201135954 六、發明說明： 【發明所屬之技術領域] 本發明係關於感測紫外線燈等所發出的紫外線的紫外 線感測器及其製造方法。 【先前技術】 照射ι外線的紫外線燈係在水或空氣的殺菌、半導體 或食品容器的洗淨、及半導體的表面改質等在產業界被多 方面使用。該紫外線燈爲消耗品’壽命爲其重要的關鍵， 由於無法連續監視紫外線燈，因此事前訂定出壽命時間來 作爲規格。但是，各個燈的壽命並非爲固定，會有與所設 定的時間產生偏差的情形。若實際壽命比設定時間爲短時 ’由於在不會發生紫外線的情況下直接使用，因此無法獲 得所希望的淨化、殺菌等效果，若實際壽命比設定時間爲 長時’會發生儘管尙可使用卻被更換的浪費。 對於如上所示之問題，以往係有藉由感測紫外線燈所 照射的紫外線，來連續性監視紫外線燈的感測器。以該感 測器而言，係有使用矽者。但是，由於測定波長的選擇係 使用濾光片來進行，因此在波長254nm附近的短波長範圍 中，受光頻譜的頻寬寬，而且發生因濾光片的紫外線所造 成的劣化及伴隨其之受光波長的移位，而必須進行供感度 調整之用的校正。此外，基於濾光片周邊零件等的耐熱性 問題，會有通常在40°C以下使用的限制，因此必須頻繁進 行測定場所的調整。 -5- 201135954 此外’以檢測波長28 0nm以下之紫外線的 ’已知有光電管。該光電管係已被實用化作爲 亮滅的感測器，主要被使用在工業爐等大型燃 動控制用的火焰感測器，但是使用光電管的感 非爲太陽盲光（solar-blind)的感測器，因此 易變動、耐熱性亦低的問題。 相對於此，係有一種使用氧化鎵單結晶基 性及耐久性佳的太陽盲光的紫外線感測器及紫 測器（參照例如專利文獻1及2 )。 〔先前技術文獻〕 〔專利文獻〕 〔專利文獻1〕日本特開2009-130012號公〗 〔專利文獻2〕日本特開2009-200222號公! 【發明內容】 (發明所欲解決之課題） 上述氧化鎵單結晶基板係將其單結晶的晶 與（100)面（以下稱爲a面）呈平行的面進行 該a面以化學機械硏磨法（CMP: Chemical Polishing)進行鏡面硏磨而加工成晶圓狀，藉 。但是，氧化鎵單結晶基板係當將該單結晶c 切片時，以前工程而言，進行將單結晶切成圓 亦即，將單結晶中的a面進行切片的工程係成 感測器而言 感測火焰之 燒裝置的自 測器由於並 具有感度容 板，且耐熱 外線用光檢 錠以線鋸將 切片，且將 Mechanical 此予以製作 3的a面進行 片的工程。 爲以線鋸將 -6- 201135954 該切成圓片的狀態的單結晶之中與a面呈平行的面進行切 片的工程。氧化鎵單結晶基板係藉由上述切成圓片及切片 工程予以製作’因此會有該製造工程耗費時間等成本的問 題。 本發明係爲解決上述問題點所硏創者，目的在提供使 用低成本的氧化鎵單結晶基板的紫外線感測器及其製造方 法。 (解決課題之手段） 爲了解決上述問題點’本發明之紫外線感測器係具有 :將相對氧化鎵單結晶的成長方向呈正交的正交面或由該 正交面呈預定角度傾斜的面作爲受光面的氧化鎵單結晶基 板；形成在前述氧化鎵單結晶基板的第1表面的第1電極； 及形成在包含前述受光面的前述氧化鎵單結晶基板的第2 表面的第2電極。 此外，本發明之紫外線感測器之製造方法係以相對氧 化鎵單結晶的成長方向呈正交的正交面或由該正交面呈預 定角度傾斜的面成爲切斷面的方式，將前述氧化鎵單結晶 以預定厚度進行切斷而將切斷面作爲受光面，在前述氧化 鎵單結晶基板的第1表面形成第1電極，在包含前述受光面 的前述氧化鎵單結晶基板的第2表面形成第2電極。 (發明之效果） 藉由本發明，可將低成本的氧化鎵單結晶基板使用在 201135954 紫外線感測器。 【實施方式】 以下針對本發明之實施形態，一面參照圖示， 以說明。 (實施形態1 ) 首先，使用第1圖及第2圖，針對實施形態1之 感測器加以說明。第1圖係顯示實施形態1之紫外線 之構成的剖視圖。第2圖係顯示封裝體內的感測器 上視圖。如第1圖所示，紫外線感測器1係用以檢測 的感測器晶片10被收容在封裝體2內而由外部雰圍 密封。 封裝體2係具有：平板狀的芯柱（stem ) 22 ; 在芯柱22之上，在中央具有開口部20之屬於圓筒構 蓋23 ;及由帽蓋23的內部覆蓋開口部20的窗構件21 部係以氮或氬等惰性氣體所塡充、或者被維持在真 〇 芯柱22及帽蓋23若由被收容在內部的感測器晶 靜電遮蔽的觀點來看，較佳爲導電性，例如由不銹 、黃銅、鐵、鎳等金屬所形成。 窗構件2 1係由對作爲檢測對象的紫外線具有透 材料所形成，可使用對波長254iim附近的紫外線分 較大透射率的藍寶石或石英（水晶）。該窗構件21 一面加 紫外線 感測器 晶片的 紫外線 氣予以 及設置 件的帽 ，其內 空狀態 片10的 鋼、鋁 光性的 別具有 係以由 -8 - 201135954 封裝體2內覆蓋開口部2〇的方式使用環氧接著劑24而被接 著在帽蓋23的內側。 感測器晶片10係具有氧化鎵（Ga203 )單結晶基板（ 以下稱爲單結晶基板）u，在該單結晶基板1 1的上面（第 1表面）形成有與該單結晶基板1 1作肯特基接觸的紫外線 檢測用的肖特基電極1 2。此外，與肖特基電極1 2相對應而 在單結晶基板1 1的下面（第2表面）形成有與該單結晶基 板U作歐姆接觸的歐姆電極1 3。 單結晶基板1 1係在其表面形成有絕緣層1 1 a，在絕緣 層1 la之下形成有導電層1 lb。單結晶基板1 1係在表面形成 有絕緣層1 1 a，藉此可進行因紫外線激發所發生的電子電 洞對的分離。肖特基電極1 2係以與該絕緣層1 1 a相接，歐 姆電極1 3係以與導電層1 1 b相接的方式所形成。此外，在 單結晶基板1 1上形成有配線用的焊墊電極1 4。 單結晶基板Π的第1表面及肯特基電極1 2的上面係接 收紫外線等之光的受光面12r，以該受光面12ι•與封裝體2的 窗構件2 1相對向的方式在芯柱22設置有感測器晶片1 〇 ^ 被使用在單結晶基板1 1的氧化鎵單結晶係在波長2 8 0 nm以下的波長領域具有選擇性感度，尤其在波長254nm附 近具有較大感度。氧化鎵單結晶係能帶間隙爲4.7〜4.9eV 之較寬大（wide)的氧化物半導體，因此成爲太陽盲光（ solar-blind )的感測器。此外，氧化鎵單結晶係熔點較高 ，爲1 74〇°C，已經爲氧化物，因此不會有因氧化所造成的 劣化之虞，耐熱性、安定性及耐久性佳。 -9 - 201135954 作爲紫外線感測器1的端子的引線3 1及32係由芯柱22 被拉出至封裝體2的外部，與電流/電壓源3分別作電性連 接。引線3 1的一端3 la係透過直徑25 μηι的Au接合導線25而 與焊墊電極1 4作電性連接。此外，引線3 1係透過玻璃26而 被固定在芯柱22。引線32係藉由與芯柱22相連接，而與感 測器晶片10的歐姆電極13作電性連接。其中，接合導線25 的直徑係設爲2 5 μπι，但是並非限定於此。 此外，如第2圖所示，單結晶基板1 1係被設置在芯柱 22的中央，在其中央形成有肖特基電極12。其中，肖特基 電極1 2係形成爲從單結晶基板1 1至焊墊電極1 4中的半圓部 分。在實施形態1中，單結晶基板1 1及肖特基電極1 4係形 成爲圓狀，但是亦可爲橢圓等形狀。 在實施形態1中，以受光面1 2ι•而言，使用相對氧化鎵 單結晶的成長方向呈正交的正交面（以下將該手法稱爲手 法A )。以下針對手法A，參照第3圖詳加說明。 第3圖係用以說明手法a的模式圖。第3圖所示之a、b 及c的箭號係表示氧化鎵單結晶的結晶方位。 如第3圖所示，正交面4 1係氧化鎵單結晶的晶錠4的成 長方向’亦即’相對c軸方向呈正交的面。藉由將該正交 面4 1作爲受光面丨2r加以使用，可僅在將晶錠4切成圓片的 工程中切出單結晶基板1 1。若將a面42作爲受光面12r時， 藉由進行將晶錠4切成圓片，再進行切片的二階段工程， 來切出氧化鎵單結晶基板。因此，與將a面42作爲受光面 12r的情形相比，藉由將正交面41作爲受光面i2r，可達成 -10 - 201135954 作業時間的縮短及收率的提升。此外，若將受光面1 2r作 鏡面硏磨時，由於a面42爲分開面，因此會有藉由分開而 使其表面以鱗狀剝落的情形，但是在正交面4 1並不會有以 鱗狀剝落的情形，可縮短作業時間。 在此，在實施形態1中，係將正交面4 1作爲受光面1 2r ，但是亦可將由正交面41呈預定角度傾斜的面作爲受光面 1 2 r。例如，若將屬於具有由成長方向傾斜1 4 °左右的分開 性的面的（001 )面（以下稱爲c面）作爲受光面12r•時， 藉由利用分開，可輕易地由晶錠4切出單結晶基板1 1。離 分開面的傾斜並不會大幅影響感測器性能。因此，若考慮 到由晶錠4切出單結晶基板1 1的作業性時，受光面i 2r係以 由正交面41在-20°〜20°的範圍內傾斜的面爲佳，朝正交面 41與c面所成角度方向’在-15°〜15°的範圍內傾斜的面爲 較好，尤其以正交面41及與c面呈平行的面爲佳。 以下針對使用手法A的紫外線感測器的製作方法，參 照第4圖，按每個工程加以說明。第4圖係用以說明實施形 態1之紫外線感測器之製作工程的模式圖。 (S 1 :氧化鎵單結晶的育成及單結晶基板的切出） 以純度4 N的氧化鎵粉末作爲原料而封入在橡皮管，藉 由靜水壓衝壓來進行壓縮成形。壓縮成形後，使用電氣爐 ，以丨4〇0〜16001、10〜40小時，在大氣雰圍氣進行燒成 ’將所得的氧化鎵燒結體作爲原料棒。使用該原料棒，藉 由F Z ( F1 〇 a t i n g Ζ ο n e )法來育成氧化鎵單結晶。以單結晶 -11 - 201135954 成長條件而言，以成長速度爲5〜30mm / h、雰圍氣爲乾 燥空氣、壓力爲1氣壓的條件來進行。 將所育成的氧化鎵單結晶的晶錠4黏貼在載玻片，以 與氧化鎵單結晶的成長方向呈垂直的正交面成爲切斷面的 方式進行切斷，而得單結晶基板Π。單結晶基板1 1的厚度 係以0.6〜0.7mm爲佳。 (S2 :氧化鎵單結晶基板的硏磨） 將所切出的單結晶基板1 1黏貼在硏磨夾具，在#1000 的硏磨盤硏削10〜30分鐘，藉此進行出面。接著，以Ιμηι 鑽石及CMP，分別硏磨20〜60分鐘、較佳爲20〜30分鐘， 藉此而得鏡面。 (S 3 :單結晶基板的退火） 將硏磨後的單結晶基板1 1以丙酮、乙醇洗淨，且進行 熱處理。該熱處理係基於在單結晶基板1 1形成因光激發所 發生的載子分離所需的絕緣層1 1 a，使殘留在結晶成長後 的氧化鎵單結晶內的缺陷復原的目的下進行》熱處理係在 氧雰圍氣中，以1 100°C進行3〜24小時。在比3小時爲更短 的時間，結晶性的恢復不充分，即使施加比24小時還要長 的處理時間，則會大致飽和而在特性上並沒有改變。之所 以使用氧，是爲了補充氧化鎵單結晶育成時所發生的氧缺 損之故。 在單結晶基板11退火後，藉由未圖示的以下S4〜S6工 -12- 201135954 程，在該單結晶基板1 1形成電極。 (S4 :歐姆電極的形成） 在將Ti蒸鍍30〜100nm、較佳爲蒸鍍50〜80nm後，將 A1蒸鍍100〜300nm、較佳爲蒸鍍150〜300nm，而形成A1 /Ti的歐姆電極13。電極尺寸係設爲1〜、較佳爲3 〜4mm φ。其中，尺寸愈大，接觸電阻愈小。蒸鍍係使用 例如電子束蒸鑛法（EB蒸鍍法）。 (S 5 :焊墊電極的形成） 在單結晶基板1 1上製作配線用的焊墊電極1 4。焊墊電 極 14 的尺寸爲 0_05 〜1·5ιηηιφ，將 Cr(13a)以 10 〜50nm、 較佳爲30〜50nm，將Au或Pt以30〜lOOnm、較佳爲50〜 lOOnm蒸鑛在單結晶基板11上。 (S6 :肯特基電極的形成） 以肖特基電極12的材料而言，由於爲n型半導體，因 此使用屬於被設爲工作函數較大的金屬的Au、Pt等。將Ni 蒸鑛2〜5nm、較佳爲2nm後，將Au或Pt蒸鍍6〜10nm，製 作Au/ Ni或Pt/ Ti之具有半透明或透明的透光性的金屬薄 膜的電極。其中，在金屬係除了未插入Ni層的Au、Pt、甚 至 Au、Pt以外，亦可使用 Al、Co、Ge、Sn、In、W、Mo、 Cr、Cu等。在蒸鍍係使用例如EB蒸鍍法。 之所以蒸鍍Ni，係因爲若爲Au或Pt單體，與基板的密 -13- 201135954 接性較差，因此插入較薄的N i層而改善密接性之故。此形 成爲受光面12r，此時的電極尺寸爲1〜5ιηιηφ、較佳爲1〜 2 mm φ。尺寸愈大，愈會造成受光面擴大。 (實施例1 ) 按照上述S 1〜6工程，進行紫外線感測器1的製作。 將純度4N的氧化鎵粉末封入在橡皮管，藉由靜水壓衝 壓來進行壓縮成形。壓縮成形後，使用電氣爐，以1 5 0 0 °C 、4〇小時，在大氣雰圍氣中進行燒成，使用所得的氧化鎵 燒結體，藉由FZ法來育成氧化鎵單結晶。以單結晶成長條 件而言，以成長速度爲2 0mm/ h、雰圍氣爲乾燥空氣、壓 力爲1氣壓的條件來進行。 將所得的氧化鎵單結晶的晶錠4黏貼在載玻片，以與 氧化鎵單結晶的成長方向呈垂直的正交面成爲切斷面的方 式進行切斷，而得單結晶基板1 1。單結晶基板1 1的厚度係 形成爲0.65mm。 將所切出的單結晶基板1 1黏貼在硏磨夾具，在#1 000 的硏磨盤硏削20分鐘，藉此進行出面《硏磨夾具係使用武 藏野電子股份有限公司製的MA-200D。接著，以Ιμπι鑽石 及CMP分別硏磨30分鐘而藉此獲得鏡面。 將所硏磨的單結晶基板1 1以丙酮、乙醇進行洗淨，在 氧雰圍氣中以1 1 0 0 °C、4小時進行熱處理。 接著，按照上述S4〜6工程，進行電極的製作。單結 晶基板1 1中的第2表面的歐姆電極13係在將Ti蒸鍍7 5nm後 -14- 201135954 ’將A1蒸鍍3〇〇nm。另一方面，單結晶基板11中的第i表面 的肖特基電極I2係在將Ni蒸鍍2nm後，將Au蒸鑛8nm。配 線用的焊墊電極1 4係形成爲1 m m φ，將C r蒸鍍5 Ο n m後，將 Au蒸鍍l〇〇nm。此時，實質的受光部係成爲由肖特基電極 部分去除焊墊電極後的部分。 爲了保護形成有肖特基電極1 2的單結晶基板1 1，預先 藉由環氧接著劑將窗構件21接著在帽蓋23，而且使用設有 引線3 1〜3 2的封裝體2 ’進行封裝。此時，在封裝體2內係 塡充有惰性氣體。 對所封裝的紫外線感測器1的引線3 1〜3 2，如第1圖所 示連接電流/電壓源3，可藉由測定電流値來進行紫外線 的檢測。 備妥複數個如上所示所製作的紫外線感測器1，對複 數個紫外線感測器1 ’使用岩崎電氣股份有限公司製的低 壓水銀燈QGL200U-2 1 ’連續照射波長254nm的紫外線，進 行感測器感度的測定。此外，藉由CV測定，測定出複數個 紫外線感測器1的靜電電容〔F〕。在C V測定係使用三和 電氣計器股份有限公司製的Μ 1 _ 4 9 0 K。測定結果顯示於第 5圖。 第5圖係顯示實施例1之紫外線感測器的感測器感度與 靜電電容的曲線圖。第5圖所示之斜線範圍係顯示在受光 面使用a面的紫外線感測器的特性範圍。 在受光面使用a面的紫外線感測器係可藉由使紫外線 感測器的製作條件改變，使感測器感度改變至χ丨〇·2〜 -15- 201135954 1.0x10 _4〔 A/W〕、使靜電電容改變至 ι〇χ1〇·ΐ2〜 1〇χ1〇-9 〔F〕。如第5圖所示’確認出紫外線感測器i係具有與在 受光面使用a面的紫外線感測器爲同等的感測器特性（感 測器感度、靜電電容等）。 此外’在受光面1 2ι•使用a面的紫外線感測器所使用的 單結晶基板係可由晶錠切出25〜30枚左右。另一方面，單 結晶基板11係可由晶鍵4切出50〜60枚左右。 (實施形態2 ) 在實施形態1中’係使用上述手法A來製作紫外線感測 器’但是除了手法A以外’可藉由將以下手法b〜E適當組 合’來製作具有成本減低及良好的感測器特性的紫外線感 測器》 手法B :調整在育成氧化鎵單結晶時所使用的原料粉 末中的不純物fi，來育成氧化鎵單結晶的手法》 手法C :調整單結晶基板的表面狀態的手法。 手法D:使用超音波焊料器（超音波焊鐵），在單結 晶基板上形成歐姆電極，而且進行該單結晶基板與芯柱的 晶粒接合的手法。 手法E :使用IBS法來形成肖特基電極的手法。 關於將上述手法A〜E全部加以組合的紫外線感測器的 構成，參照第6圖加以說明。 第6圖係顯示實施形態2之紫外線感測器之構成的剖視 圖。在第6圖中，與第1圖相同元件符號者係表示與第1圖 -16- 201135954 所示者爲相同者，且省略重複說明。 紫外線感測器100係具有感測器晶片110來取代第1圖 所示之感測器晶片1 0，具有封裝體1 02來取代封裝體2。感 測器晶片1 1 0係具有單結晶基板1 1 1來取代第1圖所示之單 結晶基板1 1。此外’感測器晶片1 1 0係具有肖特基電極u 2 來取代肖特基電極12，具有歐姆電極113來取代歐姆電極 13 ° 封裝體102係在藉由低熔點玻璃124而將窗構件21接著 在帽蓋23方面與封裝體2有所不同，單結晶基板Hi係在原 料粉末中的Si濃度及Fe濃度與硏磨損傷受到調整方面與單 結晶基板1 1有所不同。肖特基電極1 1 2係在使用IB S法而形 成在單結晶基板111上方面與肖特基電極12有所不同，歐 姆電極1 1 3係在使用超音波焊料器而藉由低熔點金屬而形 成在單結晶基板11上方面與歐姆電極13有所不同。 接著’使用上述紫外線感測器1 00之構成，詳加說明 手法B〜E。首先，說明手法B。 手法B係藉由調整原料粉末中的不純物量，來育成載 子密度較低的氧化鎵單結晶的手法。所調整的不純物列舉 有Si與Fe。 結晶中的載子密度與感測器感度係具有密切的關係， 以載子密度的範圍而言，係以1 x 1 〇17〜1 X 1 01 8爲佳，以 1.47χ1017 〜2·51χ1018 爲較佳，以 2〜5xl017cm·3 爲特佳。 若載子密度過高時，氧化鎵單結晶雖然導電性高，但是結 晶性會降低，在應用在感測器晶片1 1 〇時，不僅無法獲得 201135954 較高的感測器感度，暗電流亦會增加。另一方面，若載子 密度過低時，氧化鎵單結晶係結晶中的電阻會變高，而成 爲不易流通光電流者。201135954 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to an ultraviolet sensor for sensing ultraviolet rays emitted from an ultraviolet lamp or the like and a method of manufacturing the same. [Prior Art] Ultraviolet lamps that emit ultraviolet rays are used in many fields in the industry, such as sterilization of water or air, cleaning of semiconductors or food containers, and surface modification of semiconductors. This ultraviolet lamp is a key to the consumable life. Since the ultraviolet lamp cannot be continuously monitored, the life time is set in advance as a specification. However, the life of each lamp is not fixed and may deviate from the set time. If the actual life is shorter than the set time, 'Because it is used directly without ultraviolet rays, the desired purification and sterilization effects cannot be obtained. If the actual life is longer than the set time, it will occur, although it can be used. It was wasted being replaced. In the past, there has been a sensor for continuously monitoring an ultraviolet lamp by sensing ultraviolet rays irradiated by an ultraviolet lamp. In the case of this sensor, there is a use. However, since the selection of the measurement wavelength is performed using a filter, in the short wavelength range around the wavelength of 254 nm, the frequency spectrum of the light receiving spectrum is wide, and deterioration due to ultraviolet rays of the filter and light reception therewith occur. The shift of the wavelength, and the correction for the sensitivity adjustment must be performed. Further, there is a limitation that the heat resistance of the peripheral parts of the filter or the like is usually used at 40 ° C or lower. Therefore, it is necessary to frequently adjust the measurement site. -5- 201135954 Further, a photocell is known as 'detecting ultraviolet rays having a wavelength of 280 nm or less'. This photocell system has been put into practical use as a sensor for light-off, and is mainly used in a flame sensor for large-scale combustion control such as an industrial furnace, but the feeling of using a photocell is not a sense of solar-blind. The detector is therefore subject to problems such as easy change and low heat resistance. On the other hand, there is an ultraviolet sensor and a violet detector which use solar blind light having a single crystal of alumina and excellent durability (see, for example, Patent Documents 1 and 2). [PRIOR ART DOCUMENT] [Patent Document 1] [Patent Document 1] JP-A-2009-130012 [Patent Document 2] JP-A-2009-200222 [Abstract] (Problems to be Solved by the Invention) The gallium oxide single crystal substrate is obtained by mirror-honoring the a-plane by a chemical mechanical honing method (CMP: Chemical Polishing) by forming a crystal of a single crystal on a plane parallel to a (100) plane (hereinafter referred to as a plane). Wafer-like, borrowed. However, in the gallium oxide single crystal substrate, when the single crystal c is sliced, in the prior art, the single crystal is cut into a circle, that is, the engineering system is formed by slicing the a surface of the single crystal. The self-tester that senses the flame-burning device has a sensible-capacity plate, and the heat-resistant outer wire is sliced by a wire saw with a wire saw, and the a-side of the mechanically-made 3 is subjected to sheet work. It is a work of cutting a surface parallel to the a-plane among the single crystals which are cut into the wafer by the wire saw -6-201135954. The gallium oxide single crystal substrate is produced by the above-described dicing and slicing process. Therefore, there is a problem that the manufacturing process takes time and the like. The present invention has been made in an effort to solve the above problems, and an object thereof is to provide an ultraviolet sensor using a low-cost gallium oxide single crystal substrate and a method of manufacturing the same. (Means for Solving the Problem) In order to solve the above problem, the ultraviolet sensor of the present invention has an orthogonal surface that is orthogonal to the growth direction of the single crystal of gallium oxide or a surface that is inclined at a predetermined angle from the orthogonal surface. a gallium oxide single crystal substrate as a light receiving surface; a first electrode formed on the first surface of the gallium oxide single crystal substrate; and a second electrode formed on the second surface of the gallium oxide single crystal substrate including the light receiving surface. Further, the method for producing the ultraviolet sensor of the present invention is such that the surface orthogonal to the growth direction of the single crystal of gallium oxide or the surface inclined at a predetermined angle from the orthogonal surface serves as a cut surface. The gallium oxide single crystal is cut at a predetermined thickness, and the cut surface is used as a light receiving surface, and a first electrode is formed on the first surface of the gallium oxide single crystal substrate, and the second surface of the gallium oxide single crystal substrate including the light receiving surface is formed. A second electrode is formed on the surface. (Effect of the Invention) According to the present invention, a low-cost gallium oxide single crystal substrate can be used in the 201135954 ultraviolet sensor. [Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings. (Embodiment 1) First, a sensor according to Embodiment 1 will be described using Figs. 1 and 2 . Fig. 1 is a cross-sectional view showing the configuration of ultraviolet rays in the first embodiment. Figure 2 shows the top view of the sensor inside the package. As shown in Fig. 1, the sensor wafer 10 for detecting by the ultraviolet sensor 1 is housed in the package 2 and sealed by an external atmosphere. The package 2 has a flat-shaped stem 22; on the stem 22, a cylindrical cover 23 having an opening 20 at the center; and a window covering the opening 20 by the inside of the cap 23 The member 21 is filled with an inert gas such as nitrogen or argon, or is maintained in the true core column 22 and the cap 23. It is preferably electrically conductive from the viewpoint of electrostatic shielding of the sensor crystal contained therein. Properties, for example, are formed of metals such as stainless steel, brass, iron, and nickel. The window member 21 is formed of a transparent material for ultraviolet rays to be detected, and sapphire or quartz (crystal) having a large transmittance for ultraviolet rays in the vicinity of the wavelength 254iim can be used. The window member 21 is provided with a UV gas of the ultraviolet sensor wafer and a cap of the mounting member, and the steel and aluminum of the inner state sheet 10 are covered with an opening portion of the package body -8 - 201135954 The 2-inch method is followed by the epoxy adhesive 24 to be carried on the inside of the cap 23. The sensor wafer 10 has a gallium oxide (Ga203) single crystal substrate (hereinafter referred to as a single crystal substrate) u, and a single crystal substrate 1 is formed on the upper surface (first surface) of the single crystal substrate 1 as Kent. The Schottky electrode 1 2 for ultraviolet detection of the base contact. Further, an ohmic electrode 13 which is in ohmic contact with the single crystal substrate U is formed on the lower surface (second surface) of the single crystal substrate 11 in correspondence with the Schottky electrode 1 2 . The single crystal substrate 11 has an insulating layer 11 a formed on its surface, and a conductive layer 1 lb is formed under the insulating layer 1 la. The single crystal substrate 11 is formed with an insulating layer 11 a on the surface, whereby separation of electron hole pairs by ultraviolet excitation can be performed. The Schottky electrode 12 is connected to the insulating layer 11a, and the ohmic electrode 13 is formed in contact with the conductive layer 11b. Further, a pad electrode 14 for wiring is formed on the single crystal substrate 11. The first surface of the single crystal substrate 及 and the upper surface of the Kent base electrode 12 are light receiving surfaces 12r that receive light such as ultraviolet rays, and the light receiving surface 12 ι is opposed to the window member 21 of the package 2 in the stem 22 The sensor wafer 1 is provided. The gallium oxide single crystal used in the single crystal substrate 1 has a selective sensitivity in a wavelength region of a wavelength of 280 nm or less, and particularly has a large sensitivity in the vicinity of a wavelength of 254 nm. The gallium oxide single crystal system has a wide oxide semiconductor with a gap of 4.7 to 4.9 eV, and thus becomes a solar-blind sensor. Further, the gallium oxide single crystal has a high melting point of 1,74 〇 ° C and is already an oxide, so that there is no deterioration due to oxidation, and heat resistance, stability, and durability are good. -9 - 201135954 Leads 3 1 and 32 which are terminals of the ultraviolet sensor 1 are pulled out from the stem 22 to the outside of the package 2, and are electrically connected to the current/voltage source 3, respectively. One end 3 la of the lead 3 1 is electrically connected to the pad electrode 14 through a Au bonding wire 25 having a diameter of 25 μm. Further, the lead wire 3 is fixed to the stem 22 through the glass 26. The lead 32 is electrically connected to the ohmic electrode 13 of the sensor wafer 10 by being connected to the stem 22. Here, the diameter of the bonding wire 25 is set to 2 5 μm, but is not limited thereto. Further, as shown in Fig. 2, the single crystal substrate 11 is provided in the center of the stem 22, and the Schottky electrode 12 is formed in the center thereof. Among them, the Schottky electrode 12 is formed as a semicircular portion from the single crystal substrate 11 to the pad electrode 14. In the first embodiment, the single crystal substrate 11 and the Schottky electrode 14 are formed in a circular shape, but may be in the shape of an ellipse or the like. In the first embodiment, the light-receiving surface 1 2 ι• is an orthogonal surface which is orthogonal to the growth direction of the single crystal of gallium oxide (hereinafter referred to as the method A). The following is a detailed description of the method A, which is described in detail with reference to FIG. Figure 3 is a schematic diagram for explaining the method a. The arrows of a, b, and c shown in Fig. 3 indicate the crystal orientation of the single crystal of gallium oxide. As shown in Fig. 3, the orthogonal surface 41 is a plane in which the growth direction of the ingot 4 of the gallium oxide single crystal is orthogonal to the c-axis direction. By using the orthogonal surface 4 1 as the light receiving surface 丨 2r, the single crystal substrate 11 can be cut only in the process of cutting the ingot 4 into a wafer. When the a-plane 42 is used as the light-receiving surface 12r, a gallium oxide single-crystal substrate is cut out by performing a two-stage process of cutting the ingot 4 into a wafer and then slicing. Therefore, compared with the case where the a-plane 42 is used as the light-receiving surface 12r, by using the orthogonal surface 41 as the light-receiving surface i2r, it is possible to achieve a shortening of the working time of -10 - 201135954 and an improvement in the yield. Further, when the light-receiving surface 1 2r is mirror-honed, since the a-surface 42 is a separate surface, there is a case where the surface thereof is peeled off by a scale, but the orthogonal surface 4 1 does not have In the case of flaking, the working time can be shortened. Here, in the first embodiment, the orthogonal surface 41 is defined as the light receiving surface 1 2r. However, the surface inclined by the orthogonal surface 41 at a predetermined angle may be used as the light receiving surface 1 2 r. For example, when the (001) plane (hereinafter referred to as the c-plane) having a surface having a separation of about 1 4° in the growth direction is used as the light-receiving surface 12r•, it is easy to use the ingot 4 by using the separation. The single crystal substrate 11 was cut out. Tilting from the separation surface does not significantly affect sensor performance. Therefore, in consideration of the workability of cutting out the single crystal substrate 11 by the ingot 4, the light receiving surface i 2r is preferably a surface inclined by the orthogonal surface 41 in the range of -20° to 20°. It is preferable that the surface of the intersection surface 41 and the c-plane is inclined in the range of -15° to 15° in the angular direction ′, and it is particularly preferable that the orthogonal surface 41 and the surface parallel to the c-plane are used. Hereinafter, a method of manufacturing the ultraviolet sensor using the method A will be described with reference to Fig. 4 for each project. Fig. 4 is a schematic view for explaining the manufacturing process of the ultraviolet sensor of the embodiment 1. (S1: development of single crystal of gallium oxide and cutting of single crystal substrate) Gallium oxide powder having a purity of 4 N was used as a raw material, and sealed in a rubber tube, and compression-molding was carried out by hydrostatic pressing. After the compression molding, the electric furnace was used to calcine in an atmospheric atmosphere at 〇4〇0 to 1601, 10 to 40 hours. The obtained gallium oxide sintered body was used as a raw material rod. Using the raw material rod, a gallium oxide single crystal was grown by the F Z (F1 〇 a t i n g Ζ ο n e ) method. The single crystal -11 - 201135954 growth condition is carried out under the conditions of a growth rate of 5 to 30 mm / h, an atmosphere of dry air, and a pressure of 1 atmosphere. The ingot 4 of the gallium oxide single crystal thus grown is adhered to a glass slide, and is cut so that the orthogonal surface perpendicular to the growth direction of the single crystal of gallium oxide is cut into a cut surface, thereby obtaining a single crystal substrate. The thickness of the single crystal substrate 11 is preferably 0.6 to 0.7 mm. (S2: Honing of Gallium Oxide Single Crystal Substrate) The cut single crystal substrate 1 1 was adhered to a honing jig, and the honing disc of #1000 was boring for 10 to 30 minutes to carry out the surface. Next, the Ιμηι diamond and CMP are respectively honed for 20 to 60 minutes, preferably 20 to 30 minutes, thereby obtaining a mirror surface. (S3: Annealing of Single Crystal Substrate) The honed single crystal substrate 1 1 was washed with acetone and ethanol, and heat-treated. This heat treatment is based on the fact that the insulating layer 11 a required for the carrier separation by photoexcitation is formed on the single crystal substrate 1 1 and the defect remaining in the gallium oxide single crystal after crystal growth is restored. It was carried out in an oxygen atmosphere at 1 to 100 ° C for 3 to 24 hours. In a shorter period of time than 3 hours, the recovery of crystallinity is insufficient, and even if a treatment time longer than 24 hours is applied, it is substantially saturated and does not change in characteristics. Oxygen is used to supplement the oxygen deficiency that occurs when gallium oxide single crystals are grown. After the single crystal substrate 11 is annealed, an electrode is formed on the single crystal substrate 1 1 by the following S4 to S6 -12 to 201135954 (not shown). (S4: Formation of ohmic electrode) After depositing Ti to 30 to 100 nm, preferably 50 to 80 nm, A1 is deposited by vapor deposition of 100 to 300 nm, preferably 150 to 300 nm, to form A1/Ti. Ohmic electrode 13. The electrode size is set to 1 to, preferably 3 to 4 mm φ. Among them, the larger the size, the smaller the contact resistance. For the vapor deposition, for example, an electron beam evaporation method (EB vapor deposition method) is used. (S5: Formation of Pad Electrode) A pad electrode 14 for wiring was formed on the single crystal substrate 1 1. The pad electrode 14 has a size of 0_05 〜1·5 ηηηιφ, and Cr (13a) is 10 to 50 nm, preferably 30 to 50 nm, and the Au or Pt is distilled at 30 to 100 nm, preferably 50 to 100 nm in a single crystal. On the substrate 11. (S6: Formation of Kent base electrode) Since the material of the Schottky electrode 12 is an n-type semiconductor, Au, Pt, or the like belonging to a metal having a large work function is used. After Ni is distilled from 2 to 5 nm, preferably 2 nm, Au or Pt is vapor-deposited to 6 to 10 nm to prepare an electrode of a semi-transparent or transparent translucent metal film of Au/Ni or Pt/Ti. Among them, in the metal system, in addition to Au, Pt, or even Au or Pt in which the Ni layer is not inserted, Al, Co, Ge, Sn, In, W, Mo, Cr, Cu, or the like may be used. For the vapor deposition system, for example, an EB vapor deposition method is used. The reason why Ni is deposited is because if it is an Au or a Pt monomer, the adhesion to the substrate is inferior to that of the substrate, so that a thin Ni layer is inserted to improve the adhesion. This shape becomes the light receiving surface 12r, and the electrode size at this time is 1 to 5 ιηιηφ, preferably 1 to 2 mm φ. The larger the size, the more the light-receiving surface will expand. (Example 1) The production of the ultraviolet sensor 1 was carried out in accordance with the above S 1 to 6 works. Gallium oxide powder having a purity of 4 N was sealed in a rubber tube, and compression molding was carried out by hydrostatic pressing. After compression molding, it was fired in an atmosphere using an electric furnace at 1,500 ° C for 4 hours, and a gallium oxide single crystal was grown by the FZ method using the obtained gallium oxide sintered body. In the single crystal growth condition, the growth rate was 20 mm/h, the atmosphere was dry air, and the pressure was 1 atmosphere. The ingot 4 of the obtained single crystal of gallium oxide is adhered to a glass slide, and is cut in such a manner that the orthogonal surface perpendicular to the growth direction of the single crystal of gallium oxide is cut into a cut surface, and a single crystal substrate 11 is obtained. The thickness of the single crystal substrate 11 was formed to be 0.65 mm. The cut single crystal substrate 1 1 was adhered to a honing jig, and the honing disc of #1 000 was boring for 20 minutes, and the honing jig was used for the MA-200D manufactured by Musashino Electronics Co., Ltd. Next, the mirror was obtained by honing for 30 minutes with Ιμπι diamond and CMP, respectively. The honed single crystal substrate 1 1 was washed with acetone and ethanol, and heat-treated at 1,100 ° C for 4 hours in an oxygen atmosphere. Next, the electrode was produced in accordance with the above S4 to 6 works. The ohmic electrode 13 on the second surface of the single crystal substrate 11 is vapor-deposited by 3 〇〇 nm from -14 to 201135954 ' after vapor deposition of Ti at 75 nm. On the other hand, the Schottky electrode I2 on the i-th surface of the single crystal substrate 11 was vapor-deposited by 8 nm after Ni was vapor-deposited for 2 nm. The pad electrode 14 for wiring is formed to be 1 m m φ, and after C r is deposited by 5 Ο n m , Au is vapor-deposited by 10 nm. At this time, the substantial light receiving portion is a portion where the pad electrode is removed by the Schottky electrode portion. In order to protect the single crystal substrate 1 1 on which the Schottky electrode 12 is formed, the window member 21 is previously attached to the cap 23 by an epoxy adhesive, and the package 2' provided with the leads 3 1 to 3 2 is used. Package. At this time, the inside of the package 2 is filled with an inert gas. The lead wires 3 1 to 3 2 of the packaged ultraviolet sensor 1 are connected to the current/voltage source 3 as shown in Fig. 1, and the ultraviolet rays can be detected by measuring the current 値. A plurality of ultraviolet sensors 1 prepared as described above are prepared, and a plurality of ultraviolet sensors QGL200U-2 1 ' made by Iwasaki Electric Co., Ltd. are used to continuously irradiate ultraviolet rays having a wavelength of 254 nm for sensing. Determination of the sensitivity of the device. Further, the electrostatic capacitance [F] of the plurality of ultraviolet sensors 1 was measured by CV measurement. In the C V measurement system, Μ 1 _ 4 9 0 K manufactured by Sanhe Electric Meter Co., Ltd. was used. The measurement results are shown in Fig. 5. Fig. 5 is a graph showing the sensor sensitivity and electrostatic capacitance of the ultraviolet sensor of Example 1. The oblique line range shown in Fig. 5 shows the characteristic range of the ultraviolet sensor using the a-plane on the light-receiving surface. The ultraviolet sensor using the a-plane on the light-receiving surface can change the sensitivity of the sensor to χ丨〇·2~ -15- 201135954 1.0x10 _4[ A/W] by changing the manufacturing conditions of the ultraviolet sensor. Change the electrostatic capacitance to ι〇χ1〇·ΐ2~1〇χ1〇-9 [F]. As shown in Fig. 5, it was confirmed that the ultraviolet sensor i has the same sensor characteristics (sensor sensitivity, electrostatic capacitance, etc.) as the ultraviolet sensor using the a surface on the light receiving surface. Further, the single crystal substrate used in the ultraviolet sensor using the a surface on the light receiving surface 1 2 ι can be cut out from the ingot by about 25 to 30 pieces. On the other hand, the single crystal substrate 11 can be cut out by the crystal key 4 by about 50 to 60 pieces. (Embodiment 2) In the first embodiment, the ultraviolet sensor is manufactured by using the above-described method A. However, in addition to the method A, the following methods b to E can be appropriately combined to produce a cost reduction and a good feeling. Ultraviolet sensor for measuring the characteristics of the detector. Method B: Adjusting the impurity in the raw material powder used in the single crystal of gallium oxide to develop a single crystal of gallium oxide. Method C: Adjusting the surface state of a single crystal substrate technique. Manipulation D: An ohmic electrode is formed on a single crystal substrate using an ultrasonic solder (ultrasonic soldering iron), and a method of bonding the single crystal substrate to the crystal grain of the stem is performed. Manipulation E: The method of forming a Schottky electrode using the IBS method. The configuration of the ultraviolet sensor in which all of the above-described methods A to E are combined will be described with reference to Fig. 6. Fig. 6 is a cross-sectional view showing the configuration of the ultraviolet sensor of the second embodiment. In the sixth embodiment, the same components as those in the first embodiment are denoted by the same reference numerals as those in the first embodiment of Fig. 16-201135954, and the overlapping description will be omitted. The ultraviolet sensor 100 has a sensor wafer 110 in place of the sensor wafer 10 shown in Fig. 1, and has a package 102 instead of the package 2. The sensor wafer 110 has a single crystal substrate 11 1 instead of the single crystal substrate 11 shown in Fig. 1. In addition, the sensor wafer 110 has a Schottky electrode u 2 instead of the Schottky electrode 12, and has an ohmic electrode 113 instead of the ohmic electrode 13 °. The package 102 is attached to the window member by the low-melting glass 124. 21 is different from the package 2 in terms of the cap 23, and the single crystal substrate Hi differs from the single crystal substrate 1 in the Si concentration and the Fe concentration in the raw material powder and the adjustment of the honing damage. The Schottky electrode 1 1 2 is different from the Schottky electrode 12 in that it is formed on the single crystal substrate 111 by the IB S method, and the ohmic electrode 1 13 is made of a low melting point metal using an ultrasonic soldering iron. The ohmic electrode 13 is different in that it is formed on the single crystal substrate 11. Next, using the configuration of the above ultraviolet sensor 100, the descriptions B to E will be described in detail. First, explain the technique B. Manipulation B is a method of cultivating a single crystal of gallium oxide having a low carrier density by adjusting the amount of impurities in the raw material powder. The adjusted impurities are listed as Si and Fe. The density of the carrier in the crystal has a close relationship with the sensitivity of the sensor. In terms of the range of the carrier density, it is preferably 1 x 1 〇 17~1 X 1 01 8 and is 1.47 χ 1017 〜 2·51 χ 1018. Preferably, it is particularly preferably 2 to 5 x 10 17 cm·3. If the carrier density is too high, the gallium oxide single crystal has high conductivity, but the crystallinity is lowered. When applied to the sensor wafer 1 〇, it is not possible to obtain a higher sensitivity of the sensor of 201135954, and the dark current is also Will increase. On the other hand, when the carrier density is too low, the electric resistance in the gallium oxide single crystal system becomes high, and it becomes a problem that light current does not easily flow.
Si係與所育成的氧化鎵單結晶的載子密度有密切關係 者，原料粉末中的Si濃度係以1〜50ppm爲佳，以5〜30pprn 爲較佳，其中以6〜29ppm爲特佳。載子密度係呈與Si濃度 成正比增加的傾向，若S i濃度過高時，氧化鎵單結晶的結 晶性會降低，並不適於應用在感測器晶片1 1 0。另一方面 ，若S i濃度過低時，氧化鎵單結晶中的電阻會變高，在應 用在感測器晶片1 1 〇時所流通的光電流會變小。The Si system is closely related to the carrier density of the single crystal of the grown gallium oxide. The Si concentration in the raw material powder is preferably 1 to 50 ppm, more preferably 5 to 30 pprn, and particularly preferably 6 to 29 ppm. The carrier density tends to increase in proportion to the Si concentration. When the Si concentration is too high, the crystallinity of the gallium oxide single crystal is lowered, which is not suitable for use in the sensor wafer 110. On the other hand, if the concentration of Si is too low, the electric resistance in the single crystal of gallium oxide becomes high, and the photocurrent which flows when applied to the sensor wafer 1 1 变 becomes small.
Fe係對氧化鎵單結晶的導電性造成影響者，原料粉末 中的Fe濃度係以3ppm以下爲佳，其中以lppm以下爲特佳 。此係因爲當由使用含有大量Fe的原料粉末的氧化鎵單結 晶製作感測器晶片1 1 〇時，因在氧化鎵單結晶中被取入Fe ，導電性會顯著降低，而不易反應紫外線之故。 接著，針對手法C，參照第7圖加以說明。 第7圖係顯示背特基電極之表面狀態的圖。在第7圖中 ，與第6圖相同元件符號者係顯示與第6圖所示者爲相同者 ，且省略重複說明。 手法c係當將單結晶基板1 1 1進行鏡面硏磨時，在單結 晶基板1 π的第1表面設置如第7圖所示之預定的硏磨損傷 ，藉此在肖特基電極112形成時產生凹凸而將肯特基電極 112形成爲網目狀的手法。該手法C係達成使因肯特基電極 1 1 2與單結晶基板1 1 1的熱膨脹率的不同所產生的應力緩和 -18- 201135954 ，抑制因肖特基電極112的斷線等而起的感測器感度降低 ，且使感測器晶片1 1 0的耐久性提升的效果。 硏磨損傷係以長度ο·1〜〇.5mm、寬幅0.02〜〇.丨111111的 尺寸爲佳。此外，硏磨損傷係在1 m m2的範圍內以1 0〜1 0 0 個爲佳，以30〜60個爲特佳。若硏磨損傷超過該尺寸時、 或硏磨損傷數超過該範圍時’會有在肖特基電極113產生 斷線的可能性。另一方面，若硏磨損傷未達該尺寸時、或 硏磨損傷數未達該範圍時，上述應力的緩和會顯著降低。 形成在單結晶基板Π 1之第1表面的肖特基電極1 1 2的 作用在收集藉由紫外線激發所發生的載子，作爲光電流而 取出。若焊墊電極1 4及接合導線2 5藉由肖特基電極1 1 2而 與在肖特基電極Π 2的正下方發生載子的部位作電性連接 ’載子係可作爲光電流而取出至外部。因此，即使在單結 晶基板111的第1表面有因硏磨損傷所造成的凹凸，且隨著 該凹凸而肖特基電極112形成爲網目狀的情形下，亦只要 未發生斷線，作爲肖特基電極112的功能並不會大幅受損 。其中’硏磨損傷並不是僅限定於單結晶基板n丨的第1表 面’亦可同樣地設在單結晶基板n丨的第2表面等單結晶基 板1 1 1的表面。 接著，說明手法D。 手法系使用超音波焊料器而藉由低熔點金屬而在單 結晶基板1 1 1上形成歐姆電極u 3的手法。 糸外線感測器係受到紫外線等光，藉由將所發生的光 電流作爲訊號而取出來進行動作。爲了將該光電流有效率 -19- 201135954 地取出至外部，必須使歐姆電極低電阻化。藉由蒸鑛所形 成的歐姆電極係歐姆電極間的電阻爲1〜10〔kQ〕左右。 藉由在將該電極蒸鍍後的氧化鎵單結晶基板施行燒結處理 ，可改善歐姆電極間的電阻。 但是，藉由手法D，在歐姆電極113形成時使用超音波 焊料器與低熔點金屬，藉此可形成具有與進行上述燒結處 理者同等的電阻的歐姆電極113。此外，無須進行需要如 蒸鍍般的真空製程的方法，即可在短時間內在單結晶基板 111形成歐姆電極113，因此達成成本減低的效果。例如， 在低熔點金屬係使用超音波焊料、In、Sn等。 此外，形成歐姆電極1 1 3，並且使用低熔點金屬來進 行單結晶基板11 1與芯柱22的晶粒接合。由於在晶粒接合 材料未含有有機物，因此可藉由紫外線照射來抑制因有機 物的分解或分解物所造成的紫外線吸收，而達成可進行更 爲短波長的紫外線的檢測的效果。 此外，藉由低熔點玻璃1 2 4而使窗構件2 1與帽蓋2 3相 接著’與使用低熔點金屬的晶粒接合同樣地，達成抑制因 有機物所造成的紫外線吸收的效果。 接著，說明手法E。 手法E係使用IB S法而將A u或P t蒸鍍在單結晶基板11 1 上’藉此形成肖特基電極112的手法。 蒸鍍在氧化鎵單結晶基板上的金屬薄膜的密度愈高， 對紫外線的耐久性愈高，但是若爲Au或Pt單體，與氧化鎵 單結晶基板的密接性較差。在該手法中，由於以更強的力 -20- 201135954 將Au或Pt供給至基板，因此使用與平常的濺鏟法相比爲2 位數以上的高真空進行成膜的IBS法。藉由進行利用IBS法 所爲之高真空下的濺鍍，此時被供予至粒子的運動能量幾 乎不會受損而可堆積在預定的基板，因此獲得密接性及密 度高的薄膜。因此，使用IBS法而在單結晶基板1 1 1上作爲 金屬薄膜而成膜出緻密的肖特基電極1 1 2，藉此可比使用 例如EB蒸鍍法所成膜的電極獲得密度及密接性較高的電極 ，亦可改善耐久性。 在實施形態2中，係將B〜E與上述手法a適當組合而 製作紫外線感測器，但是即使爲將手法B與C、手法B與D 、手法C與E等2個手法加上手法A的紫外線感測器，亦可 達成成本減低、感測器感度提升等效果。其中，將手法B 〜D、手法C〜E、手法B、C及E等3個手法加上手法a的紫 外線感測器，相較於將2個手法加上手法A的紫外線感測器 ’在成本減低、感測器感度提升方面爲較佳，以在手法A 全部加上手法B〜E的紫外線感測器爲特佳。 接著，針對紫外線感測器1 0 0的製作方法，參照第8圖 ，按照每個製作工程來作說明。 第8圖係用以說明實施形態2之紫外線感測器之製作工 程的模式圖。 (S 1 1 :氧化鎵單結晶的育成及結晶基板的切出） 以Si濃度30ppm以下、Fe濃度lppm以下的純度4N的氧 化鎵粉末作爲原料而封入在橡皮管，藉由靜水壓衝壓來進 • 21 - 201135954 行壓縮成形。壓縮成形後，使用電氣爐，以1400〜1 600°C 、10〜40小時，在大氣雰圍氣中進行燒成，將所得的氧化 鎵燒結體作爲原料棒。使用該原料棒，藉由FZ ( Floating Zone )法來育成氧化鎵單結晶。以單結晶成長條件而言， 以成長速度爲5〜30mm/h、雰圍氣爲乾燥空氣、壓力爲1 氣壓的條件來進行。 將所育成的氧化鎵單結晶的晶錠1 04黏貼在載玻片， 以與氧化鎵單結晶的成長方向呈垂直的正交面成爲切斷面 的方式進行切斷，將其作爲單結晶基板1 1 1。單結晶基板 1 1 1的厚度係以0.6〜0.7mm爲佳。 (S12.氧化嫁單結晶基板的硏磨） 將所切出的單結晶基板1 1 1黏貼在硏磨夾具，以# 1 0 0 0 的硏磨盤硏削1 0〜3 0分鐘而藉此進行出面。接著，以1 μηΐ 鑽石及CMP分別硏磨20〜60分鐘、較佳爲20〜30分鐘，藉 此獲得鏡面。此時，以長度0.1〜0.5 mm、寬幅0.02〜0.1mm 的尺寸設置硏磨損傷，在1mm2的範圍內設置30〜60個。 (S 1 3 :單結晶的退火） 將經硏磨的單結晶基板1 1 1以丙酮、乙醇洗淨，且進 行熱處理。該熱處理係與實施形態1同樣地，基於在單結 晶基板111形成藉由光激發所發生的載子分離所需的絕緣 層111 a，使殘留在結晶成長後的氧化鎵單結晶內的缺陷復 原的目的下進行。熱處理係在氧雰圍氣中以1100 °C、3〜 -22- 201135954 24小時進行。此外，若爲使用Si濃度較低的原料粉末等所 備妥的載子密度較低的單結晶基板時，並不需要供載子分 離之用的絕緣層，因此可省略該工程。 (S14:歐姆電極的形成） 以加熱至200〜2 50°C的超音波焊料器將低熔點金屬5 附在單結晶基板1 1 1的第2表面。之後，較佳爲以1〜5秒左 右超音波供予振動，藉此弄破低熔點金屬5的表面氧化膜 ，低熔點金屬5在單結晶基板1 1 1內擴散，藉此將歐姆電極 1 1 3形成在單結晶基板1 1 1的第2表面全體或一部分。其中 ，尺寸愈大，接觸電阻愈小。 (S15:對芯柱的安裝） 在加熱至300〜350度的熱板之上置放金屬製區塊6a， 在區塊6a上置放設有引線31〜32的芯柱22且將其加熱。將 形成有歐姆電極1 1 3的單結晶基板1 Π ’以歐姆電極1 1 3與 芯柱22相接的方式設置在芯柱22上。由單結晶基板1 1 1之 上以銷組合、載玻片等均一加壓，將單結晶基板1 1 1中的 第2表面的低熔點金屬5同樣擴展，來進行單結晶基板1 1 1 與芯柱22的晶粒接合。將芯柱22由區塊6a取下來進行冷卻 ，藉此完成單結晶基板1 1 1對芯柱22的安裝。 (S 1 6 :焊墊電極的形成） 製作打線接合用的焊墊電極1 4。以焊墊電極用遮罩6b -23- 201135954 將單結晶基板1 1 1上方覆蓋，藉由E B蒸鍍法形成焊 14。焊墊電極14的尺寸爲0.05〜，將Crl4i 〜100nm、較佳爲30〜50nm’將Au或Ptl4b設爲80 -、較佳爲9 0〜1 0 0 n m，而蒸鍍在單結晶基板1 1 1的| (S 1 7 :肖特基電極的形成） 以肖特基電極1 1 2的材料而言，與實施形態1同 使用屬於被設爲工作函數較大的金屬的Au、Pt等。 密接性及耐久性高的肖特基電極112蒸鍍在單結晶基 上，因此以肖特基電極用遮罩6c覆蓋單結晶基板1 方，藉由IBS法將Au或Pt蒸鍍6〜l〇nm，製作屬於具 明或透明的透光性的金屬薄膜的肯特基電極112。 在金屬，除了 Au、Pt以外，亦可使用Al、Co、Ge、 、W、Mo、Cr、Cu等。 在藉由IBS法所爲之肖特基電極的形成中，係 子能量、離子電流密度、離子種、濺鍍時的動作真 堆積速度等作爲形成條件。肖特基電極11 2的形成 的條件係以離子能量爲5 00〜1 200eV、離子電流密| lOmA/cm2、離子種爲Ar離子、動作真空度爲4〜5 、堆積速度爲0.1〜〇.2nm/ sec爲佳。例如，以該等 形成的厚度l〇nm的Au薄膜係波長200nm下的紫外線 爲6 0%左右。此形成爲受光面1 2r，此時的電極尺·" 5ιηηιφ '較佳爲1〜2ιηΓηφ。其中，電極尺寸愈大’ 墊電極 設爲10 1 5 0 nm 5 1表面 樣地， 由於將 基板1 1 1 1 1的上 有半透 其中， Sη、In 列舉離 空度、 所使用 卖爲1〜 X103Pa 條件所 透射率 t爲1〜 受光面 -24- 201135954 愈大，但是伴隨於此，受到一定照度的紫外線時的損傷亦 愈大。 (實施例2 ) 按照第8圖所示的S11〜S17工程，製作複數個Si濃度 及Fe濃度不同的紫外線感測器。 將Si濃度6ppm、Fe濃度lppm的純度4N的氧化鎵粉末 封入至橡皮管，藉由靜水壓衝壓而進行壓縮成形。壓縮成 形後，使用電氣爐，以1 5 0 0 °C、4 0小時，在大氣雰圍氣中 進行燒成，使用所得的氧化鎵燒結體，藉由FZ法而育成氧 化鎵單結晶。以單結晶成長條件而言，以成長速度爲 20mm/h、雰圍氣爲乾燥空氣、壓力爲1氣壓的條件來進 行。 將所得的氧化鎵單結晶的晶錠1 04黏貼在載玻片，以 正交面41成爲切斷面的方式進行切斷，將其作爲單結晶基 板1 1 1。單結晶基板1 1 1的厚度係形成爲0.65mm。 將所切出的單結晶基板1 1 1黏貼在硏磨夾具，以# 1 〇〇〇 的硏磨盤硏削20分鐘而藉此進行出面。在硏磨夾具係使用 武藏野電子股份有限公司製的MA-200D。接著，以Ιμηι鑽 石及CMP分別硏磨30分鐘而藉此獲得鏡面。 將經硏磨的單結晶基板1 1 1以丙酮、乙醇洗淨’在氧 雰圍氣中以1 1 〇 〇 °C、4小時進行熱處理。 以加熱至225度的超音波焊料器將低熔點金屬5附在單 結晶基板1 1 1的第2表面，3秒左右以超音波供予振動’藉 -25- 201135954 此將歐姆電極1 1 3形成在單結晶基板1 1 1的第2表面的一部 分。 在加熱至325度的熱板之上置放金屬製區塊6a，在區 塊6a上置放芯柱22且將其加熱，將形成有歐姆電極i 13的 單結晶基板1 1 1設置在芯柱22上。由單結晶基板1 1 1之上以 載玻片均一加壓，進行單結晶基板1 1 1與芯柱22的晶粒接 合。晶粒接合後，將芯柱22由區塊6a取下而進行冷卻》 冷卻後，以焊墊電極用遮罩6b覆蓋單結晶基板1 1 1上 方，藉由EB蒸鍍法形成焊墊電極14。焊墊電極14的尺寸爲 0.775mm<i)，將 Crl4a 以 50nm、Aul4b 以 100nm蒸鍍在單結 晶基板1 1 1的第1表面。 藉由IBS法，在包含受光面12r的單結晶基板111的第1 表面蒸鍍l〇nm的Au，而形成肖特基電極112。以IBS法的 條件而言，將離子能量設爲850 eV、離子電流密度設爲5.5 mA/cm2、離子種設爲Ar離子、動作真空度設爲4.5xlO_3Pa 、堆積速度設爲0.15nm/ sec。此外，電極尺寸係設爲 2mm φ 。此時，實質的受光部係成爲由肖特基電極112的 2ηιιηφ去除焊墊電極14的0.775mm<i)的部分。 爲了保護形成有肖特基電極1 1 2的單結晶基板1 1 1，預 先藉由低熔點玻璃124而將窗構件21接著在帽蓋23，而且 使用設有.引線3 1〜3 2的封裝體1 02來進行封裝。此時，在 封裝體1 02內係塡充惰性氣體，而製作出紫外線感測器 1 00A。 接著，使用Si濃度及Fe濃度分別不同的複數原料粉末 -26- 201135954 ’以與紫外線感測器i 00 A同樣的方法來分別製作紫外線感 測器B〜G。 對紫外線感測器1 00 A〜G ’使用岩崎電氣股份有限公 司製的低壓水銀燈QGL200U-21，連續照射波長254nm的紫 外線，確認是否作爲紫外線感測器來進行動作。結果，紫 外線感測器1 0 〇 F及G並未作爲紫外線感測器來進行動作。 第9圖係顯示實施例2之紫外線感測器之不純物濃度的 曲線圖。第9圖所示之A〜G的各個係顯示紫外線感測器 1 00 A〜G的各個。此外’斜線範圍係表示作爲紫外線感測 器而進行動作的Fe濃度的範圍。如第9圖所示，未作爲紫 外線感測器來進行動作的紫外線感測器丨〇 〇 F及G係F e濃度 爲14及1 8PPm。由此可確認出斜線範圍外的Fe濃度14以上 者並未作爲紫外線感測器來進行動作。此外，可確認出Fe 濃度3以下者係作爲紫外線感測器來進行動作。 接著’使用Bio-Rad Laboratories股份有限公司製的霍 耳效應測定裝置’測定出紫外線感測器丨〇 〇 A〜C所具有的 感測器晶片的比電阻〔Qcm〕與載子密度〔cm·3〕。測定 結果顯示於第10及1 1圖。 第1 〇圖係顯示實施例2之紫外線感測器之不純物濃度 、比電阻及載子密度的表。第1 1圖係顯示實施例2之紫外 線感測器的比電阻、載子密度及S i濃度的相關的曲線圖。 如第1 〇及1 1圖所示，隨著Si濃度的增加，載子密度及 比電阻會減少’隨著S i濃度的減少，載子濃度及比電阻會 增加’因此可確認出原料粉末中所含有的S i濃度係與所育 -27- 201135954 成的氧化鎵單結晶的載子密度有密切關係。 (實施例3 ) 按照第8圖所示之S 1 1〜S 1 7工程，製作複數個單結晶 基板表面狀態不同的紫外線感測器。 使用Si濃度4ppm、Fe濃度lppm的原料粉末，其他係以 與實施例2同樣的條件來製作複數個紫外線感測器。其中 ，在製作複數個紫外線感測器時，在以第8圖的S 1 2所示的 氧化鎵單結晶基板硏磨工程中，複數個紫外線感測器各個 係以單結晶基板111上的平均單位面積的硏磨損傷數爲不 同的方式來調整硏磨損傷。 對所製作出的硏磨損傷數不同的複數個紫外線感測器 ，使用岩崎電氣股份有限公司製的低壓水銀燈QGL200U-2 1，連續照射波長254nm的紫外線，測定出其感測器感度 、及該感測器感度降低至1 〇%爲止的時間。測定結果顯示 於第12圖》 第1 2圖係顯示實施例3之紫外線感測器的感測器感度 、感測器的壽命及硏磨損傷的個數的相關的曲線圖。 如第1 2圖所示’隨著硏磨損傷數的增加’感測器感度 會輕微減少’但是並未發現較大的降低。由此可確認出’ 硏磨損傷數對_測器感度所造成的影響並不大。另一方面 ，可確認出感測器感度降低至1 0%爲止的時間係隨著硏磨 損傷數的增加而變長。此係因爲若按照因硏磨損傷所造成 的單結晶基板1 1 1中的第1表面的凹凸’肖特基電極1 1 2被 -28- 201135954 形成爲網目狀時，可緩和隨著紫外線照射的熱的影響之故 。其中，若在未設置硏磨損傷的平坦的單結晶基板1 1 1均 —形成肖特基電極112時，雖然感測器感度較高，但是AU 等金屬薄膜會因熱而凝集，會有造成感度降低的可能性。 (實施例4 ) 按照第8圖所示的S11〜S17工程，製作複數個歐姆電 極的形成方法不同的紫外線感測器。 使用Si濃度4ppm、Fe濃度lppm的原料粉末，其他係以 與實施例2同樣的條件來製作紫外線感測器1 ooh。其中， 在製作紫外線感測器1 0 0 Η時，在以第8圖的S 1 4所示的歐姆 電極形成工程中，使用黑田技術股份有限公司製的 Cerasolzer (註冊商標）作爲低熔點金屬而在單結晶基板 1 1 1上形成歐姆電極1 1 3。 此外，取代在第8圖的S 1 4中所示的歐姆電極形成工程 ’製作出使用在實施形態1的S 4中所示的歐姆電極形成工 程的紫外線感測器1 001與紫外線感測器〗00 j。 紫外線感測器1001係具有在該歐姆電極形成工程中， 藉由Τι / A1的蒸鍍而形成歐姆電極的氧化鎵單結晶基板。 此外，紫外線感測器1 〇〇1係使用^接合而將該氧化鎵單結 晶基板安裝在芯柱2 2。 紫外線感測器1 00〗係具有在該歐姆電極形成工程中， 藉由Ti/ Α丨的蒸鍍而形成歐姆電極而且施行燒結處理（ 800°C · 2分鐘）的氧化鎵單結晶基板。此外，紫外線感測 -29- 201135954 器1 00J係使用銀膠而將該氧化鎵單結晶基板安裝在芯柱22 〇 使用菊水電子工業股份有限公司製的二極體曲線描繪 器來測定出該等紫外線感測器1 00H〜J的歐姆電極間的電 阻。此外，對紫外線感測器100H〜J，使用上述SLUV-6連 續照射波長254nm的紫外線而測定出感測器感度。測定結 果顯示於第13及14圖。 第1 3圖係顯示實施例4之紫外線感測器中之歐姆電極 間之電阻的曲線圖。第1 4圖係顯示實施例4之紫外線感測 器中之感測器感度的曲線圖。第13及14圖所示之Η〜J的各 個係分別顯示紫外線感測器1 0 0Η〜J。 如第1 3圖所示’藉由蒸鍍所形成的紫外線感測器〗〇〇1 的歐姆電極係電阻爲1〜1 OkQ，施行燒結處理的紫外線感 測器100J的歐姆電極爲1〇〜1〇〇〇Ω。因此，藉由燒結處理 ，可確認電阻改善，但是由於工程增加，因而造成成本上 升。另一方面，可確認出紫外線感測器丨〇〇H的歐姆電極 1 1 3係具有與紫外線感測器丨〇 0 j同等的電阻。 此外，如第1 4圖所示，可確認出紫外線感測器丨〇〇;係 感測器感度優於紫外線感測器1001，紫外線感測器1〇叫係 感測器感度優於紫外線感測器〗〇〇J。由此可知，藉由使用 超曰波焊料，歐姆特性獲得改善，結果，作爲紫外線感測 器所取出的光電流會增加。 (實施例5 ) -30- 201135954 按照 電極的形 使用 以與實施 此外 ，製作出 工程的紫 對紫 續照射波 結果顯不 第15 度的曲線 1 00K及 L 如第 3 X 1 〇·3 〜 1 X 1 0·3〜丨 來形成肯 本發 他各種形 僅爲例示 請專利範 者，屬於 、替代及 其中 第8圖所示之S11〜S17工程，製作複數個肖特基 成方法爲不同的紫外線感測器。The Fe system affects the conductivity of the single crystal of gallium oxide, and the Fe concentration in the raw material powder is preferably 3 ppm or less, and particularly preferably 1 ppm or less. This is because when the sensor wafer 11 is made of a single crystal of gallium oxide using a raw material powder containing a large amount of Fe, since Fe is taken in a single crystal of gallium oxide, the conductivity is remarkably lowered, and it is not easy to react with ultraviolet rays. Therefore. Next, the method C will be described with reference to FIG. Fig. 7 is a view showing the surface state of the back electrode. In the seventh embodiment, the same components as those in the sixth embodiment are denoted by the same reference numerals as those in the sixth embodiment, and the overlapping description will be omitted. When the single crystal substrate 1 1 1 is mirror-honed, the predetermined honing damage as shown in FIG. 7 is set on the first surface of the single crystal substrate 1 π, whereby the Schottky electrode 112 is formed. When the unevenness is generated, the Kent base electrode 112 is formed into a mesh shape. In the method C, the stress relaxation caused by the difference in the coefficient of thermal expansion between the Kent base electrode 11 1 and the single crystal substrate 11 1 is achieved, and the feeling due to the disconnection of the Schottky electrode 112 is suppressed. The sensitivity of the detector is lowered, and the durability of the sensor wafer 110 is improved. The honing damage is preferably a size of ο·1 to 〇.5 mm and a width of 0.02 to 〇.丨111111. In addition, the honing damage is preferably in the range of 1 m m2, preferably 10 to 100, and particularly preferably 30 to 60. If the honing damage exceeds this size, or the number of honing damages exceeds this range, there is a possibility that the Schottky electrode 113 is broken. On the other hand, if the honing damage does not reach the size or the number of honing damages does not reach this range, the relaxation of the above stress is remarkably lowered. The Schottky electrode 1 1 2 formed on the first surface of the single crystal substrate Π 1 functions to collect a carrier generated by ultraviolet excitation and take it out as a photocurrent. If the pad electrode 14 and the bonding wire 25 are electrically connected to the portion where the carrier is generated directly under the Schottky electrode 2 by the Schottky electrode 112, the carrier system can serve as a photocurrent. Take it out to the outside. Therefore, even if the first surface of the single crystal substrate 111 has irregularities due to honing damage, and the Schottky electrode 112 is formed in a mesh shape with the unevenness, as long as no disconnection occurs, The function of the special base electrode 112 is not greatly impaired. The 'honing damage is not limited to the first surface of the single crystal substrate n丨', and may be similarly provided on the surface of the single crystal substrate 11 such as the second surface of the single crystal substrate n丨. Next, the technique D will be described. The technique is to form an ohmic electrode u 3 on the single crystal substrate 11 by using a low-melting metal using an ultrasonic solder. The external line sensor receives light such as ultraviolet light and operates by taking out the generated photocurrent as a signal. In order to take this photocurrent efficiency -19-201135954 to the outside, it is necessary to reduce the resistance of the ohmic electrode. The electric resistance between the ohmic electrodes of the ohmic electrode formed by the steaming is about 1 to 10 [kQ]. The electric resistance between the ohmic electrodes can be improved by performing a sintering treatment on the gallium oxide single crystal substrate after vapor deposition of the electrode. However, by the method D, the ultrasonic solder and the low melting point metal are used in forming the ohmic electrode 113, whereby the ohmic electrode 113 having the same electric resistance as that of the above sintering process can be formed. Further, the ohmic electrode 113 can be formed on the single crystal substrate 111 in a short time without performing a vacuum process such as vapor deposition, so that the cost reduction effect is achieved. For example, ultrasonic solder, In, Sn, or the like is used for the low melting point metal. Further, the ohmic electrode 1 13 is formed, and the low-melting-point metal is used to perform grain bonding of the single crystal substrate 11 1 and the stem 22 . Since the crystal graining material does not contain an organic substance, it is possible to suppress ultraviolet light absorption by decomposition or decomposition of the organic substance by ultraviolet irradiation, thereby achieving an effect of detecting ultraviolet rays having a shorter wavelength. Further, the window member 21 and the cap 23 are brought into close contact with the low-melting-point metal by the low-melting glass 1 2 4, and the effect of suppressing ultraviolet absorption by the organic substance is achieved. Next, the technique E will be described. The method E is a method of forming the Schottky electrode 112 by vapor-depositing A u or P t on the single crystal substrate 11 1 by the IB S method. The higher the density of the metal thin film deposited on the gallium oxide single crystal substrate, the higher the durability against ultraviolet rays. However, if it is an Au or Pt monomer, the adhesion to the gallium oxide single crystal substrate is inferior. In this method, since Au or Pt is supplied to the substrate with a stronger force -20-201135954, the IBS method of forming a film with a high vacuum of 2 or more digits compared with the usual sputtering method is used. By performing sputtering under a high vacuum by the IBS method, the kinetic energy supplied to the particles at this time is hardly damaged and can be deposited on a predetermined substrate, so that a film having high adhesion and high density is obtained. Therefore, a dense Schottky electrode 112 is formed as a metal thin film on the single crystal substrate 1 1 1 by the IBS method, whereby density and adhesion can be obtained by using an electrode formed by, for example, EB vapor deposition. Higher electrodes also improve durability. In the second embodiment, B to E and the above-described method a are appropriately combined to produce an ultraviolet sensor, but the technique is added to the two methods of the methods B and C, the methods B and D, and the methods C and E. The UV sensor can also achieve the effects of reduced cost and improved sensor sensitivity. Among them, the three methods of the methods B to D, the method C to E, the methods B, C, and E are added to the ultraviolet sensor of the method a, compared to the ultraviolet sensor that adds the two methods to the method A. It is preferable in terms of cost reduction and sensor sensitivity improvement, and it is particularly preferable to add a UV sensor of the method B to E in all of the methods A. Next, a method of manufacturing the ultraviolet sensor 100 will be described with reference to Fig. 8 for each production process. Fig. 8 is a schematic view for explaining the manufacturing process of the ultraviolet sensor of the second embodiment. (S 1 1 : crystallization of gallium oxide single crystal and cutting of crystal substrate) A gallium oxide powder having a Si concentration of 30 ppm or less and a Fe concentration of 1 ppm or less is used as a raw material, and is sealed in a rubber tube by hydrostatic pressing. Progress • 21 - 201135954 Line compression forming. After compression molding, it was fired in an air atmosphere at 1400 to 1 600 ° C for 10 to 40 hours in an electric furnace, and the obtained gallium oxide sintered body was used as a raw material rod. Using the raw material rod, a gallium oxide single crystal was grown by the FZ (Floating Zone) method. The single crystal growth conditions were carried out under the conditions of a growth rate of 5 to 30 mm/h, an atmosphere of dry air, and a pressure of 1 atmosphere. The ingot 104 of the single crystal of the oxidized monocrystal is adhered to the glass slide, and is cut so that the orthogonal plane perpendicular to the growth direction of the single crystal of the gallium oxide is cut into a cut surface, and this is used as a single crystal substrate. 1 1 1. The thickness of the single crystal substrate 1 1 1 is preferably 0.6 to 0.7 mm. (S12. Honing of the oxidized marryle crystal substrate) The cut single crystal substrate 11 1 is adhered to the honing jig, and the honing disc of #1 0 0 0 is boring for 10 to 30 minutes. Come out. Next, the mirror is obtained by honing with 1 μηΐ of diamond and CMP for 20 to 60 minutes, preferably 20 to 30 minutes, respectively. At this time, honing damage is set in a size of 0.1 to 0.5 mm in length and 0.02 to 0.1 mm in width, and 30 to 60 are provided in a range of 1 mm 2 . (S 1 3 : Annealing of Single Crystal) The honed single crystal substrate 11 was washed with acetone and ethanol, and heat-treated. In the heat treatment, in the same manner as in the first embodiment, the insulating layer 111 a required for the carrier separation by photoexcitation is formed on the single crystal substrate 111, and the defects remaining in the gallium oxide single crystal after the crystal growth are restored. The purpose is carried out. The heat treatment was carried out in an oxygen atmosphere at 1100 ° C, 3 to -22-201135954 for 24 hours. Further, in the case of using a single crystal substrate having a low carrier density which is prepared by using a raw material powder having a low Si concentration or the like, an insulating layer for separating the carriers is not required, and thus the construction can be omitted. (S14: Formation of ohmic electrode) The low melting point metal 5 was attached to the second surface of the single crystal substrate 1 1 1 by an ultrasonic soldering device heated to 200 to 2 50 °C. Thereafter, it is preferable that the ultrasonic wave is supplied with vibration for about 1 to 5 seconds, thereby breaking the surface oxide film of the low melting point metal 5, and the low melting point metal 5 is diffused in the single crystal substrate 11 to thereby form the ohmic electrode 1 1 3 is formed on the entire surface or a part of the second surface of the single crystal substrate 1 1 1 . Among them, the larger the size, the smaller the contact resistance. (S15: Mounting of the stem) The metal block 6a is placed on the hot plate heated to 300 to 350 degrees, and the stem 22 provided with the leads 31 to 32 is placed on the block 6a and heated . The single crystal substrate 1 Π ' on which the ohmic electrode 1 1 3 is formed is disposed on the stem 22 so that the ohmic electrode 1 1 3 is in contact with the stem 22 . The single crystal substrate 1 1 1 and the low-melting-point metal 5 on the second surface of the single crystal substrate 1 1 1 are uniformly pressurized by a pin combination, a slide glass, or the like uniformly on the single crystal substrate 1 1 1 to carry out the single crystal substrate 1 1 1 and The die of the stem 22 is joined. The stem 22 is removed from the block 6a for cooling, thereby completing the mounting of the single crystal substrate 11 to the stem 22. (S 1 6 : Formation of pad electrode) A pad electrode 14 for wire bonding was produced. The single crystal substrate 11 1 is covered with a pad electrode cover 6b-23-201135954, and a solder 14 is formed by E B vapor deposition. The size of the pad electrode 14 is 0.05 〜, and Crl4i 〜100 nm, preferably 30 〜50 nm', Au or Ptl4b is set to 80 -, preferably 90 0 to 1 0 0 nm, and evaporated on the single crystal substrate 1 (1: 7: formation of a Schottky electrode) In the case of the material of the Schottky electrode 1 1 2, Au, Pt, or the like belonging to a metal having a large work function is used in the same manner as in the first embodiment. . Since the Schottky electrode 112 having high adhesion and durability is deposited on a single crystal base, the single crystal substrate is covered with the Schottky electrode mask 6c, and the Au or Pt is vapor-deposited by the IBS method. At 〇nm, a Kent base electrode 112 belonging to a metal film having a clear or transparent light transmissive property is produced. In addition to Au and Pt, Al, Co, Ge, W, Mo, Cr, Cu, or the like can be used as the metal. In the formation of the Schottky electrode by the IBS method, the energy of the ray, the ion current density, the ion species, the true stacking speed at the time of sputtering, and the like are used as forming conditions. The Schottky electrode 11 2 is formed under the conditions of an ion energy of 500 00 to 1 200 eV, an ion current density of 10 OmA/cm 2 , an ion species of Ar ion, an operating vacuum of 4 to 5, and a deposition speed of 0.1 to 〇. 2nm/sec is preferred. For example, the Au film having a thickness of 10 nm formed by the above has an ultraviolet ray at a wavelength of 200 nm of about 60%. This is formed as the light receiving surface 1 2r, and the electrode scale ·" 5ιηηιφ ' at this time is preferably 1 to 2ιηΓηφ. Among them, the larger the electrode size is, the pad electrode is set to 10 1 50 nm 5 1 surface sample, because the substrate 1 1 1 1 1 is semi-transparent, Sη, In lists the air separation, and the sold price is 1 ~ X103Pa Condition Transmittance t is 1 ~ The light-receiving surface -24-201135954 is larger, but with this, the damage to the ultraviolet light with a certain illuminance is also increased. (Example 2) According to the steps S11 to S17 shown in Fig. 8, a plurality of ultraviolet sensors having different Si concentrations and Fe concentrations were produced. A gallium oxide powder having a Si concentration of 6 ppm and a Fe concentration of 1 ppm and having a purity of 4 N was sealed in a rubber tube, and compression molding was carried out by hydrostatic pressing. After compression molding, it was fired in an atmosphere using an electric furnace at 1,500 ° C for 40 hours, and a gallium oxide single crystal was grown by the FZ method using the obtained gallium oxide sintered body. The single crystal growth conditions were carried out under the conditions of a growth rate of 20 mm/h, an atmosphere of dry air, and a pressure of 1 atmosphere. The obtained ingot of the monocrystal gallium oxide 104 was adhered to a glass slide, and the orthogonal surface 41 was cut so as to be a cut surface, and this was used as a single crystal substrate 1 1 1 . The thickness of the single crystal substrate 11 1 was formed to be 0.65 mm. The cut single crystal substrate 1 1 1 was adhered to a honing jig, and the honing disc of # 1 硏 was boring for 20 minutes to carry out the surface. The MA-200D manufactured by Musashino Electronics Co., Ltd. was used for the honing jig. Next, the mirror was obtained by honing for 30 minutes with Ιμηι diamond and CMP, respectively. The honed single-crystal substrate 11 1 was washed with acetone and ethanol, and heat-treated at 1 1 〇 ° C for 4 hours in an oxygen atmosphere. The low-melting-point metal 5 is attached to the second surface of the single-crystal substrate 1 1 1 by an ultrasonic soldering device heated to 225 degrees, and is supplied with ultrasonic waves for about 3 seconds. 'By -25, 35, 35,954, this ohmic electrode 1 1 3 A part of the second surface of the single crystal substrate 11 1 is formed. A metal block 6a is placed over the hot plate heated to 325 degrees, the stem 22 is placed on the block 6a and heated, and the single crystal substrate 11 formed with the ohmic electrode i 13 is placed in the core On column 22. The crystal grains of the single crystal substrate 1 1 1 and the stem 22 are bonded by uniformly pressurizing the slide glass on the single crystal substrate 1 1 1 . After the die bonding, the stem 22 is removed by the block 6a and cooled. After cooling, the pad electrode 6b is covered with the mask electrode 6b over the single crystal substrate 1 1 1 , and the pad electrode 14 is formed by EB evaporation. . The size of the pad electrode 14 was 0.775 mm < i), and Crl4a was deposited on the first surface of the single crystal substrate 11 1 at 50 nm and Aul4b at 100 nm. The Schottky electrode 112 is formed by vapor-depositing 10 nm of Au on the first surface of the single crystal substrate 111 including the light receiving surface 12r by the IBS method. According to the conditions of the IBS method, the ion energy was 850 eV, the ion current density was 5.5 mA/cm2, the ion species was Ar ion, the operating vacuum was 4.5 x 10 3 Pa, and the deposition rate was 0.15 nm/sec. In addition, the electrode size is set to 2 mm φ . At this time, the substantial light-receiving portion is a portion where 0.775 mm < i) of the pad electrode 14 is removed by 2 ι η η of the Schottky electrode 112. In order to protect the single crystal substrate 1 1 1 in which the Schottky electrode 1 1 2 is formed, the window member 21 is previously attached to the cap 23 by the low melting point glass 124, and a package provided with the leads 3 1 to 32 is used. Body 102 is packaged. At this time, an inert gas was filled in the package 012 to produce a UV sensor 100A. Then, the ultraviolet sensor B to G were separately produced in the same manner as the ultraviolet sensor i 00 A using a plurality of raw material powders -26 to 201135954 ' in which the Si concentration and the Fe concentration were different. For the ultraviolet sensor 1 00 A to G ', a low-pressure mercury lamp QGL200U-21 manufactured by Iwasaki Electric Co., Ltd. was used, and an ultraviolet ray having a wavelength of 254 nm was continuously irradiated to confirm whether or not it was operated as an ultraviolet sensor. As a result, the ultraviolet sensors 1 0 〇 F and G did not operate as ultraviolet sensors. Fig. 9 is a graph showing the impurity concentration of the ultraviolet sensor of Example 2. Each of the systems A to G shown in Fig. 9 shows each of the ultraviolet sensors 1 00 A to G. Further, the 'slash range' indicates the range of the Fe concentration that operates as an ultraviolet sensor. As shown in Fig. 9, the ultraviolet sensor 丨〇 F and G system F e concentrations that are not operated as the ultraviolet sensor are 14 and 18 ppm. From this, it was confirmed that the Fe concentration outside the oblique line range of 14 or more did not operate as an ultraviolet sensor. Further, it was confirmed that the Fe concentration of 3 or less was operated as an ultraviolet sensor. Next, the specific resistance (Qcm) and the carrier density [cm· of the sensor wafers of the ultraviolet sensors 丨〇〇A to C were measured using a Hall Effect measuring device manufactured by Bio-Rad Laboratories Co., Ltd. 3]. The results of the measurements are shown in Figures 10 and 11. Fig. 1 is a table showing the impurity concentration, specific resistance and carrier density of the ultraviolet sensor of Example 2. Fig. 1 is a graph showing the correlation between the specific resistance, the carrier density, and the Si concentration of the ultraviolet sensor of Example 2. As shown in Figures 1 and 1 1 , as the Si concentration increases, the carrier density and specific resistance decrease. 'As the concentration of S i decreases, the carrier concentration and specific resistance increase.' Therefore, the raw material powder can be confirmed. The concentration of S i contained in the system is closely related to the carrier density of gallium oxide single crystal formed by the breeding of -27-201135954. (Example 3) An ultraviolet sensor having a different surface state of a plurality of single crystal substrates was produced in accordance with S 1 1 to S 1 7 shown in Fig. 8. A plurality of raw material powders having a Si concentration of 4 ppm and a Fe concentration of 1 ppm were used, and a plurality of ultraviolet sensors were produced under the same conditions as in Example 2. Wherein, when a plurality of ultraviolet sensors are produced, in the gallium oxide single crystal substrate honing process shown by S 1 2 in FIG. 8, the plurality of ultraviolet sensors are each averaged on the single crystal substrate 111. The number of honing injuries per unit area is a different way to adjust for honing damage. The low-pressure mercury lamp QGL200U-2 manufactured by Iwasaki Electric Co., Ltd. was used to continuously irradiate ultraviolet rays having a wavelength of 254 nm to a plurality of ultraviolet sensors having different number of honing damages, and the sensitivity of the sensor was measured. The time until the sensitivity of the sensor is reduced to 1 〇%. The measurement results are shown in Fig. 12, and Fig. 12 is a graph showing the correlation between the sensor sensitivity of the ultraviolet sensor of Example 3, the life of the sensor, and the number of honing damages. As shown in Fig. 2', the sensor sensitivity is slightly reduced as the number of honing damage increases', but no significant decrease was observed. From this, it can be confirmed that the influence of the number of honing damages on the sensor sensitivity is not large. On the other hand, it was confirmed that the time until the sensor sensitivity was reduced to 10% became longer as the number of honing damages increased. In this case, when the unevenness of the first surface of the single crystal substrate 1 1 1 caused by the honing damage is formed into a mesh shape by -28-201135954, the ultraviolet ray irradiation can be alleviated. The heat of the cause. In the case where the flat single crystal substrate 1 1 1 is not provided with honing damage, the Schottky electrode 112 is formed, although the sensitivity of the sensor is high, the metal film such as AU may agglomerate due to heat, which may cause The possibility of reduced sensitivity. (Example 4) An ultraviolet sensor having a different method of forming a plurality of ohmic electrodes was produced in accordance with the steps S11 to S17 shown in Fig. 8. The raw material powder having a Si concentration of 4 ppm and a Fe concentration of 1 ppm was used, and the ultraviolet sensor 1 ooh was produced under the same conditions as in the second embodiment. In the ohmic electrode forming process shown by S 14 in Fig. 8, the ERA electrode forming process shown in S1 4 of Fig. 8 is used as a low melting point metal in the production of the ultraviolet sensor. On the single crystal substrate 1 1 1 , an ohmic electrode 1 13 is formed. Further, instead of the ohmic electrode forming process shown in S 14 of FIG. 8 , the ultraviolet sensor 1 001 and the ultraviolet sensor using the ohmic electrode forming process shown in S 4 of the first embodiment are fabricated. 〗 00 j. The ultraviolet sensor 1001 has a gallium oxide single crystal substrate in which an ohmic electrode is formed by vapor deposition of Τι / A1 in the ohmic electrode forming process. Further, the ultraviolet sensor 1 〇〇 1 is mounted on the stem 22 by using a bonding of the gallium oxide single crystal substrate. The ultraviolet sensor 1 00 is a gallium oxide single crystal substrate in which an ohmic electrode is formed by vapor deposition of Ti/Α丨 and a sintering treatment (800 ° C · 2 minutes) is performed in the ohmic electrode formation process. In addition, UV-sensing -29-201135954 00J uses silver paste to mount the gallium oxide single crystal substrate on the stem 22, and uses a diode curve plotter manufactured by Kikusui Electronics Co., Ltd. to measure these. The resistance between the ohmic electrodes of the UV sensor 1 00H~J. Further, for the ultraviolet sensors 100H to J, the sensor sensitivity was measured by continuously irradiating the ultraviolet light having a wavelength of 254 nm with the SLUV-6. The results of the measurements are shown in Figures 13 and 14. Fig. 13 is a graph showing the electric resistance between the ohmic electrodes in the ultraviolet sensor of Example 4. Fig. 14 is a graph showing the sensitivity of the sensor in the ultraviolet sensor of Example 4. The respective systems of Η~J shown in Figs. 13 and 14 respectively display ultraviolet sensors 100 Η to J. As shown in Fig. 3, the ohmic electrode resistance of the ultraviolet sensor formed by vapor deposition is 1 to 1 OkQ, and the ohmic electrode of the ultraviolet sensor 100J subjected to the sintering treatment is 1 〇. 1 〇〇〇 Ω. Therefore, it is confirmed by the sintering treatment that the electric resistance is improved, but the cost is increased due to an increase in engineering. On the other hand, it was confirmed that the ohmic electrode 1 1 3 of the ultraviolet sensor 丨〇〇H has the same resistance as the ultraviolet sensor 丨〇 0 j . In addition, as shown in Fig. 14, it can be confirmed that the ultraviolet sensor is better than the ultraviolet sensor 1001, and the sensitivity of the ultraviolet sensor 1 is better than the ultraviolet sensor. Measure 〗 〇〇 J. From this, it is understood that the ohmic characteristics are improved by using the ultra-wave solder, and as a result, the photocurrent taken out as the ultraviolet sensor is increased. (Example 5) -30- 201135954 According to the shape of the electrode and the implementation, in addition, the result of the process of producing the purple-violet illuminating wave of the project is not the 15th degree curve 1 00K and L as the 3 X 1 〇·3 〜 1 X 1 0·3~丨 to form Kenben's various forms are only examples of patents, belonging to, replacing and S11~S17 projects shown in Figure 8, making a plurality of Schottky methods different. UV sensor.
Si濃度24ppm、Fe濃度ippm的原料粉末，其他係 例2同樣的條件來製作紫外線感測器1 〇〇κ。 ’取代第8圖的S 1 7中所示的肖特基電極形成工程 使用在實施形態1的S 6中所示之肖特基電極形成 外線感測器100L。 外線感測器1 0 0 Κ及L，使用上述Q g L 2 0 0 U - 2 1來連 長254nm的紫外線’而測定出感測器感度。測定 於第1 5圖。 圖係顯示實施例5之紫外線感測器中之感測器感 圖。第1 5圖所示之K及l係表示紫外線感測器 〇 1 5圖所示，紫外線感測器丨〇〇L係感測器感度爲 6xl〇-4 ( A / W ]，但是紫外線感測器100K爲 5xl〇·3 [ A/ W )。因此，可確認出若使用IBS法 特基電極112時，感測器感度會提升。 明在未脫離其要旨或主要特徵的情形下，可以其 式來實施。因此，前述實施形態1〜2的所有內容 ，並不能作限定性解釋。本發明的範圍係藉由申 圍所示者，在說明書本文中並未作任何限制。再 申請專利範圍的均等範圍的所有變形、各種改良 改質均在本發明的範圍內。 ，申請專利範圍所記載的正交面例如爲前述實施 -31 - 201135954 形態1及2中的正交面4 1 ’受光面例如爲受光面1 2 r。氧化 鎵單結晶基板例如爲氧化鎵單結晶基板1 1及1 1 I，封裝體 例如爲封裝體2及封裝體1 0 2。第1表面例如爲氧化鎵單結 晶基板1 1及1 1 1的第1表面’第2表面例如爲氧化鎵單結晶 基板Π及1 1 1的第2表面。低熔點金屬例如爲低熔點金屬5 ’低熔點玻璃例如爲低熔點玻璃1 2 4。窗構件例如爲窗構 件2 1，第1及第2端子例如爲引線3 1及3 2。 【圖式簡單說明】 第1圖係顯示實施形態1之紫外線感測器之構成的剖視 圖。 第2圖係顯示封裝體內之感測器晶片的上視圖。 第3圖係用以說明手法A的模式圖》 第4圖係用以說明實施形態1之紫外線感測器之製作工 程的模式圖。 第5圖係顯示實施例1之紫外線感測器之感測器感度與 靜電電容的曲線圖。 第6圖係顯示實施形態2之紫外線感測器之構成的剖視 圖。 第7圖係顯不肯特基電極的表面狀態圖。 第8圖係用以說明實啤形態2之紫外線感測器之製作工 程的模式圖》 第9圖係顯示實施例2之紫外線感測器之不純物濃度的 曲線圖 -32- 201135954 第1 0圖係顯示實施例2之紫外線感測器之不純物濃度 、比電阻及載子密度的表。 第1 1圖係顯示實施例2之紫外線感測器之比電阻、載 子密度及S i濃度的相關的曲線圖。 第1 2圖係顯示實施例3之紫外線感測器的感測器感度 、感測器的壽命及硏磨損傷的個數的相關的曲線圖.。 第1 3圖係顯不實施例4之紫外線感測器中之歐姆電極 間的電阻的曲線圖。 第1 4圖係顯示實施例4之紫外線感測器中之感測器感 度的曲線圖。 第1 5圖係顯示實施例5之紫外線感測器中之感測器感 度的曲線圖。 【主要元件符號說明】 I、 100 :紫外線感測器 2 ' 102 :封裝體 3:電流/電壓源 4、104 :晶錠 5 :低熔點金屬 6a·區塊 6b:焊墊電極用遮罩 6c:肖特基電極用遮罩 1 〇、11 〇 :感測器晶片 II、 Π 1 :氧化鎵單結晶基板 -33- 201135954 1 1 a、1 1 1 a :絕緣層 1 1 b :導電層 12、112:肖特基電極 1 2r :受光面 1 3、1 1 3 :歐姆電極 1 4 :焊墊電極 14a ： Cr 14b: Au或 Pt 20 :開口部 21 :窗構件 2 2 :芯柱 23 :帽蓋 2 4 :接著劑 25 :接合導線 2 6 :玻璃 3 1、3 2 :引線 3 1 a :引線3 1的一端 4 1 :正交面 4 2: a 面 124 :低熔點玻璃 -34-The raw material powder having a Si concentration of 24 ppm and a Fe concentration of i ppm was produced under the same conditions as in the other example 2 to produce an ultraviolet sensor 1 〇〇κ. The Schottky electrode forming process shown in S17 of Fig. 8 is used instead of the Schottky electrode forming external line sensor 100L shown in S6 of the first embodiment. The external line sensors 1 0 0 Κ and L, and the sensitivity of the sensor was measured by using the above Q g L 2 0 0 U - 2 1 to connect the ultraviolet ray of 254 nm. It is measured in Figure 15. The figure shows the sensor sense in the ultraviolet sensor of Example 5. The K and l series shown in Fig. 5 show the ultraviolet sensor 〇15, and the sensitivity of the ultraviolet sensor 丨〇〇L sensor is 6xl〇-4 (A / W), but the ultraviolet sensation The detector 100K is 5xl 〇 · 3 [ A / W ). Therefore, it can be confirmed that when the IBS method base electrode 112 is used, the sensitivity of the sensor is improved. It can be implemented in a manner that does not deviate from its gist or main features. Therefore, all of the above-described first to second embodiments are not to be construed as limiting. The scope of the present invention is intended to be limited by the scope of the invention. All modifications, variations and modifications of the equivalent scope of the claims are intended to be within the scope of the invention. The orthogonal surface described in the patent application range is, for example, the above-described embodiment -31 - 201135954. The orthogonal surface 4 1 ' in the first and second embodiments is a light-receiving surface 1 2 r. The gallium oxide single crystal substrate is, for example, a gallium oxide single crystal substrate 1 1 and 1 1 I, and the package is, for example, a package 2 and a package 1 0 2 . The first surface is, for example, the first surface of the gallium oxide single crystal substrate 1 1 and the 119. The second surface is, for example, a gallium oxide single crystal substrate Π and a second surface of 191. The low melting point metal is, for example, a low melting point metal. The low melting point glass is, for example, a low melting point glass 1 24 . The window member is, for example, the window member 2 1, and the first and second terminals are, for example, leads 31 and 32. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a cross-sectional view showing the configuration of an ultraviolet sensor according to a first embodiment. Figure 2 shows a top view of the sensor wafer within the package. Fig. 3 is a schematic view for explaining the method of the method A. Fig. 4 is a schematic view for explaining the manufacturing process of the ultraviolet sensor of the first embodiment. Fig. 5 is a graph showing the sensor sensitivity and electrostatic capacitance of the ultraviolet sensor of Example 1. Fig. 6 is a cross-sectional view showing the configuration of the ultraviolet sensor of the second embodiment. Figure 7 is a diagram showing the surface state of the ungrounded electrode. Fig. 8 is a schematic view for explaining the manufacturing process of the ultraviolet sensor of the real beer form 2. Fig. 9 is a graph showing the impurity concentration of the ultraviolet sensor of the embodiment 2 - 32 - 201135954 Fig. 1 0 A table showing the impurity concentration, specific resistance, and carrier density of the ultraviolet sensor of Example 2 was shown. Fig. 1 is a graph showing the correlation between the specific resistance, the carrier density, and the Si concentration of the ultraviolet sensor of Example 2. Fig. 12 is a graph showing the correlation of the sensor sensitivity of the ultraviolet sensor of Example 3, the life of the sensor, and the number of honing damages. Fig. 1 is a graph showing the electric resistance between the ohmic electrodes in the ultraviolet sensor of the fourth embodiment. Fig. 14 is a graph showing the sensitivity of the sensor in the ultraviolet sensor of Example 4. Fig. 15 is a graph showing the sensitivity of the sensor in the ultraviolet sensor of Example 5. [Main component symbol description] I, 100: UV sensor 2 ' 102 : Package 3: Current/voltage source 4, 104: Ingot 5: Low melting point metal 6a · Block 6b: Pad electrode mask 6c : Schottky electrode mask 1 〇, 11 〇: sensor wafer II, Π 1 : gallium oxide single crystal substrate -33- 201135954 1 1 a, 1 1 1 a : insulating layer 1 1 b : conductive layer 12 , 112: Schottky electrode 1 2r : light receiving surface 1 3, 1 1 3 : ohmic electrode 1 4 : pad electrode 14a : Cr 14b: Au or Pt 20 : opening portion 21 : window member 2 2 : stem 23 : Cap 2 4 : Adhesive 25 : Bonding wire 2 6 : Glass 3 1 , 3 2 : Lead 3 1 a : One end of lead 3 1 4 1 : Orthogonal surface 4 2: a Face 124: Low melting glass - 34-