201211290 六、發明說明： 【發明所屬之技術領域】 本發明係關於-種㈣乾及其製造方法，更具體而言 係關於一種銦靶及其製造方法。 【先前技術】 姻係用於Cu- in— Ga— Se系（CI(}S系）薄膜太陽電 池之光吸收層形成用之濺鍍靶。 以彺，銦靶主要係以熔解鑄造法來製造。 於日本特公昭63 — 4482G號公報（專利文獻1)中記載 下方法·於支持板形成銦薄膜後，使銦流入該薄膜上， 進仃鑄造’藉此與支持板形成為一體。 士又於日本特開2〇1〇__ 24474號公報（專利文獻2)中 :載有如下方法：將規定量之銦原料投人加熱之铸模中， 其熔解，去除懸浮於表面之氧化銦，進行冷卻而獲得铸 然後研削所得之缚錠表面而得到銦乾，此時並非一欠 ?:定量之銦原料投入鑄模中，而是分成多次投入，去除 人生成之>谷液表面的氧化銦，然後 進行表面研削而獲得銦靶。 ^之‘旋 專利文獻1 .日本特公昭63 - 44 820號公報 專利文獻2 .日本特開2010 - 24474號公報 【發明内容】 速度：而則炫解-造法來製造銦乾時，若冷卻 ;靶内部會產生空孔，因而有 常放電之問題。另一士 力雌鍍時產生異 万面，若減小冷卻速度，則會有結晶 201211290 粒徑變大、賤鍍速率變小之問題。又，由 法所製造的銦靶，亦有濺 μ解鑄以 率隨時間改變之問Γ 鐘速率不同，或錢錄速 因此’本發明之課題在於提供—種可抑制異常放電之 ’同時可維持高濺料率之銦減其製造方法。 本發明人為了.+. π n 解、上述课螭，經潛心研究合 =:::卻，且表面相較於内部更快冷卻，因而根據： ' 3構成乾之銦晶粒之粒徑必然不JS]。並且發現 因此而發生濺鍍面内路 改變的問題。 賤錢速率不同、或賤鑛速率隨時間 因此’可知於轉#造峰後，藉由壓延使平均粒徑 =J至疋程度’可減少中央部之結晶粒徑相對於輕外圍 4的不同、或者内部之結晶粒徑相對於表面的不同 由Μ延可使内部之空孔被遲扁而縮小。且可知以上述方^ 所獲得之銦無可抑制異t放電並同時維持高难鑛速率。 基於以上見解而完成之本發明於一態樣令，係一種鋼 靶其表面之平均結晶粒徑為10mm以下而自平行於塵延 方向之剖面觀察到的晶粒，垂直於_方向之方向的 粒從相對於平行於壓延方向之方向的平均粒徑之比 以上m7,且孔徑心,以上之空孔為1個八 下0 本發明之銦靶於一實施形態中，靶表面之 徑為2〇mm以下。 大日日粒 4 201211290 部之ΐ?:之銦靶於另—實施形態中，存在於靶表面中央 結晶的平均粒徑相對於存在於乾表面外圍部…的 平均粒徑之比為10〜18。 固Ρ之一曰的 平行且通過= = 一實施形態中，存在於與無表面 相對於存=表Γ::Γ的切割面之結晶的平均粒徑 ' 之、，°晶的平均粒徑之比為1 . 〇〜丨8。 發月於另一態樣中’係一種銦靶之製造方法，豆 含有對銦原料進行輯铸造後進行冷軋之步驟。、201211290 VI. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates to a seed (four) dry and a method of manufacturing the same, and more particularly to an indium target and a method of manufacturing the same. [Prior Art] The marriage system is used for a sputtering target for forming a light-absorbing layer of a Cu-in-Ga-Se system (CI(}S-based) thin film solar cell. The germanium and indium targets are mainly produced by a melt casting method. Japanese Patent Publication No. Sho 63-4-482G (Patent Document 1) describes a method in which an indium film is formed on a support sheet, and indium is poured into the film to form a tantalum casting, thereby forming a unitary body with the support sheet. In Japanese Laid-Open Patent Publication No. Hei. No. 24474 (Patent Document 2), there is a method in which a predetermined amount of indium raw material is poured into a heating mold, which is melted to remove indium oxide suspended on the surface. After cooling, the surface of the obtained ingot is obtained by grinding and then obtained, and the indium is not dried. In this case, the indium raw material is put into the mold, but is divided into multiple inputs to remove the indium oxide formed on the surface of the human liquid. Then, the surface is ground to obtain an indium target. The patent is disclosed in Japanese Patent Application Laid-Open No. Hei. No. Hei. No. 2010-24474. When making indium to make indium, if it is cold There is a problem that the inside of the target will have voids, so there is a problem of frequent discharge. Another stone force produces a different surface when it is plated. If the cooling rate is reduced, there will be a problem that the crystal size of 201211290 becomes larger and the plating rate becomes smaller. In addition, the indium target manufactured by the method also has a different rate of time due to the change of the sputtering rate, or the speed of the recording. Therefore, the object of the present invention is to provide a kind of "can suppress abnormal discharge" The invention can maintain the high sputter rate of indium minus the manufacturing method. The present inventors have studied the combined =::: but the surface is cooled faster than the inside for the .+. π n solution, the above-mentioned lesson, and thus according to: '3 The particle size of the dry indium grains is inevitably not JS.> And the problem of the internal path change of the sputtering surface is found. The rate of money is different, or the rate of the ore is time-dependent. By calendering, the average particle diameter = J to the degree of ' can reduce the difference in the crystal grain size of the central portion with respect to the light periphery 4, or the difference in the internal crystal grain size with respect to the surface. It is reduced by being flattened, and it can be known that the above side ^ The obtained indium does not inhibit the iso-t discharge while maintaining a high difficult ore rate. The present invention based on the above findings is a state in which a steel target has an average crystal grain size of 10 mm or less from parallel to dust. The grain observed in the cross section of the direction, the ratio of the grain perpendicular to the _ direction from the average particle diameter in the direction parallel to the rolling direction is m7, and the pore diameter, the upper hole is one octave In one embodiment, the indium target of the invention has a diameter of 2 〇 mm or less on the surface of the target. Days of the granules 4 201211290 ΐ : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 铟 铟The ratio of the diameter to the average particle diameter existing in the outer periphery of the dry surface is 10 to 18. The parallelism of one of the solids and the pass of = = in one embodiment, the average particle diameter of the crystal of the crystal with respect to the surface of the cut surface of the surface: The ratio is 1. 〇~丨8. In another aspect, the invention is a method for producing an indium target, and the bean contains a step of cold-rolling the indium raw material after casting. ,
料溶m之銦靶之製造方法於-實施形態中’在將銦原 才4这解鑄造後，進行A 1。_以下。 冷軋直至乾表面之平均結晶粒徑在 ^發月之銦靶之製造方法於-實施形態中，於熔解鑄 w銦原料時’以3〜7Gt/分之冷卻速度進行冷卻，接著， 以合計50〜80q/。之軋縮率進行冷軋。 根據本發明，可獲得一種可抑制異常放電之產生，並 同時可維持高賤鍍速率之銦乾。 … 【實施方式】 本發明之銦把具有如下特徵：乾表面之平均結晶粒徑 击广m以下，而自平行於壓延方向之剖面觀察到的晶粒， 一直於壓延方向之方向（板厚方向）@平均粒徑相對於平 仃於壓延方向之方向的平均粒徑之比（以下亦稱作「剖面 ^比」Ο為（M以上、未達仏通常，㈣解鑄造法製 作銦铸錠時，若欲避免於銦铸❹產生空孔，則必須以— 定程度緩慢之冷卻速度進行冷卻，此情形時之平均結晶粒 201211290 徑會變大至約10mm以上左右。此稽曰 種大、、〇 B日粒徑會使得濺鍍 之成膜速度變小。於本發明中藉由對銦鑄鍵進行冷乳，而 對結晶結構施加物理力’利用滑動差排等之作用而縮小結 晶粒徑。又’施加於結晶結構物 ^初理力會作用至銦鑄錠内 部，而亦可縮小銦铸錠内部之咭 晶粒徑。另一方面，由於 藉由冷軋而將結晶粒徑壓扁於妬 构於板厚方向，故剖面縱橫比會 小於壓延前。 結晶粒徑可藉由冷乾時之軋縮率進行調節，越提高軋 縮率則結晶粒徑越小，❹縱橫比越小。只要結晶粒徑變 小則成膜速度會相應地變大，但只要為某程度以下之粒 徑，則即使粒徑變得更小，由其引起之成膜速度的增大效 果會逐漸遞減，相對於此，㈣料粒徑之壓延次數所花 費的夫會逐漸增加。因此，銦乾表面之平均結晶粒徑較 佳為卜6賴’更佳為卜3随。又，剖面縱橫比，較佳為 〇·1〜0_6，更佳為〇 }〜〇 5。 於本發明中，銦纪表面之平均結晶粒徑係利用以下之 法進行測疋。以弱酸稍微蝕刻靶表面，或於表面擦附碳 粉使晶界容易|見之後，將靶表面中央部之任意 5mm 50mm 圍及把表面外圍部之任意25咖心咖範圍 的區域作為測疋對象區域，肖由目測，計算該區域内之晶 粒的個數（N)。橫跨存在於區域邊界之晶粒視A 0.5個。 藉由將測定對象區域之面積（S= 1250mm2)除以晶粒之個 (、）算出曰曰粒之平均面積（s )。將晶粒假設為球，用 下式算出平均結晶粒徑（A )。 201211290 1/2 A= 2 ( s/ π ) 於本發明Φ .. ，d面縱橫比係利用以下之方 定。沿著板厚方A m ^ w _ 左硬订測 向對靶進行切割，使切割面通過靶丨〇 心抽1 5且平并认防. ^ Ύ ;1延方向。以弱酸稍微蝕刻露 16，或於切到而,< μ 山刀割面 ° 6 “附碳粉而使晶界容易看見之後， 意2〇個晶粒作為敎對象，藉由目測，算出與壓延方向平 的平均粒徑(a)、及與壓延方向垂直之方向（板 旱向）y的平均粒徑（b)，從而求出剖面(b/ + ^ =方向平行之方向的平均粒徑，係當將各晶粒可 延方向平行的最長線段作為各晶粒之與壓延方 向平行之方向的粒徑時測定對象之晶粒的平均值。 與壓延方向垂直之方向的平均粒徑’係當將各晶粒可 包圍之與壓延方向垂直的最長線段作為各晶粒之與壓延方 向垂直之方向的粒徑時測定對象之晶粒的平均值。 本發明之銦乾於較佳之實施形態中，最大結晶粒徑為 20麵以下。除了控制乾整體之平均結晶粒徑以外，亦將最 :結晶粒徑控制為2〇_以下，藉此，由於可使結晶粒徑之 分佈之不均減少’故可使濺鑛之成膜速度的變化變小，且 可排除成膜速度特別慢之區域。最大結晶粒徑較佳為15_ 以下，更佳為l〇mm以下，例如為5〜1〇職。 於本發明中，銦乾之最大結晶粒徑係利用以下之方法 進行測定。 7於上述之平均粒徑測定時測定對象面積内的晶粒中 最大晶粒的面積（smax)’係假設晶粒為球，用下式算出最 201211290 大粒徑（B )。 1/2 B = 2 ( smax/ π 本發明之姻靶於較佳之實施形態中，存在於乾表 央部之結晶的平均粒徑相對於存在於靶表面外圍部之、纟士 = 的平均粒徑之比為在通常之熔解鑄造法中， 速度較外圍部慢的中央部其結晶粒徑較容易變大，作=部 本發明’可於外圍部與中央部縮小結晶粒徑之差。藉2據 可使濺鍍速率之平面均一性獲得提高。其結果，獏厚之 面均一性獲得提高。該比較佳為1.〇〜16，更佳為ι〇〜ΐ4' 於本發明中，所謂「把表面中央部」係定義為乾表面 内包括把表面之中心點並佔紙總面積之一半面積的區域， 且為具有與靶表面形狀相似之形狀的區域。因此，例如， 於半徑R之圓板狀紙之情形時，所謂中央部係指半徑 Γ義=1。因此’於本發明中所謂「％表面外圍部」則是 =為：表面内除中央部以外的區域。圖1示意性地顯示 靶表面中央部與靶表面外圍部。 於本發明中，存在於靶表面中央部之結晶的平均粒徑 ’於存在於靶表面外圍部之結晶的平均粒徑之比，係利 用以下之方法進行測定。 用以：先，存在於靶表面中央部之結晶的平均粒徑，係利 用以下之方法進行測定。 :弱醆稍微蝕刻靶表面，或於表面擦附碳粉使晶界容 後將乾表面中央部之任意25mmx50mm的範圍作 為測定對象區域，由曰、al ^ ”已固作 曰由目測，計算該區域内之晶粒的個數 8 201211290 (N ) °橫跨存在於區域邊界之晶粒視為〇 5個。藉由將測 定對象區域之面積= 1250mm2)除以晶粒之個數（Ν)， 算出晶粒之平均面積（s )。將晶粒假設為球，用下式算出平 均結晶粒徑（A )。 A= 2 ( s/ π )丨" 另一方面，存在於靶表面外圍部之結晶的平均粒徑， 係利用以下之方法進行測定。 以弱酸稍微蝕刻靶表面，或於表面擦附碳粉使晶界容 易看見之後’將靶表面外圍部之任意之25mmx5〇mm的範圍 作為測定對象區域，藉由目測，計算該區域内之晶粒的個 數（N·)。橫跨存在於區域邊界之晶粒視為μ㈣。藉由將 /則疋對象區域之面積（s = 125〇_2)除以晶粒之個數（Ν)， 算出晶粒之平均面積⑴。將晶粒假設為球， 均結晶粒徑（A )。 、异十 A = 2 ( s/ τΓ ) 1/2 根據用上述方法痛^圼+十+ λ 平均粒徑、與存在Si 乾表…部之結晶的 在於輕表…部之！=圍部之結晶的平均粒徑，存 外圍部之結晶的平的平均粒徑相對於存在於乾表面 部之姓曰的二 徑之比係利用（存在於無表面中央 口Ρ之、.ΌΒ日的平均粒徑）/ 、 平均粒徑）求得。 〖在於靶表面外圍部之結晶的 本發明之銦乾於較佳之實施形 平行且通過靶中心軸 " 存在於與靶表面 内部）之結晶的 ^面（以下亦稱作「靶 均^相料存m表面之結晶的 201211290 平均粒徑之比為10〜18。於 度較乾表面慢之内部其結晶㈣較造法中，冷卻速 明，可於…與内部縮小結晶粒徑之差。藉：根據本發 骑鑛速率隨時間改變。該比較佳為’可減少 1.4。圓2不意性地顯示靶表面與靶内部。 ”” 1.0〜 :本發明中，存在於與乾表面平 中央點的切割面之結晶的平均粒 ：乾中、軸之 結晶的平均粒徑之比，係利用以下之方法進乾表面之 首先，利用上述方法測定峰表面之仃測定。 桩荽土輕表面之平均結晶粒徑。 接者，求出存在於與靶表面平 央點的切割面之a 通過1^中心轴之令 η, 的千均粒徑。其係藉由切片機切出哕 切割面後’利用與測定銦靶表面 出該 同的方法，、>丨$ + β 干$日日粒徑之方法相 的方法/則疋该切割面之平均結晶粒徑。 "根據用上述方法獲得之存在於革巴表面之結 徑、與存在於餘表面平行且通過Μ心轴之中央點^ 割：之：晶的平均粒徑，存在於與乾表 =之央點的切割面之結晶的平均粒徑相 = 2之結：的平均粒徑之比係利用（存在於乾表面之結； 中水均粒位）/(存在於與靶表面平行且通過靶中心軸之 中央點的切割面之結晶的平均粒徑）求得。 本發明之銦乾於較佳之實施形態中，于孔徑5〇”以上 ^孔為1個"以下。存在於取内部之空孔，尤其是孔 ^ 5 0 // m 以上之大空：f丨合士'丸他 孔會成為濺鍍時產生異常放電之原 因’故較佳為盡可能地減少。根據本發明，由於在冷乾時 10 201211290 將炫解鑄造時存在於㈣之空孔壓扁，因此可縮小空孔。 孔控50…上之空孔為i個八m3以下。孔徑以… 上之空孔較佳為0.5個/cm3以下’更佳為〇 3個/ 下，例如為0〜0.3個/cm3。 於本發明中，孔徑5〇…上之空孔的數量係利用電 子掃描式超音波探傷器進行敎。㈣放置於上述裝置之 探傷器水槽内，於頻帶…⑼驗、脈衝重複頻率則z、 掃描速度6〇mm/min下進行測定’才艮據所獲得之圖像，計 數孔徑50" m以上之空孔’再根據測定對象乾之體積求出 空孔之個數比例。此處’所謂孔徑係以包圍圖像之孔的最 小圓之直徑來定義。 接著，依序說明本發明之銦乾之製造方法的較佳例。 首先’使作為原料之銦熔解，並流人至鑄模巾。使用之原 料銦若含有雜質，則會導致由該原料製作之太陽電池的轉 換效率降低，因此較佳為具有高純度，例如可使用純度99 質量❶/。以上之銦。然後’冷卻至室溫，形成銦鑄錠。 即使於該階段中產生空孔或結晶粒徑大，由於可藉由 其後之冷軋予以控制，故並無太大問題，但是由於用以增 大乳縮率之壓延需要花費工夫’故於熔解鑄造之階段，較 佳為預先將空孔減少至某種程度，並將結晶粒徑縮小至某 種程度。 冷卻速度可為利用空氣之自然冷卻（約丨〇〇C /分），於 熔解鑄造之階段重視預先抑制鑄錠内之空孔時，可儘量緩 慢地例如以9。(： /分以下，較佳為以8〇c /分以下進行冷 201211290 卻。然而，若太慢，則由於本次無法充分地獲得超音波振 動之結晶粗大化抑制效果，故較佳設為3〇c /分以上，更佳 設為5。〇/分以上。另一方面，於重視防止結晶粒徑成長 時，則儘量快速地冷卻。尤其，可藉由冷軋使空孔縮小許 多，故藉由高速冷卻獲得微細之晶粒並獲得高濺鍍速率係 有利於提升靶整體之特性。例如可設為2〇t/分以上，較 佳可設為50°C /分。然而，若太快，則由於本次用以壓扁 空孔所需之壓延加工會變得麻煩，故較佳為最大以7〇<5(： / 分進行冷卻。 對於冷卻速度之調整，當減小冷卻速度之情形時，可 利用加熱器等對鑄模進行加熱保溫，相反地，於增大冷卻 速度之情形時，可利用將冷卻水供給至鑄模周邊之水冷等 方法來進行》此處之冷卻速度係以（銦之熔解溫度—25<t ) / (開始冷卻後，銦之溫度自熔解溫度降低至25C>c之時間） 來計算。 然後，對所獲得之銦鑄錠進行冷軋，直至整個平均結 晶粒徑在10mm以下，並視需要進行形狀加工或表面研磨而 製成銦把。《軋時之合計的軋縮率越A，則結晶粒徑越微 細化，結晶粒徑之不均越減少，空孔越縮小，剖面縱橫比 亦越小。就製造本發明之銦靶而言，合計的軋縮率較佳為 50〜80°/° ’更佳為60〜70%。 靶之厚度並無特別限制，可視使用裝置或使用目的作 適备0又疋，通常為3〜20mm左右，典型為5〜l〇mm左右。 可視所要之靶的厚度來調節冷軋前之銦铸錠的厚度或冷軋 12 201211290 的軋縮率。 以上述方式所獲得之銦靶，可適合使用作為CIGS系薄 膜太陽電池用光吸收層用之賤錄乾。 [實施例] 以下表不本發明之實施例與比較例，該等實施例係為 了能更好地理解本發明及其優點而提供者，並無限定本發 明之意圖。 使經160 °C熔解之銦原肖（純纟5N )流人至内徑 25〇mm、深度50酿之鑄模的内部後，以表i所記載之冷： 速度冷部至至溫（25 C)’形成圓盤狀之銦鑄錠（直徑 25〇mmx厚度30_)。接著，於表1所記載之條件下對_ 錠進行冷軋’藉此獲得發明例及比較例之各銦托。 將所獲知之銦1^加工成直徑2Q4mmx厚度職利用 ^方法測定如下之A〜F之特性值。a〜D之測定，係對 =面^表面研削且利用酸進行㈣而使粒界變得容易 觀察之後’進行藉由目測翻家„ , . 日狀械察、尺寸敎，E之測定係使 用日本K_k職㈣份有限公司製造 探傷系統PA-101。結果示於表卜 “式超曰波 A :每個部位之平均結晶粒徑 B :靶表面最大結晶粒徑 Γ 表1·中存央在於:表广中央部之結晶的平均粒徑(表1中以 平均粒徑(、丄二！目Γ存在於乾表面外圍部之結晶的 .衣中以表面外圍」表示）之比 D·存在於與乾表面平行且通過乾中心軸之中央點的切 13 201211290 割面之結晶的平均粒徑（表1中以「内部」表示）相對於 存在於乾表面之結晶的平均粒徑（表丨中以「整個表面」 表示）之比 E :孔徑5 0以m以上之空孔的個數比例 F :剖面縱橫比 又’表2表示利用ANELVA製造之SPF- 3 13H濺鍍裝 置’於開始濺鍍前之腔室内的真空壓力lxl〇_4pa、濺鑛時 之壓力0.5Pa、氬濺鑛氣體流量5SCCM、減:鍵功率650W下， 對該等發明例及比較例之銦靶濺擊分鐘後之結果。於表 2中’表示有初期之濺鍍速率及經過3.3K：whr濺鍍後之濺鍍 速率、濺鍍時之異常放電次數。 濺鑛速率係根據成膜時間與表面形貌儀測得之膜厚的 結果算出，異常放電次數則是利用目測之方法來測定。 由表1及表2可知： 比較例1中’由於沒有進行壓延，故結晶粒徑大，濺 鍍速率慢。又，由於晶粒之每個部位的不均皆大，故濺鍍 速率隨時間改變的情形亦很嚴重。 比較例2中’雖進行了壓延，但由於冷卻速度過大， 故無法充分地抑制空孔。因此，異常放電之次數多。 比較例3中’雖進行了壓延，但由於冷卻速度過小， 故結晶粒徑變大。 於發明例1〜3中，可知隨著提升冷卻速度，結晶粒徑 變小’每個部位之不均亦變小，可維持高濺鍍速率。又， 可知雖然空孔的數量隨著提升冷卻速度而會些微增加，但 14 201211290 可充分地抑制異常放電。 15 201211290 (X| 0.27 0.31 0.23 0.26 0.28 0.30 ω 〇 Ο CN 〇 〇 〇 CO c5 ο Q 00 1.17 1.55 1.19 1.16 1.06 u Ο) 00 〇 1.17 m 25.0 CN CN oo 00 CN A :每個部位之平均粒徑（mm) 1内部| 22.0 18.3 OS 00 00 iri 00 |整個表面| 12.0 <Ν 00 〇 in r-； |表面中央 16.0 rn 13.8 od 寸 00 表面外圍 (N 00 ο — 'O 合計之軋縮率 (%) o ν〇 冷卻速度 rc/分） 10 (自然冷卻） 100 (高速冷卻） 1 (低速冷卻） 3 (低速冷卻） 10 (自然冷卻） 70 (高速冷卻） d 比較例1 比較例2 比較例3 發明例1 發明例2 發明例3 201211290 [表2]A method for producing an indium target in which m is dissolved is in the embodiment. After the indium precursor 4 is melt-cast, A 1 is performed. _the following. The method for producing the indium target in the cold rolling until the dry surface is in the form of the indium target in the embodiment, and is cooled at a cooling rate of 3 to 7 Gt/min when melting the indium raw material, and then 50~80q/. The rolling reduction rate is cold rolled. According to the present invention, it is possible to obtain an indium dry which can suppress the generation of abnormal discharge while maintaining a high ruthenium plating rate. [Invention] The indium of the present invention is characterized in that the average crystal grain diameter of the dry surface is less than m, and the crystal grains observed from the cross section parallel to the rolling direction are always in the direction of the rolling direction (plate thickness direction). The ratio of the average particle diameter to the average particle diameter in the direction of the rolling direction (hereinafter also referred to as "section ratio" is ( (M or more, not up to 仏, (4) when ingot casting is performed by the casting method) If you want to avoid creating voids in the indium cast, you must cool it at a slower cooling rate. In this case, the average crystal grain 201211290 will increase to about 10 mm or more. The particle size of 〇B is such that the film formation speed of the sputtering becomes small. In the present invention, by applying cold strength to the indium cast bond, physical force is applied to the crystal structure, and the crystal grain size is reduced by the action of sliding difference or the like. Also, 'applied to the crystal structure ^ initial force will act on the inside of the indium ingot, but also reduce the crystal grain size inside the indium ingot. On the other hand, the crystal grain size is flattened by cold rolling Yu Yu is in the direction of the plate thickness, The aspect ratio of the section will be smaller than that before rolling. The crystal grain size can be adjusted by the rolling reduction rate during cold drying. The higher the rolling reduction ratio, the smaller the crystal grain size, and the smaller the aspect ratio of the warp. The film speed will be correspondingly increased, but as long as the particle diameter is a certain degree or less, even if the particle diameter becomes smaller, the effect of increasing the film formation speed caused by it gradually decreases, whereas (4) the particle size of the material The number of times of calendering will gradually increase. Therefore, the average crystal grain size of the indium dry surface is preferably a better one. Further, the aspect ratio of the cross section is preferably 〇·1 to 0_6. More preferably 〇}~〇5. In the present invention, the average crystal grain size of the indium surface is measured by the following method: slightly etching the target surface with a weak acid, or rubbing the surface with a carbon powder to make the grain boundary easy | After seeing, any 5mm 50mm circumference of the center of the target surface and any 25 coffee area of the outer surface of the surface are used as the measurement target area, and the number of crystal grains in the area is calculated by visual inspection (N). Across the boundary of the area, the grain is A 0.5. The area (S = 1250 mm 2 ) of the measurement target area was divided by the number of grains (the) to calculate the average area ( s ) of the particles. The crystal grains were assumed to be spheres, and the average crystal grain size (A ) was calculated by the following formula. 201211290 1/2 A= 2 ( s / π ) In the present invention Φ .. , d aspect ratio is determined by the following formula: along the plate thickness A m ^ w _ left hard to determine the direction of the target cutting, Let the cut surface pass through the target to draw 1 5 and flatten it out. ^ Ύ ;1 direction of extension. Slightly etch the exposed 16 with weak acid, or cut it, < μ knives cut surface ° 6 "with toner After making the grain boundary easy to see, it is possible to calculate the average particle diameter (a) which is flat with the rolling direction and the direction perpendicular to the rolling direction (board dry direction) y by visual inspection. The diameter (b) is used to obtain the cross-section (b/ + ^ = the average particle diameter in the direction parallel to the direction, and the longest line segment in which the respective crystal grains can be parallel-oriented is used as the grain in the direction parallel to the rolling direction of each crystal grain. The average value of the crystal grains of the object is measured at the time of the diameter. The average particle diameter in the direction perpendicular to the rolling direction is the average value of the crystal grains to be measured when the longest line segment which is surrounded by the respective crystal grains and which is perpendicular to the rolling direction is the particle diameter in the direction perpendicular to the rolling direction of each crystal grain. . The indium of the present invention is dried in a preferred embodiment, and has a maximum crystal grain size of 20 or less. In addition to controlling the average crystal grain size of the dry whole, the crystal grain size is also controlled to be 2 〇 _ or less, whereby the film formation speed of the splash can be made due to the uneven distribution of the crystal grain size. The change is small, and the area where the film formation speed is particularly slow can be excluded. The maximum crystal grain size is preferably 15 Å or less, more preferably 10 Å or less, for example, 5 to 1 〇. In the present invention, the maximum crystal grain size of indium dry is measured by the following method. 7 In the measurement of the average particle diameter described above, the area (smax) of the largest crystal grain in the crystal grain in the measurement target area is assumed to be a sphere, and the maximum particle diameter (B) of 201211290 is calculated by the following formula. 1/2 B = 2 ( smax / π In the preferred embodiment, the average particle diameter of the crystal present in the central portion of the dry surface is relative to the average particle of the gentleman = present in the peripheral portion of the target surface. The ratio of the diameter is such that in the normal melting casting method, the crystal grain size is more likely to be increased in the central portion which is slower than the peripheral portion, and the difference between the crystal grain size in the outer portion and the central portion of the present invention can be reduced. According to the invention, the uniformity of the plane of the sputtering rate can be improved. As a result, the uniformity of the surface of the crucible is improved. The comparison is preferably 1.〇~16, more preferably ι〇~ΐ4'. In the present invention, the so-called The "center portion of the surface" is defined as a region in the dry surface that includes a center point of the surface and occupies one-half of the total area of the paper, and is a region having a shape similar to the shape of the target surface. Therefore, for example, at a radius R In the case of a disk-shaped paper, the central portion means a radius =1 = 1. Therefore, the "% surface peripheral portion" in the present invention is a region other than the central portion in the surface. The central portion of the target surface and the peripheral portion of the target surface are displayed. In the invention, the ratio of the average particle diameter of the crystal present in the central portion of the target surface to the average particle diameter of the crystal present in the peripheral portion of the target surface is measured by the following method: First, it is present in the center of the target surface. The average particle diameter of the crystals of the part is measured by the following method: weakly rubbing the surface of the target slightly, or rubbing the surface of the carbon powder to make the grain boundary, and the range of 25 mm x 50 mm at the center of the dry surface is used as the measurement target region. From 曰, al ^ ” has been fixed 目 by visual inspection, calculate the number of grains in the area 8 201211290 (N ) ° across the grain boundary of the area is considered to be 〇 5. By measuring the object The area of the area = 1250 mm 2 ) divided by the number of crystal grains (Ν), and the average area (s ) of the crystal grains was calculated. The crystal grains were assumed to be spheres, and the average crystal grain size (A ) was calculated by the following formula: A = 2 ( s / π ) 丨 " On the other hand, the average particle size of the crystal present in the peripheral portion of the target surface is determined by the following method: slightly etching the target surface with a weak acid, or adhering the carbon powder to the surface to make the grain boundary easy After seeing 'outside the target surface The range of 25 mm x 5 〇 mm of any part is used as the measurement target area, and the number of crystal grains (N·) in the area is calculated by visual inspection. The crystal grains existing across the boundary of the area are regarded as μ (four). Then, the area of the target region (s = 125 〇 2) is divided by the number of grains (Ν), and the average area of the crystal grains (1) is calculated. The crystal grains are assumed to be spheres, and the average crystal grain size (A). Ten A = 2 ( s / τ Γ ) 1/2 According to the above method, the average particle size of the pain + 十 + 10 + λ, and the crystal of the Si dry surface... is in the light part... The average particle diameter, the ratio of the average average particle diameter of the crystal of the peripheral portion to the ratio of the two diameters of the surname 存在 existing in the dry surface portion (existing in the surface of the surface without the surface, the average particle diameter of the day) / , average particle size) obtained. The indium of the present invention which is crystallized at the outer periphery of the target surface is dried in a preferred embodiment and is parallel to the target axis and exists in the surface of the target surface (hereinafter also referred to as "target" The ratio of the average particle size of the 201211290 crystal of the m surface is 10 to 18. The crystal is slower than the dry surface. (4) Compared with the method, the cooling is quick, and the difference between the crystal grain size and the inside can be reduced. : According to the present invention, the rate of riding is changed with time. The comparison is preferably 'can be reduced by 1.4. Circle 2 unintentionally shows the target surface and the inside of the target." 1.0~: In the present invention, it exists at a flat central point with the dry surface. The average particle size of the crystal of the cut surface: the ratio of the average particle diameter of the crystal in the dry and the axis, which is obtained by the following method. First, the measurement of the peak surface is carried out by the above method. The average crystal of the light surface of the pile bauxite The particle size is obtained by determining the thousand-average particle diameter of the a-cut surface of the cutting surface which is at the center point of the target surface and passing through the center axis of the target, which is cut by the slicer. The same side as the surface of the indium target is measured ,, >丨$ + β method of dry day diameter method method / then average crystal grain size of the cut surface. "According to the diameter and presence of the surface of the leather obtained by the above method The remaining surface is parallel and passes through the center point of the Μ mandrel: the average particle size of the crystal, the average particle size of the crystal at the cut surface of the central surface of the dry table = 2: the average particle The ratio of the diameters is obtained by using (the knot existing on the dry surface; the average water content in the middle water) / (the average particle diameter of the crystal which is present in the cut surface parallel to the target surface and passing through the center point of the target central axis). The indium is dried in the preferred embodiment, and the pore size is 5 〇" or more and the pore size is 1 " below. It exists in the pores of the inside, especially the pores of the pores above 0.50 // m: f丨合士'The pill hole may become the cause of abnormal discharge during sputtering', so it is preferable to reduce it as much as possible. According to the present invention, since the pores existing in (4) are flattened at the time of drying, 10 201211290, The hole can be narrowed. The hole on the hole control 50... is less than eight m3. The hole is made of holes on ... Preferably, it is 0.5/cm3 or less, more preferably 3/down, for example, 0 to 0.3/cm3. In the present invention, the number of holes in the aperture 5〇 is detected by electronic scanning ultrasonic inspection. (4) placed in the flaw detector sink of the above device, in the frequency band ... (9) test, pulse repetition frequency z, scan speed 6 〇 mm / min to determine the image obtained by the data, counting aperture 50 & quot The aperture of m or more is determined by the volume of the measurement target dry volume. Here, the so-called aperture diameter is defined by the diameter of the smallest circle surrounding the aperture of the image. Next, the present invention will be described in order. A preferred embodiment of the method for producing indium dry. First, 'indium as a raw material is melted and flowed to a mold towel. If the raw material indium contains impurities, the conversion efficiency of the solar cell produced from the raw material is lowered, so that it is preferably of high purity, for example, a purity of 99 ❶/. The above indium. Then 'cooled to room temperature to form an indium ingot. Even if voids are formed at this stage or the crystal grain size is large, since it can be controlled by cold rolling thereafter, there is no problem, but it takes time to increase the calendering rate. In the stage of melt casting, it is preferred to reduce the pores to some extent in advance and to reduce the crystal grain size to some extent. The cooling rate can be a natural cooling by air (about /C / min), and when it is important to preliminarily suppress the voids in the ingot during the stage of the melt casting, it can be as slow as possible, for example, 9. (: / / or less, it is preferable to carry out the cold 201211290 at 8 〇 c / min or less. However, if it is too slow, the crystal coarsening suppression effect of ultrasonic vibration cannot be sufficiently obtained this time, so it is preferable to set it 3 〇 c / min or more, more preferably 5 〇 / min or more. On the other hand, when it is important to prevent the crystal grain size from growing, it is cooled as quickly as possible. In particular, the voids can be reduced by cold rolling. Therefore, obtaining fine crystal grains by high-speed cooling and obtaining a high sputtering rate are advantageous for improving the characteristics of the entire target. For example, it can be set to 2 〇 t / min or more, preferably 50 ° C / min. Too fast, since the calendering process required for flattening the voids becomes troublesome, it is preferable to perform cooling at a maximum of 7 〇 < 5 (: / min. For the adjustment of the cooling rate, when reducing In the case of the cooling rate, the mold may be heated and insulated by a heater or the like. Conversely, when the cooling rate is increased, the cooling rate may be obtained by a method such as cooling the water supplied to the periphery of the mold. Is (indium melting temperature -25 < t ) / ( After the initial cooling, the temperature of the indium is reduced from the melting temperature to the time of 25 C > c. Then, the obtained indium ingot is cold-rolled until the entire average crystal grain size is 10 mm or less, and shape processing is performed as needed. Or the surface is ground to form indium. When the total rolling reduction ratio at the time of rolling is increased, the crystal grain size is finer, the unevenness of crystal grain size is reduced, and the porosity is smaller, and the aspect ratio of the cross section is also smaller. In the production of the indium target of the present invention, the total rolling reduction ratio is preferably 50 to 80°/°', more preferably 60 to 70%. The thickness of the target is not particularly limited, and may be suitably used depending on the device or the intended use. 0 疋, usually about 3~20mm, typically about 5~l〇mm. The thickness of the indium ingot before cold rolling or the rolling reduction of cold rolling 12 201211290 can be adjusted according to the thickness of the desired target. The obtained indium target can be suitably used as a light-absorbing layer for a CIGS-based thin film solar cell. [Examples] The following Tables are not intended to illustrate the examples and comparative examples of the present invention. Understand the present invention and its advantages, and The intention of the present invention is limited. The indium precursor (pure 纟 5N ) melted at 160 ° C is flowed to the inside of the mold having an inner diameter of 25 〇 mm and a depth of 50, and the cold is described in Table i: To the temperature (25 C)', a disk-shaped indium ingot (diameter: 25 mm × thickness 30 Å) was formed. Then, _ ingot was cold-rolled under the conditions described in Table 1 to obtain invention examples and comparative examples. Indium ingots are obtained. The obtained indium 1^ is processed into a diameter of 2Q4 mmx. The characteristic values of A to F are measured by the method. The measurement of a to D is performed by grinding the surface of the surface and using acid (4). After making the grain boundary easy to observe, 'by performing visual inspection, „ , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , The results are shown in the table "Expression of super-wave A: average crystal grain size of each part B: maximum crystal grain size of the target surface Γ Table 1 · The center of the deposit is: the average particle size of the crystal in the central part of the table (Table 1 In the case of the average particle diameter (indicated by the presence of the surface of the dry surface, the outer surface of the dry surface is expressed as the surface periphery), the ratio D· exists in parallel with the dry surface and passes through the center point of the dry central axis. 13 201211290 The ratio of the average particle size of the crystals of the cut surface (indicated by "inside" in Table 1) to the average particle diameter of the crystals present on the dry surface (expressed as "the whole surface" in the surface) E: the pore diameter 5 0 Ratio of the number of holes above m: F: Aspect ratio of the section and 'Table 2 shows the vacuum pressure of the SPF-3 13H sputtering apparatus manufactured by ANELVA's chamber before the start of sputtering. lxl〇_4pa, splashing The pressure at the time of 0.5 Pa, the argon splash gas flow rate of 5 SCCM, and the reduction: the key power of 650 W, the results of the splash of the indium target of the inventive examples and the comparative examples. In Table 2, 'the initial sputtering rate is indicated. And after 3.3K: whr sputtering, the sputtering rate, the number of abnormal discharges during sputtering. The ore rate is calculated based on the film formation time and the film thickness measured by the surface topographer, and the number of abnormal discharges is measured by a visual method. It can be seen from Table 1 and Table 2: In Comparative Example 1, 'because no rolling is performed Therefore, the crystal grain size is large, and the sputtering rate is slow. Moreover, since the unevenness of each portion of the crystal grains is large, the sputtering rate changes with time is also serious. In Comparative Example 2, although calendering is performed, However, since the cooling rate is too large, the pores are not sufficiently suppressed. Therefore, the number of abnormal discharges is large. In Comparative Example 3, although the rolling was performed, the cooling rate was too small, so that the crystal grain size became large. In 3, it can be seen that as the cooling rate is increased, the crystal grain size becomes smaller. 'The unevenness of each portion is also small, and the high sputtering rate can be maintained. Moreover, it is known that although the number of voids increases slightly with the increase of the cooling rate. However, 14 201211290 can sufficiently suppress abnormal discharge. 15 201211290 (X| 0.27 0.31 0.23 0.26 0.28 0.30 ω 〇Ο CN 〇〇〇CO c5 ο Q 00 1.17 1.55 1.19 1.16 1.06 u Ο) 00 〇1.17 m 25.0 C N CN oo 00 CN A : average particle size per part (mm) 1 internal | 22.0 18.3 OS 00 00 iri 00 | entire surface | 12.0 <Ν 00 〇in r-; | surface center 16.0 rn 13.8 od 00 Surface periphery (N 00 ο — 'O total rolling reduction (%) o ν 〇 cooling rate rc / min) 10 (natural cooling) 100 (high speed cooling) 1 (low speed cooling) 3 (low speed cooling) 10 (natural cooling 70 (high-speed cooling) d comparative example 1 comparative example 2 comparative example 1 invention example 2 invention example 3 201211290 [Table 2]
No. ' -—---- —.— 〜〜_濺鑛速率（nm/min) 異常放電次數 比較例1 期 37η 經過3.3KWhr濺鍍後 比較例1 比較你 6j1 420 〇 28 發明例"1 ~~~ ^385 350 「0 發明作ΪΤ^ ^96 393 0 發明作? __400 395 〇 _402_ --2_ 【圖式簡單說明】 圖卜係顯示靶表面之中央部與靶表面之外圍部的示意 =2,係顯示靶表面與靶内部的示意圖 圖3:，定剖面縱橫比時平： X及垂直於壓延方向之方向y的示意圖。 方向 【主要元件符號說明】 10 銦靶 11 把表面中央部 12 乾表面外圍部 13 靶表面 14 靶内部 15 中心軸 16 切割面 17No. '----------~~_ Splash rate (nm/min) Abnormal discharge times comparison example 1 stage 37η After 3.3KWhr sputtering comparison example 1 Compare you 6j1 420 〇28 invention example" 1 ~~~ ^385 350 "0 invention ΪΤ ^ ^96 393 0 invention __400 395 〇 _402_ --2_ [Simple diagram of the diagram] The diagram shows the central part of the target surface and the outer surface of the target surface Schematic = 2, showing the schematic of the target surface and the inside of the target. Figure 3: Schematic diagram of the aspect ratio: X and the direction perpendicular to the direction of the rolling direction y. Direction [Main component symbol description] 10 Indium target 11 Center of surface Part 12 dry surface peripheral portion 13 target surface 14 target interior 15 central axis 16 cutting surface 17