[go: up one dir, main page]

TW528816B - Method for heat treating silicon wafer - Google Patents

Method for heat treating silicon wafer Download PDF

Info

Publication number
TW528816B
TW528816B TW89122008A TW89122008A TW528816B TW 528816 B TW528816 B TW 528816B TW 89122008 A TW89122008 A TW 89122008A TW 89122008 A TW89122008 A TW 89122008A TW 528816 B TW528816 B TW 528816B
Authority
TW
Taiwan
Prior art keywords
wafer
silicon
area
heat treatment
field
Prior art date
Application number
TW89122008A
Other languages
Chinese (zh)
Inventor
Hiroshi Koya
Hisashi Furuya
Yoji Suzuki
Yukio Muroi
Takaaki Shiota
Original Assignee
Mitsubishi Material Silicon
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP11686299A external-priority patent/JP3855531B2/en
Priority claimed from JP21375199A external-priority patent/JP4003351B2/en
Priority claimed from JP33532899A external-priority patent/JP4107628B2/en
Application filed by Mitsubishi Material Silicon filed Critical Mitsubishi Material Silicon
Application granted granted Critical
Publication of TW528816B publication Critical patent/TW528816B/en

Links

Landscapes

  • Crystals, And After-Treatments Of Crystals (AREA)

Abstract

The present invention is related to a process where a polycrystalline silicon layer with a thickness of 0.1-1.6 μm is formed on the rear surface of the silicon wafer by chemical vapor deposition (CVD) method and oxygen concentration of a silicon wafer, with which crystal oriented particles (COPs) and interstitial-type large dislocations (L/Ds) are not produced on a wafer surface is not larger than 1.2x10<SP>18</SP> atoms/cm<SP>3</SP> (old ASTM). The above silicon wafer is subjected to a thermal treatment at 1,000 DEG C ± 30 DEG C for 2-5 hours in an oxygen atmosphere, and successively, is subjected to a thermal treatment at 1,130 DEG C ± 30 DEGC for 1-16 hours. More specifically, oxidation induced stacking faults (OSFs) are made obvious in the central part of the silicon wafer. Therefore, when a conventional thermal treatment for making OSF obvious is carried out, the polycrystalline silicon wafer produced through above mentioned method is free of OSF and COP. It makes oxide deposited uniformly over the whole surface of the wafer and obtains uniform gettering effects, without variation between the circumferential edge and central part of the wafer.

Description

528816 A7 B7 五、發明説明(1 ) 〈發明所屬之技術領域〉 本發明係關於一種由丘克拉斯基法(C20C hr a 1 ski me hod, K下簡稱CZ法)製造之半導體電路製造用矽晶圓 之熟處理方法,用於該方法之晶圓及由該方法處理之晶圓。 〈習知之技術〉 近年,在半導體電路之製造上被視為影響產率之原因 可舉:構成氧化引發疊層缺陷(Oxidation Induced Stacking F a ii 11,K下簡稱0 S F )的核之氧析出物之微小缺陷,或起 因於结晶之粒子(Crystal Originated Particle, Μ 下稱 COP),或侵入型轉位(Interstitial-type Large Dislocation, M下稱L/D)等之存在。OSF係在结晶成長 時搆成其核之微小缺陷被導入矽晶塊中而在製造半導體裝 置之氧化程序等時顯現,構成裝置之漏電流增加等不良原 因。又,使用氨水及過氧化氫之混合液洗淨鏡面研磨後之 矽晶圓時,在晶圓表面會形成洞孔(Pit),利用測粒子器 檢測此晶圓時,此洞孔會與本來固有之粒子一起被檢出。 上逑之洞孔係起因於结晶,為使其與本來固有之粒子區別 ,特稱為COP。此種晶圓表面之洞孔之COP乃構成電氣特 性,例如氧化膜之時間相關(經時)絕緣破壞特性(Time Dependent Dielectric Breakdown,簡稱 TDDB),氧化膜 耐壓特性(Time Zero Dielectric Breakdown,簡稱 TZDB) 。又,晶圓表面有COP存在時,會在裝置之配線時發生段 差,構成斷線之原因,同時亦會在元件分離部分構成漏電 原因,降低製品產率。 -4 一 本紙張尺度適用中國國家標準(CNS ) A4規格(210X 297公釐) JI----Ί---- (請先閱讀背面之注意事項再填寫本頁) -、1Τ #·. 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明説明(2 ) L/D有時被稱為轉位藤(dislocation cluster),或 由於具有此缺陷之矽晶圓浸漬於主含氫氟酸之選擇蝕刻液 中時會產生蝕刻洞孔,因而亦被稱為轉位洞孔(dislocation Pit)。L/D亦會使漏電特性及隔絕特性等電氣特性之劣化。 由上所逑,製造半導體積體電路所用之矽晶圓必·須儘 可能的減低0 S F、C 0 P及L / D。 關於不發生0SF及轉位簇(L/D)等之無缺陷矽單晶之 製造方法已見揭示於日本特開平8- 3 30316號及特開平11-iS 9 3 號公 報中。 特開平 8 - 3 3 0 3 1 6 號 之方法 為一種 利用低 速育成矽單晶,使Μ矽晶圓之狀態熱氧化處理時,環圈狀 的發生之0SF會在晶圓中心部消失且可將L/D從晶圓之全 面去除之方法。 但,依此方法製造無缺陷之矽單晶時,矽單晶之提拉 速度之範圍及軸方向之结晶內溫度梯度(temperature gradation)之範圍均較狹窄,隨著提拉之矽單晶的直徑 之增大,增加無缺陷矽單晶製造之困難度,且有時因提拉 速度之變動等會使0 S F不Μ環圈狀而Μ塊狀發生於晶圓中 心部。此0SF如上所述會導致漏電特性之惡化,因此矽單 晶之改良製法為一般所殷求。 曰本特開平1 1 - 1 3 9 3號公報揭示之方法則為一種Κ空 孔型點缺陷之凝聚體及格子間矽型點缺陷之凝聚體不存在 之完整無缺,即完美領域(perfect domain)作為[ρ]時, 從矽融液提拉由此完整無缺領域[P]構成之矽單晶之方法 。從此種晶塊(ingot)切出之矽晶圓即由完整無缺領域[p] -5 - 本紙張又度適用中國國家標準(CNS ) A4規格(210X 297公釐) (請先閲讀背面之注意事項再填寫本買) 訂 經濟部智慧財產局員工消費合作社印製 A7 B7528816 A7 B7 V. Description of the invention (1) <Technical Field to which the Invention belongs> The present invention relates to a silicon for semiconductor circuit manufacturing manufactured by the C20Chr a 1 ski me hod (CZ method for short). A method for ripening a wafer, a wafer used in the method, and a wafer processed by the method. 〈Knowledgeable Technology〉 In recent years, the factors that have been regarded as affecting productivity in the manufacture of semiconductor circuits are as follows: Oxygen precipitation that constitutes Oxidation Induced Stacking F a ii 11, K hereinafter referred to as 0 SF The existence of small defects in the material, or the origin of crystalline particles (M) (hereinafter referred to as COP), or invasive transposition (Interstitial-type Large Dislocation (M, referred to as L / D)). OSF is caused by small defects that form its core during the crystal growth, which are introduced into the silicon ingot, and appear during the oxidation process of manufacturing semiconductor devices. In addition, when a mirror-polished silicon wafer is cleaned with a mixed solution of ammonia and hydrogen peroxide, a hole (Pit) will be formed on the wafer surface. When the wafer is detected by a particle detector, the hole will be the same as the original Inherent particles are detected together. The hole system of the upper part is due to crystallization. In order to distinguish it from the inherent particles, it is called COP. The COP of the holes on the surface of such wafers constitutes electrical characteristics, such as time-dependent (time-dependent) dielectric breakdown characteristics (TDDB) of oxide films, and time zero dielectric breakdown characteristics (oxidized films) TZDB). In addition, when a COP exists on the wafer surface, a step difference may occur during the wiring of the device, which may cause a disconnection. At the same time, it may cause a leakage current in the component separation part, which reduces the product yield. -4 A paper size is applicable to China National Standard (CNS) A4 specification (210X 297 mm) JI ---- Ί ---- (Please read the precautions on the back before filling this page)-, 1Τ # ·. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 B7 V. Description of the invention (2) L / D is sometimes called a dislocation cluster, or because a silicon wafer with this defect is immersed in the main hydrogen Etching holes are generated in the selective etching solution of hydrofluoric acid, so it is also called dislocation Pit. L / D also deteriorates electrical characteristics such as leakage characteristics and isolation characteristics. From the above, the silicon wafers used in the fabrication of semiconductor integrated circuits must be reduced to 0 S F, C 0 P, and L / D as much as possible. The manufacturing methods of non-defective silicon single crystals, such as 0SF and transposed clusters (L / D), have not been disclosed in Japanese Patent Laid-Open Nos. 8-3 30316 and 11-iS 93. The method of Japanese Patent Application Laid-Open No. 8-3 3 0 3 1 6 is a method of growing silicon single crystals at a low speed to thermally oxidize the state of the M silicon wafer. The ring-shaped 0SF will disappear at the center of the wafer and may be removed. A method to remove L / D from the wafer. However, when a defect-free silicon single crystal is manufactured by this method, the range of the pulling speed of the silicon single crystal and the range of the temperature gradation within the crystal in the axial direction are narrower. The increase of the diameter increases the difficulty of manufacturing a defect-free silicon single crystal, and sometimes the 0 SF is not ring-shaped and the M-block shape occurs at the center of the wafer due to changes in the pulling speed. As described above, this 0SF may cause deterioration of the leakage characteristics, and therefore, an improved method for manufacturing a silicon single crystal is generally desired. The method disclosed in Japanese Patent Laid-Open Publication No. 1 1-1 3 9 3 is the completeness of the existence of a kind of K-hole type point defect aggregates and inter-lattice silicon-type point defect aggregates, which is the perfect domain. ) As [ρ], a method of pulling a silicon single crystal composed of a complete and unbroken field [P] from a silicon melt. The silicon wafer cut out from this ingot is from the complete field [p] -5-This paper is also applicable to the Chinese National Standard (CNS) A4 specification (210X 297 mm) (Please read the note on the back first) Please fill in this item to buy) Order printed by the Intellectual Property Bureau of the Ministry of Economic Affairs Consumer Cooperatives A7 B7

528816 五、發明說明(3 ) 構成。完整無缺領域[P】係介在於格子間矽型點缺陷佔_ 勢之領域[I]及在矽單晶晶塊內空孔型點缺陷佔優勢之讀 域[V】之間。由此完蝥無缺領域[P ]構成之矽晶圓係藉由 設定V/G值(fflffl2/分Ό),使晶塊之提拉速度為VUm/分)、 矽融液及晶塊之界面附近之晶塊垂直方向之溫度梯度為G (υ /mm)時,實施熱氧化處理時圈狀地發生之〇SF會在晶 圓中心部消失之條件下製取。 另外,有些半導體製造商要求不具有〇SF、COP及L/d 且又具有可將裝置工程(d e v i c e ρ γ 〇 c e s s )產生之金屬污 染物除氣(Gotte ring)性能之矽晶圓。不具備充分之除氧 性能之晶圓在裝置工程受金屬之污染時,會發生接合漏姨 (junction leakage)或由於金屬不純物之陷阱能级( trap level)引起裝置之動作不良等而降低製品之產率。 故為解決此問題,一般要求一種在裝置製造商之裝置;Cg 之熱處理時可發揮固有除氣(intrinsic gettering, 簡稱I G )效果之矽晶圓。 但,從上逑之完蝥無缺領域[P ]構成之晶塊切出之0 晶圓,雖然不具有OSF、COP及L/D,但在裝置製造工程&gt; 熱處理時不一定會在晶圓的內部發生氧析出,结果有不能 獲得充分之IG效果之缺點。 傳統上,在裝置工程為了使晶圓發揮IG效果之熱處理 法係一種在晶圓內部預先形成缺陷或故意的添加不純物之 方法。由此方法處理之砂晶圓可在晶圓上預先形成之缺陷 周邊吸收其後製程發生之污染。因此,可K防止在製作裝 一 6 - 本紙張尺度適用中國國家標準(CNS)A4規格(210x 297公釐) (請先閱讀背面之注意事項再填寫本頁) ---------tr----------線·. 經濟部智慧財產局員工消費合作社印製 528816 經濟部智慧財產局員工消費合作社印製 Α7 Β7 五、發明說明(4 ) 置之晶圓表面之附近領域發生缺陷或污染。 另外,近年由於裝置之高集聚化(封装密度之增高), 裝置工程之熱處理溫度有1 0 Q 〇°C以下之低溫化傾向,故配 合此低溫化,其前工程之I G處理亦被要求低溫化。 為此,在日本特開平8 - 4 5 9 4 5號專利中有揭示一種包 括:將矽單晶晶塊切出而經研磨拋光後之矽晶圓在5 0 0 - 8 0 0 °C保持0.5-2G小時而於晶圓內導入氧析出核(precipitation nucli of oxygen)之工程;將此含有氧析出核之矽晶圓 從室溫急速加熱至8 G (3 - 1 0 0 0 °C而保持0 . 5 - 2 G分鐘之工程; 將該急速加熱保持0 · 5 - 2 0分鐘之矽晶圓放冷到室溫之工 -程;以及將該放冷之矽晶圓從5 0 ϋ - 7 fl Q°C K 2 - 1 Q /分之速度 加熱至8 0卜1 1 0 0 υ而在該溫度下保持2 - 4 8小時之工程之 I G處理法。 依此處理法時,當Κ上逑溫度條件急速加熱時,不僅 晶圓的表面,晶圓的內部亦會暫時變為熱平衡濃度Μ下, 使格子間矽原子變為缺乏狀態,從而提供氧析出核易於安 定成長之環境。同時,為補足上述格子間之缺乏矽原子而 變成安定狀態,於晶圓的表面發生格子間矽原子之生成, 並使生成之格子間矽原子向晶圓內部擴散。於是,格子間 矽原子呈缺乏狀態之晶圓的表面附近即由格子間矽原子之 生成變為飽和狀態,而使氧析出核開始消失。但,由於晶 圓表面生成之格子間矽原子擴散至內部需要一段時間,於 是愈是遠離晶圓表面之内部深處愈保持長久之易供氧析出 核成長之環境◦因此,愈接近晶圓表面,氧析出核之密度 -7 - 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 11^-----^---· ^-------- ^ · --------I 、(請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 528816 A7 __________ B7 五、發明說明(5 ) 愈低,同時熱處理時間(0.5-20分鐘)愈長,氧析出核, 即不形成缺陷之層(denuded zone, Μ下簡稱DZ)之厚度 愈大。此外,在8 0 0 - 1 0 0 0 之範圍,溫度愈度,格子間砂 原子之擴散係數愈大,因此DZ厚度在短時間變大。 急速加熱並於室溫放冷後再將晶圓加熱至800-11001 時,急速加熱存活之晶圓内部之氧析出核會成長為氧析出 物而成為安定之IG源。本文中,將此氧析出物稱為BMD( Bulk Micro Defect)0 但,上逑之IG處理法,由於需要在研磨拋光旋後之砂 晶圓,在5 0 G - 8 0 0 υ保持5 - 2 0小時而將氧析出核導入晶圓 内之步驟作為生成I G源之前處理,進而於急速加熱後需實 施熱處理使晶圓内部之氧析出核成長為BMD。因此,有在 晶圓的狀態實施多次熱處理之缺失存在。 &lt;發明之概要〉 本發明之第1目的為提供一種即使在實施傳統之0SF 顯化熱處理(OSF_fflanifesting heat treatment)時會在 中心部集中發生顯化之晶圓亦可防止因熱氧化而發生0 S F K及不含C 0 P之矽晶圓的熱處理方法。 本發明之第2目的為提供一種在晶圓之全面實施均句 之氧析出,而在晶圓周緣部及晶圓中心部之間獲得無偏差 均勻之除氣效果(gettering effect)之具有多晶矽層之 矽晶圓的製造方法。 本發明之第3目的為提供一種即使是由領域[P v ]及 領域[P 1】之混合領域構成之濃度0 · 8 X 1 0 1 8〜1 . 4 X 1 0 1 8 -8- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) (請先閱讀背面之注意事項再填寫本頁) 裝·-- 訂------·!· 528816 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明(6 原子/ c m 3 (舊A S Τ Μ )之晶塊切出之矽晶圓,亦不具有點缺 陷之凝集體且能充份產生氧析出核,且可藉裝置製造工程 之熱處理發揮IG效果之矽晶圓之熱處理方法。 本發明之第4目的為提供一種不需要實施氧施主去除 處理(oxygen donor killer treatment)工程之矽晶圓的 熱處理方法。 本發明之第5目的為提供一種可使在矽晶圓狀態之熱 處理次數少,在9 5 0 °C Μ下之熱處理即可發揮高I G效果之 矽晶圓的熱處理方法。 本發明之第6目的為提供一種,由此處理法製取之可 發揮高IG效果之矽晶圓。 本發明之第7目的為提供可製取此種可發揮高IG效果 之矽晶圓之單晶晶塊。 〈達成發明之目的之手段&gt; 本發明之第1觀點為提供一種含有藉化學的氣相堆積 (M下簡稱CVD)法,在矽晶圓面内無發生起因於结晶之粒 子(COP)及侵入型轉位(L/D)之氧濃度i.2xl〇iaat〇ffis/ cm3以下(舊ASTM)之矽晶圓的背面,在67〇υ±3〇υ之溫 度形成厚度0·1-1·6//ιη之多晶矽層之步驟;及將上述之帶 有多晶矽層之矽晶圓置於氧氣氛,在丨0 〇 t 土 3 Q它之溫度 下實施2-5小時之熱處理之後,再在U3(rc 士 3(rc之溫度 下實施卜1 6小時之處理之步驟之矽晶圓的熱處理方法,其 特徵乃在於形成該多晶矽層之前之矽晶圓在實施上逑之熱 處理後會在晶圓中心部發生氧化引發堆積缺陷(〇 s F )之顯 - 9- 紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公爱)— ·---l· --------- ------ (請先閱讀背面之注意事項再填寫本頁) 528816 Α7 Β7 經濟部智慧財產局員工消費合作社印製 五、發明說明(7 ) 化。 本發明之第2觀點為提供一種在矽晶圓面内無發生起 因於结晶之粒子(COP)及侵入型轉位(L/D)之氧濃度1.2 X 10 1 8 atoms/cm3 Μ下(舊ASTM)之矽晶圓的背面形成有厚 度0· 1-1 .6u m之多晶·矽層之帶有多晶矽層之矽晶圓,其特 徵乃在形成上逑多晶矽層之前之矽晶圓係置於氧氣氛,在 1 0 Q G °C 土 3 Q °C之溫度下實施2 - 5小時之熱處理之後再在 113 ο υ之溫度下實施1 - 1 6小時之熱處理時會在晶圓中心 部發生氧化引發堆積缺陷(0SF)之顯化。 本發明之第1及第2觀點之矽晶圓係在其中心部會顯 現0SF的條件下依丘克拉斯基(CZ)法製取之晶圓,其中 心部具有較多之氧析出核,其他部分殆不具有氧析出核。 同時,該中心部Κ外之部分不含C 0 Ρ。於此種矽晶圓的背 面形成多晶矽層時,在CVD過程會在晶圓全面形成BMD。 结果可在晶圓的全面實行均勻之氧析出,發揮晶圓中心部 與其他部分之間全無偏差之均勻I G效果。 本發明之第3觀點為,將矽單晶晶瑰内之格子間矽型 點缺陷佔優勢之領域作為「I」,將空孔型點缺陷佔優勢 之領域作為[V ],將格子間矽型點缺陷之凝聚體及空孔型 點缺陷之凝聚體完全不存在之無缺陷領域作為「Ρ」,將 鄰接上述領域[I]且屬於上述缺陷領域[Ρ]之形成浸入型 轉位之未滿最低格子間矽濃度之領域作為[Ρ Π ,及將鄰接 上逑領域「V」且屬於上述無缺陷領域[Ρ]之可形成COP或 F P D之空孔濃度Μ下之領域作為[P v ]時,由上逑無缺陷 -10- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) (請先閱讀背面之注意事項再填寫本頁) ----訂---------線 經濟部智慧財產局員工消費合作社印製 528816 A7 ___________ B7 五、發明說明(8 ) 領域[P ]構成之晶塊切出之無點缺陷之凝聚體存在之矽晶 圓的熱處理方法。 此熱處理方法之特徵點為將上述領域[P v ]及領域[P i ] 之混合領域構成且酸濃度為0.8X 1018〜1.4X c πι3 (舊A S T Μ )之矽單晶晶塊提拉,而將上逑晶塊切出之矽 晶圓置於氮、氬、氫、氧或其混合氣體之氣氛下,在6 〇 0 _ 8501¾傑持30-90分鐘。 本發明之第4觀點為將由上述領域[Pv]及領域[Pl ] 之混合領域構成旦氧濃度為0.8X HP8〜1.3X l〇i8at〇IRS/ c πι3 (舊A S T Μ )之矽單晶晶塊提拉,而將上逑晶塊切出之矽 晶圓置於氮、氩、氫、氧或其混合氣體之氣氛下,在6 0 0 _ 8 5 0 υ保持1 2 0 - 2 5 0分鐘。 本發明之第5觀點為將由上逑領域[Ρν]及領域[Pl ] 之混合領域構成之氧濃度為0.8X 1018〜1.4X l(]i8atc)nis/ cm3 (舊ASTM)之矽單晶晶塊提拉,而將上逑晶塊切出之石夕 晶圓置於氮、氬、氫、氧或其混合氣體之氣氛下,Μ丨 1 5 0 t /秒之昇溫速度由1 1 5 0 υ加熱至1 2 0 0 °C,並在1 1 5 〇 _ 1 2 0 G t保持Q - 3 Q秒鐘。 依本發明之上述第3 - 5觀點之方法,即使晶塊之氧冑 度為 0.8Χ1018〜1.4xi018atoms/cin3 (舊 ASTM)時,若石夕 覃晶圓係由領域[Pv]及領域[P]之混合領域構成時,$ 上逑條件對由該晶塊切出之矽晶圓進行熱處理,即在結帛 成長時亦可使未導入氧析出核之領域[Pi]顯現氧析出核, 同時在结晶成長時有導入氧析出核之領域[P v ]顯琨氧析 _ 1 1 _ 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) r L----------------- ^:ί------Aw (請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7 五、發明說明(9 ) 出核密度之增高。因此,在半導體裝置製造商之裝置製造 工程時對經過上述熱處理後之晶圓進行熱處理時,上述氧 析出核會成長成為BMD,故即使是由領域[Pv】及領域[PJ 之混合領域構成之晶圓亦可使晶圓全面具有I G效果。 本發明之第6觀點為提供一種包含:從矽融液提拉矽 簞晶塊之步驟及由該晶塊製作矽晶圓之步驟,以及將該矽 晶圓從室溫分K上之昇溫速度急速加熱至650-950 °C,並在該溫度下保持ϋ. 5-30分鐘之步驟之矽晶圓的熱處 理方法,而其特徵為該矽晶圓受熱氧化處理時會在晶圓總 面積之25¾ Κ上處發生OSF,且含有不發生轉位之BMD IX 10s 〜3x 107 個 /cib3〇 本發明之第6觀點之方法為使用含有上逑比率之OS F 領域及所述密度之BMD之晶圓時,可以省略Μ往之於晶圓 内導入氧析出核之前熱處理步驟及使氧析出核成長為BMD 之成長步驟,對從晶塊切出而研削研磨後之晶圓直接依上 述條件急速加熱即可發揮高IG效果。 次依實施形態詳细說明本發明於下: Α.第1實施形態 本發明之第1-4實施形態之矽晶圓係依CZ法,從熱區 爐内之矽融液,根據Voronkov理論之所定提拉速度輪廓( p u 11 i n g - u p s p e e d p r 〇 f i 1 e)提拉晶塊(i n g 〇 t)後,從此 晶塊切製而得。 通常,依CZ法從熱區爐內之矽融液提拉矽單晶之晶塊 時,會發生點缺陷(point defect)及點缺陷之凝聚體( -12- $紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) (請先閱讀背面之注意事項再填寫本頁) Φ 訂------«'線· 經濟部智慧財產局員工消費合作社印製 經濟部智慧財產局員工消費合作社印制π 528816 A7 B7 五、發明說明(10 ) agglomerates:三維缺陷)等砂單晶之缺陷。點缺陷有兩 種一般的形態,即空孔型缺陷(vacancy point defect)及 格子間砂型缺陷(interstitial point defect)。空孔型 缺陷為有一個矽原子從矽结晶格子之正常位置脫離,此種 空孔導致空孔型缺陷。另外,在结晶格子點K外之位置( interstitial site)有矽原子存在時,導致格子間矽型 點缺陷。點缺陷通常形成於矽融液(熔融矽)及晶塊(固體 矽)之接觸面。但,將晶塊繼續提拉(pull up)時,其接 觸面部分於提拉之同時會開始冷卻。在冷卻期間,空孔型 點缺陷或格子間矽型點缺陷會因擴散互相合併形成空孔型 點缺陷之凝聚體或格子間矽型點缺陷之凝聚體。換言之, 凝聚體係點缺陷合併形成之三維構造物。 空孔型點缺陷之凝聚體除上述之COP之外,尚含有 L S T D ( L a s e r S c a 11 e r i n g T 〇 m 〇 g r a p h D e f e c t s )或 F P D ( Flow Pattern Defects)等缺陷,而格子間矽型點缺陷之 凝聚體則含有前逑之L/D缺陷。FPD為將由晶塊(ingot), 切割製取之矽晶圓在無攪拌下施予Secco蝕刻(使用K2 Cr2〇7 :5 (U H F :純水=4 4 g : 2 0 0 0 c c : 1 0 0 0 c c之混合液之蝕刻叫 做S e c c o e t c h i n g }時顯現之特異之流線譜(f 1 o w p a 11 e r η )之痕跡源。L S T D為將紅外線照射於矽單晶内時發生具 有與矽不同之折射率之散射光之根源。528816 V. Description of invention (3) Composition. The complete and non-defective field [P] lies between the field [I] where inter-lattice silicon-type point defects account for the potential and the reading region [V] where void-type point defects are dominant in silicon single crystal blocks. The complete silicon wafer [P] is completed by setting the V / G value (fflffl2 / minutes) so that the pulling speed of the crystal block is VUm / minute, the interface of the silicon melt and the crystal block. When the temperature gradient in the vertical direction of the nearby crystal block is G (υ / mm), 〇SF generated in a circle when the thermal oxidation treatment is performed will be prepared under the condition that the center portion of the wafer disappears. In addition, some semiconductor manufacturers require silicon wafers that do not have OSF, COP, and L / d, and have the performance of de-aeration (Gotte ring) of metal contaminants generated by device engineering (d e v i c e ρ γ o c e s s). When wafers without sufficient oxygen removal performance are contaminated by metal during device engineering, junction leakage or poor operation of the device due to trap levels of metal impurities can reduce the product's performance. Yield. Therefore, in order to solve this problem, a silicon wafer that can exert an intrinsic gettering (referred to as I G) effect during the heat treatment of the Cg is generally required. However, the 0 wafers cut out from the ingots formed by the complete field [P] of the upper part, although they do not have OSF, COP, and L / D, will not necessarily be on the wafer during the device manufacturing process &gt; heat treatment. Oxygen evolution occurs in the electrode, and as a result, there is a disadvantage that a sufficient IG effect cannot be obtained. Traditionally, the heat treatment method in device engineering for the wafer to exert the IG effect is a method of forming defects or intentionally adding impurities in the wafer. Sand wafers processed in this way can absorb contamination from subsequent processes around the pre-formed defects on the wafer. Therefore, it is possible to prevent the installation of 6-this paper size is applicable to the Chinese National Standard (CNS) A4 specification (210x 297 mm) (Please read the precautions on the back before filling this page) -------- -tr ---------- line ·. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 Β7 V. Description of the invention (4) Wafer surface Defects or contamination in nearby areas. In addition, in recent years, due to the high concentration of the device (increasing the packaging density), the heat treatment temperature of the device project has a tendency to be lower than 10 Q 0 ° C. Therefore, in conjunction with this low temperature, the IG treatment of the previous project is also required to be low temperature. Into. For this reason, Japanese Patent Application Laid-Open No. 8-4 5 9 4 5 discloses a method including: cutting a silicon single crystal block and polishing and polishing the silicon wafer at a temperature of 500-800 ° C. 0.5-2G hours to introduce a precipitation nucli of oxygen into the wafer; this silicon wafer containing oxygen precipitation nuclei is rapidly heated from room temperature to 8 G (3-1 0 0 0 ° C and The process of maintaining 0.5-2 G minutes; the process of cooling the rapidly heated silicon wafer to room temperature for 0.5-20 minutes; and the process of cooling the cooled silicon wafer from 50 0 ϋ -7 fl Q ° CK 2-1 Q / min. Heating to 80 0 1 1 0 0 υ and keeping at this temperature for 2-4 8 hours of IG processing method. According to this processing method, when Κ When the heating temperature is rapidly heated, not only the surface of the wafer, but also the interior of the wafer will temporarily change to a thermal equilibrium concentration M, which will cause the silicon atoms in the lattice to become deficient, thereby providing an environment where the oxygen precipitation nuclei can grow easily and stably. In order to make up for the lack of silicon atoms between the grids and become stable, the generation of silicon atoms between the grids occurs on the surface of the wafer, and makes the The inter-lattice silicon atoms diffuse into the wafer. Therefore, near the surface of the wafer in which the inter-lattice silicon atoms are deficient, the generation of inter-lattice silicon atoms becomes saturated, and the oxygen nucleation begins to disappear. As silicon atoms generated between the lattices on the wafer surface diffuse into the interior, it takes a while, so the farther away from the interior of the wafer surface, the longer the environment where the oxygen can grow out of the nucleus and grow longer. Therefore, the closer to the wafer surface, the more oxygen Density of precipitation core-7-This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm) 11 ^ ----- ^ --- · ^ -------- ^ ·- ------- I 、 (Please read the notes on the back before filling this page) Printed by the Intellectual Property Bureau Employee Consumer Cooperative of the Ministry of Economic Affairs 528816 A7 __________ B7 V. The invention description (5) the lower the heat treatment time ( 0.5-20 minutes), the longer the oxygen precipitation nucleus, that is, the thickness of the denuded zone (DZ for short), the greater the thickness. In addition, the temperature is in the range of 800-1000, The larger the diffusion coefficient of the sand atoms between the lattices, the DZ thickness changes in a short time. When heating rapidly and cooling the wafer to 800-11001 after cooling at room temperature, the oxygen precipitation nucleus inside the rapidly heated wafer will grow into oxygen precipitates and become a stable IG source. Oxygen precipitates are called BMD (Bulk Micro Defect). However, the IG treatment method of the upper part requires grinding and polishing of the post-processed sand wafer. It is maintained for 5-20 hours at 50 G-8 0 0 υ. The step of introducing the oxygen precipitation nuclei into the wafer is processed before generating the IG source, and further, after rapid heating, heat treatment is required to grow the oxygen precipitation nuclei inside the wafer into BMD. Therefore, there is a lack of performing multiple heat treatments in the state of the wafer. &lt; Summary of the invention &gt; The first object of the present invention is to provide a wafer which can be concentratedly developed in the center portion even when the conventional 0SF flaring heat treatment (OSF_fflanifesting heat treatment) is performed. Heat treatment method for SFK and C 0 P-free silicon wafers. A second object of the present invention is to provide a polycrystalline silicon layer having a uniform degassing effect (gettering effect) between the wafer peripheral edge portion and the wafer center portion in the full implementation of uniform oxygen precipitation. Silicon wafer manufacturing method. The third object of the present invention is to provide a concentration of 0. 8 X 1 0 1 8 to 1. 4 X 1 0 1 8 -8- even if it is composed of a mixed field of a field [P v] and a field [P 1]. Paper size applies to Chinese National Standard (CNS) A4 (210 X 297 g) (Please read the precautions on the back before filling out this page) Installation ------ Order ------...! 528816 A7 B7 Ministry of Economic Affairs Printed by the Intellectual Property Bureau's Consumer Cooperatives. 5. Silicon wafers cut out from the crystal block of the invention description (6 atoms / cm 3 (old AS T M)). They do not have point defect aggregates and can fully generate oxygen precipitation nuclei. A heat treatment method for a silicon wafer that can exert the IG effect through the heat treatment of a device manufacturing process. A fourth object of the present invention is to provide a heat treatment method for a silicon wafer that does not require an oxygen donor killer treatment process. A fifth object of the present invention is to provide a heat treatment method for a silicon wafer that can reduce the number of heat treatments in the state of a silicon wafer and can exhibit a high IG effect by heat treatment at 950 ° C Μ. 6 The purpose is to provide a method, which can be obtained by the processing method can play a high IG Fruity silicon wafer. The seventh object of the present invention is to provide a single crystal ingot that can produce such a silicon wafer that exhibits a high IG effect. <Means for achieving the object of the invention> The first aspect of the present invention is Provided is a chemical vapor deposition (hereinafter referred to as CVD) method, which does not cause oxygen concentration i.2xl in the silicon wafer surface due to crystalline particles (COP) and intrusive indexing (L / D). a step of forming a polycrystalline silicon layer with a thickness of 0.1-1.6 // ιη at a temperature of 67 ° ± 3 ° at the back of a silicon wafer below iaat〇ffis / cm3 (old ASTM); and A silicon wafer with a polycrystalline silicon layer is placed in an oxygen atmosphere, and then heat-treated at a temperature of 0 ° C to 3 ° C for 2-5 hours, and then implemented at a temperature of U3 (rc ± 3 (rc) for 16 hours. The heat treatment method of the silicon wafer in the processing step is characterized in that the silicon wafer before forming the polycrystalline silicon layer undergoes a heat treatment at the center of the wafer and causes an accumulation of oxidation defects (0 s F) at the center of the wafer. -9- The paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 public love) — · --- l · --------- ------ (please first Read the notes on the back and fill in this page again) 528816 Α7 Β7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. The description of the invention (7). The second aspect of the present invention is to provide a cause that does not occur in the silicon wafer surface. A silicon wafer having a thickness of 0 · 1-1.6u is formed on the back surface of the silicon wafer at a crystal concentration of COP and invasive translocation (L / D) at an oxygen concentration of 1.2 X 10 1 8 atoms / cm3 M (former ASTM). Silicon wafers with a polycrystalline silicon layer with a thickness of m are characterized in that the silicon wafer before the formation of the upper polycrystalline silicon layer is placed in an oxygen atmosphere at a temperature of 10 QG ° C to 3 Q ° C After the heat treatment is performed for 2 to 5 hours, and then the heat treatment is performed at a temperature of 113 ο υ for 1 to 16 hours, oxidation will occur at the center of the wafer to cause the accumulation of defects (0SF). The silicon wafers according to the first and second aspects of the present invention are wafers prepared by the Chuklaski (CZ) method under the condition that 0SF appears at the center of the silicon wafer. Part of the tritium does not have an oxygen precipitation nucleus. At the same time, the part outside the central part K does not contain C 0 P. When a polycrystalline silicon layer is formed on the back surface of such a silicon wafer, a BMD will be formed on the wafer in its entirety during the CVD process. As a result, uniform oxygen precipitation can be performed throughout the wafer, and the uniform I G effect without any deviation between the center portion of the wafer and other parts can be exerted. According to a third aspect of the present invention, the area where the inter-lattice silicon-type point defects are dominant in the silicon single crystal crystal is "I", the area where the void-type point defects are dominant is [V], and the inter-lattice silicon is The non-defective areas where the aggregates of type point defects and the aggregates of void type point defects do not exist at all are regarded as "P". The area with the lowest inter-lattice silicon concentration is taken as [P Π, and the area under the pore density M that can form COP or FPD adjacent to the upper area "V" and belongs to the above-mentioned non-defective area [P] is taken as [P v] At the time, there is no defect from the top -10- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) (Please read the precautions on the back before filling this page) ---- Order --- ------ Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 ___________ B7 V. Description of the invention (8) Silicon wafers with point-free defect aggregates cut out of the crystals formed in the field [P] The heat treatment method. The characteristic point of this heat treatment method is to pull up a silicon single crystal block composed of a mixed field of the above-mentioned fields [P v] and [P i] and having an acid concentration of 0.8X 1018 ~ 1.4X c π3 (former AST Μ), The silicon wafer cut out from the upper crystal block is placed in an atmosphere of nitrogen, argon, hydrogen, oxygen, or a mixed gas thereof, and is held for 30-90 minutes at 6000-8501¾. The fourth aspect of the present invention is to form a silicon single crystal having a denier oxygen concentration of 0.8X HP8 ~ 1.3X l0i8at〇IRS / c πι3 (formerly AST M) from the mixed field of the above-mentioned fields [Pv] and [Pl]. The silicon wafer cut out from the upper crystal block is placed in an atmosphere of nitrogen, argon, hydrogen, oxygen, or a mixed gas thereof, and maintained at 60 2 0-8 5 0 υ 1 2 0-2 5 0 minute. A fifth aspect of the present invention is a silicon single crystal having an oxygen concentration of 0.8X 1018 to 1.4X l (] i8atc) nis / cm3 (former ASTM) composed of a mixed field of the upper field [Pν] and the field [Pl]. The wafer is pulled up, and the Shi Xi wafer cut out of the upper crystal block is placed in an atmosphere of nitrogen, argon, hydrogen, oxygen or a mixed gas thereof, and the temperature rising rate of M 1 15 0 t / s is from 1 1 5 0 υ is heated to 1 2 0 0 ° C, and maintained at Q-3 Q seconds at 1 150-12 G. According to the method of the third to fifth aspects of the present invention, even if the oxygen inclination of the crystal block is 0.8 × 1018 ~ 1.4xi018atoms / cin3 (former ASTM), the wafer of Shi Xiqin is composed of the field [Pv] and the field [P]. When the mixed field is constituted, the silicon wafer cut out from the crystal block is subjected to heat treatment under the conditions of $ uploading, that is, during the growth of the crust, the field [Pi] where no oxygen precipitation nucleus is introduced can show the oxygen precipitation nucleus, and at the same time crystallize In the field of growth, there are areas where oxygen nucleation is introduced. [P v] Obvious oxygen leaching _ 1 1 _ This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) r L ------- ---------- ^: ί ------ Aw (Please read the notes on the back before filling this page) 528816 A7 B7 V. Description of the invention (9) Increase in the core density. Therefore, in the semiconductor device manufacturer's device manufacturing process, when the wafer subjected to the above-mentioned heat treatment is heat-treated, the above-mentioned oxygen precipitation nuclei will grow into BMD, so even if it is composed of a mixed field of the field [Pv] and the field [PJ The wafer can also make the wafer fully have the IG effect. A sixth aspect of the present invention is to provide a method including: a step of pulling a silicon wafer from a silicon melt, a step of manufacturing a silicon wafer from the silicon wafer, and a heating rate of the silicon wafer from room temperature to K. Heat treatment method for silicon wafers with rapid heating to 650-950 ° C and maintaining at that temperature for 5-30 minutes, which is characterized in that the silicon wafers will OSC occurs at 25¾ K and contains BMD IX 10s ~ 3x 107 pcs / cib30 without transposition. The method of the sixth aspect of the present invention is to use a crystal containing BMD in the OS F field and the density mentioned above. When round, you can omit the heat treatment step before introducing oxygen precipitation nuclei into the wafer and the growth steps to grow the oxygen precipitation nuclei into BMD. The wafers that have been cut and polished from the crystal block can be rapidly heated according to the above conditions. Can exert high IG effect. This invention will be described in detail according to the following embodiments: A. First embodiment The silicon wafers according to the first to fourth embodiments of the present invention are based on the CZ method, from the silicon melt in the hot zone furnace, according to Voronkov theory. A predetermined pulling speed profile (pu 11 ing-upspeedpr 0fi 1 e) is obtained by pulling a crystal block (ing 〇t) and then cutting from the crystal block. Generally, when the silicon single crystal ingot is pulled from the silicon melt in the hot zone furnace according to the CZ method, point defects and aggregates of point defects (-12- $ paper size are applicable to Chinese national standards ( CNS) A4 size (210 X 297 mm) (Please read the notes on the back before filling out this page) Φ Order ------ «'Line · Intellectual Property Bureau of the Ministry of Economic Affairs Consumer Cooperatives Print Intellectual Property of the Ministry of Economic Affairs Printed by the Bureau ’s Consumer Cooperative π 528816 A7 B7 V. Description of the invention (10) Agglomerates: defects in sand single crystals. There are two general forms of point defects, namely, vacancy point defects and interstitial point defects. Void-type defects are those in which a silicon atom is detached from the normal position of a silicon crystal lattice. Such voids cause a void-type defect. In addition, when silicon atoms are present at an interstitial site outside the crystal lattice point K, silicon-type point defects between the lattices are caused. Point defects are usually formed at the contact surfaces of molten silicon (fused silicon) and crystal blocks (solid silicon). However, when the crystal block is continuously pulled up, the contact surface portion of the crystal block will start to cool while being pulled up. During cooling, void-type point defects or inter-lattice silicon-type point defects will merge with each other due to diffusion to form aggregates of void-type point defects or inter-lattice silicon-type point defects. In other words, the three-dimensional structure formed by the aggregation of point defects in the aggregation system. In addition to the above-mentioned COP, the aggregates of hole-type point defects also contain defects such as LSTD (Laser S ca 11 ering T graph graphs) or FPD (Flow Pattern Defects). The agglomerates contain the previous L / D defects. FPD is a Secco etching of silicon wafers prepared from ingots and dicing (using K2 Cr2 07: 5 (UHF: pure water = 4 4 g: 2 0 0 0 cc: 1 0 The etching source of 0 0 cc mixed solution is called S eccoetching}, which is a trace source of a unique streamline spectrum (f 1 owpa 11 er η). LSTD is a kind of refraction that is different from silicon when infrared rays are irradiated into a silicon single crystal. The source of the scattered light rate.

Voronkov之理論係藉控制V/G比率(C° /lain)而使缺 陷少之高純度晶塊成長之理論,其中VUni/分)為晶塊之 提拉速度,G為熱區搆造(hot zone structure)之晶塊 - 13- 本紙張尺度適用中國國家標準(CNS)A4規格(21ϋ X 297公釐) 一 IJ ^----^ --------- ^ · I.-------1 ^ •(請先閱讀背面之注意事項再填寫本頁)Voronkov's theory is based on controlling the V / G ratio (C ° / lain) to grow high-purity crystal blocks with few defects, where VUni / min is the pulling speed of the crystal blocks, and G is the hot zone structure (hot zone structure)-13- This paper size is in accordance with China National Standard (CNS) A4 (21ϋ X 297mm)-IJ ^ ---- ^ --------- ^ · I.- ------ 1 ^ • (Please read the notes on the back before filling this page)

經濟部智慧財產局員工消費合作社印製 與矽融液之界(接觸)面之溫度梯度。依此理論時,如第1 圖所示,KV/G作為横軸、空孔型點缺陷濃度與格子間矽 型點缺陷濃度作為縱軸而以圖式來表示V/G與點缺陷濃度 的闞係K說明空孔領域與格子間矽領域之境界係由V/G決 定。更具體的說,V/G比在臨界點Μ上時形成空孔型點缺 陷濃度佔優勢之晶塊,反之V/G比率在臨界點Κ下時形成 格子間矽型點缺陷濃度佔優勢之晶塊。 本發明第1實施形態之設定的提拉速度輪鄹(pulling-up speed profile) 係如此決定 ,使從熱區爐 (hot zone furnace)内之矽融液提拉晶塊時,提拉速度與溫度梯度 之比(V/G)嵩於防止格子間矽型點缺陷之凝聚體發生之第 一臨界比((V/Gh)且使空孔型點缺陷之凝聚體維持於第2 臨界比((V/G)2) K下,並限制於空孔型點缺陷佔儍勢之 晶塊之中央領域內。 此提拉速度輪廓係藉模擬法,根據Voronkov理論將 基準晶塊(reference ingot)軸方向實驗的切割或將基準 晶塊實驗的切割成晶圓,或將此等技術組合而決定。即, 此決定傣在模擬(simulation)之後藉確認晶塊之軸方向 切割及切割之晶圓而實行,然後反覆實行模擬。為模擬, 在所定之範圍内決定多種之提拉速度,並使多數個之基準 晶塊成長。如第2圖所示,將模擬用之提拉速度輪廓從例 如1·2βπι/分之高提拉速度(a)調蝥至例如分之低提 拉速度(c),再調整至較高提拉速度U)。上逑之低提拉速 度亦可為〇·4βιπι /分或Μ下,且提拉速度(b)及(d)之變化 -1 4 - 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公爱) iru-----------------^訂 ί------^線φ * (請先閱讀背面之注意事項再填寫本頁) 528816 經濟部智慧財產局員工消費合作社印製 A7 B7 五、發明說明(I2 ) 最好為線形(linear)。 將不同速度提拉之多數個的基準晶塊分別沿軸方向切 割。最佳之V/G比係由軸方向之切割,晶圓之確認及模擬 之结果的相關關係決定,然後決定最佳之提拉速度輪廓, 並根據此輪廓製造晶塊。實際之提拉速度輪廓係取決於所 期之晶塊的直徑、使用之特定之熱區爐及矽融液的品質等 K及包含不限定此等之多種變數。 將徐徐的降低提拉速度K連績降低V/G時之晶塊之斷 面圖描繪便可知道第3圖所示之事實。在第3圖中,[V】 表示晶塊中空孔型點缺陷佔優勢且有空孔型點缺陷之凝聚 體存在之領域,[I】表示格子間型點缺陷佔優勢且有格子 間矽型點缺陷之凝聚體存在之領域,[P】表示空孔型點缺 陷之凝聚體K及格子間型點缺陷之凝聚均不存在之完美( 理想)領域。 另外,COP或L/D等之點缺陷之凝聚體有因檢測方法 有時在檢測敏感度及檢測下限值上顯示不同之值。為此, 在本說明書中「點缺陷之凝聚體不存在J 一詞是對鏡面加 工後之矽單晶施K無攪拌Secco蝕刻後,利用光學顯微鏡 K觀察面積與蝕刻留量(etching allowance)之積作為檢 査體積(testing volume)加K觀察時,在lxi〇-3Cffl3之 檢査體積中有流動模樣(空孔型缺陷)及轉位簇(格子間矽 型點缺陷)之各凝聚體被檢测到一個之場合作為檢測下限 值(1X103個/cm3)時,稱為點缺陷之凝聚體之數在上逑 檢測下限值K下。 -15- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) —r IW------------------訂 --------&quot;•線 *(請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7 五、發明說明(13 ) 如第3圖所示,晶塊的軸方向位置Pi包含中央有空孔 型點缺陷佔優勢之領域。位置P2比位置h包含中央有小空 孔型點缺陷佔優勢之領域。位置P4包含格子間矽型點缺陷 佔優勢之環形領域及中央之完美領域。又,位置p3為在中 央既無空孔型點缺陷,在緣部分亦無格子間矽型點缺陷之 完美領域。 與此空孔型點缺陷佔優勢領域中存在之完美領域相鄰 接之極小領域是在晶圓面內無發生任何COP或L/D之領域 。但對此種矽晶圓依傳統之0SF顯化熱處理法,在氧氣氛 下ΐοοου 土 30¾溫度實行2-5小時之熱處理之後,接著在 1 1 3 0 υ 士 3 0 υ實行卜1 6小時之熱處理,則能發生〇 S F。即 如第4圖所示,在晶圓^之半徑之1/2附近處發生0SP環 圈。由此0SF環圈圍繞之空孔型點缺陷佔優勢領域有C0P 發生之傾向。對此,在晶圓^中,0SF不成環圈狀且僅發 生於晶圓之中心部。使用於第1實施例之矽晶圓即為此種 晶圓W2。此矽晶圓W2由第5圖可知,0SF非環圈狀,而是 僅發生於晶圓之中心部。此種晶圓W 2係從在〇 S F僅可發生 於中心部之一選定之提拉速度輪廓成長之晶塊切割製作。 第6圖為其平面圖。此晶圓^由於0SF不形成環圈狀,因 此不含任何C0P,同時亦不發生侵入型轉位(interstitial dislocation) 〇 對第1實Ife形態之矽晶圓,需將其晶圓內之氧濃度進 一步控制。依CZ法時,藉改變供給於熱區爐內之氫流量、 貯存熔融物之石英坩堝之回轉速度及熱區爐內之壓力等來 -1 6 - 本,紙張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐) (請先閱讀背面之注意事項再填寫本頁) \ --------訂 P--^------線 1^----- 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明說明(14 ) (請先閱讀背面之注意事項再填寫本頁) 控制晶圓内之氧濃度。晶圓内部之氧濃度需控制於1 · 2 X 1018atoms/cia3M下(舊ASTM)。為達成此氧濃度,例如需 控制氫之流量於18-150公升/分、貯存矽熔融物之石英坩 堝之回轉速度於4-9rpm,及熱區爐內之壓力於15-60氣壓 。將第1實施形態之矽晶圈之氧濃度控制於1·2Χ 10^β a toms/cm3 Κ下(舊AS ΤΜ)係為防止氧析出核之過多析出。 於依上逑條件提拉之晶塊切取之矽晶圓表面依CVD法 ,例如使用SiH4在670¾ 土 30¾之溫度形成厚度0.1-1.6 mi之多晶矽層(polysilicon layer)。多晶矽層之厚度在 O.luinM下時,效果不佳,超過時,降低生產性,, 因此可取之厚度為1.0-1.6/iia範圍。多晶矽層形成前,即 使晶圓面內之氧濃度均勻,易會在晶圓中心部發生氧析出 ,而其他部分不易發生氧析出者,藉由多晶矽層之形成得 使晶圓面内之氧析出均勻化。 因此,在半導體裝置工程中熱處理上述之具有多晶矽 層之矽晶圓時,即使在晶圓内有氧析出核存在,由於該核 不會成長,故依傳統實施0SF顯化之熱處理(OSF-manifesting h e a t t r e a t πι e n t)亦不發生 0 S F。 [B]本發明之第2實施形態 經濟部智慧財產局員工消費合作社印制衣 本發明之第2實施彤態係與第1實施形態同樣,依 Voronkov理論從砂融液提拉砂晶塊。本案實施例之所定之 提拉速度輪廓乃如第7圖所示,將其決定使在從熱區爐內 之矽融疲提拉矽晶塊時,提拉速度與溫度梯度之比(V/G) 維持於可防止發生格子間矽型點缺陷之凝聚體之第3臨界 -17- 本紙張尺度適用中國國家標準(CNS)A‘丨規格(210 χ 297公笼) 528816 A7 ___B7 _ 五、發明說明(15 ) (請先閱讀背面之注意事項再填寫本頁) tb ( ( V / G) 3 ) Μ上及將空孔型點缺陷之凝聚體限制於晶塊之 中央之空孔型點缺陷佔優勢之領域內之第4臨界比((V/G)4) 以下。第7圖中,[I]表示格子間矽型點缺陷優勢且格子 間矽型點缺陷存在之領域((V/G)3), [P]表示空孔型點缺 陷之凝聚體及格子間矽型缺陷之凝聚體不存在之完美領域 ((V/G)3)〜((V/GW)。鄰接於領域[P]之領域[V]有形成 OSF 核之領域[〇SF]((V/G)4)〜((V/G)5)存在。 此完美領域[P]進一步分類成為領域[Pd及領域[Pv] ♦ [Pi]為V/G比從上述(▽”。至臨界點之領域,而[Pv] 為V/G比從臨界點至上逑之(V/G)2之領域。即[Pd為鄰 接於領域[I]且具有可形成侵入型轉位之最低之格子間矽 型點缺陷濃度以下之格子間矽型點缺陷濃度之領域,而 [Pv]為鄰接於[v]且具有可形成〇SF之最低之空孔型點缺 陷濃度Μ下之空孔型點缺陷濃度之領域。 經濟部智慧財產局員工消費合作社印製 將提拉速度徐徐降低Κ連绩降低V/G時之晶塊斷面圖 加Κ描製,即可知第8圖所示之事實。在第8圖中分別顯 示晶塊內空孔型點缺陷佔優勢之領域[V]、格子間矽型點 缺陷佔優勢之領域,及空孔型點缺陷之凝聚體及格子間矽 型點缺陷之凝聚不存在之完美領域[Ρ]。如上所述,完美 領域[Pt]進一步分類為領域[Pd及領域[Ρν】,其中領域 [Ρν]為不形成凝聚體而有空孔型點缺陷存在之領域,領域 [Pi]則為不形成凝集體而有格子間矽型點缺陷存在之領 域。 如第8圖所示,晶塊的軸方向位置Pi是在其中央含 -1 8 - ^紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明說明(I6 ) 有空孔型點缺陷佔優勢之領域。位置P4是為有格子間矽型 點缺陷佔優勢之環圈領域及中央之完美領域。又,位置P3 為既無關連第2實施形態之在中央含有空孔型點缺陷之凝 聚體,亦無在緣部含有格子間矽型點缺陷之凝聚體,故完 全是完美領域。 ’ 又由第8圖可知,對應於位置Pi之晶圓Wi,為其中央 含有空孔型點缺陷佔優勢之領域。對應於位置P4之晶圓W4 為含有格子間矽型點缺陷佔優勢之環圈及中央之完美領域 。又,對應於位置p3之晶圓Wi為第2實施形態之晶圓,是 在其中央無空孔型點缺陷之凝聚體且在緣部亦無格子間矽 型缺陷之凝聚體,故完全是完美領域,即領域[Pv】及領 域[PU混在之領域。與此空孔型點缺陷佔優勢之領域之 完全領域相接觸之極小領域(第7圖之(V/G)4-(V/G)5)為 在晶圓面內無發生COP及LD之領域。但,對此種矽圓^ ,依習知之0SF顯化熱處理,在氧氣氛下、ΚΙΟΟΟυ 士 30 C之溫度熱處理2 - 5小時後,接著Κ 11 3 0它土 3 0 1C之溫度 熱處理卜16小時時,會生成0SF。如第1實施形態之第4 圖所示,在晶圓h之晶圓半徑之1/2附近發生0SF,由此 OSF環圈包圍之空孔型點缺陷佔優勢之領域有COP出規之 傾向。 第2實施形態之矽晶為上逑之晶圓W3,其平面圖示於 第9圖。晶圓W3 ,係擬依第2實施形態之熱處理,在晶圓W3 上形成所期密度K上之氧析出核,故其氧濃度須維持0·8 X 1018 〜1.4X 1018atoms/cia3 (舊 ASTM)。 -19- 本紙張尺度適用中國國家標準(CNS)A4規格(2〗(jx 297公釐) I I Γ --------^ · L. I I----- (請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7 五、發明說明(1*7) 次說明上逑矽之晶圓W3之熱處理於下: 此熱處理係將晶圓W3置於氮、氩、氫或氧或此等之混 合氣體的氣氛下,在6 0 0 -8 5 0 ¾保持3G-9Q分,或在6 0 0 -850¾保持120-250分而實施。加熱最好是以50-100¾ /分 之速度將晶圓導入保持於600-8501之熱處理爐中實施。 保持溫度在δϋουκ下,或保持時間30分Μ下時,氧析出 核之增加(成長)不足,於半導體裝置之製造廠商之裝置製 造工程實施處理時,無法獲得發揮IG效果所需之BMD密度 。又,保持溫度越過850Ό時,由於領域[Pi]之氧析出核 密度低,故在裝置製造工程實施熱處理時,無法獲得可發 揮IG效果所需之BMD密度。另外,保持溫度在600-850¾ 而保持溫度超過9 0分且120分K下時,由於伴隨氧析出核 形成之格子間型點缺陷之過多,發生氧析出核之析出量之 抑制。保持時間在250分K上時,生產性會降低。 上述之熱處理條件係包含於依CVD法在晶圓之背面形 成聚矽層時之熱處理條件(即,保持溫度6 5 Q υ 土 3 G °C,保 持時間5 - 3 0 0分)。因此,依CVD法在晶圓之背面形成多 晶矽層時,實施上述熱處理條件即可藉由多晶矽層之形成 達成本發明第2實施形態之目的。此時之多晶矽層之厚度 為0.1-2.0/iiu。晶圓之背面形成有多晶矽層之場合,於接 觸於多晶矽層之晶圓的背面附近形成更多量之氧析出核。 此形態之晶圓,其多晶矽層可原狀保留,或亦可浸漬於氟 酸及硝酸之混酸K水或乙酸稀釋之酸蝕刻液,或Κ0Η或 NaOH水稀釋之鹼蝕刻液中而去除多晶矽層。 一 2 0 - 本纸張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐) --------*— l· — (請先閱讀背面之注意事項再填寫本頁) 訂L-----!線| 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明說明(18 ) 實施上逑之熱處理時,可省略晶圓製程中之氧施主去 除處理步驟(oxygen donor killer treatment)。 C·本發明之第3實施形態 本發明之第3實施形態係與第丨實施形態同樣,根據 Voronkov理論從矽融液提拉矽晶塊。本發明第3實施形 態之所定之提拉速度輪廓與第2實施形態者相同。 經濟部智慧財產局員工消費合作社印製 (請先閱讀背面之注意事項再填寫本頁) 第3實施形態之矽晶圓係如第8及9圖所示之晶圓W3 。此晶圓W3依第3實施形態之熱處理使在其上發生所期密 度以上之氧析出核時,必須具有氧湄度須維持〇.8xl(P8 〜1·4Χ lOUatoms/cmM舊“TM)。此矽晶圓W3之熱處理係 將晶圓W3置於氮、氩、氫、氧或其混合氣體之氣氛中,以 10-15Dt!/秒之昇溫速度從室溫加熱至溫度1150-1200t:後 ,在該溫度範圍下保持0-30秒而實施。即,第3實施形 態之熱處理為急速加熱。在此所諝保持時間”0秒”係指僅 實施昇溫而不保持之意。加熱係將晶圓導入維持室溫之熱 處理爐,若是連續蓮轉之場合則由其餘熱保持數百度之熱 處理爐之內部,而K 10-150 t!/秒,最好K 5 0 - 1 0 0*0/秒之 速度昇溫至1150-12ϋ0ΐ!。昇溫速度l〇t!/秒Μ下時,雖然 氧析出核畲增加,但處理能力低,故不符合實用。又, 115Q°CK下時氧析出核之增加不充分,在半導體裝置製造 商之裝置製程中實施熱處理時,不能獲得發揮IG效果所必 要之BMD密度。又,保持溫度超過1200¾,或保持時間超 過30秒時,會發生例如滑動(slippage)或降低熱處理之 生產性等問題。又,昇溫速度超過1 5 G °c /秒,則會因自重 -21- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 528816 A7 B7 五 經濟部智慧財產局員工消費合作社印製 、發明說明(I9 ) 應力或面內溫度分布之偏差而發生滑動之問題。 依上述之熱處埋時可Μ省卻(減免)晶圓製程中之氧施 主去除處理步驟。 D·本發明之第4實施形態 本發明之第4實施形態係與第1實施形態同樣,根據 Voronkov理論從矽融液提拉矽晶塊。本發明第4實施形 態之所定之提拉速度輪廓與第1實施形態者相同。玆依對 應於第3圖之第10圖說明之。第10圖之各符號為對應於第 3圖者。此第4實施形態之特徵點為對應於位置P2之晶圓 W2相較於晶圓Wi,在其中央含有晶圓總面積之1/2面積( 50%)之空孔型點缺陷佔優點之領域。 與此空孔型點缺佔優勢之領域之完美領域相鄰接之小 領域係在晶圓面內無發生C0P及L/D之領域。但對此矽晶 圓,依傳統之0SF顯©化熱處理(〇SF manifesting heat treatment),在氧氣氛下KlflOOt: 土 30Ό之溫度實施2-5 小時之熱處理,繼之在11 3 Q C 土 3 D υ之溫度實施1 - 1 6小 時之熱處理時會發生0 S F。即,如第11圖所示,在晶圓W2 之周緣附近發生0SF環圈。被此〇SF環圈包圍之空孔型點 缺陷佔優勢領域有出現C0P之傾向。對此,在晶圓》2上 0SF不形成環圈狀,而在晶圓之中心部磲狀(disk shape) 的發生。使用於第4實施形態之矽晶圓為此種晶圓W2,於 晶圓之總面積之25¾ K上發生〇SF。0SF在總面積之25%以 下時,BMD之發生領域狹窄,不能充分發揮ig效果,因此 最好為5 0 - 8 0 %。 -22- ,本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) -βι ίφ ----I---I--- — 111 —--^ · 1 I I------- (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明說明(20) 第4實施形態之矽晶圓W2,如第12圖所示,其〇SP非 呈環圈狀而係由依所定之提拉速度輪廓選定在中心部顗現 成長之晶塊切製者,其平面圖如第13圖所示。由於此矽晶 圓W2$〇SF非環圈狀,故不含有C0P,同時亦不發生LD(侵 入型轉位)。製作本發明之矽晶圓之晶塊係以IX 〜 3Χ107個/ cm3之比率含有不發生轉位之BMD。因此,不必 依特開平8 - 4 5 9 4 5号公報所逑,在急速加熱前,K晶圓之 狀態實施500-8001之較低溫加熱並保持0.5-20小時,而 於晶圓內高密度的導入氧析出核。BMD密度在IX 1〇5個/ cid3M下時,K晶圓狀態實施急速加熱時難獲得充分之IG 效果。又3X 107個/ cbi3為允許BMD發生於0SF領域之最大 密度。 第4實施形態之加熱方法最好是將含有上逑比率之不 發生轉位之BMD之矽晶圓ίί2,於室溫快速的導入加熱於 650-950¾之爐中。另外,將上逑之矽晶圓W2,於室溫放 置於配設有可發生高熱之燈之高速加熱爐内,利用該燈之 熱射急速的加熱至650-950¾之方法亦可。即,第4實施 形態之熱處理亦為急速加熱。所謂急速加熱係指昇溫速度 ίου/分Μ上,最好3〇υ/分Μ上之熱處理。利用燈光照射 實施急速加熱時,可均勻加熱晶圓,因此比起置於預先加 熱之爐内者晶圓有不易發生彎翹之優點。急速加熱到達之 最終溫度若是6 5 0 t:K下時,不能使晶圓表面附近之BMD完 全消除,故無法充分確保DZ (不形成缺陷之層)。若是超過 950 °C,則在晶圓表面附近之BMD消失前會發生轉位,故 -23- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) ------------i 裝--- (請先閱讀背面之注意事項再填寫本頁) 1T. 528816 A7 B7 五、發明說明(2工) (請先閱讀背面之注意事項再填寫本頁) 亦無法充分確保DZ。為此,最好為80G-9001。又,保持 時間在0 . 5分K下時,使晶圓表面之B M D縮小之時間過短 ,不能確保晶圓表面之BMD充分消除,故難充分確保DZ。 若是超過30分,將使DZ厚度超過要求且影響生產性。為此 ,保持時間需設於0.5-30分,最好10-30分範圍。急速 加熱是在氮氣氣氛中或氧氣氣氛或大氣中實行,最好是在 氮氣氣氛中。 急速加熱後,將矽晶圓放冷至室溫即可自晶圓表面深 達l-100iuffl處形成DZ,因而可獲得比此DZ深之部分具有 BMD密度lx 105〜3X 107個/cm3之IG晶圓。 〈實施例〉 次說明本發明之實施例與比較例於下: 實施例1 經濟部智慧財產局員工消費合作社印製 將晶塊提拉使在晶塊之全長成長對應於第3圖所示位 置之P2之領域。此時為了控制晶塊内之氧濃度,維持氩之 流量為約110公升/分、儲存矽融液之石英坩堝之轉速為 5-10rpm及熱區爐內為約6 0托(torr)。將如此提拉之晶塊 切割所得之矽晶圓研磨、去角及拋光處理後,利用化學蝕 刻處理去除晶圓表面之損傷處及依CVD法,使用SifU在 680¾在晶圓的背面形成厚度1.5^ in之多晶矽層。然後施 予鏡面研磨獲得直徑8英时、厚度725/iiB之矽晶圓。 比較例1 除了不形成多晶矽層之外,其他悉依實施例1所逑方 法實施,製取矽晶圓作為比較例1。 -24- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 經濟部智慧財產局員工消費合作社印製 528816 A7 B7 五、發明說明(22) 〈比較評價1 &gt; 對於實施例1及比較例1之矽晶圓分別實施仿矽半導 裝置工程之熱處理之第2熱處理。即,於氧氣氣氛下,W 7 Ο Ο υ溫度處理該等晶圓8小時後,在1 0 Q Q Ό溫度下熱處 理12小時。然後利用FT-IR測定實施例1及比較例1之晶 圓中心部至周緣部之晶圓表面之氧濃度。熱處理前後之氧 濃度差△ [〇i]示於第15圖。 如第14及15圖所示,比較例1之熱處理前後之氧濃度 差△ [ 0 i ]從晶圓中心部向外約4 0 ffia部分有很大之變動。 但,與此相較,實施例1之熱處理前後之氧濃度差△[Oi] 從晶圓中心部向外約90niia部分僅徐徐減少,因此K晶圓 全面言大致均勻。 另外,將實施例1之另一矽晶圓及比較例1之另一矽 晶圓同樣置於1 ο ο ο υ溫度下實施4小時之熱處理後,再於 1130¾溫度下熱處理3小時(火成氧化處理,pyrogenic oxidation treatment),然後利用目視撿査有無0SF顯現 ,結果在比較例1之晶圓之中心部發現有白濁化之0SF, 但實施例1之晶圓則在晶圓面內無0SF出現。 實施例2 利用矽單晶提拉裝置提拉摻雜(dope)有直徑8吋之 硼(B)之P型矽晶塊。此晶塊具有直胴部長度1200mm,结 晶方位(100)、電阻率約1GQ cm,氧濃度1·0Χ 1018atoas/ cm3 (舊ASTM)。將提拉時之V/G由0·24®πι2/分υ連續減少 至0 · 181^2/分t,並於此同樣條件培育2條晶塊,將其中 -25- 本紙張尺度適用中0國家標準(CNS)A4規格(210 X 297公釐) -------i---rAW ^-------Γ -------- (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 528816 Α7 _______ Β7 五、發明說明(23) 之1條晶塊,如第8圖所示,沿提拉方向從其中心切斷, 並檢査各領域之位置,另1條則切出對應於第8圖之P3位 置之矽晶圓W3作為試樣。即此實施例之試樣為中心部具有 領域[Pv],其周圍有領域[Pd,而在此領域之周圍進一步 具有領域[Pv]之如第9圖所示之晶圓W3。 將從晶塊切出及鏡面研磨後之晶圓W3置於氮氣氣氛下 ,實施在650Ό保持30分之熱處理。 實施例3 除對實施例2之同一晶塊切出而經過鏡面研磨之晶圓 W3實施溫度650¾、保持時間9 0分之熱處理外,其餘悉依 實施例2實施。 簧施例4 除對鏡面研磨後之晶圓實施溫度65Dt!、保持時間 2 1 〇分之熱處理之外,其餘悉依實施例2實施。 實施例5 除對鏡面研磨後之晶圓W3實施溫度7 5 0 Ό、保持時間 6 0分之熱處理之外,其餘悉依實施例2實施。 實施例6 除對鏡面研磨後之晶圓W3實施溫度7 5 0 Ό、保持時間 9 〇分之熱處理之外,其餘悉依實Ife例2實施。 實施例7 除對鏡面研磨後之晶圓W3實施溫度85 0 Ό、保持時間 30分之熱處理之外,其餘悉依實施例2實施。 實施例8 -2 6 - 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) „-----------1TL-------線 (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 528816 Λ7 B7 五、發明說明(24) 除對鏡面研磨後之晶圓w3實施溫度85〇υ、保持時間 120分之熱處理之外,其餘悉依實施例2實施。 比較例2 對實施例2之同一晶塊切出之鏡面研磨後之晶圓W3不 實施熱處理。 比較例3 除對鏡面研磨後之晶圓W3實施溫度6 5 0 υ、保持時間 100分之熱處理之外,其餘悉依實施例2實施。 比較例4 除對鏡面研磨後之晶圓W3實施溫度750 υ、保持時間 20分之熱處理之外,其餘悉依實施例2實施。 比較例5 除對鏡面研磨後之晶圓W3實施溫度80 Gt、保持時間 100分之熱處理之外,其餘悉依實施例2實施。 〈比較評價2 &gt; 製備實施例2-8及比較例2-5之晶圓W3各4片,在此 等晶圓W3之表面滴下分別含有Fe、Cr、Ni、Cu金屬元素 之溶疲而將該等晶圓之全面分別強制污染Fe、Cr、Ni及Cu 。然後將全體晶圓W3依序在9 0 0 下熱處理2小時,在 1 0 0 0它下熱處理5小時後在8 0 0 °C熱處理1 · 5小時,使各 金鼷元素擴散在晶圓之全體(bulk)。此種污染後之熱處理 為相當於半導體製造商之裝置製程之熱處理。 為確認金屬污染之IG效果,利用蝕刻溶液(secco etching solution)將晶圓蝕刻約2/z in厚度後置於集光燈 -27- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) l·----Μ.-------r ^ ^------ (請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7 五、發明說明(25 ) 下觀察有無混濁(haze)存在。賁施例2-8及比較例2-5 之有無混濁狀況乃示於表1。另外,實施例2之光學顯微 鏡照片示於第16A-16D圖,而比較例2之光學顯微鏡照片 示於第17A-17D圖。第16A圖及第17A圖分別表示受Fe污 染之實施例2及比較例2之晶圓之四分之一。同樣地,第 l6B圖及第17B圖分別表示受Cr污染,第16C圖及第17C 瞻分別受Ni污染,而第16D圖及第17D圖分別受Cu污染之 實施例2及比較例2之晶圓之四分之一。 --裝--- (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 表1 熱處 理條件 混濁之有無 溫度(υ ) 時間(分) 領域[Ρν] 領域[Pi] 實施例2 650 30 無 無 實施例3 650 90 無 無 實施例4 650 210 無 無 實施例5 750 60 無 inr Μ 實施例6 750 90 無 無 實施例7 850 30 無 Μ 實施例8 850 120 無 -ίππ Μ 比較例2 - - inC 热 有 比較例3 650 100 inf 有 fcb較例4 750 20 無 有 比較例5 800 100 無 有 由表1、第16A-16D圖及第17A-17D圖顯示,只有在 比較例2-5之晶圓之領域[Pi]出現混濁,此乃由於比較例 -2 8 - 本纸張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐) 訂-· 528816 A7 ______ B7 五、發明說明(26 ) 2 - 5之熱處理條件時,晶圓之氧析出核密度低,污染後熱 處理不能發揮IG效果所致。實施例2-8之晶圓由於無出現 混濁,領域[Pv】及[Pi]全面有高密度之氧析出核存在, 因而發揮IG效果所致。 實施例9 ’ 使用矽單晶提拉裝置提拉直徑8时之硼(B)摻雜P型 砂晶塊。此晶塊為直胴部之長度1200fflffl、结晶方位(1〇〇) 、電阻率約 10Ω cm、氧濃度 1·〇χ 1018at〇ffls/cni3(舊 ASTM) 。晶塊係將提拉時之V/G由0·24βιβι2/分t連續減少至0.18 mm2/分υ之條件下育成,並於此同一條件培育二條晶塊, 其中之一條如第8圖所示,沿提拉方向切斷晶塊中心而檢 査各領域之位置;另一晶塊則如第8圖所示,將對應於Ρ 2 位置之晶圓W3切出作為試樣。此試樣晶圓W3如第9圖所示 ,在中心部具有領域[Ρν],在其周圍具有領域[ΡΗ,而在 該領域[Pi]之周圍進一步有領域[Ρν]。 將從晶塊切出及鏡面研磨後之晶圓W3置於氮氣氣氛下 ,從室溫以約50 °C/秒之昇溫速度加熱至1150 °C之後不在 此溫度下保持,立刻實行熱處理。 實施例10 除對晶圓W3實施熱處理時在1150¾之溫速度下保持5 秒之外,其餘悉依實施例9實施。 實施例11 除對晶圓W3實施熱處理時,在1150Ό之溫速度下保持 3 0秒之外,其餘悉依實施例9實施。 - 29- 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) (請先閱讀背面之注意事項再填寫本頁) ;裝 t.— ijm 一: 口、f ϋ· 1 an n i··— ·1 . 經濟部智慧財產局員工消費合作社印製 528816 Α7 _ _____ Β7 五、發明說明(27 ) m 12 除對晶圓W3實施熱處理之溫度改為1 2 0 0 ¾之外,其餘 悉依實施例9實施。 m 13 除對晶圓W3實施熱處理之溫度改為ΐ2〇〇υ及在此溫度 下保持5秒之外,其餘悉依實施例9實施。 實施例1 4 除對晶圓W3實施熱處理之溫度改為12()〇t!及在此溫度 下保持3Q秒之外,其餘悉依實施例9實施。 比較例6 對實廊例2之同一晶塊切出之鏡面研磨後之晶圓W3不 實施熱處理。 比較例7 除對晶圓W3實施熱處理時,在uoot!之溫度下保持5 秒之外,其餘悉依實施例9實施。 比較例8 除對晶圓W3實施熱處理時,在1100¾之溫度下保持30 秒之外,其餘悉依實施例9實施。 比較例9 除對晶圓W3實施熱處理時,在llOQt:之溫度下保持60 秒之外,其餘悉依實施例9實施。 比較例1〇 除對晶圓W3實施熱處理時,在11501C之溫度下保持60 秒之外,其餘悉依實施例9實施。 * -30- 本紙張尺度適用中國國家標準(CNS)A4規格(210 χ 297公釐) (請先閱讀背面之注意事項再填寫本頁) 裝 經濟部智慧財產局員工消費合作社印製 528816 A7 ___ ______ B7 五、發明說明(28 )Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs, Consumer Cooperative, and the temperature gradient of the boundary (contact) surface with the silicon melt. According to this theory, as shown in Figure 1, KV / G is used as the horizontal axis, and the void type point defect concentration and the inter-lattice silicon point defect concentration are used as the vertical axis to graphically represent the V / G and point defect concentration. Unit K indicates that the boundary between the void area and the inter-lattice silicon area is determined by V / G. More specifically, when the V / G ratio is at the critical point M, a crystal block with a hole-type point defect concentration is dominant, and when the V / G ratio is at the critical point K, an inter-lattice silicon-type point defect concentration is dominant. crystal. The pull-up speed profile set in the first embodiment of the present invention is determined in such a way that when pulling a crystal block from a silicon melt in a hot zone furnace, the pulling speed and The temperature gradient ratio (V / G) is the first critical ratio ((V / Gh) to prevent the occurrence of inter-lattice silicon-type point defects aggregates and maintains the void-type point defect aggregates at the second critical ratio ( (V / G) 2) K, and is limited to the central area of the silly crystal block with hole-type point defects. This pulling speed profile is based on the Voronkov theory by reference simulation. The cutting of the axial direction experiment or the cutting of the reference crystal block experiment into a wafer, or a combination of these technologies. That is, the decision is made by simulating the axial cutting and cutting of the wafer after the simulation. Then, the simulation is performed repeatedly. For the simulation, a variety of pulling speeds are determined within a predetermined range, and a plurality of reference crystal blocks are grown. As shown in FIG. 2, the pulling speed profile used for the simulation is changed from, for example, 1 · 2βπι / minute high lifting speed (a) adjusted to As mentioned low pull rate per (c), and then adjusted to a higher pulling rate U). The lower pull speed of the upper part can also be 0.4βιπι / min or M, and the change of the pull speed (b) and (d) -1 4-This paper size applies the Chinese National Standard (CNS) A4 specification ( 210 X 297 public love) iru ----------------- ^ Order ί ------ ^ 线 φ * (Please read the precautions on the back before filling in this page) 528816 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 5. The description of the invention (I2) should be linear. A plurality of reference crystal blocks pulled at different speeds are cut along the axial direction, respectively. The optimal V / G ratio is determined by the correlation between the cutting in the axial direction, the confirmation of the wafer and the results of the simulation, and then the optimal pulling speed profile is determined, and a crystal block is manufactured according to this profile. The actual pulling speed profile depends on the diameter of the ingot, the quality of the specific hot zone furnace and the silicon melt used, and K includes a variety of variables that are not limited to these. The fact that the graph shown in Fig. 3 is drawn can be obtained by drawing the sectional view of the crystal block when the K-speed is gradually lowered and the K-speed is reduced by V / G. In Figure 3, [V] represents the area where the hollow hole-type point defect is dominant and the aggregate with the hole-type point defect is present, and [I] indicates that the inter-lattice point defect is dominant and the inter-lattice silicon type is present. In the area where the aggregates of point defects exist, [P] represents the perfect (ideal) area where neither the aggregates K of void point defects nor the aggregation of inter-lattice point defects exist. In addition, there is a causal detection method for point defects such as COP and L / D. The detection sensitivity and the lower detection limit may show different values. For this reason, in this specification, the term "the absence of agglomerates of point defects J" refers to the application of K without stirring to Secco etching of silicon single crystals after mirror processing, using an optical microscope to observe the area and etching allowance. When the product is used as the test volume plus K, the aggregates with flow patterns (void defects) and translocation clusters (silicon point defects between cells) are detected in the inspection volume of lxi0-3Cffl3. When one occasion is used as the lower detection limit (1X103 pcs / cm3), the number of aggregates called point defects is below the upper detection lower limit K. -15- This paper size applies Chinese National Standard (CNS) A4 Specifications (210 X 297 mm) —r IW ------------------ Order -------- &quot; • Wire * (Please read the note on the back first Please fill in this page again for details) 528816 A7 B7 V. Description of the invention (13) As shown in Figure 3, the axial position Pi of the crystal block includes the area where the center-hole type point defect is dominant. The position P2 is greater than the position h. Areas with small hole-type point defects dominate. Position P4 contains the ring-shaped area and center where silicon-type point defects between lattices dominate. Perfect area. Position p3 is the perfect area where there are neither void point defects in the center, nor inter-lattice silicon point defects at the edge. It is adjacent to the perfect area in which the void point defects are dominant. The next smallest area is the area where no COP or L / D occurs within the wafer surface. However, this silicon wafer is subjected to the traditional 0SF explicit heat treatment method, and the temperature is 30οοου in an oxygen atmosphere for 30 to 5 hours at a temperature of 2-5 hours. After the heat treatment, followed by a heat treatment for 16 hours at 1 1 3 0 υ ± 30 0 υ, 0SF can occur. That is, as shown in Figure 4, it occurs near 1/2 of the radius of the wafer ^. 0SP ring. Therefore, the hole type point defect surrounded by the 0SF ring has a tendency of C0P to occur. In this regard, in the wafer ^, 0SF is not ring-shaped and occurs only in the center of the wafer. Use The silicon wafer in the first embodiment is such a wafer W2. As shown in FIG. 5, the silicon wafer W2 is not ring-shaped, but occurs only at the center of the wafer. Such a wafer W 2 series cut from the crystal block that grows at a pulling speed profile that can only occur in one of the center sections Figure 6 is a plan view. Because this wafer does not form a ring shape, 0SF does not contain any C0P, and no interstitial dislocation occurs. For the silicon wafer of the first real Ife form, The oxygen concentration in the wafer is further controlled. According to the CZ method, by changing the flow rate of hydrogen supplied to the furnace in the hot zone, the rotation speed of the quartz crucible storing the melt, and the pressure in the furnace in the hot zone, etc. In this paper, the paper size applies the Chinese National Standard (CNS) A4 specification (210 x 297 mm) (Please read the precautions on the back before filling in this page) \ -------- Order P-^ --- --- Line 1 ^ ----- Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 B7 V. Description of the invention (14) (Please read the precautions on the back before filling this page) Control the oxygen in the wafer concentration. The oxygen concentration inside the wafer needs to be controlled at 1 · 2 X 1018atoms / cia3M (old ASTM). To achieve this oxygen concentration, for example, the flow rate of hydrogen needs to be controlled at 18-150 liters / minute, the rotation speed of the quartz crucible storing the silicon melt is 4-9 rpm, and the pressure in the hot zone furnace is 15-60 atmospheric pressure. The oxygen concentration of the silicon crystal ring of the first embodiment is controlled at 1.2 × 10 ^ β a toms / cm3 κ (former AS TM) to prevent excessive precipitation of oxygen precipitation nuclei. The surface of the silicon wafer cut from the crystal block pulled according to the above conditions is CVD, for example, using SiH4 to form a polysilicon layer with a thickness of 0.1-1.6 mi at a temperature of 670¾ to 30¾. When the thickness of the polycrystalline silicon layer is under O.luinM, the effect is not good. When it exceeds, the productivity is reduced. Therefore, the preferred thickness is in the range of 1.0-1.6 / iia. Before the polycrystalline silicon layer is formed, even if the oxygen concentration in the wafer surface is uniform, oxygen precipitation is likely to occur at the center of the wafer, and the other parts are not prone to oxygen precipitation. The formation of the polycrystalline silicon layer allows the oxygen in the wafer surface to be precipitated. Homogenize. Therefore, when the above-mentioned silicon wafer having a polycrystalline silicon layer is heat-treated in a semiconductor device project, even if there is an oxygen precipitation core in the wafer, since the core does not grow, a conventional OSF-manifesting heat treatment (OSF-manifesting) heat treat ent) does not occur at 0 SF. [B] The second embodiment of the present invention Prints clothes for the consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs The second embodiment of the present invention is the same as the first embodiment, and the sand crystal block is pulled from the sand melt according to Voronkov theory. The pulling speed profile of the embodiment in this case is as shown in Figure 7, which is determined so that when pulling the silicon ingot from the silicon melt in the hot zone furnace, the ratio of the pulling speed to the temperature gradient (V / G) Maintained at the 3rd threshold of aggregates that can prevent the occurrence of inter-lattice silicon-type point defects. -17- This paper size applies the Chinese National Standard (CNS) A '丨 size (210 χ 297 male cage) 528816 A7 ___B7 _ 5. Description of the invention (15) (Please read the precautions on the back before filling out this page) tb ((V / G) 3) and limit the agglomerates of void-type point defects to the void-type points in the center of the crystal block Below the 4th critical ratio ((V / G) 4) in the area where defects are dominant. In Figure 7, [I] represents the advantages of inter-lattice silicon-type point defects and the inter-lattice silicon-type point defects exist ((V / G) 3), and [P] represents the aggregates and inter-lattice-type point defects. The perfect field where the aggregate of silicon-type defects does not exist ((V / G) 3) ~ ((V / GW). The field [V] adjacent to the field [P] has a field [OSF] that forms an OSF core [(OSF] (( V / G) 4) ~ ((V / G) 5) exist. This perfect domain [P] is further classified into domain [Pd and domain [Pv] ♦ [Pi] is the V / G ratio from the above (▽). To The critical point area, and [Pv] is the V / G ratio from the critical point to the upper (V / G) 2 area. That is, [Pd] is the lowest adjacent to the area [I] and has an invasive transposition. The area of inter-lattice silicon-type point defect concentration below the inter-lattice silicon-type point defect concentration, and [Pv] is a void type at the lowest void-type point defect concentration M adjacent to [v] and having the lowest SF that can be formed. The area of the concentration of point defects. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, the pulling speed is slowly reduced, and the cross section of the crystal block is reduced when V and G are reduced. The fact shown in Figure 8 can be known. .In the crystal block shown in Figure 8 Areas where void-type point defects are dominant [V], areas where inter-lattice silicon-type point defects are dominant, and areas where aggregates of void-type point defects and inter-lattice silicon-type point defects do not exist [P] As mentioned above, the perfect field [Pt] is further classified into a field [Pd and a field [Pν], where the field [Pν] is a field that does not form aggregates and has void point defects, and the field [Pi] is not Areas where aggregates are formed and inter-lattice silicon-type point defects are present. As shown in Figure 8, the axial position Pi of the crystal block contains -1 8-^ in the center of the paper. The Chinese standard (CNS) A4 specification applies. (210 X 297 mm) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 B7 V. Description of the invention (I6) The area with hole type dot defects is dominant. Position P4 is the area with silicon grid type point defects. The superior circle area and the center perfect area. In addition, the position P3 is neither agglomerates containing void-type point defects in the center in the second embodiment, nor agglomerates containing silicon-type point defects between cells. Body, so it is a perfect realm. As can be seen from Fig. 8, the wafer Wi corresponding to the position Pi is an area in which a hole type dot defect is dominant in the center. The wafer W4 corresponding to the position P4 is a circle containing silicon point defects between the lattices. And the perfect area in the center. In addition, the wafer Wi corresponding to the position p3 is the wafer of the second embodiment, which is a condensate with no void-type point defects in its center and no inter-lattice silicon-type defects in its edges. The aggregate is completely perfect, that is, the field [Pv] and the field [PU mixed field. The very small field that touches the complete field where the void point defect is dominant (Figure 7 (V / G ) 4- (V / G) 5) is an area where COP and LD do not occur in the wafer surface. However, for this kind of silicon round ^, conventionally known as 0SF explicit heat treatment, heat treatment in an oxygen atmosphere at a temperature of ΚΙΟΟΟυ ± 30 C for 2 to 5 hours, followed by a heat treatment of κ 11 3 0 other soil 3 0 1C. 16 At hours, 0SF is generated. As shown in Fig. 4 of the first embodiment, 0SF occurs near 1/2 of the wafer radius of the wafer h, and thus the area where the hole-type point defects surrounded by the OSF ring dominates tends to be COP. The silicon crystal of the second embodiment is a wafer W3 with a top surface, and its plan view is shown in FIG. Wafer W3 is intended to undergo the heat treatment according to the second embodiment to form an oxygen precipitation nucleus at a desired density K on wafer W3. Therefore, its oxygen concentration must be maintained at 0 · 8 X 1018 to 1.4X 1018 atoms / cia3 (formerly ASTM ). -19- The size of this paper applies to China National Standard (CNS) A4 (2) (jx 297 mm) II Γ -------- ^ · L. I I ----- (Please read the back first Please note this page and fill in this page again) 528816 A7 B7 V. Description of the invention (1 * 7) The heat treatment of the wafer W3 on silicon is as follows: This heat treatment is to place the wafer W3 under nitrogen, argon, hydrogen or oxygen Or in the atmosphere of such mixed gas, it is implemented by maintaining 3G-9Q points at 600-850-5, or 120-250 points at 600-850¾. The heating is preferably 50-100¾ / min. The wafer is introduced into a heat-treating furnace maintained at 600-8501 at a high speed. When the temperature is maintained at δϋουκ or at a holding time of 30 minutes, the increase (growth) of oxygen precipitation nuclei is insufficient, and it is used by the semiconductor device manufacturer's device. During the manufacturing process, the BMD density required for the IG effect cannot be obtained. When the holding temperature exceeds 850 ° F, the density of the oxygen nucleus in the field [Pi] is low, so it cannot be obtained during the heat treatment of the device manufacturing process. BMD density required for IG effect. In addition, keep the temperature at 600-850¾ and keep the temperature over 90 minutes and 120 At K, there are too many inter-lattice-type point defects accompanying the formation of oxygen precipitation nuclei, which inhibits the precipitation of oxygen precipitation nuclei. When the retention time is above 250 minutes K, the productivity decreases. The above heat treatment conditions are included in Heat treatment conditions for forming a polysilicon layer on the back of a wafer by CVD method (ie, holding temperature 6 5 Q υ soil 3 G ° C, holding time 5-300 minutes). Therefore, CVD method When a polycrystalline silicon layer is formed on the back surface, the above-mentioned heat treatment conditions can be used to achieve the purpose of the second embodiment of the present invention by forming the polycrystalline silicon layer. The thickness of the polycrystalline silicon layer at this time is 0.1-2.0 / iiu. The polycrystalline silicon layer is formed on the back surface of the wafer. In this case, a larger amount of oxygen precipitation nuclei are formed near the back surface of the wafer that is in contact with the polycrystalline silicon layer. In this form, the polycrystalline silicon layer can be left as it is, or it can be immersed in mixed acid K water or fluoric acid and nitric acid or Polycrystalline silicon layer is removed from acid etching solution diluted with acetic acid or alkaline etching solution diluted with K0Η or NaOH water.-20-This paper size applies to China National Standard (CNS) A4 (210 x 297 mm) --- ----- * — l · (Please read the precautions on the back before filling this page) Order L -----! Line | Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs's Consumer Cooperatives 528816 A7 B7 V. Description of the invention (18) When the heat treatment of the upper part is performed, The oxygen donor killer treatment step in the wafer process can be omitted. C. Third Embodiment of the Present Invention The third embodiment of the present invention is the same as the first embodiment, and is melted from silicon based on Voronkov theory. Lift the silicon block. The predetermined pulling speed profile of the third embodiment of the present invention is the same as that of the second embodiment. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs (please read the precautions on the back before filling out this page) The silicon wafer of the third embodiment is the wafer W3 shown in Figures 8 and 9. When the wafer W3 is heat-treated in accordance with the third embodiment to cause oxygen nucleation above the desired density to occur thereon, the wafer W3 must have an oxygen saturation of 0.8xl (P8 to 1.4 × lOUatoms / cmM old "TM). The heat treatment of this silicon wafer W3 is to place the wafer W3 in an atmosphere of nitrogen, argon, hydrogen, oxygen, or a mixed gas thereof, and heat it from room temperature to a temperature of 1150-1200 t at a temperature increase rate of 10-15 Dt! / S: after In this temperature range, the temperature is maintained for 0-30 seconds. That is, the heat treatment of the third embodiment is rapid heating. The retention time "0 seconds" here means that the temperature is only increased without being maintained. The heating means The wafer is introduced into a heat treatment furnace that maintains room temperature. In the case of continuous lotus rotation, the inside of the heat treatment furnace that maintains hundreds of degrees by the remaining heat, and K 10-150 t! / S, preferably K 5 0-1 0 0 * 0 The rate of heating per second is 1150-12ϋ0 !. At a heating rate of 10t! / Sec, although the oxygen precipitation nuclei increase, but the processing capacity is low, it is not practical. Also, the oxygen precipitation nuclei at 115Q ° CK The increase is insufficient, and the IG effect cannot be obtained when heat treatment is performed in the device manufacturing process of a semiconductor device manufacturer. The necessary BMD density. In addition, when the holding temperature exceeds 1200¾ or the holding time exceeds 30 seconds, problems such as slippage or reduction in heat treatment productivity may occur. In addition, the heating rate exceeds 15 G ° c / second, Because of its own weight-21- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 528816 A7 B7 Five printed by the Intellectual Property Bureau of the Ministry of Economic Affairs Consumer Cooperatives, the invention statement (I9) stress or in-plane The problem of sliding occurs due to the deviation of the temperature distribution. The oxygen donor removal processing step in the wafer manufacturing process can be eliminated (reduced) during the thermal processing as described above. D. Fourth embodiment of the present invention. Fourth embodiment of the present invention. It is the same as the first embodiment, and the silicon ingot is pulled from the silicon melt based on Voronkov theory. The predetermined pulling speed profile of the fourth embodiment of the present invention is the same as that of the first embodiment. This is illustrated in Fig. 10. Each symbol in Fig. 10 corresponds to that in Fig. 3. The characteristic point of this fourth embodiment is that wafer W2 corresponding to position P2 contains wafers in the center compared to wafer Wi. 1 of the total area / 2 area (50%) area where the hole type point defect has the advantage. The perfect area where the hole type point has the advantage is the small area adjacent to the surface where no C0P and L / The field of D. But for this silicon wafer, according to the traditional 0SF manifestation heat treatment (〇SF manifesting heat treatment), Klf100t under an oxygen atmosphere: heat treatment at a temperature of 30 ° C for 2-5 hours, followed by 11 3 When QC soil 3 D υ is heat-treated for 1 to 16 hours, 0 SF occurs. That is, as shown in FIG. 11, a 0SF ring occurs near the periphery of the wafer W2. C0P tends to appear in the hole-type point defects surrounded by this 0SF ring. In response, 0SF does not form a ring shape on the wafer 2 but a disk shape occurs at the center of the wafer. The silicon wafer used in the fourth embodiment is such a wafer W2, and 0SF occurs at 25¾ K of the total area of the wafer. When 0SF is less than 25% of the total area, the occurrence area of BMD is narrow and the ig effect cannot be fully exerted. Therefore, it is preferably 50-80%. -22-, this paper size applies Chinese National Standard (CNS) A4 specification (210 X 297 mm) -βι ίφ ---- I --- I --- — 111 —-- ^ · 1 I I-- ----- (Please read the notes on the back before filling this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 B7 V. Description of the invention (20) Silicon wafer W2 of the fourth embodiment, as described in Section As shown in FIG. 12, the OSP is not in the shape of a ring and is selected by the ingot growing in the center according to the predetermined pulling speed profile. The plan view is shown in FIG. 13. Since the silicon wafer W2 $ SF is not ring-shaped, it does not contain COP, and LD (invasive transposition) does not occur at the same time. The ingot used to make the silicon wafer of the present invention contains BMD that does not undergo indexing at a ratio of IX to 3 × 107 pieces / cm3. Therefore, it is not necessary to use it according to JP-A No. 8-4 5 9 4 5. Before the rapid heating, the state of the K wafer should be heated at a low temperature of 500-8001 and maintained for 0.5-20 hours. The introduced oxygen precipitates the nucleus. When the BMD density is IX 105 / cid3M, it is difficult to obtain a sufficient IG effect when the K wafer is heated rapidly. Another 3X 107 / cbi3 is the maximum density that allows BMD to occur in the 0SF field. In the heating method according to the fourth embodiment, it is preferable that a silicon wafer 含有 2 containing BMD which does not undergo indexing with a lift ratio is rapidly introduced into a furnace heated at 650-950 ° C at room temperature. In addition, the upper silicon wafer W2 can be placed at room temperature in a high-speed heating furnace equipped with a lamp that can generate high heat, and the method can be used to rapidly heat the lamp to 650-950¾. That is, the heat treatment of the fourth embodiment is also rapid heating. The so-called rapid heating refers to heat treatment at a heating rate of υ / min, preferably 30 oxi / min. When rapid heating is performed by light irradiation, the wafer can be heated uniformly. Therefore, it has the advantage that the wafer is less likely to warp than a wafer placed in a preheated furnace. If the final temperature reached by the rapid heating is 650 t: K, the BMD near the wafer surface cannot be completely eliminated, so DZ (layers without defects) cannot be fully ensured. If it exceeds 950 ° C, indexing will occur before the BMD near the surface of the wafer disappears, so -23- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ------ ------ i Install --- (Please read the precautions on the back before filling this page) 1T. 528816 A7 B7 V. Invention Description (2 works) (Please read the precautions on the back before filling this page) Nor can DZ be fully ensured. For this reason, it is best to be 80G-9001. In addition, when the holding time is at 0.5 minute K, the time for shrinking the B M D on the wafer surface is too short, and the BMD on the wafer surface cannot be sufficiently eliminated, so it is difficult to fully ensure the DZ. If it exceeds 30 minutes, the thickness of DZ will exceed requirements and productivity will be affected. For this reason, the holding time needs to be set in the range of 0.5-30 minutes, preferably in the range of 10-30 minutes. The rapid heating is carried out in a nitrogen atmosphere or an oxygen atmosphere or the atmosphere, preferably in a nitrogen atmosphere. After rapid heating, cool the silicon wafer to room temperature to form a DZ from the wafer surface to a depth of l-100iuffl, so that the part deeper than this DZ has a BMD density of lx 105 ~ 3x 107 pieces / cm3. Wafer. <Examples> Examples and comparative examples of the present invention are described below: Example 1 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, pulling the crystal block to grow over the full length of the crystal block, corresponding to the position shown in FIG. 3 The field of P2. At this time, in order to control the oxygen concentration in the crystal block, the flow rate of argon was maintained at about 110 liters / minute, the rotation speed of the quartz crucible storing the silicon melt was 5-10 rpm, and the inside of the hot zone furnace was about 60 torr. After grinding, chamfering and polishing the silicon wafer obtained by cutting the pulled crystal block, the chemical etching process is used to remove the damage on the surface of the wafer and CVD method is used to form a thickness of 1.5 on the back of the wafer using SifU at 680¾ ^ In polycrystalline silicon layer. Then, mirror polishing was performed to obtain a silicon wafer having a diameter of 8 inches and a thickness of 725 / iiB. Comparative Example 1 Except that a polycrystalline silicon layer was not formed, other methods were implemented according to the method described in Example 1, and a silicon wafer was prepared as Comparative Example 1. -24- This paper size is in accordance with Chinese National Standard (CNS) A4 (210 X 297 mm) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 A7 B7 V. Description of the invention (22) <Comparative Evaluation 1 &gt; For implementation The silicon wafers of Example 1 and Comparative Example 1 were subjected to the second heat treatment of the heat treatment of the silicon-like semiconductor device engineering. That is, in an oxygen atmosphere, the wafers were processed at a temperature of W 7 Ο υ for 8 hours, and then thermally processed at a temperature of 10 Q Q Ό for 12 hours. Then, the FT-IR was used to measure the oxygen concentration on the wafer surface from the central portion to the peripheral portion of the wafers of Example 1 and Comparative Example 1. The oxygen concentration difference Δ [0i] before and after the heat treatment is shown in FIG. As shown in Figs. 14 and 15, the oxygen concentration difference Δ [0 i] before and after the heat treatment of Comparative Example 1 varies greatly from the center portion of the wafer to about 40 ffia. However, compared to this, the oxygen concentration difference Δ [Oi] before and after the heat treatment in Example 1 decreases only approximately 90 niia from the center of the wafer, so that the K wafer is substantially uniform. In addition, another silicon wafer of Example 1 and another silicon wafer of Comparative Example 1 were also subjected to a heat treatment at a temperature of 1 ο ο ο υ for 4 hours, and then heat-treated at a temperature of 1130 ¾ for 3 hours (ignition). Oxidation treatment (pyrogenic oxidation treatment), and then visually inspected for the presence of 0SF. As a result, a white clouded 0SF was found in the center of the wafer of Comparative Example 1, but the wafer of Example 1 was free of 0SF. appear. Example 2 A silicon single crystal pulling device was used to pull up a P-type silicon crystal block doped with boron (B) having a diameter of 8 inches. The crystal block has a length of 1200 mm, a crystal orientation (100), a resistivity of about 1 GQ cm, and an oxygen concentration of 1.0 × 1018atoas / cm3 (formerly ASTM). V / G during pulling is continuously reduced from 0 · 24®πι2 / min to 0 · 181 ^ 2 / min t, and two crystal ingots are cultivated under the same conditions, among which -25- this paper size is applicable 0 National Standard (CNS) A4 specification (210 X 297 mm) ------- i --- rAW ^ ------- Γ -------- (Please read the Please fill out this page again) Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 Α7 _______ Β7 V. One of the crystal blocks of the invention description (23), as shown in Figure 8, cut off from its center along the pull direction, And check the position of each field, the other one cut out the silicon wafer W3 corresponding to the position P3 in Figure 8 as a sample. That is, the sample of this embodiment is a wafer W3 as shown in FIG. 9 with a domain [Pv] in the center and a domain [Pd] around it, and a domain [Pv] further around this domain. The wafer W3 cut out from the crystal block and mirror-polished is placed in a nitrogen atmosphere, and heat-treated at 650 ° F for 30 minutes. Example 3 Except that the same wafer in Example 2 was cut out and the mirror-polished wafer W3 was heat-treated at a temperature of 650 ° and a holding time of 90 minutes, the rest was implemented in accordance with Example 2. Spring Example 4 Except that the wafer after mirror polishing is subjected to a heat treatment at a temperature of 65 Dt! And a holding time of 210 minutes, the rest is implemented according to Example 2. Embodiment 5 Except that the wafer W3 after mirror polishing is heat-treated at a temperature of 750 ° F and a holding time of 60 minutes, the rest is implemented according to Embodiment 2. Example 6 The wafer W3 after mirror polishing was heat-treated at a temperature of 750 ° F. and a holding time of 90 °. Embodiment 7 Except that the wafer W3 after mirror polishing is heat-treated at a temperature of 85 ° F. and a holding time of 30 minutes, the rest is implemented in accordance with Embodiment 2. Example 8-2 6-This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) „----------- 1TL ------- line (please first Read the notes on the back and fill in this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 Λ7 B7 V. Description of the invention (24) Except for the wafer w3 with a temperature of 85 ° and a holding time of 120 minutes Except for the heat treatment, the rest is implemented according to Example 2. Comparative Example 2 The wafer W3 after mirror polishing of the same crystal block cut out in Example 2 is not subjected to heat treatment. Comparative Example 3 except for wafer W3 after mirror polishing Except for the heat treatment at a temperature of 6 5 0 υ and a holding time of 100 minutes, the rest is implemented in accordance with Example 2. Comparative Example 4 Except for performing a heat treatment at a temperature of 750 υ and a holding time of 20 minutes on the wafer W3 after mirror polishing, the rest It is implemented according to Example 2. Comparative Example 5 Except that the wafer W3 after mirror polishing is heat-treated at a temperature of 80 Gt and a holding time of 100 minutes, it is implemented according to Example 2. <Comparative Evaluation 2 &gt; Preparation Example 4 each of wafers W3 of 2-8 and Comparative Example 2-5, and the surface of these wafers W3 It contains Fe, Cr, Ni, Cu and other metal elements to dissolve the wafers and forcefully pollute Fe, Cr, Ni, and Cu. Then, the entire wafer W3 is sequentially heat-treated at 900 for 2 hours. Heat treatment at 1000 ° C for 5 hours and heat treatment at 800 ° C for 1 · 5 hours to diffuse the gold element into the entire bulk of the wafer. The heat treatment after such contamination is equivalent to semiconductor manufacturing In order to confirm the IG effect of metal contamination, the wafer is etched with a secco etching solution to a thickness of about 2 / z in thickness and placed in a spotlight -27- This paper size applies Chinese national standards ( CNS) A4 specification (210 X 297 mm) l · ---- M .------- r ^ ^ ------ (Please read the precautions on the back before filling this page) 528816 A7 B7 V. Description of the invention (25) The presence or absence of haze is observed. The presence or absence of haze in Examples 2-8 and Comparative Examples 2-5 is shown in Table 1. In addition, an optical microscope photograph of Example 2 is shown in Figs. 16A-16D, and an optical microscope photograph of Comparative Example 2 is shown in Figs. 17A-17D. Figs. 16A and 17A respectively show contamination with Fe. A quarter of the wafers of Example 2 and Comparative Example 2. Similarly, Figures 16B and 17B show contamination with Cr, Figures 16C and 17C are contaminated with Ni, and Figures 16D and 17B Figure 17D shows a quarter of the wafers of Example 2 and Comparative Example 2 contaminated with Cu, respectively. --Install --- (Please read the notes on the back before filling this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs Table 1 Heat treatment conditions turbidity temperature (υ) time (minutes) field [Ρν] field [ Pi] Example 2 650 30 No No Example 3 650 90 No No Example 4 650 210 No No Example 5 750 60 No Inr Μ Example 6 750 90 No No Example 7 850 30 No M Example 8 850 120 None -ίππ Μ Comparative Example 2--InC Thermal Yes Comparative Example 3 650 100 inf Yes fcb Comparative Example 4 750 20 No Yes Comparative Example 5 800 100 No Yes Table 1, 16A-16D and 17A-17D Only in the area [Pi] of the wafer of Comparative Example 2-5, turbidity occurs. This is because Comparative Example-2 8-This paper size applies the Chinese National Standard (CNS) A4 specification (210 x 297 mm). Order- · 528816 A7 ______ B7 V. Description of the invention (26) Under the heat treatment conditions of 2-5, the oxygen density of the wafer is low, and the heat treatment after pollution cannot exert the IG effect. Since the wafer of Example 2-8 did not appear turbid, the areas [Pv] and [Pi] had high-density oxygen precipitation nuclei in their entirety, which caused the IG effect. Example 9 'A boron (B) -doped P-type sand crystal block with a diameter of 8 was pulled using a silicon single crystal pulling device. The crystal block has a length of 1200 fflffl, a crystal orientation (100), a resistivity of about 10 Ω cm, and an oxygen concentration of 1.0 × 1018 at 0ffls / cni3 (former ASTM). The crystal block was grown under the condition that the V / G during pulling was continuously reduced from 0 · 24βιβι2 / min t to 0.18 mm2 / min, and two crystal blocks were grown under the same condition, one of which is shown in Figure 8 , Cut the center of the crystal block in the pulling direction to check the position of each field; the other crystal block is cut out as a sample as shown in FIG. 8 to the wafer W3 corresponding to the P 2 position. As shown in FIG. 9, this sample wafer W3 has a region [Pν] in the center, a region [PΗ] around it, and a region [Pν] further around the region [Pi]. The wafer W3 cut out from the crystal block and mirror-polished is placed in a nitrogen atmosphere, heated from a room temperature to a temperature of about 50 ° C / s to a temperature of 1150 ° C, and then not maintained at this temperature, and immediately heat-treated. Embodiment 10 Except that the heat treatment is performed on the wafer W3 at a temperature of 1150 ° for 5 seconds, the rest is implemented according to Embodiment 9. Embodiment 11 Except that the wafer W3 is heat-treated, it is maintained at a temperature of 1150 ° F. for 30 seconds, and the rest is implemented according to Embodiment 9. -29- This paper size is in accordance with Chinese National Standard (CNS) A4 (210 X 297 mm) (please read the precautions on the back before filling this page); install t.— ijm 1: mouth, f ϋ · 1 an ni ··· · 1. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 528816 Α7 _ _____ Β7 V. Description of the invention (27) m 12 Except for the temperature at which the heat treatment of wafer W3 is changed to 1 2 0 0 ¾ The rest is implemented according to Embodiment 9. m 13 is implemented in accordance with Embodiment 9 except that the temperature for heat-treating wafer W3 is changed to ΐ200υ and maintained at this temperature for 5 seconds. Example 1 4 Except that the temperature for heat-treating the wafer W3 was changed to 12 () t! And maintained at this temperature for 3Q seconds, the rest was implemented in accordance with Example 9. Comparative Example 6 Wafer W3 after mirror polishing of the same crystal block cut out of Example 2 was not subjected to heat treatment. Comparative Example 7 Except that the wafer W3 was heat-treated, it was kept at the temperature of uoot! For 5 seconds, and the rest was implemented according to Example 9. Comparative Example 8 Except that the wafer W3 was heat-treated, it was maintained at a temperature of 1100 ° for 30 seconds, and the rest was implemented according to Example 9. Comparative Example 9 Except that the wafer W3 was heat-treated, it was maintained at a temperature of 110 ° C. for 60 seconds, and the rest was implemented according to Example 9. Comparative Example 10 Except that the wafer W3 was heat-treated, it was maintained at a temperature of 11501C for 60 seconds, and the rest was implemented according to Example 9. * -30- This paper size is in accordance with China National Standard (CNS) A4 (210 χ 297 mm) (Please read the precautions on the back before filling out this page) Printed by the Intellectual Property Bureau Employee Consumer Cooperative of the Ministry of Economic Affairs 528816 A7 ___ ______ B7 V. Description of Invention (28)

Jbt 例 11 除對晶圓W3實施熱處理時,在1 2 0 0 ¾之溫度下保持60 秒之外,其餘悉依實施例9實施。 例 12 除對晶圓W3實施熱處理時,在125〇υ之溫度下保持5 秒之外,其餘悉依實施例9實施。 比較例1 3 除對晶圓W3實施熱處理時,在ΐ25〇υ之溫度下保持30 秒之外,其餘悉依實施例9實施。 &lt;比較評價3 &gt; 模仿半導體裝置製造商之裝置製程之熱處理,對實施 例9-14及比較例6-13之晶圓分別實施於氮氣氛下,在 8 0 0 Ό保持4小時,然後於氧氣氛下,在1 0 0 0 t保持1 6小 時之熱處理。熱處理後,將各晶圓劈開,藉萊特蝕刻液( Wright etching solution)對晶圓表面實施選擇性蝕刻 後,利用光學顯微鏡觀察,測定晶圓表面至深度3 5 0 /i m之 領域[Pi]及相當於領域[Pi]之部分之BMD面積密度及有 無滑移(slip)。结果如表2所示。 (請先閱讀背面之注意事項再填寫本頁) -·裝 tri-----1、. 經濟部智慧財產局員工消費合作社印製 -31- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 528816 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明(29 ) 表2 熱處 理條件 BMD面積密度(cm2) 有無滑移 溫度(Ο 時間(分) 領域[Pv】 領域[Pi] 實施例9 1150 0 3.6X 104 3·5Χ ΙΟ4 無 實施例10 1150 5 2.4X 10* 2.3Χ ΙΟ4 無 實施例11 1150 30 1.2X 104 1.0Χ ΙΟ4 無 實施例12 1200 0 532.0X lfl4 411.ΟΧ ΙΟ4 無 實施例13 1200 5 412.OX 104 356.ΟΧ ΙΟ4 無 實施例14 1200 30 37.7Χ 104 77.3Χ ΙΟ4 無 比較例6 無處理 40.ΟΧ ΙΟ4 0·1Χ ΙΟ4 無 比較例7 1100 5 1.0Χ ΙΟ4 0·1Χ ΙΟ4 無 比較例8 1100 30 2·2Χ ΙΟ4 0·1Χ ΙΟ4 無 比較例9 1100 60 2·2Χ ΙΟ4 0·1Χ ΙΟ4 無 比較例10 1150 60 0·5Χ ΙΟ4 0.1Χ ΙΟ4 有 比較例11 1200 60 125.ΟΧ ΙΟ4 0.5Χ ΙΟ4 有 比較例12 1250 5 73·5Χ ΙΟ4 68.5Χ ΙΟ4 有 比較例13 1250 30 65.4Χ ΙΟ4 58.8Χ ΙΟ4 有 由表2可知,比較例6-11之晶圓之相當於領域[Pi】 之部分之BMD面積密度均未達到可表現IG效果之BMD面積 密度(即IX 1〇4個/ cm2,最好2X 104個/ cm2)。雖然比較例 12及13之領域[Pv]及相當於領域[Pi]之部分之BMD面積 密度超過2X 104個/ cb2,但發生滑移(slip)。比較例1〇及 11之晶圓亦有滑移。對此,實施例9、10、12-14之晶圓 之領域[Pv】及相當於領域[Pi]之部分之BMD面積密度超 過2X1G4個/cm2,且無發生滑移,尤其實施例12-14之晶 圓具有更高之BMD面積密度。另外,實施例11之晶圓雖然 -32- 本纸張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐) ί. *&quot;&quot;·裝 广訂卜----- — .S. ,(請先閱讀背面之注意事項再填寫本頁) 528816 A7Jbt Example 11 Except that the wafer W3 is heat-treated, it is maintained at a temperature of 12 00 ¾ for 60 seconds, and the rest is implemented according to Example 9. Example 12 Except that the heat treatment was performed on the wafer W3, the temperature was maintained at 125 ° C for 5 seconds, and the rest was implemented according to Example 9. Comparative Example 1 Except that the wafer W3 was heat-treated, it was maintained at a temperature of ΐ25〇υ for 30 seconds, and the rest was implemented according to Example 9. &lt; Comparative evaluation 3 &gt; The wafers of Examples 9-14 and Comparative Examples 6-13 were respectively subjected to a heat treatment in a device process of a semiconductor device maker under a nitrogen atmosphere and maintained at 800 ° F. for 4 hours, and then Heat treatment in an oxygen atmosphere at 1000 t for 16 hours. After the heat treatment, the wafers were cleaved, and the wafer surface was selectively etched with a Wright etching solution, and then observed with an optical microscope to measure the wafer surface to a depth of 3 50 / im [Pi] and The BMD area density corresponding to the area [Pi] and the presence or absence of slip. The results are shown in Table 2. (Please read the notes on the back before filling this page)-· 装 tri ----- 1 、. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs-31- This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) 528816 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (29) Table 2 Heat treatment conditions BMD area density (cm2) With or without slip temperature (0 Time (minutes) Field [Pv ] Field [Pi] Example 9 1150 0 3.6X 104 3 · 5 × ΙΟ4 Without Example 10 1150 5 2.4X 10 * 2.3 × ΙΟ4 Without Example 11 1150 30 1.2X 104 1.0 × ΙΟ4 Without Example 12 1200 0 532.0X lfl4 411.0 × ΙΟ4 No Example 13 1200 5 412.OX 104 356.ΟΧ ΙΟ4 No Example 14 1200 30 37.7 × 104 77.3 × IO4 No Comparative Example 6 No treatment 40.〇Χ ΙΟ4 0 · 1 × ΙΟ4 No Comparative Example 7 1100 5 1.0 × ΙΟ4 0 · 1Χ ΙΟ4 No comparative example 8 1100 30 2 · 2Χ ΙΟ4 0 · 1Χ ΙΟ4 No comparative example 9 1100 60 2 · 2Χ ΙΟ4 0 · 1Χ IO04 No comparative example 10 1150 60 0 · 5 × ΙΟ4 0.1 × ΙΟ4 Yes Comparative Example 11 1200 60 125.〇Χ ΙΟ4 0.5χ ΙΟ4 Comparative example 12 1250 5 73 · 5 × IO4 68.5 × ΙΟ4 Comparative example 13 1250 30 65.4 × ΙΟ4 58.8 × ΙΟ4 It can be known from Table 2 that the BMD area corresponding to the area [Pi] of the wafer of Comparative Example 6-11 The density does not reach the BMD area density that can express the effect of IG (ie IX 104 / cm2, preferably 2X 104 / cm2). Although the areas [Pv] of Comparative Examples 12 and 13 and equivalent areas [Pi] Part of the BMD area density exceeds 2X 104 pcs / cb2, but slippage occurs. The wafers of Comparative Examples 10 and 11 also have slippage. In response, the wafers of Examples 9, 10, 12-14 The area [Pv] and the area equivalent to the area [Pi] have a BMD area density of more than 2 × 1G4 pieces / cm2, and no slippage occurs. In particular, the wafers of Examples 12-14 have a higher BMD area density. In addition, the embodiments Although the wafer of 11-32- This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 x 297 mm) ί. * &Quot; &quot; · Broadcasting book ----- — .S., (Please read the precautions on the back before filling this page) 528816 A7

五、發明說明(30 ) BMD面積密度低於2X 104個/ cm2,但晶圓内之析出分布均 句。 實施例15 從矽融液提拉矽單晶晶塊時,如第1圖所示,將提拉 速度與溫度梯度之比(V/G)保持於臨界點以上之 (V/G)2之間,使在氧氣氛下及以looot溫度處理2小時之 後再以11QQ°C處理12小時之晶圓會在其總面積之252發生 0SF。此晶塊之全長對應於第1〇圖所示之位置p2。從此提 拉之晶塊切出晶圓並加K研磨一去角後,利用化學蝕刻處 理去除晶圓表面之傷部獲得鏡面矽晶片。 將該鏡面晶片,Κ3〇υ/分之昇溫速度,從室溫加熱 至850¾,並在此溫度保持5分後放冷至室溫。 實施例1 6 除提拉晶塊使在晶圓之總面積之50¾發生0SF之外, 與實施例15同樣將加工後之晶圓施以850¾、5分之熱處 理〇 實施例17 除提拉晶塊使在晶圓之總面積之8Π發生0SF之外, 與實施例15同樣將加工後之晶圓施K85G°C、0.5分之熱 處理。 實施例18 除提拉晶塊使在晶圓之總面積之8 0 %發生0 S F之外, 與實施例15同樣將加工後之晶圓施K850P、5分之熱處 理0 -33- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) (請先閱讀背面之注意事項再填寫本頁) 裝-------11-訂··------- 經濟部智慧財產局員工消費合作社印製 528816 A7 ----- B7 經濟部智慧財產局員工消費合作社印製 五、發明說明(31) 實施例1 9 除提拉晶塊使在晶圓之總面積之80S!發生OSF之外, 與實施例15同樣將加工後之晶圓施K85Qt:、10分之熱處 理。 實施例2 0 除提拉晶塊使在晶圓之總面積之80¾發生OSF之外, 與實施例15同樣將加工後之晶圓施Κ85〇υ、20分之熱處 理。 實施例2 1 除提拉晶塊使在晶圓之總面積之80%發生0SF之外, 與實施例15同樣將加工後之晶圓施Κ850Ό、30分之熱處 理。 實施例22 除提拉晶塊使在晶圓之總面積之80%發生OSF之外, 與實施例15同樣將加工後之晶圓施K700t:、5分之熱處 理。 實施例2 3 除提拉晶塊使在晶圓之總面積之80%發生0SF之外, 與實施例15同樣將加工後之晶圓施以8001C、5分之熱處 理。 實施例2 4 除提拉晶塊使在晶圓之總面積之80%發生0SF之外, 與實施例15同樣將加工後之晶圓施K950t!、5分之熱處 理。 -34 - (請先閱讀背面之注意事 4 項再填 裝 寫本頁) --! *訂--------Γ f 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 528816 經濟部智慧財產局員工消費合作社印製 A7 _ B7_-__ 五、發明說明(32 ) 比較例1 4 除提拉晶塊使在晶圓之總面積之15%發生OSF之外, 與實施例15同樣將加工後之晶圓施以850Ό、5分之熱處 理。 比較例1 5 除提拉晶塊使在晶圓之總面積之80X發生OSF之外, 與實施例15同樣將加工後之晶圓施M64G1C、5分之熱處 理。 比較例16 除提拉晶塊使在晶圓之總面積之80%發生OSF之外, 與實施例15同樣將加工後之晶圓施ΚΙΟΟΟΌ、5分之熱處 理。 比較例1 7 除提拉晶塊使在晶圓之縴面積之80%發生0SF之外, 與實施例15同樣將加工後之晶圓施以850Ό、4 0分之熱處 理。 〈比較評價4 &gt; 將實施例15-24及比較例14-17之各晶圓劈開,繼之 藉萊德蝕刻液對晶圓表面實施選擇性蝕刻後,利用光學顯 微鏡觀察测定DZ之寬度及晶圓表面深度25〇αΐΒ之BMD密度 。測定结果示於表3。實施例18之急速加熱後之晶圓內之 BMD之50,000倍放大顯微鏡照片示於第18圖。 -35 - 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) I -----— It ----- (請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7 五、發明說明(33 ) 表3 0SF領域 之總面積 比率U) IG熱處理條件 BMD密度 (X 106/cm3) DZ寬度 (u ffl) 溫度(υ) 時間(分) 實施例15 25 850 5 2.6 40 實施例16 50 850 5 3.4 40 實施例Π 80 850 0.5 10.0 15 實施例18 80 850 5 10.0 35 實施例19 80 850 10 11.0 45 實施例20 80 850 20 10.0 65 實施例21 80 850 30 12.0 85 實施例22 80 700 5 23.0 20 實施例23 80 800 5 22.0 35 實施例24 80 950 5 24.0 55 比較例14 15 850 5 1.0 Μ下 100K 上 比較例15 80 640 5 20.0 0 比較例16 80 1000 5 5.0 100K 上 比較例Π 80 850 40 12.0 100K 上 (請先閱讀背面之注意事項再填寫本頁) 裝 經濟部智慧財產局員工消費合作社印製 由表3可知,於IG熱處理後,由於比較例14之OSF領 域太小,僅佔全晶片表面之15¾,故BMD密度未達可發揮 IG效果之10s/cm3之譜。至於比較例15則由於熱處理溫度 太低,即640¾,因此在晶圓表面未形成DZ。比較例16之 熱處理過高,即100P,因此形成必要Μ上寬度之02。再 者,比較例17由於熱處理時間過長,即40分,因此亦形成 必要Κ上寬度之DZ。對此,實施例15-24之矽晶圓,由於 其BMD密度具有可發揮IG效果之lDs〜l〇7/cm3之譜,尤其 0SF領域為80%之實施例17-22,其BMD密度為l〇7/cai3之 -36- ~1~~----訂---------. 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 經濟部智慧財產局員工消費合作社印製 528816 A7 ___B7 _ 五、發明說明(34 ) 譜,其中熱處理時間10-30分之實施例19-21及熱處理溫 度9 5 0 ¾之實施例24,顯示獲得4 5 -8 5 /i m之大寬度之DZ。 另由第18圖之顯微鏡照片可知,急速加熱處理後之晶 圓中存在之BMD有發生轉位(dislocation)。 圖式之簡單說明 第1圖為根據Voronkov理論顯示本發明第1實施形 態之V/G比及空孔型點缺陷或格子間矽型缺陷濃度之關係 圖; 第2圖為顯示決定所期之提拉速度輪廓之提拉速度之 變化之特性圖·, 第3圖為顯示本發明第1實施形態之基準晶塊之空孔 型點缺陷佔優勢之領域、格子間矽型點缺陷佔優勢之領域 及完美領域之X線斷層影像; 第4圖為顯示在對應於第3圖之位置Pi之矽晶圓Wi有 出現0SF之狀態之圖; 第5圖為通過對應於第3画之位置Pit晶塊之軸中心 沿軸方向切割之斷面圖; 第6圖為顯示在對應於第3圖之位置卩2之矽晶圓#2之 中心部有出現0SF之狀態之圖; 第7圖為根據Voronkov理論顯示本發明第2及3實 施形態之V/G比及空孔型點缺陷或格子間矽型缺陷濃度之 關係圖; 第8圖顯示本發明第2及第3實施形態之基準晶塊之 -37- 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) &quot; 1*111111 ^1-11 I · 11111FI (請先閱讀背面之注意事項再填寫本頁) 528816 經濟部智慧財產局員工消費合作社印製 A7 B7 五、發明說明(35 ) 空孔型點缺陷佔優勢之領域、格子間矽型點缺陷佔優勢之 領域及完美領域之X線斷層影像; 第9圖為顯示對應於第8圖之位置p3之矽晶圓W3之平 面圖; 第10圖為本發明第4實施形態之基準晶塊之空孔型點 缺陷佔優勢之領域、格子間矽型點缺陷佔優勢之領域及完 _領域之X線斷層影像; 第11圖為顯示在對應於第3圖之位置h之矽晶圓^有 出現OSF之狀態之圖; 第12圖為通過對應於第10圖之位置卩2之晶塊之軸中心 之沿軸方向切割之斷面圖; 第13圖為顯示在對應於第10圖之位置卩2之矽晶圓^之 中心部有出現OSF之狀況圖; 第14圖為顯示實施例1及比較例1之各矽晶圓在模仿 半導體裝置工程之熱處理之第1熱處理前後之晶圓面內之 △ [ 〇 i】之狀況之圖; 第15圖為顯示實施例1及比較例1之各矽晶圓在模仿 半導體裝置工程之熱處理之第2熱處理前後之晶圓面內之 △ [ 〇 i ]之狀況之圖; 第16A圖為使實施例2之晶圓》2受Fe之污染,令Fe擴 散至晶圓之全體內後有無發生混濁之光學顯微鏡照片; 第16B圖為使實施例2之晶圓W2受Cr之污染,令Cr擴 散至晶圓之全體内後有無發生混濁之光學頸微鏡照片; 第16C圖為使實施例2之晶圓^受Ni之污染,令Ni擴 -38- 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) ^--------tT-------— (請先閱讀背面之注意事項再填寫本頁) 528816 A7 B7_ 五、發明說明(36) 散至晶圓之全體内後有無發生混濁之光學顯微鏡照片; 第16D圖為使實施例2之晶圓¥2受(:11之污染,令Cu擴 散至晶圓之全體內後有無發生混濁之光學顯微鏡照片; 第17A圖為使比較例2之晶圓¥2受卩6之污染,令.Fe擴 散至晶圓之全體內後有無發生混濁之光學顯微鏡照片; 第17B圖為使實施例2之晶圓》2受以之污染,令Cr擴 散至晶圓之全體内後有無發生混濁之光學顯微鏡照片; 第17C圖為使實施例2之晶圓W2受Ni之污染,令Ni擴 散至晶圓之全體內後有無發生混濁之光學顯微鏡照片; 第17D圖為使實施例2之晶圓¥2受(:11之污染,令Cu擴 散至晶圓之全體內後有無發生混濁之光學顯微鏡照片; 第ίδ圓為顯示急速加熱處理之晶圓内之氧析出物(BMD) 之狀況之顯微鏡照片。 -*—·」1-1#-裝-------卜訂--------卜—up. (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印制衣 本紙張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐)V. Description of the invention (30) The BMD area density is lower than 2X 104 pcs / cm2, but the precipitation distribution in the wafer is uniform. Example 15 When pulling a silicon single crystal ingot from a silicon melt, as shown in FIG. 1, the ratio of the pulling speed to the temperature gradient (V / G) is maintained at (V / G) 2 or more above the critical point. At the same time, a wafer with an oxygen atmosphere and a looot temperature of 2 hours and then processed at 11QQ ° C for 12 hours will produce 0SF at 252 of its total area. The full length of this crystal block corresponds to the position p2 shown in FIG. 10. After pulling out the wafer from the pulled crystal block and grinding it with K to remove corners, the surface of the wafer was removed by chemical etching to obtain a mirror silicon wafer. This mirror wafer was heated from room temperature to 850 ° C at a temperature rising rate of κ30 / min, and kept at this temperature for 5 minutes, and then cooled to room temperature. Example 1 6 Except that the ingot was pulled to make 0SF in 50¾ of the total area of the wafer, the processed wafer was subjected to 850¾ and 5/5 heat treatment in the same manner as in Example 15. Example 17 In addition to the 0SF occurring in 8Π of the total area of the wafer, the processed wafer was heat-treated at K85G ° C and 0.5 minutes in the same manner as in Example 15. Example 18 Except that the ingot was pulled so that 0 SF occurred in 80% of the total area of the wafer, the processed wafer was subjected to K850P and a heat treatment of 5/5 in the same manner as in Example 15. Applicable to China National Standard (CNS) A4 specification (210 X 297 mm) (Please read the precautions on the back before filling out this page). ------- 11-Order ·· ------- Economy Printed by the Employees 'Cooperatives of the Ministry of Intellectual Property Bureau 528816 A7 ----- B7 Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the Invention (31) Example 1 9 Except for lifting the crystal block to make the total area on the wafer Except for the occurrence of OSF at 80S !, the processed wafer was subjected to K85Qt: and 10-minute heat treatment in the same manner as in Example 15. Example 20 0 Except that the ingot was pulled to cause OSF to occur in 80¾ of the total area of the wafer, the processed wafer was subjected to a heat treatment of K85 and 20 minutes in the same manner as in Example 15. Example 2 1 Except that the ingot was pulled to cause 0SF to occur in 80% of the total area of the wafer, the processed wafer was treated with K850Κ and heat-treated for 30 minutes in the same manner as in Example 15. Example 22 Except that the ingot was pulled so that OSF occurred in 80% of the total area of the wafer, the processed wafer was subjected to K700t :, 5 minutes heat treatment in the same manner as in Example 15. Example 2 3 Except that the ingot was pulled to cause 0SF to occur in 80% of the total area of the wafer, the processed wafer was heat-treated at 8001C and 5 minutes in the same manner as in Example 15. Example 2 4 Except that the ingot was pulled so that 0SF occurred in 80% of the total area of the wafer, the processed wafer was treated with K950t! For 5 minutes in the same manner as in Example 15. -34-(Please read 4 notes on the back before filling and writing this page)-! * Order -------- Γ f This paper size is applicable to China National Standard (CNS) A4 (210 X 297 mm) 528816 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 _ B7 _-__ V. Description of the Invention (32) Comparative Example 1 4 Except for pulling the crystal block to cause OSF to occur in 15% of the total area of the wafer In the same manner as in Example 15, the processed wafer was heat-treated at 850 ° F and 5 minutes. Comparative Example 15 Except that the ingot was pulled to cause OSF to occur at 80X of the total area of the wafer, the processed wafer was treated with M64G1C for 5 minutes in the same manner as in Example 15. Comparative Example 16 Except that the ingot was pulled to cause OSF to occur in 80% of the total area of the wafer, the processed wafer was heat-treated in the same manner as in Example 15 for 5 minutes. Comparative Example 17 Except that the ingot was pulled to cause 0SF to occur in 80% of the fiber area of the wafer, the processed wafer was subjected to heat treatment at 850 ° F and 40 minutes in the same manner as in Example 15. <Comparative Evaluation 4> Each of the wafers of Examples 15-24 and 14-17 was cleaved, and then the wafer surface was selectively etched by Ryder etchant, and then the width and DZ of the DZ were measured with an optical microscope observation. BMD density of wafer surface depth of 25 〇αΐΒ. The measurement results are shown in Table 3. A 50,000-times magnification microscope photo of the BMD in the wafer after rapid heating in Example 18 is shown in FIG. 18. -35-This paper size applies to Chinese National Standard (CNS) A4 (210 X 297 mm) I -----— It ----- (Please read the precautions on the back before filling this page) 528816 A7 B7 V. Description of the invention (33) Table 3 Total area ratio in the 0SF field U) IG heat treatment conditions BMD density (X 106 / cm3) DZ width (u ffl) Temperature (υ) Time (minutes) Example 15 25 850 5 2.6 40 Example 16 50 850 5 3.4 40 Example Π 80 850 0.5 10.0 15 Example 18 80 850 5 10.0 35 Example 19 80 850 10 11.0 45 Example 20 80 850 20 10.0 65 Example 21 80 850 30 12.0 85 Implementation Example 22 80 700 5 23.0 20 Example 23 80 800 5 22.0 35 Example 24 80 950 5 24.0 55 Comparative Example 14 15 850 5 100K at 1.0 M Comparative Example 15 80 640 5 20.0 0 Comparative Example 16 80 1000 5 5.0 100K The above comparative example Π 80 850 40 12.0 100K (Please read the precautions on the back before filling this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs It can be seen from Table 3 that after the IG heat treatment, due to the OSF of Comparative Example 14, The field is too small, accounting for only 15¾ of the total wafer surface, so the BMD density is not up to IG effect volatilization of 10s / cm3 of the spectrum. As for Comparative Example 15, the DZ was not formed on the wafer surface because the heat treatment temperature was too low, that is, 640 °. The heat treatment of Comparative Example 16 was too high, i.e., 100P, and thus formed 02 with a necessary width. In addition, Comparative Example 17 had a long heat treatment time, i.e., 40 minutes, and thus a DZ having a necessary width over K was formed. In this regard, the silicon wafers of Examples 15-24 have a BMD density of lDs to 107 / cm3, which can exert the IG effect. In particular, Examples 17-22 in which the 0SF field is 80% have a BMD density of l〇7 / cai3 of -36- ~ 1 ~~ ---- order ---------. This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) Ministry of Economic Affairs Printed by the Intellectual Property Bureau's Consumer Cooperatives 528816 A7 ___B7 _ 5. The invention description (34) spectrum, in which Example 19-21 of heat treatment time 10-30 minutes and Example 24 of heat treatment temperature 9 5 0 ¾, shows 4 5 -8 5 / im DZ with large width. It can also be seen from the micrograph of Fig. 18 that the BMD existing in the crystal circle after the rapid heating treatment has been dislocated. Brief description of the drawings. Figure 1 is a diagram showing the relationship between the V / G ratio and the concentration of void-type point defects or inter-lattice silicon-type defects according to Voronkov theory in the first embodiment of the present invention. The characteristic diagram of the variation of the pulling speed in the pulling speed profile. Figure 3 shows the areas where the void type point defects are dominant in the reference crystal block of the first embodiment of the present invention, and the silicon type point defects between the lattices are dominant. X-ray tomography image of the realm and the perfect realm; Fig. 4 is a diagram showing a state where 0SF appears on the silicon wafer Wi at the position Pi corresponding to Fig. 3; Fig. 5 is a Pit through the position corresponding to the third picture Sectional view of the axial center of the crystal block cut along the axial direction; FIG. 6 is a diagram showing a state where 0SF appears in the center portion of the silicon wafer # 2 corresponding to the position 卩 2 in FIG. 3; FIG. 7 is Voronkov theory shows the relationship between the V / G ratio and the concentration of void-type point defects or inter-lattice silicon-type defects in the second and third embodiments of the present invention; FIG. 8 shows the reference crystals of the second and third embodiments of the present invention. Block No. -37- This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) &quot; 1 * 111111 ^ 1-11 I · 11111FI (Please read the notes on the back before filling out this page) 528816 Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Invention Description (35) X-ray tomographic images of the area where the hole type point defect is dominant, the area where the silicon type point defect is dominant, and the perfect field; Figure 9 is a plan view showing the silicon wafer W3 corresponding to the position p3 of Figure 8; Fig. 10 is an X-ray tomographic image of a field in which a hole type point defect is dominant in a reference crystal block in the fourth embodiment of the present invention, a field in which inter-lattice silicon type point defects are dominant, and a complete field. Fig. 11 is a display The silicon wafer at position h corresponding to FIG. 3 has a state where OSF appears; FIG. 12 is a cross-section cut along the axial direction through the axis center of the crystal block corresponding to position 卩 2 in FIG. 10 FIG. 13 is a diagram showing a state where OSF appears at the center of the silicon wafer ^ corresponding to position 10 in FIG. 10; FIG. 14 is a diagram showing each silicon wafer in Example 1 and Comparative Example 1 In the wafer surface before and after the first heat treatment imitating the heat treatment of semiconductor device engineering △ [〇i] status diagram; Figure 15 shows the silicon wafers of Example 1 and Comparative Example 1 in the wafer surface before and after the second heat treatment imitating the semiconductor device engineering heat treatment △ [〇i] Fig. 16A is an optical microscope photograph of whether the wafer of Example 2 was contaminated with Fe, and Fe was diffused into the entire wafer with or without turbidity; Fig. 16B is a diagram of Example 2 Wafer W2 is contaminated with Cr, and whether there is an optical neck micromirror photo of turbidity after Cr diffuses into the entire wafer; FIG. 16C is a view showing that the wafer of Example 2 is contaminated with Ni, and Ni expands to -38. -This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm) ^ -------- tT --------- (Please read the precautions on the back before filling in this (Page) 528816 A7 B7_ V. Description of the invention (36) Optical micrograph of turbidity after scattering to the whole wafer; Figure 16D shows the wafer ¥ 2 of Example 2 contaminated by (: 11, Cu Optical micrograph of turbidity after diffusion into the entire wafer; Figure 17A shows the wafer ¥ 2 of Comparative Example 2 contaminated with 卩 6 Let the Fe diffuse into the entire wafer and see if there is any optical turbidity; Figure 17B is the contamination of the wafer of Example 2 and the contamination caused by the diffusion of Cr into the entire wafer. Photograph of an optical microscope; FIG. 17C is an optical microscope photograph of whether the wafer W2 of Example 2 is contaminated with Ni, and Ni has diffused into the entire wafer with or without turbidity; FIG. 17D is a crystal of Example 2 Circle ¥ 2 is contaminated by (: 11, whether there is an optical microscope photo of turbidity after Cu has diffused into the entire wafer; circle δ is the status of oxygen precipitates (BMD) in wafers subjected to rapid heat treatment. Micrograph. -* — · "1-1 #-装 ------- 卜 定 -------- 卜 —up. (Please read the notes on the back before filling out this page) Intellectual Property Bureau, Ministry of Economic Affairs The paper size of the printed clothing paper of the employee consumer cooperative is applicable to the Chinese National Standard (CNS) A4 (210 x 297 mm)

Claims (1)

528816 經濟部智慧財產局員工消費合作社印製 A9 B9 C9 D9 第89122008號專利申請案 民國91年&quot;月对曰呈 修正申請專利範圍 &lt;請先閱讀背面之注意事項再行繪製) 1 . 一種矽晶圓之熱處理方法,包括: 於晶圓面内無發生起因於結晶之粒子及侵入型轉位之 氣濃度1.2&gt;&lt;101831:〇11^/(!1113下(舊&amp;5了《)之矽晶圓的背面, 依化學的氣相堆積法形成厚度〇 . 1- 1 , 6 ^ m之多晶矽層之步 驟; 將上述之帶有多晶矽層之矽晶圓置於氧氣氛下,以 lOOOCtSOC之溫度實施2-5小時之熱處理後,繼續以 11301 土 3010之溫度實施1-16小時之熱處理之步驟;其 特徵乃在: 灯 該矽晶圓傺在形成上述之多晶砂層之前,對其實施上 述之熱處理時,會在晶圓中心部顯現氧化引發堆積缺陷者。 2 . —種帶有多晶矽層之矽晶圓之製造方法,·它偽藉 化學的氣相堆積法,於晶圓面内無發生起因於結晶之粒子 及侵人型轉位之氧濃度1·2Χ 1018atoros/cin3以下(舊ASTM) 之矽晶圓的背面形成有厚度0.1-1·6μ in之多晶矽層之方法 ,其特歡在於: 該矽晶圓係在形成上述之多晶矽層之前,置於氧氣氛 下,以1000¾ 士 30¾之溫度實施2-5小時之熱處理後,繼 續以1130¾ 土 30t:之溫度實施1-16小時之熱處理時,會 在晶圓中心部顯現氧化引發堆積缺陷者。 -1- 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 528816 B9 C9 D9 3. —種矽晶圓之熱處理方法,該晶圓係從一種由完 美領域[P]形成之晶塊切出者,該方法包括: X請先閲讀背面之注意事項再行繪製) 從矽融液中提拉由領域[Pv]及領域[PJ之混合領域 構成且具有氧濃度0.8X1018〜1·4Χ1018(舊ASTM)之矽 單晶塊;及 從該晶塊切出矽晶圓之後,將該矽晶圓置於氮、氬、 氫、氧或其混合物之氣氛中,在600-850 t:之溫度下保持 3 0-90分;其中領域[P]為格子間矽型點缺點之凝聚體及 空孔型點缺陷之凝聚體皆不存在之完美領域;領域[Pd為 鄰接於領域[I]及屬於上述之完美領域[P]且具有可形成 侵人型轉位之最低之格子間矽濃度以下之格子間矽濃度之 領域; 上述領域[I]為在矽單晶塊内,格子間矽型點缺陷佔 優勢之領域; 領域[V ]為空孔型點缺陷佔優勢之領域;而 領域[Pv]為鄰接於上述之領域[V]及屬於上述之完 美領域[P]且具有可形成C0P或FPD之空孔濃度以下之空 孔濃度之領域。 經濟部智慧財產局員工消費合作社印製 矽積 在堆 中相 其氣 ,學 法化 方用 理使 處傺 熱理 之處 項熱 3 之 第時 圍層 範矽 利晶 專多 請成 申形 如面 背 4,之 圓 。 晶法 5 完 由 —1 種 1 從 偽 圓 晶 該 法 方 311 理 處 熱 之 圓 晶 矽 be 種 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 528816 A9 B9 C9 D9 美領域[P ]形成之晶塊切出者,該方法包括: 從矽融液中提拉由領域[Pv]及領域[Pi]之混合領域 構成且具有氧濃度〇·δΧΐ〇18〜1·4Χ1018(舊ASTM)之砂 單晶塊;及 從該晶塊切出矽晶圓之後,將該矽晶圓置於氮、鐘 氫、氧或其混合物之氣氛中,在600-850 ”之溫度下保持 1 20-250分;其中領域[Ρ]為格子間矽型點缺點之凝聚體 及空孔型點缺陷之凝聚體皆不存在之完美領域;領域[Pl] 為鄰接於領域Π]及屬於上述之完美領域[Ρ]且具有可形 成侵入型轉位之最低之格子間矽濃度以下之格子間矽濃度 之領域; 上述領域[I]為在矽單晶塊内,格子間矽型點缺陷佔 優勢之領域; 領域[V]為空孔型點缺陷佔優勢之領域;而 領域[Ρν]為鄰接於上述之領域[V]及屬於上述之完 美領域[Ρ]且具有可形成C0P或FPD之空孔濃度以下之空 濃度之領域 C請先閲讀背面之注意事項再行繪製) 、τ 線 經濟部智慧財產局員工消費合作社印製 6. 如申請專利範圍第5項之熱處理方法,其中在矽 晶圓之背面形成多晶矽層時之熱處理偽使用化學氣相堆積法。 7. —種矽晶圓之熱處理方法,該晶圓偽從一種由完 美領域[P]形成之晶塊切出者,該方法包括: 3 一人一 本紙張尺度適用中國國家標準(CNS ) Α4規格(210X29*7公釐) 528816 A9 B9 C9 D9 從矽融疲中提拉由領域[Pv]及領域[Pi]之混合領域 構成且具有氧濃度0.8X1018〜1.4X1018(舊ASTM)之矽 單晶晶塊;及 從該晶塊切出矽晶圓之後,將該矽晶圓置於氮、氬、 氫、氣或其涯合物之氣氛中,以10-150T:/秒之舁溫速度 從室溫加熱至1150-1200*0,並在此溫度範圍保持0_30秒 ;其中領域[P ]為格子間矽型點缺點之凝聚體及空孔型點 缺陷之凝聚體皆不存在之完美領域;領域[Pd為鄰接於領 域[I]及屬於上述之完美領域[P]且具有可形成侵入型轉 位之最低之格子間矽濃度以下之格子間矽濃度之領域; 上述領域[I]為在矽單晶晶塊内,格子間矽型點缺陷 佔優勢之領域; 領域[V]為空孔型點缺陷佔優勢之領域;而 領域[Pv]為鄰接於上述之領域[V]及屬於上述之完 美領域[P]且具有可形成C0P或FPD之空孔濃度以下之空 孔濃度之領域。 8 . —種矽晶圓之熱處理方法,包括: 從矽融液提拉矽單晶塊之步驟; 從該晶塊切製矽晶圓之步驟; 以101/分以上之昇溫速度,從室溫將該矽晶圓急速 加熱至650_950t:,並於該溫度範圍保持0,5-30分之步驟; 其特徵在於:該晶圓受熱氧化處理時,在晶圓總面積之25 %以上部分發生氧化引發堆積缺陷,且不發生轉位之情形 —^ — 本紙張尺度適用中國國家標隼(CNS ) A4規格(210X297公釐) ---------#1. 0請先閱讀背面之注意事項再行繪製) 訂_ 經濟部智慧財產局員工消費合作社印製 ---- 528816 A9 B9 C9 D9 下産生lx 1015〜3X 107個/cra3之氧析出物。 9·如申請專利範圍第8項之熱處理方法,其中該砂 單晶塊偽從矽融液提拉,使由其製得之矽晶圓受熱氧化處 理時,會在該矽晶圓之總面積之25%以上區域發生氣化引 發堆積缺陷及IX 1〇15〜3Χ 1〇7個/ Cffl3之無發生轉位之氣 析出物。 J---?---1.-- (請先閲讀背面之注意事項再行繪製) -線_ 經濟部智慧財產局員工消費合作社印製 一 5- 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐)528816 Printed by the Consumer Property Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, A9 B9 C9 D9 Patent Application No. 89122008, Republic of China &quot; Monthly Revision, Revised Application Scope of Patent &lt; Please read the precautions on the back before drawing) 1. The heat treatment method for silicon wafers includes the following: No crystal particles and intrusive indexing gas concentrations occur within the wafer surface 1.2 &gt; &lt; 101831: 〇11 ^ / (! 1113 下 (旧 &amp; 5 了The step of forming a polycrystalline silicon layer with a thickness of 0.1-1.6 m in accordance with the chemical vapor deposition method on the back of the silicon wafer; placing the above-mentioned silicon wafer with a polycrystalline silicon layer under an oxygen atmosphere, After performing heat treatment at a temperature of 100OCtSOC for 2-5 hours, the process of heat treatment at a temperature of 11301 to 3010 is continued for 1-16 hours; its characteristics are as follows: before forming the polycrystalline sand layer described above, When the above-mentioned heat treatment is performed on the wafer, those who cause accumulation defects due to oxidation will appear at the center of the wafer. 2. A method for manufacturing a silicon wafer with a polycrystalline silicon layer, which uses a chemical vapor deposition method to crystallize No occurrence in a circular plane The method of forming a polycrystalline silicon layer with a thickness of 0.1-1 · 6 μin on the back of a silicon wafer with crystalline particles and an oxygen-converting oxygen concentration of 1.2 × 1018atoros / cin3 (older ASTM) is particularly advantageous. : The silicon wafer is subjected to a heat treatment at a temperature of 1000 ¾ ± 30 ¾ for 2-5 hours under the oxygen atmosphere before the formation of the above polycrystalline silicon layer, and then a heat treatment of 1130 ¾ to 30 t: is continued for 1 to 16 hours. At the time, the defects caused by oxidation and accumulation at the center of the wafer will appear. -1- This paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm) 528816 B9 C9 D9 3. — A heat treatment method for silicon wafers, This wafer is cut out from a crystal block formed by the perfect field [P]. The method includes: X Please read the precautions on the back before drawing.) Pull from the silicon melt. [Pv] and field [Single silicon ingot composed of a mixed field of PJ and having an oxygen concentration of 0.8X1018 ~ 1.4 × 1018 (former ASTM); and after a silicon wafer is cut out from the ingot, the silicon wafer is placed in nitrogen, argon, and hydrogen In the atmosphere of oxygen, oxygen or mixture thereof, at 600-850 t: The temperature is maintained between 30 and 90 minutes; the domain [P] is a perfect domain where neither the interstitial silicon-type point defect aggregates nor the void-type point defect aggregates exist; the domain [Pd] is adjacent to the domain [I] And the area which belongs to the above-mentioned perfect area [P] and has the lowest inter-lattice silicon concentration that can form an intrusive translocation; the above-mentioned area [I] is in the silicon single crystal block, the inter-lattice silicon The field where the point defect is dominant; the field [V] is the field where the hole type point defect is dominant; and the field [Pv] is adjacent to the above field [V] and belongs to the above perfect field [P] and has a formability The area of pore density below the pore density of COP or FPD. The Intellectual Property Bureau of the Ministry of Economic Affairs ’s consumer co-operative society printed silicon to accumulate in the stack, and learned how to use the tactics to deal with the heat. The third layer of the Fan Si Lijing, please apply for it. Such as the face back 4, the circle. Crystal method 5 Finished by—1 kind 1 from pseudo-round crystal The method 311 heat-treated round crystal silicon be kind This paper size is applicable to China National Standard (CNS) A4 specification (210X297 mm) 528816 A9 B9 C9 D9 US field [P] The formed crystal block is cut out, and the method includes: pulling up from a silicon melt solution composed of a mixed field of a field [Pv] and a field [Pi] and having an oxygen concentration of 0 · δ × ΐ18 ~ 1 · 4 × 1018 ( Old ASTM); and after cutting out the silicon wafer from the crystal block, the silicon wafer is placed in an atmosphere of nitrogen, bell hydrogen, oxygen or a mixture thereof, and maintained at a temperature of 600-850 " 1 20-250 points; the field [P] is a perfect field in which the aggregates of silicon-type point defects between the lattices and the aggregates of void-type point defects do not exist; the field [Pl] is adjacent to the field Π] and belongs to the above The perfect area [P] and the area with the lowest inter-lattice silicon concentration that can form intrusive translocations. The above-mentioned area [I] is that in the silicon single crystal block, the inter-lattice silicon point defects account for Area of advantage; Area [V] is the area where void point defects are dominant; and Area [Ρν] Adjacent to the above-mentioned field [V] and the perfect field [P] described above and have a void concentration below the void concentration that can form C0P or FPD, please read the precautions on the back before drawing), τ line economy Printed by the Ministry of Intellectual Property Bureau's Consumer Cooperatives 6. If the heat treatment method of item 5 of the patent scope is applied, the chemical vapor deposition method is used for the heat treatment when a polycrystalline silicon layer is formed on the back of the silicon wafer. 7. — Silicon wafer Heat treatment method, the wafer is pseudo cut out from a crystal block formed by the perfect field [P], the method includes: 3 one person one paper size applies Chinese National Standard (CNS) Α4 specification (210X29 * 7 mm) 528816 A9 B9 C9 D9 is a silicon single crystal ingot composed of a mixed domain of the domain [Pv] and the domain [Pi] and having an oxygen concentration of 0.8X1018 to 1.4X1018 (formerly ASTM) from silicon melting; and from the crystal After the silicon wafer is cut out, the silicon wafer is placed in an atmosphere of nitrogen, argon, hydrogen, gas, or a compound thereof, and is heated from room temperature to 1150-1200 at a temperature of 10-150 T: / s. * 0, and stay in this temperature range for 0_30 seconds; where The domain [P] is a perfect domain where neither agglomerates of silicon-type point defects between the lattices nor agglomerates of void-type point defects exist; the domain [Pd] is adjacent to the domain [I] and belongs to the above-mentioned perfect domain [P] and The area with the lowest inter-lattice silicon concentration that can form intrusive indexing; the above-mentioned area [I] is the area where inter-lattice silicon-type point defects are dominant in silicon single crystal blocks; V] is the area where the hole-type point defect is dominant; and the area [Pv] is the space adjacent to the above-mentioned area [V] and the perfect area [P] described above, and has a space below the concentration of pores that can form COP or FPD. Area of pore concentration. 8. A method for heat treatment of a silicon wafer, including: a step of pulling a silicon single crystal block from a silicon melt; a step of cutting a silicon wafer from the crystal block; a temperature rising rate of 101 / min or more from room temperature The silicon wafer is rapidly heated to 650_950t: and maintained at the temperature range of 0,5-30 minutes; characterized in that when the wafer is subjected to thermal oxidation treatment, oxidation occurs in more than 25% of the total area of the wafer Causes stacking defects without indexing— ^ — This paper size applies to China National Standard (CNS) A4 (210X297 mm) --------- # 1. 0 Please read the back Note for re-drawing) Order _ Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs ---- 528816 A9 B9 C9 D9 produces lx 1015 ~ 3X 107 oxygen precipitates / cra3. 9. The heat treatment method according to item 8 of the scope of patent application, wherein the sand single crystal block is pseudo-pulled from the silicon melt, so that when the silicon wafer prepared by the silicon wafer is subjected to thermal oxidation treatment, the total area of the silicon wafer will be In more than 25% of the regions, gasification-induced accumulation defects and IX 1015 ~ 3 × 107 / Cffl3 gas deposits without translocation occurred. J ---? --- 1 .-- (Please read the precautions on the back before drawing) -Line _ Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5- This paper size applies to Chinese National Standards (CNS) A4 specifications (210X297 mm)
TW89122008A 1999-04-23 2000-10-20 Method for heat treating silicon wafer TW528816B (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP11686299A JP3855531B2 (en) 1999-04-23 1999-04-23 Silicon wafer with polysilicon layer and method for manufacturing the same
JP21375199A JP4003351B2 (en) 1999-07-28 1999-07-28 IG processing method
JP33532899A JP4107628B2 (en) 1999-11-26 1999-11-26 Pre-heat treatment method for imparting IG effect to silicon wafer
JP33532799 1999-11-26

Publications (1)

Publication Number Publication Date
TW528816B true TW528816B (en) 2003-04-21

Family

ID=28457847

Family Applications (1)

Application Number Title Priority Date Filing Date
TW89122008A TW528816B (en) 1999-04-23 2000-10-20 Method for heat treating silicon wafer

Country Status (1)

Country Link
TW (1) TW528816B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8920560B2 (en) 2006-11-06 2014-12-30 Sumco Corporation Method for manufacturing epitaxial wafer

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8920560B2 (en) 2006-11-06 2014-12-30 Sumco Corporation Method for manufacturing epitaxial wafer

Similar Documents

Publication Publication Date Title
KR100226374B1 (en) Silicon Wafer Manufacturing Method
TWI236506B (en) Silicon single crystal wafer having few defects wherein nitrogen is doped and a method for producing it
US8529695B2 (en) Method for manufacturing a silicon wafer
JP4460671B2 (en) Silicon semiconductor substrate and manufacturing method thereof
JP2007186376A (en) Epitaxial wafer and method for manufacturing epitaxial wafer
JP3731417B2 (en) Method for producing silicon wafer free of agglomerates of point defects
JP2001217251A (en) Method of heat-treating silicon wafer
JP4131077B2 (en) Silicon wafer manufacturing method
JP2004043256A (en) Silicon wafer for epitaxial growth, epitaxial wafer, and method of manufacturing the same
US6547875B1 (en) Epitaxial wafer and a method for manufacturing the same
CN101016651A (en) Annealed wafer and manufacturing method of annealed wafer
JP4107628B2 (en) Pre-heat treatment method for imparting IG effect to silicon wafer
TW528816B (en) Method for heat treating silicon wafer
JP4750916B2 (en) Method for growing silicon single crystal ingot and silicon wafer using the same
JP2004165489A (en) Epitaxial silicon wafer, its manufacturing method and semiconductor device
TW516092B (en) Thermal treatment method of silicon wafer and silicon wafer treated by the same
US6818197B2 (en) Epitaxial wafer
JP3903655B2 (en) IG processing method of silicon wafer
JP2001089294A (en) Method of continuously pulling up silicon single crystal free from agglomerates of point defects
TW201909246A (en) Wafer manufacturing method and wafer
JP4615785B2 (en) Method for producing epitaxial wafer having no epitaxial layer defect using nitrogen-added substrate
JP3855531B2 (en) Silicon wafer with polysilicon layer and method for manufacturing the same
JP4577320B2 (en) Silicon wafer manufacturing method
JP4144163B2 (en) Epitaxial wafer manufacturing method
JP3855527B2 (en) Heat treatment method for silicon wafer

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees