軸。若將光朿點Bs之寬度（短轴）設為Wb、 第一冷卻點之寬度設為Wcsl、第二冷卻點之寬度設8為 Wes,，則該等寬度之大小關係為Wcsi<WdWm。 具體而言，將第-冷卻點CS1之寬度w⑶設為光束 點BS之寬度^之1/4〜3/4左右，更佳為μ左右。例如 將光束點BS之寬度（短軸）設定為2贿，則第一冷卻點 c S 1之短軸設為1 mm進行喷射。 又，將第二冷卻點CS2之寬度Wcs2設為與光束點Bs 之寬度^相同’或略寬於％。例如將光束，點bs之寬度 (紐軸）設定為2mm，則第二冷卻點⑶之短轴設為3随 左右進行噴射。 又’於安裝框架15透過上下移動調節機構17安裝有 刀輪18。該刀輪18係在於玻璃基板A之端緣形成初始龜 裂（觸發點）日夺，使台座7 —面沿χ方向移動一面暫時下 16 200914385 降後使用。 又’於基板加工裝置 搭载有可檢測刻印於玻璃 暴板A上之用於定位之對 像機20所檢測出之對準：位機20 ’可自藉由攝 位置並進行^ “之位置切分割預定線P之 繼而’就上述基板加工裝 將玻璃基板A載置於旋轉台==動作進行說明。 耩由攝像機20檢測出刻印於玻璃基板八之對準標記，並 根據其檢測結果，建立分割 丁 』了角疋深Ρ方疋轉台12以及滑動 口之位置關係。之後’調整旋轉台12以及滑動台2之 位置:以使分割預定線Ρ到達與刀輪18相對之位置。 4而’為了於破璃基板Α之端部形成初始龜m 而使旋轉台12返回初始位置（於圖1中為右端），且於 降下刀輪18之狀態下將旋轉台12^x方向（於圖i中為 自右向左）移動，藉此將刀輪18壓接於玻璃基板a之側 端，开> 成初始龜裂（TR )。 形成初始龜裂（TR)後，再次使旋轉台12返回初始 位置’並且自雷射裴置13射出雷射光束，自第—冷卻^ 嘴1 6a以及第二冷卻喷嘴1 6b喷射冷媒。 於此狀態下，使旋轉台12往_χ方向（於圖！中為自 右向左）移動。藉此，沿著玻璃基板Α之分割預定線‘ρ， 使光束點BS、第一冷卻點CS1、第二冷卻點CS2按該順 序受到掃描。 μ 針對如下狀態進行說明，即，藉由以上之動作，产著 玻璃基板Α之分割預定線Ρ並以初始龜裂TR為起點形成 17 200914385 W痕，此時生成於玻璃基板A之 好穿_、人" 衣展因第一冷卻點CS 1以 及弟一冷部點CS2之冷卻 巷板内部之狀態。 圖4係說明沿著分割預定 P精由光束點BS進行基 以及藉由第一冷卻.點CS1、第二冷卻點⑶進行 二❿各點上之寬度方向（短軸方向）之溫度分布變化、 應力變化的示意圖。圖4 ( a ) # ^ ^ ）係表不藉由光束點BS進行 加熱’以及藉由第一冷卻點、筮_ ” 第一冷部點進行冷卻之區域 之位置關係之俯視圖。 如圖所示’設以光束點bs進行加熱後，第— 冷部點C S 1進行冷卻之前方i斤虎 別万近處位置上之正交於分割預定 線P之剖面為D-D，剖面，以笛、人，、 ^ 』曲以第一冷部點CS1進行冷卻時正 父於分割預定線P之剖面A F p，立丨& 』囱马E-E剖面，以第二冷卻點CS2 進行冷卻時正交於分割預定線p之剖面為Μ剖面。 k. 圖4 ( b )係D-D剖面上之基板表面之溫度分布圖以 及表示基板内部之溫度分布與熱應力之示意圖。於以光束 點BS進行加熱後，以第一冷卻點⑶進行冷卻之前方近 處之D-D’剖面上，由於光束點Bs之熱能強度分布於中央 處為最大’故基板表面之溫度分布呈現具有以分割預定線 P為中心向上凸出之溫度尖峰之分布。又，就D_D，剖面上 之基板内部之溫度分布以及熱應力而言，由於以分割預定 線P為中心且自基板表面側放射狀散熱，故其結果形成半 圓狀之溫度分布，且於基板表面附近受到壓縮應力。 圖4 ( c )係E-E，剖面上之基板表面以及基板内部之溫 度分布圖，以及表示基板内部之溫度分布與熱應力的示意 圖。於已以第一冷卻點CS1進行冷卻之E_E，剖面上，由於 18 200914385 基板表面上之受第-冷卻點⑶冷卻之寬度被縮小至小於 以光束，點BS加熱之寬度，因此，溫度分布變化為於溫产 尖峰之中央處形成低谷（溫度極小區域），故變為分:^ 個尖峰之溫度分布。其結果，於分割預定線p之上方且接 近表面處，局部產生較大之熱應力（拉伸應力），故於表 面形成裂痕。另一方面，由於基板内部係剛以光束點 進行局部加熱，故自基板表面傳遞過來之熱量並未充分傳 〆 遞至基板内部，故於基板内部不太會產生溫度上升。其結 " 果，僅產生於基板表面附近之較大之熱應力使基板表面附 近確實形成裂痕，但由於裂痕未能進入基板内部，故其結 果為形成較淺之裂痕。 圖4 (d)係F-F，剖面上之基板表面以及基板内部之溫 度分布圖，以及表示基板内部之溫度分布與熱應力的示意 圖。於以第二冷卻點CS2進行冷卻之F_Fi剖面上，由於基 板表面上受第二冷卻點CS2冷卻之寬度與受光束點加 熱之寬度相同，或者寬於受光束點BS加熱之寬度，因此， I 溫度尖峰整體受到冷卻’故溫度尖峰消失，於整個表面上 形成較大之熱應力（拉伸應力）。另一方面，於基板内部 殘存有自表面傳遞而來之熱量，其結果，基板内部與基板 表面上大範圍地產生較大之溫度差。該溫度差使基板表面 產生拉伸應力，且於基板内部產生壓縮應力，藉此，因周 圍之應力分布而產生使先前所形成之較淺裂痕裂開之力， 發揮使較淺之裂痕深入之作用。 圖5係表示於沿著分割預定線P之方向之剖面上之裂 痕產生狀態的示意圖。 19 200914385 如至目前為止之說明’於受光束點BS加熱之區域Ll 中’雖於表面附近受到垂直於紙面之方向之壓縮應力之作 用，但並未產生裂痕。 於受第一冷卻點CS1冷卻之區域L2中，於表面附近 產生垂直於紙面之方向之局部拉伸應力（參照圖4 ( c ))， 形成較淺之裂痕S2。 進而於受第二冷卻點CS2冷卻之區域L3中，藉由基 板内部之垂直於紙面之方向的壓縮應力與基板表面之垂直 於紙面之方向的拉伸應力’發揮使形成於表面附近之較淺 的裂痕S2進一步進入之力之作用（參照圖4(d))，從 而形成深入之裂痕S3。 如此，於第一階段穩定且確實形成較淺之裂痕，以此 為觸發點而於第二階段使裂痕深入，藉此即使以高於目前 為止之速度掃描光束點B S等時，亦可穩定地形成深入之 裂痕。具體而言’即使對以往1〇〇 mm/秒以上之速度便無 法穩定地形成裂痕之玻璃基板，亦可以720 mm/秒以下之 速度穩定地形成裂痕。 例如’照射C〇2雷射進行劃線加工時，以往僅能於240 W且500 mm/秒以下之掃描速度進行劃線加工，但藉由使 用本發明，即便對於相同之基板將掃描速度加速至550 mm/ 秒亦可進行加工。其結果，可高速地實現交叉切割。而且， 最大720 mm/秒之高速裂痕形成亦已經實驗確認。 進而’利用本發明可使裂痕比先前更深入。其結果， 以往雖使用折斷裝置沿著裂痕（劃線）藉由按壓折斷棒折 斷基板’而根據本發明，則無需進行此種折斷步驟亦可分 20 200914385 割基板，且可實現無折斷步驟之整體切割。 本發明可應用於對玻璃基板等脆性材料基板之裂痕形 成，進而用於分割。 【圖式簡單說明】 圖1係使用本發明之加工方法之基板加工裝置之概略 構成圖。 圖2係表示雷射光束之熱能強度分布之一例之圖。 fThe controllability is not sufficient, and it is difficult to accurately align the crack position with the division line. The reason is that the 'assisted cooling point mitigates the excess thermal shock as described above, so that the energy originally lost due to thermal shock is used for crack formation, so that weak tensile stress is generated on the surface of the substrate, but the crack is also caused. The second cooling point is formed after the passage of the main cooling point. Therefore, both the auxiliary cooling point and the main cooling point will affect the location of the crack formation position. Therefore, due to the shape of the two cooling points or the refrigerant injection pressure, the distance between the two cooling points, and the cooling of the II 2 ^ The scanning speed of the point is reduced, and the position of the crack is subtly changed. SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a processing method for forming cracks with higher stability than conventionally, and which can be deepened and which can accurately form the cracks formed along the predetermined line. In order to solve the above problems, the processing method of the present invention utilizes a predetermined line of δ and 疋 division on a substrate of a brittle material, so that the beam spot of the laser beam is heated at a temperature lower than the softening temperature along the dividing pre-t line (4). a heating step of the substrate, and a processing method for forming a cracked brittle material substrate along the dividing line by performing a cooling step of moving the trace of the beam spot along the scan by 200914385 with a cooling point formed by the refrigerant jet And the steps are continuously carried out (a) the first cooling step and (b) the second cooling: 4 'the above U) the first cooling step is such that the first cooling is formed to be smaller than the width of the beam spot Point tight = beam point for relative (four), thereby making the shallow crack W, the above (b second cooling: the step is to expand the width of the cooling point to be larger than the width of the beam point to form a second cooling point along The scan of the first cooling point is scanned for relative movement, thereby allowing the previously formed shallow crack to enter in the direction. According to the present invention, along the dividing line set on the substrate, Moving the beam point (ie, moving the beam point relative to the substrate, or moving the substrate relative to the beam spot), thereby locally heating the substrate. After heating, the relative movement is reduced to a width less than the beam spot width in the manner of following the beam spot. The first cooling point is subjected to a quenching treatment, and the surface material passing through the H point just after the movement is subjected to the hot state. Therefore, the quenching treatment is performed in the vicinity of the first cooling point, and is only near the top of the dividing line and near the surface. A large thermal stress (tensile stress) is locally generated, so that a crack is accurately formed along the line to be divided, x, and a crack is formed. Then γ is formed by expanding the width of the cooling point to be larger than the width of the beam point. The second · cooling 'point, the area where the temperature of the integral cooling beam spot rises. At this time, the cooling area becomes larger, and the larger tensile stress is applied to the whole portion cooled by the second cooling point. The first cooling step has been formed. In areas with shallow cracks, a larger tensile stress is applied, and as a result, shallower cracks will penetrate deeper in the depth direction. 200914385 is the first cooling The step of forming a shallow crack to cut the predetermined line to the exact crack, so as to form a crack along the position of the minute Chisaki, starting from the shallow crack formed previously, and making it deep into the cold step, according to the present invention, The shallower fissures are different: the cooling point of the piece is continuously deep in the shape of the crack. The position of the soap is stable in the position of the mark, and it can be better to make the width of the beam point reach the maximum in the center of the beam point. Therefore, the center of the beam spot is heated to the maximum, and the temperature difference between the center of the beam point is maximized by the first cooling, so that the concentrated portion of the thermal stress (stretching =) is the center of the beam spot, that is, the dividing line, so & cutting the pre-twisting line to accurately form a shallow crack. In the above invention, the first cooling step is preferably such that the first cooling point just passes through the in-plane temperature distribution perpendicular to the dividing line to sandwich the dividing line Forming a pair of high temperature regions on the left and right sides and cooling so that the yy degree of the segmentation line is lower than the temperature range of the above-mentioned temperature range, and by the pair of temperature regions The thermal stress generated by the temperature difference from the extremely small temperature region described above forms a shallow crack. Thereby, the temperature distribution perpendicular to the plane of the planned dividing line becomes a shape in which the extremely small temperature region is sandwiched between the two temperature peaks (high temperature region), and thus the thermal stress (tensile stress) is concentrated in a region having a very small temperature, which is at a temperature Very small areas form cracks accurately. In the above invention, the refrigerant may be contained in the refrigerant, and the first cooling point is formed by spraying a mist jet that sprays the refrigerant that atomizes the water, and the 12th 200914385 second cooling point is sprayed by a large range of water. After gasification, self-cooling is formed by vaporization cooling of the refrigerant lower than the above-mentioned atomizer. Here, the term "atomized refrigerant" refers to a process in which a refrigerant containing water is concentratedly sprayed toward a small area and is ejected from a nozzle in a state where it is difficult to vaporize, so that self-cooling phenomenon caused by vaporization of moisture does not occur and is ejected. The state of the refrigerant. When the atomized refrigerant is used, since the self-cooling does not occur, the temperature of the refrigerant does not decrease, and the smaller area can be concentrated and cooled. The term "gasification cooling" refers to a method in which a refrigerant containing f water is sprayed so as to diffuse water, and a self-cooling phenomenon caused by gasification of a water jet in a refrigerant from a nozzle nozzle in a state where it is easy to vaporize, and a refrigerant temperature Reduce and cool with a low temperature refrigerant. With gasification cooling, it is difficult to centrally cool a small area, but it can cool a wide area at a low temperature. With the present invention, it is easy to carry out a partial cold portion by means of atomized refrigerant, and on the other hand, it is strongly cooled by a gasification-cooled refrigerant. In the above invention, preferably, the relative moving speed of the beam spot, the first cooling point, and the second I cooling point is from 10 mm / sec to 72 〇 mm / sec. + according to the present invention, although there are differences in thickness, material, and processing method, by continuously performing the steps of forming a shallow crack and forming a deep crack, even if ΐθθ claws - seconds to 72 〇 ... seconds The speed is increased, and cracks can be surely formed along the line to be divided, so that cracks can be formed at a high speed. [Embodiment] Hereinafter, embodiments of the present invention will be described based on the drawings. The figure is a schematic configuration diagram of a substrate processing apparatus LSI according to the embodiment of the present invention. Here, a case where the glass substrate is processed will be described as an example. 13 200914385 The substrate processing unit Ls is described in the horizontal frame A. A pair of guides along the σ is arranged in pairs. The front and rear directions of the paper (hereinafter referred to as the Υ direction) are provided in the door 2 of the two guide rails 3 and 4, and the slide table 2 of the κ 彺 私 。 。. The lead screw 5 is provided with a lead screw 5 on the screw (4), and the motor (: a Γ the above-mentioned sliding ... fixing member 6 and the positive and negative rotation by the guide 3 4 = complex moving lead screw 5) 4 is set to move in the γ direction. The right side moving table 2 is provided with a water platform seat 7 which reciprocates along the guide 8 to the left Vi of Fig. 1 (hereinafter referred to as "direction". Guide: MU Betong screw has a lead screw 10 that rotates with the motor 9. The screw is rotated by the screw, and the yoke is rotated in the X direction. The reverse rotation causes the pedestal 7 to reciprocate along the guide rail 8. The gantry & 7 is provided with a rotary table 12 that is rotated by a rotating mechanism 11, and a brittle material substrate such as a surface substrate A is attached to the spirometer 2 in a horizontal shape g. The glass (10) is inversely used for cutting A smaller single 1 mother board is provided. The rotation mechanism U rotates the rotary table 12 about a vertical axis so that it can rotate at an arbitrary rotation angle with respect to the reference position. Further, the glass substrate AH is fixed to the rotation by the suction chuck. Above the turntable 12, the laser mount 13 and the optical holder 14 are held on the mounting frame 15. ', the field device 13 'As a defensive use of a brittle material substrate, a conventional laser device can be used. Specifically, excimer laser, YAG laser-oxidized stone anti-laser or carbon monoxide laser is used. Processing on glass substrate A Preferably, a carbon dioxide laser that oscillates light of a wavelength at which the energy absorption efficiency of the glass material is large is used. 14 200914385 A laser beam emitted from the laser device 13 has a beam spot of a predetermined shape The optical holder 14 in which the lens optical system for adjusting the beam shape is assembled is irradiated onto the glass substrate a. Regarding the shape of the beam spot, the shape of the long axis (elliptical shape, oblong shape, etc.) can be high along the division line. Although it is excellent in heating efficiently, the shape of the beam spot is not particularly limited as long as it can be heated at a temperature lower than the softening temperature. In the present embodiment, a beam spot having an elliptical shape is formed. 2 shows the plan view (a) of the elliptical beam spot to be irradiated and the thermal energy intensity distribution (b) in the width direction (short axis direction). The width of the beam spot bS The heat energy intensity distribution, such as a Gaussian distribution, is distributed to the left and right symmetrical and the center of the beam point is the largest thermal energy. When the beam spot Bs moves on the substrate, the trace passing through the central portion of the beam spot BS is heated to the highest Degree (but below the softening temperature). Regarding the thermal energy intensity distribution in the long axis direction of the beam spot B s , it can be either a Gaussian distribution or a thermal energy intensity distribution with different distribution shapes for increasing the heating rate. 15. The proximity optical holder 14 is provided with a first cooling nozzle 16a and a second cooling nozzle 16b. The refrigerant is sprayed by the cooling nozzles i^, (6). The refrigerant can use cooling water, compressed air, helium, dioxide, etc. In the present embodiment, a mixed fluid of cooling water and compressed air is sprayed. & The refrigerant ejected from the first cooling nozzle 16a is irradiated to the end portion in the longitudinal direction of the beam spot BS of the glass substrate A from the optical holder ,, and a first cooling point (3) is formed on the surface of the glass substrate A. The refrigerant injected from the first cooling nozzle 16a is sprayed by the nozzle so as to be misted. 15 200914385 The refrigerant injected from the second cooling nozzle 丨6b faces the end portion of the first cooling point cs^, or a portion partially overlapping the first cooling point, or slightly away from the rear end of the first cooling point CS1, on the glass substrate The surface of the eight forms a second cooling point CS2. Since the refrigerant injected from the second cooling nozzle 16b is sprayed by the nozzle in a wide range of diffusion, the moisture can be cooled. Fig. 3 is a plan view showing the positional relationship and the dimensional relationship between the beam spot BS, the first cooling, the point CS1, and the second cooling point CS2 formed along the dividing line p on the glass substrate A. The beam spot BS, the first cooling point CS1, and the second cooling point cs2 are each an elliptical shape or an ellipse having a long axis and the long axis direction is coaxial with the divided predetermined line P, respectively. When the width (short axis) of the pupil point Bs is Wb, the width of the first cooling point is Wcsl, and the width of the second cooling point is 8 Wes, the magnitude relationship of the widths is Wcsi < WdWm. Specifically, the width w(3) of the first cooling point CS1 is set to be about 1/4 to 3/4 of the width of the beam spot BS, and more preferably about μ. For example, if the width (short axis) of the beam spot BS is set to 2 bribes, the short axis of the first cooling point c S 1 is set to 1 mm for ejection. Further, the width Wcs2 of the second cooling point CS2 is set to be equal to or slightly wider than the width ^ of the beam spot Bs. For example, when the width of the beam and the point bs (the axis) is set to 2 mm, the short axis of the second cooling point (3) is set to 3 to be ejected with the left and right. Further, the cutter wheel 18 is attached to the mounting frame 15 via the vertical movement adjustment mechanism 17. The cutter wheel 18 is formed by forming an initial crack (trigger point) on the edge of the glass substrate A, and moving the pedestal 7 in the χ direction while temporarily lowering it. Further, the substrate processing apparatus is equipped with an alignment detected by the camera 20 for locating the glass slab A for positioning: the positioner 20' can be rotated by the position of the camera. The division of the predetermined line P is followed by the description of the operation of placing the glass substrate A on the rotary table == in the above substrate processing apparatus. 对准 The alignment mark imprinted on the glass substrate 8 is detected by the camera 20, and the detection result is established based on the detection result. The positional relationship between the turret and the slewing ring 12 and the sliding port is adjusted. Then, the position of the rotating table 12 and the sliding table 2 is adjusted so that the predetermined dividing line Ρ reaches the position opposite to the cutter wheel 18. In order to form the initial turtle m at the end of the glass substrate, the rotary table 12 is returned to the initial position (the right end in FIG. 1), and the rotary table 12 is in the direction of the lowering of the cutter wheel 18 (in FIG. The middle is moved from the right to the left, whereby the cutter wheel 18 is crimped to the side end of the glass substrate a, and the initial crack (TR) is opened. After the initial crack (TR) is formed, the rotary table 12 is again made. Return to the initial position 'and emit laser light from the laser device 13 The refrigerant is ejected from the first cooling nozzle 16a and the second cooling nozzle 16b. In this state, the rotary table 12 is moved in the _χ direction (from right to left in the figure!). The predetermined dividing line 'ρ of the glass substrate , is used to scan the beam spot BS, the first cooling point CS1, and the second cooling point CS2 in this order. μ The following state is explained, that is, the glass is produced by the above operation The substrate Α is divided into a predetermined line Ρ and the initial crack TR is used as a starting point to form a 17 200914385 W mark, which is formed on the glass substrate A. _, person " clothing exhibition due to the first cooling point CS 1 and the brother one cold part The state inside the cooling lane of the point CS2. Fig. 4 is a diagram showing the basis of the beam point BS along the division predetermined P and the width direction of each point by the first cooling point CS1 and the second cooling point (3) Schematic diagram of temperature distribution change and stress change (in the short axis direction). Fig. 4 (a) # ^ ^) is not heated by the beam spot BS' and by the first cooling point, 筮_" A top view of the positional relationship of the area where the cooling is performed. As shown in the figure below, after the heating of the beam point bs, the section of the first cold portion CS 1 is cooled, and the section orthogonal to the dividing line P is DD, the profile is When the flute, person, and ^ 曲 are cooled by the first cold spot CS1, the father is on the section AF p of the dividing line P, and the 囱 丨 & 』 囱 EE section is cooled by the second cooling point CS2. The section intersecting the planned dividing line p is a Μ section. k. Figure 4 (b) is a temperature profile of the substrate surface on the D-D profile and a schematic representation of the temperature distribution and thermal stress inside the substrate. After heating by the beam spot BS, on the D-D' section near the front where the first cooling point (3) is cooled, since the thermal energy intensity of the beam spot Bs is distributed at the center, the temperature distribution is presented. There is a distribution of temperature spikes that protrude upward from the predetermined dividing line P. Further, in the case of D_D, the temperature distribution inside the substrate on the cross section and the thermal stress are radially radiated from the substrate surface side centering on the division planned line P, and as a result, a semicircular temperature distribution is formed, and the surface is formed on the substrate surface. Compressed stress is nearby. Fig. 4 (c) is a diagram showing the temperature distribution of the surface of the substrate and the inside of the substrate on the cross section of E-E, and the temperature distribution and thermal stress inside the substrate. In the E_E which has been cooled by the first cooling point CS1, the width of the surface of the substrate on the surface of the substrate is reduced to less than the width of the beam, the heating of the point BS, and therefore the temperature distribution changes. In order to form a trough (a very small temperature region) at the center of the peak of the warmth, it becomes a temperature distribution of the peaks. As a result, a large thermal stress (tensile stress) is locally generated above the adjacent planned line p and near the surface, so that cracks are formed on the surface. On the other hand, since the inside of the substrate is locally heated by the beam spot, the heat transferred from the surface of the substrate is not sufficiently transferred to the inside of the substrate, so that temperature rise is less likely to occur inside the substrate. The result is that only a large thermal stress generated near the surface of the substrate causes a crack to form near the surface of the substrate, but since the crack does not enter the inside of the substrate, the result is a shallow crack. Fig. 4(d) is a view showing the temperature distribution of the substrate surface and the inside of the substrate on the cross section of F-F, and the temperature distribution and thermal stress inside the substrate. In the F_Fi section cooled by the second cooling point CS2, since the width of the substrate cooled by the second cooling point CS2 is the same as the width of the beam spot heating, or wider than the width of the beam spot BS, I The temperature peak is cooled as a whole, so the temperature spike disappears and a large thermal stress (tensile stress) is formed on the entire surface. On the other hand, heat transferred from the surface remains in the inside of the substrate, and as a result, a large temperature difference is generated in a large extent in the inside of the substrate and the surface of the substrate. The temperature difference causes tensile stress on the surface of the substrate, and compressive stress is generated inside the substrate, whereby a force for cracking the previously formed shallow crack is generated due to the surrounding stress distribution, and the shallow crack is further exerted. . Fig. 5 is a view showing a state in which cracks are generated in a cross section along the direction of dividing the predetermined line P. 19 200914385 As described so far, "in the region L1 heated by the beam spot BS", although it is subjected to a compressive stress perpendicular to the direction of the paper surface in the vicinity of the surface, no crack is generated. In the region L2 cooled by the first cooling point CS1, a local tensile stress in a direction perpendicular to the plane of the paper is generated in the vicinity of the surface (refer to Fig. 4 (c)) to form a shallow crack S2. Further, in the region L3 cooled by the second cooling point CS2, the tensile stress in the direction perpendicular to the paper surface of the substrate and the tensile stress in the direction perpendicular to the paper surface of the substrate surface are made shallower in the vicinity of the surface. The crack S2 further acts as a force (see Fig. 4(d)), thereby forming a deep crack S3. In this way, the first stage is stable and does form a shallow crack, which serves as a trigger point to make the crack deep in the second stage, thereby stably stabilizing the beam spot BS or the like even at a speed higher than the current speed. Form deep cracks. Specifically, even in the case of a glass substrate in which cracks cannot be stably formed at a speed of 1 mm/sec or more in the past, cracks can be stably formed at a speed of 720 mm/sec or less. For example, when the X-ray C照射2 laser is used for scribing, it is conventionally possible to perform scribing processing only at a scanning speed of 240 W and 500 mm/sec or less, but by using the present invention, the scanning speed is accelerated even for the same substrate. Processing is also possible up to 550 mm/sec. As a result, cross cutting can be realized at high speed. Moreover, high-speed crack formation of up to 720 mm/sec has also been experimentally confirmed. Further, the use of the present invention allows cracks to be deeper than before. As a result, in the related art, the breaking device is used to break the substrate by pressing the breaking bar along the crack (dash line). According to the present invention, the substrate can be cut without breaking the substrate, and the non-breaking step can be realized. Overall cutting. The present invention can be applied to the formation of cracks on a brittle material substrate such as a glass substrate, and is further used for division. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic configuration diagram of a substrate processing apparatus using the processing method of the present invention. Fig. 2 is a view showing an example of a thermal energy intensity distribution of a laser beam. f
圖3係表示光束點、第一冷卻點、第二冷卻點之位置 關係等之俯視圖。 圖4係說明以光束點進行基板加熱，以及以第一冷卻 點CS 1、第二冷卻點CS2進行冷卻時各點上之寬度方向（短 軸方向）之溫度分布變化、應力變化之示意圖。 圖5係表示沿著分割預定線之方向之剖面上之裂痕產 生狀態的示意圖。 / 6係說明以光束點進行基板加熱，其後進行輔助 P時各點上之寬度方向（短軸方向）之溫度分布變化、 應力變化之示意圖 【主要元件符號說明】 A 玻璃基板（脆性材料基板） BS 光束點 CS1 第一冷卻點 CS2 第二冷卻點 P 分割預定線 S2 較淺裂痕 S3 深入之裂痕 21 200914385 2 滑動台 12 旋轉台 13 雷射裝置 14 光學保持器 16a 第一冷卻喷嘴 16b 第二冷卻喷嘴Fig. 3 is a plan view showing a positional relationship between a beam spot, a first cooling point, and a second cooling point. Fig. 4 is a view showing the temperature distribution change and the stress change in the width direction (short axis direction) at each point when the substrate is heated by the beam spot and the first cooling point CS 1 and the second cooling point CS2 are cooled. Fig. 5 is a view showing a state in which a crack is generated in a section along a direction in which a predetermined line is divided. / 6 is a schematic diagram showing the temperature distribution change and the stress change in the width direction (short axis direction) at each point when the auxiliary point P is performed by the beam spot. [Main element symbol description] A glass substrate (brittle material substrate) BS beam spot CS1 first cooling point CS2 second cooling point P dividing line S2 shallower crack S3 deep cracking 21 200914385 2 sliding table 12 rotating table 13 laser device 14 optical holder 16a first cooling nozzle 16b second Cooling nozzle
22twenty two