TW486745B - Silicon wafer and method of valuating silicon wafer and method of manufacturing silicon wafer - Google Patents

Silicon wafer and method of valuating silicon wafer and method of manufacturing silicon wafer Download PDF

Info

Publication number
TW486745B
TW486745B TW89117204A TW89117204A TW486745B TW 486745 B TW486745 B TW 486745B TW 89117204 A TW89117204 A TW 89117204A TW 89117204 A TW89117204 A TW 89117204A TW 486745 B TW486745 B TW 486745B
Authority
TW
Taiwan
Prior art keywords
nitrogen
atoms
doped
silicon
wafers
Prior art date
Application number
TW89117204A
Other languages
Chinese (zh)
Inventor
Satoshi Komiya
Shiro Yoshino
Masayoshi Danbata
Kouichirou Hayashida
Original Assignee
Komatsu Denshi Kinzoku Kk
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 JP24118599A external-priority patent/JP3479001B2/en
Application filed by Komatsu Denshi Kinzoku Kk filed Critical Komatsu Denshi Kinzoku Kk
Application granted granted Critical
Publication of TW486745B publication Critical patent/TW486745B/en

Links

Landscapes

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

Abstract

This invention provides a method for producing a silicon wafer having characteristics sufficient for semiconductor devices while doping nitrogen. In the method for producing a silicon ingot by pulling up a silicon single crystal by a Czochralski method, the silicon ingot is produced by pulling up the silicon single crystal under a condition such that the silicon ingot having a part in which the concentration of nitrogen is 5*10<SP>13</SP> to 1*10<SP>15</SP> atom/cm<SP>3</SP> is formed by nitrogen doping and then silicon wafers, each of which contains nitrogen in a concentration of 5*10<SP>13</SP> to 1*10<SP>15</SP> atoms/cm<SP>3</SP> and is used in non-oxidizing heat treatment, are cut out.

Description

五、發明說明(1) 【發明所屬之技術領域j 本务明係關於一種滴 卜 氫氣熱處理)等的非气^ ;例用虱氛圍高溫熱處理(稱為 矽晶圓之製造方法。羊匕性熱處理的半導體元件製造用之 【習知技術】 在用蔡氏法(以下,摇兔 圓)中,存在著例如作)製造的石夕晶圓(cz —石夕晶 孔的空隙缺陷等。$種為二 FPD、cop檢測出的引起空 響,為了除去對最終製品的品質有壞影 熟處理或風亂退火)等的熱處理。實際上,已知進行 氫處理的C Ζ ~石々曰貝丨不上 已知進彳丁 的空隙缺陷消失曰f i ::::為LSTD等檢測出的引起空孔 特公平3-80338號公報^優良的氧化膜耐壓特性(曰本專利 ® L、“二ϋ存在著氫氣熱處理的效果僅限於晶圓的最表 的缺^4、’所以著眼於缺陷尺寸越小,用氫氣熱處理 :的溫度,輸Ρ速度加快,使缺陷尺寸細微= ^上,仃氫氣熱處理,使該氫氣熱處理的效果可達到更深 部的提案(、日本專利特開平1 0-2089 87號公報)。 此外’為了使該方法適應結晶直徑的增大,提出了使支 配引起缺陷的點缺陷(空孔)濃度的V/G(V :上拉速度、溶 點附f的結晶軸方向溫度斜度)最佳化,可得到優質的晶 圓的提案(日本專利1〇一26〇666號公報)。 可是’最近,提出了作為關於缺陷尺寸縮小化的其他的 11 I 1 I 1 ! 1 111 i 111 1 1 89117204.ptd ------------ 第4頁 五、發明說明(2) 手段,通過 小,可製造 10-98047 號 【發明所欲 可是,從 晶的長度方 佈。 進而,當 溫退火效果 充分的高溫 總之,通 晶圓,對於 氮的晶圓結 際上,還不 本發明就 氮製造矽晶 圓的製造條 處理後 &lt; 製 圓。 【解決問題 本發明人 究結果’發 的半導體元 發明。 添加氮 ,但是 退火條 過添加 能否適 果能否 清楚。 是鑒於 圓的方 件,上 造作為 在CJ石夕單晶生長時添加氣,使缺陷的尺寸變 適宜退火的晶圓的報告(日本專利特開平 公報)。 解決之問題】 向結晶中添加氮時,由於偏析現象在結 口氣氣/辰度變化,受其影響引起缺陷不均勻分 日守’確實可減少缺陷尺寸,容易得到高 ’在另一方面卻增加了缺陷密度,在不 件下’反而會引起晶圓質量的惡化。 上述的氮(摻入氮)使缺陷尺寸縮小化的 應產品化,尚不能充分地被驗證,摻入 作為半導體元件製造用的晶圓使用,實 以上課題而進行的,其目的在於在摻入 法中’驗證可製造具有充分特性的矽晶 述石夕晶圓可作為半導體元件使用,經熱 半導體元件用的具有適宜特性的摻氮晶 之手段】 鑒於上述問題,對於生長條件進行仔細研 件、^在進行熱處理後也可製造具有尖端技術 ^ 特性的矽晶圓製造條件,從而完成了本V. Description of the invention (1) [Technical field to which the invention belongs] The subject matter relates to a non-aqueous gas treatment such as a dip-hydrogen heat treatment; for example, a lice atmosphere high-temperature heat treatment (referred to as a manufacturing method of silicon wafers). [Known Techniques] for Manufacturing Heat-Treatment Semiconductor Elements In the Chua's method (hereinafter, shaking rabbit circle), there are, for example, Shi Xi wafers (cz — void gap defects in Shi Xi crystal pores) manufactured by the method. In order to remove the airborne noise detected by two FPDs and cop, in order to remove the heat treatment such as bad quality treatment or wind-annealing) on the quality of the final product. In fact, the C Zn ~ Shizhuangbei known to undergo hydrogen treatment is not as good as the void defect that is known to be introduced. Fi :::: It is caused by hollow holes detected by LSTD etc. ^ Excellent withstand voltage characteristics of the oxide film ("Patent ® L", "The effect of hydrogen heat treatment on the second electrode is limited to the most apparent defects of the wafer ^ 4, 'so focusing on the smaller the size of the defect, the use of hydrogen heat treatment: The temperature and the P-transmission speed are accelerated, so that the defect size is fine. The hydrogen heat treatment can make the effect of the hydrogen heat treatment reach deeper (Japanese Patent Laid-Open No. 10-2089 87). This method adapts to the increase in crystal diameter, and proposes to optimize the V / G (V: pull-up speed, temperature gradient of the melting point with f) of the concentration of point defects (voids) that dominate the defect, Proposal to obtain a high-quality wafer (Japanese Patent No. 10-2660666). However, recently, 11 I 1 I 1! 1 111 i 111 1 1 89117204 has been proposed as a reduction in defect size. ptd ------------ Page 4 V. Invention Description 2) By means of small, it is possible to manufacture No. 10-98047 [as long as the invention wants, but the length of the crystal is square cloth. Furthermore, when the temperature annealing effect is sufficient and the high temperature is short, the wafer is passed, and for the nitrogen wafer junction, The present invention does not yet process the manufacturing of silicon wafer wafers after nitrogen processing <circle making. [Solving the problem The present inventors have investigated the invention of the semiconductor element. Adding nitrogen, but whether the annealing strips are suitable for addition is clear whether the results are clear. In view of the rounded square pieces, Sangsho reports as a wafer that adds gas during the growth of CJ Shiyu single crystals to make the size of defects suitable for annealing (Japanese Patent Laid-Open Publication). Problems to be solved] Adding to the crystal When nitrogen is used, due to the segregation phenomenon in the atmosphere gas / temperature change, affected by the uneven defect caused by the daybreaking "really reduce the size of defects, easy to get high" on the other hand increased the density of defects, under no circumstances 'On the contrary, it will cause the deterioration of wafer quality. The above-mentioned nitrogen (doped with nitrogen) has reduced the size of defects, and cannot be fully verified. The use of wafers for the manufacture of components is carried out in accordance with the above issues. The purpose is to verify that the silicon crystal wafers with sufficient characteristics can be used as semiconductor elements in the incorporation method. Means of nitrogen-doped crystals with suitable characteristics] In view of the above-mentioned problems, carefully studied the growth conditions, and ^ can also produce silicon wafer manufacturing conditions with cutting-edge technology ^ characteristics after heat treatment.

五、發明說明(3) ---—- ^具體地說,本發明者們,用對於摻入氮的晶圓(摻氮 的二圓)測定閘極氧化膜耐壓特性,評估晶圓特性,但是 在該過程中,發現摻氮晶圓與在非氧化性氛圍下進行敎疋處 理後不進行摻氮的晶圓比較,TZDB(Time Zer〇 …、V. Description of the invention (3) ------ ^ Specifically, the inventors used the nitrogen doped wafer (nitrogen doped two circles) to measure the withstand voltage characteristics of the gate oxide film and evaluate the wafer characteristics. However, in this process, it was found that nitrogen-doped wafers were compared with wafers that were not doped with nitrogen after erbium treatment in a non-oxidizing atmosphere. TZDB (Time Zer0 ...,

Dielectric Breakd〇wn)試驗的結果趨於良好,但在高噥 度地摻氮時,TDDB(Time Dependent Dielectric &quot; ^ Breakdown)試驗的結果,出現異常。 =外,本發明者們得到了如下的見解,即TZDB試驗的結 果疋,良好,T D D B試驗的結果是否出現異常,都與摻氮晶 圓的氮濃度有關,但同時,TZDB試驗的結果,根據進行= =化性熱處理的氣體種類(例如,氫氣或氬氣)而不同,另 一方面,可推斷TDDB試驗的結果發現異常的氮濃度, 嫩性熱處理的氣體種類有關,而大致一定,又不,、 這二見解對於完成本發明有極大的貢獻,本發明者在本 2 ^出在非氧化性氛圍下熱處理後含有成為出現異 吊 乂馱結果的濃度的4 X 1 014原子/cm3以下濃度的氮的 ,氮晶圓,在作為半導體元件製造用的非氧化性熱處= 矽^圓是適宜的,基本上將其作為申請專利範圍的内容。 在通過閘極氧化膜耐壓良品率(G〇〇評估摻氮晶圓時, :月在表面上顯示充分的氫熱處理效果,但其效果與摻氮 =無關,在對某種程度的深度部分評估時,氧化膜耐壓良 品率=於氮濃度有依存性,對摻氮量有上限和下限,在作 為半‘體元件用製品使用時,氮濃度必須在規定的範圍 内。對於完成本發明,該見解也作出極大的貢獻。The results of the Dielectric Breakdwn test are tending to be good, but when nitrogen is added at a high degree, the results of the TDDB (Time Dependent Dielectric &quot; ^ Breakdown) test are abnormal. In addition, the present inventors have obtained the following insights: the results of the TZDB test are good, good, and whether the results of the TDDB test are abnormal are related to the nitrogen concentration of the nitrogen-doped wafer. The type of gas (for example, hydrogen or argon) that undergoes chemical heat treatment varies. On the other hand, it can be inferred that the results of the TDDB test found abnormal nitrogen concentrations, and the type of gas for the tender heat treatment is related, but it is roughly constant without These two insights have greatly contributed to the completion of the present invention. In the present invention, after the heat treatment in a non-oxidizing atmosphere, the present inventors have a concentration of 4 X 1 014 atoms / cm3 or less, which is the concentration at which the results of heterogeneous contamination occur. Nitrogen and nitrogen wafers are suitable in the non-oxidizing heat place used for semiconductor device manufacturing = silicon wafers, which are basically included in the scope of patent application. When evaluating the nitrogen-doped wafer by the gate oxide film withstand voltage yield (GOO), the month shows a sufficient hydrogen heat treatment effect on the surface, but its effect has nothing to do with nitrogen doping =. During the evaluation, the yield rate of the oxide film pressure is dependent on the nitrogen concentration, and there is an upper limit and a lower limit on the amount of nitrogen doping. When used as a product for semi-solid components, the nitrogen concentration must be within the specified range. For the completion of the present invention This insight has also contributed greatly.

486745 五、發明說明(4) '---486745 V. Description of the invention (4) '---

另外,在上述的日本專利1 〇 — 9 8 0 4 7號公報中,有”氮濃 度至少是1 XI 0“原子/cm3&quot;之說,但對其依據沒有任 開。 A 更具體地,本發明提供了以下的方法及半導體元件製造 用的非氧化性熱處理用矽晶圓。 k (1)半導體元件製造用的非氧化性熱處理用石夕晶圓,其 氮濃度在5 X 1 〇13原子/cm3〜1 X 1 〇15原子/cm3的範圍内、優 選的是在5 X1013原子/cm3〜8 X1014原子/cm3的範圍内、更 優選的是在5 X 1 〇13原子/cm3〜4 X 1 Ο&quot;原子/cm3的範圍内、 隶優選的是在1 X 1 〇14原子/cm3〜4 X 1 原子/cm3的範圍 内。 即’在僅考慮T Z D B的閘極氧化膜财壓時,作為良品的範 圍,其氮濃度在5 X 1013原子/cm3〜1 X 1015原子/cm3的範圍 内(T Z D B的閘極氧化膜耐壓良品率在9 〇 %以上的範圍内)、 優選的是其氮濃度在1 X 1 0&quot;原子/cm3〜8 X 1 〇&quot;原子/cm3的 範圍内(TZDB的閘極氧化膜耐壓良品率在95%以上的範圍 内)’此外’在考慮定電流TDDB時,必須是4 X 1 〇14原子 /cm3以下。因此,其結果,良品的範圍分四個階段,即5 X1013原子/cm3〜1 X1015原子/cm3的範圍内、5 χΐ〇ΐ3原子 /cm3〜8 X 1 014原子/cm3的範圍内、5 X 1 〇ΐ3原子/cm3〜4 X 1014原子/cm3的範圍内、1 X ι〇ΐ4原子/cm3〜4 X 1〇14原子/cm3 的範圍内,它們可根據要製造的產品適宜選擇。 另外,作為熱處理用矽晶圓的種類,有在氫氣氛圍下用 於熱處理的氫氣熱處理用矽晶圓、在氬氣氛圍下用於熱處In addition, in the aforementioned Japanese Patent Publication No. 10 to 98 0 47, there is a statement that "the nitrogen concentration is at least 1 XI 0" atoms / cm3 &quot;, but there is no basis for it. A More specifically, the present invention provides the following method and a silicon wafer for non-oxidizing heat treatment for manufacturing a semiconductor device. k (1) A non-oxidative heat treatment stone wafer for semiconductor device manufacturing, whose nitrogen concentration is in the range of 5 X 1 〇13 atoms / cm3 to 1 X 1 〇15 atoms / cm3, preferably 5 X 1013 In the range of atoms / cm3 to 8 X1014 atoms / cm3, more preferably in the range of 5 X 1 〇13 atoms / cm3 to 4 X 1 0 &quot; atoms / cm3, preferably in the range of 1 X 1 〇14 atoms / cm3 ~ 4 X 1 atoms / cm3. That is, when only considering the gate oxide film pressure of TZDB, the nitrogen concentration in the range of good products is in the range of 5 X 1013 atoms / cm3 to 1 X 1015 atoms / cm3 (TZDB gate oxide film withstand voltage good products) Rate is in the range of 90% or more), and its nitrogen concentration is preferably in the range of 1 X 1 0 &quot; atoms / cm3 ~ 8X1 〇 &quot; atoms / cm3 (TZDB gate oxide film withstand voltage yield rate In the range of 95% or more) In addition, when considering the constant current TDDB, it must be 4 X 1014 atoms / cm3 or less. Therefore, as a result, the range of good products is divided into four stages, that is, 5 X1013 atoms / cm3 to 1 X1015 atoms / cm3, 5 χΐ〇53 atoms / cm3 to 8 X 1 014 atoms / cm3, 5 X In the range of 1 × 3 atom / cm3 to 4 X 1014 atoms / cm3, and in the range of 1 × 4 atom / cm3 to 4 X 1014 atoms / cm3, they can be appropriately selected according to the product to be manufactured. In addition, as the types of silicon wafers for heat treatment, there are silicon wafers for hydrogen heat treatment which are used for heat treatment in a hydrogen atmosphere, and heat treatment under argon atmosphere.

89117204.ptd 第7頁 五、發明說明(5) ί 用矽晶圓、或在氣氣和氬氣的混合氣體的 少曰ΐ ΐ Ϊ 的混合氣體熱處理用石夕晶圓等,為了減 的層的結晶缺陷的非氧化性加熱處理(無氧狀態下 火處理)全包括在本發明的範圍内。 (j)在用蔡氏法上拉單晶矽製造矽晶棒的方法中,在摻 簌ίί ’形成氮濃度是5X1 〇13原子/CIfl3〜4X1 014原子/cm3 1件下,上拉單晶石夕製造石夕晶棒的方法。特別是在 :,法上拉單晶矽製造矽晶棒的方法中,在摻入氮後, 件Ϊ乳濃度ΐ1 X 1 〇14原子/Cm3〜4 X 1 014原子W範圍的條 曰,上拉皁晶矽製造非氧化性熱處理用矽晶圓製作用石夕 曰曰棒的方法。 ^為本發明的一個實驗例,上述的半導體元件製造用的 理用矽晶圓是用蔡氏法(CZ法)製造的。此時,用蔡氏 沄'广氮後,使其一部分或全體的氮濃度成為5 乂“^〜工 10原子/cm3,摻入後上拉單晶矽,製造矽晶棒,從該 矽晶棒切出氮濃度為5 ^(^〜丄xl〇15原子/cm3的部分,優 選的是1 Xl〇“〜8xl0“原子/cm3的範圍部分,作成半導體 凡件製造用的非氧化性熱處理用矽晶圓,特別是作成氫氣 熱1理用矽晶圓或氬氣退火用矽晶圓。此外,在使用CZ方 法日守也可以採用在炫融液中加入磁場的方式(M c Z法)。 —^入氮的方法’可以使用在結晶生長時,向通入爐内的 氬氣2混入氮的方法、在原料熔融液中溶解氮化矽後,在 上拉單晶中導入氮原子的方法等現在公知的所有方法以及 將來可發現的一切方法。89117204.ptd Page 7 V. Description of the invention (5) ί Use silicon wafers, or 石 气 Ϊ mixed gas heat treatment Shi Xi wafers, etc. in order to reduce the layers The non-oxidizing heat treatment (fire treatment in the absence of oxygen) of crystal defects is all included in the scope of the present invention. (j) In the method of manufacturing silicon ingots by pulling up single crystal silicon using the Chua's method, single crystals are pulled up under the condition that the concentration of nitrogen is 5X1 〇13 atoms / CIfl3 ~ 4X1 014 atoms / cm3. Xi manufacturing method of Shi Xi crystal rod. In particular, in the method of manufacturing silicon ingots by pulling up single-crystal silicon, after doping nitrogen, the pieces of milk concentration ΐ1 X 1 〇14 atoms / Cm3 ~ 4 X 1 014 atoms W range, The method of producing a silicon wafer for non-oxidizing heat treatment by drawing soap crystal silicon is a good method. ^ This is an experimental example of the present invention. The above-mentioned physical silicon wafer for semiconductor device manufacturing is manufactured by the Chua method (CZ method). At this time, after using Chua's Nitrogen, the nitrogen concentration of a part or the whole was made to be 5 乂 "~~ 10 atoms / cm3, and after doping, single crystal silicon was pulled up to produce a silicon ingot. From this silicon ingot, A portion having a nitrogen concentration of 5 ^ (^ ~ 丄 x1015 atoms / cm3, preferably in a range of 1 × 10 "to 8xl0" atoms / cm3 is cut out, and silicon for non-oxidizing heat treatment for manufacturing semiconductor parts is prepared. Wafers, especially silicon wafers for hydrogen thermal treatment or silicon wafers for argon annealing. In addition, the CZ method can also be used to add magnetic fields to the melt solution (M c Z method). — ^ The method of introducing nitrogen can be a method of mixing nitrogen into the argon gas 2 introduced into the furnace during crystal growth, a method of dissolving silicon nitride in a raw material melt, and introducing a nitrogen atom into a pull-up single crystal. All the methods now known and all methods that will be discovered in the future.

89117204.ptd 第8頁 486745 五 發明說明(6) 作為本發明之實施態樣,可以舉出以下的方法。 對於本發明關於的半導體元件製造用的非氧 用的石夕晶圓(上述的非氧化性熱處理用的石夕晶圓),ς處理 :熱處理或者1氣退火而t★的半導體元件 二氫 圓。 J石夕晶 :慮假設的兀件菁命’調整摻入氮量的半導體元 用的矽晶圓。 1卞衣造 摻入氮的晶圓評估方法,其特徵是通過測定 ί理晶圓上的假設的元件壽♦,進行判定摻入該氮Π圓 疋否可以作為半導體元件製造用的晶圓使用。 -:?上述曰曰圓的方法’其特徵是測定上述晶圓的假設的 兀件哥命方法,是TDDB試驗法。 w 7版又 y假設的元件”是指進行tddb 的:比的元件結構,所謂”假設的 :二:89117204.ptd Page 8 486745 V. Description of the Invention (6) As an embodiment of the present invention, the following methods can be cited. For the non-oxygen-containing Shi Xi wafers for the manufacturing of semiconductor elements according to the present invention (the aforementioned Shi Xi wafers for non-oxidizing heat treatment), the heat treatment or the 1-gas annealing of the semiconductor device . J Shi Xijing: Considering the hypothesis of the components, silicon wafers used to adjust the amount of nitrogen doped semiconductor elements. 1. A method for evaluating a wafer doped with nitrogen, which is characterized by measuring the assumed element lifetime on a physical wafer and determining whether the nitrogen doped can be used as a wafer for semiconductor device manufacturing. . -:? The above-mentioned round method 'is characterized by the TDDB test method, which is a method for measuring the hypothesis of the above wafers. w 7th edition and y hypothetical element "refers to the element structure of tddb: ratio, the so-called" hypothetical: two:

兀件結構的壽命。 了 ,疋相邊類比日J 【實驗例】 」:為實驗例’本發明人等對於從各種 :曰“夕切出石夕晶圓’進行鏡面研磨 &lt;單晶之生長〉度方向的缺陷效果進行了調查。 結晶是以作為摻入劑 口 曰曰 方位&lt;10 0〉、左右缺P :加了爛的直扭200_、P型、、1 〇.13〜〇.4_2/1^11。(:'、〇£3 因素的口^及¥&gt;^2分別是 的(v :上拉速度、Gl .囡、:0 c/min的範圍條件下生長 丄.固液介面附近的溫度斜率、G2:;The life of the component structure. [Experimental Example]: As an experimental example, the present inventors etc. performed mirror polishing from various kinds of: "Xi cut out Shixi wafer" &lt; growth of single crystal> degree defect effect The investigation was carried out. Crystallization was based on the orientation of the dopant, &lt; 10 0 &gt;, left and right P: rotten straight twist 200_, P type, 1 10.13 ~ 〇.4_2 / 1 ^ 11. (: ', 〇 £ 3 factors, ^ and ¥ &gt; ^ 2 are respectively (v: pull-up speed, Gl. 囡,: growth in the range of 0 c / min 丄. Temperature slope near the solid-liquid interface G2 :;

486745486745

&lt;氮之摻入&gt; X 1013 加氮使氮濃度達到4 9 使用不添加氮的結晶。 #為比較,也 ,的添加是通過將氮氣導入CZ爐内的方法和將在表 &lt;氮濃度之定量&gt; —氮濃度的定量是通過用SIMS對於可用SIMS測定的部分 行測定來定量的,對於不能用s I MS測定的部分可通過^ s I MS測定過的部分用規定的計算公式算出來定量的^仗 更具體地,用SIMS(Cameca制、IMF-6F型)對於具有比 氮的定量檢測下限值(〜1 XI 〇14原子/cm3)充分大的/濃度的 結晶體最尾部(固化率約90%進行定量。 彳又 此時,以未摻氮的F Z法生長的石夕結晶作為對比,確巧、對 於k進入試樣更換時因大氣中的殘留氮氣對檢測值沒有干 擾。用已知的標準試樣校正通過將離子注入氮氣形成的^ 氣濃度進行定量。 通過以下的計算公式計算出在不能用s I MS測定的位置的 氮濃度&lt; Incorporation of nitrogen &gt; X 1013 Nitrogen was added to achieve a nitrogen concentration of 4 9. A crystal was used without adding nitrogen. #For comparison, also, the addition is by introducing nitrogen into the CZ furnace and the table &lt; quantitative nitrogen concentration &gt;-the quantitative nitrogen concentration is quantified by using SIMS to measure a part of the available SIMS measurements For the part that cannot be measured by s I MS, it can be quantified by ^ s I MS measured by using a predetermined calculation formula. More specifically, SIMS (made by Cameca, IMF-6F type) has a specific nitrogen content. Quantitative detection lower limit value (~ 1 XI 〇14atoms / cm3) of the most large / concentrated crystal (the solidification rate is about 90%) is quantified. 彳 At this time, the stone grown by the FZ method without nitrogen doping. The crystallization is for comparison. It is true that there is no interference with the detection value due to the residual nitrogen in the atmosphere when k enters the sample replacement. The known standard sample is used to correct the concentration of ^ gas formed by implanting ions into nitrogen. Calculation formula to calculate nitrogen concentration at locations that cannot be measured by s I MS

Cs=kCo(l -L)h C s ·結晶中的雜貪 &gt;辰度、C 〇 ·初期的k 液中的雜質、農 度、L :固化率、K :在本技術領域經常使用的氮的偏析係Cs = kCo (l -L) h C s · Impurities in crystals> Chen degree, C 0 · Impurities in the initial k liquid, agronomy, L: curing rate, K: often used in the technical field Nitrogen segregation

89117204.ptd 第10頁 486745 五、發明說明(8) 數,Yatsurugi, et· al,從(1973)J. Electrochem· s〇c 120.975·的文獻取0.0007。 上述的計算公式是使用雜質固有的偏析係數K表示熔融 液矽中的氮氣雜質進入結晶的濃度。 上述的公式顯示偏析現象的公式。由於偏析現象,纟士曰 、、口 曰曰 生長的同時,雜質濃度(氮濃度)增加,在結晶體最尾部達 到高濃度,超過S I MS氮的定量檢測下限值(〜1 X 1 〇14原子 /cm3) ° 由此,在本實施例中,在結晶體最尾部(固化率約9〇%用 SIMS計測氮濃度,使用該值從上式計算出各結晶位置的濃 &lt;晶圓之熱處理&gt; 對,這樣摻氮定量的晶圓,施以氫氣熱處理(氫氣退火) 退火。氫氣退火條件及氬氣退火,是進行一般的 火後的二^小時Λ處理,y分別在岡彳氫氣熱處理後及氬氣退 後除去表:3和上=Ϊ Ϊ除去表層3 # m的狀態及氬氣退火 (G0I)。層3 V&quot;1的狀悲下,研究閘極氧化膜耐壓特性 &lt;風氣熱處理(氫氣退火)晶圓之評估〉 L間極氧化膜耐壓(G0I)] 閑極氧化膜耐壓是在晶圓上形, ^ ^ ^ L1 ν,Λ Λ 1 時,α曰间士 所原d # m後的閘極氧化膜耐壓 曰曰圓表面研磨3 /z m程度後,在豆上报# 加電壓進行測定。 X傻在其上形成M0S結構,89117204.ptd Page 10 486745 V. Description of the invention (8) The number, Yatsurugi, et. Al., 0.0007 from the literature of (1973) J. Electrochem. Soc 120.975. The above calculation formula uses the intrinsic segregation coefficient K of the impurities to indicate the concentration of nitrogen impurities in the molten silicon entering the crystal. The above formula shows the formula of the segregation phenomenon. Due to the segregation phenomenon, the concentration of impurities (nitrogen concentration) increased at the same time as the growth of the priests and mouths, reaching a high concentration at the end of the crystal, exceeding the lower limit of quantitative detection of nitrogen by SI MS (~ 1 X 1 〇14 atom / cm3) ° Therefore, in this example, the nitrogen concentration was measured by SIMS at the tailmost part of the crystal (the curing rate was about 90%), and the concentration of each crystal position was calculated from the above formula &lt; The heat treatment of the wafer &gt; Yes, such a nitrogen-dosed wafer is subjected to hydrogen heat treatment (hydrogen annealing) annealing. Hydrogen annealing conditions and argon annealing are performed after two hours of ordinary fire Λ treatment. And argon gas removal table: 3 and above = Ϊ Ϊ Remove the surface layer 3 # m state and argon annealing (G0I). Under the condition of layer 3 V &quot; 1, study the pressure resistance characteristics of the gate oxide film &lt; air Evaluation of heat-treated (hydrogen-annealed) wafers> Withstand voltage of the inter-electrode oxide film (G0I)] The withstand voltage of the idler oxide film is formed on the wafer. When ^ ^ ^ L1 ν, Λ Λ 1, α After the original d # m gate oxide film withstand voltage is about 3 / zm after round surface grinding, # Packets applied voltage was measured. X Pink M0S structure formed thereon,

486745 五、發明說明(9) [TZDB(Time Zero Dielectric Breakd〇wn)試驗] 首先,在用TZDB測定閘極氧化膜耐壓時,對於用該實驗 例採用的測定方法,作為刪極採用i 〇mm2的聚矽電極,用 施加分步電壓法加電壓。氧化膜厚是25nm。進而測定溫度 是室溫(2 5 °C )、判定耐壓電流是丨〇 # a。 其結果’如圖1所示,剛氫氣處理後的氧化膜耐壓良品 率,=與氮濃度和生長條件有關,幾乎為1〇〇%。 ”可是,當進行氫氣處理除去表層3//11]後,檢查氧化膜耐 壓時,如圖2所不表明,氮濃度增加到6 χ j 〇u原子/cm3的 同時,良品率增加,但若達到其以上,良品率反而降低。 因此,若隨著元件的高集成化,考慮增加重要度的晶圓 表層(元件活性層)的完全性,從表層3 “η深度的氧化膜耐 壓良品率至少必須維持在9 〇 %。 尸k圖2表明’為了將氧化膜耐壓良品率維持在9 〇 %以上, 氮濃度必須在5x 1(F〜丨x 1〇15原子/cm3的範圍内,為了達 到產Π 口的更彳笑良的氧化膜耐壓良品率在g $ %以上,必須將 氮濃度設定在1 X 1〇Η〜8 χ 1〇14原子/cm3的範圍内。 [定電流TDDB(Time Dependent Dielectric Breakdown)試 驗] =電流TDDB是通過加在元件上的電壓變化可以看到經過 規,的時間後氧化膜的破壞以使電流保持一定。測定時, 與氧化膜耐壓的檢測相同地形成M〇s結構。 測定時’對於用實驗例採用的測定方法,作為電極,採 用1mm2的聚矽電極。氧化膜厚是25nm。進而,測定溫度是486745 V. Description of the invention (9) [TZDB (Time Zero Dielectric Breakdwn) test] First, when the gate oxide film withstand voltage is measured with TZDB, the measurement method used in this experimental example uses i 〇 as the delete pole. A polysilicon electrode of mm2 is applied with a step voltage method. The oxide film thickness is 25 nm. Furthermore, the measurement temperature was room temperature (25 ° C), and it was determined that the withstand current was 丨 〇 # a. As a result, as shown in Fig. 1, the yield rate of the oxide film immediately after the hydrogen treatment was almost 100% depending on the nitrogen concentration and growth conditions. "However, when the hydrogen layer was removed to remove the surface layer 3 // 11], and the pressure resistance of the oxide film was checked, as shown in Fig. 2, the nitrogen concentration increased to 6 x j ou atoms / cm3, and the yield rate increased, but If it is more than this, the yield rate will decrease. Therefore, if the integration of the wafer surface layer (element active layer) with increasing importance is considered with the high integration of the device, the oxide film withstand voltage from the surface layer 3 "η depth is good. The rate must be maintained at least 90%. Figure 2 shows that 'in order to maintain the good yield rate of the oxide film above 90%, the nitrogen concentration must be in the range of 5 × 1 (F ~ 丨 x1015 atoms / cm3. The good yield rate of Xiaoliang's oxide film is more than g $%, and the nitrogen concentration must be set within the range of 1 X 1OΗ to 8 x 1014 atoms / cm3. [Constant current TDDB (Time Dependent Dielectric Breakdown) test ] = Current TDDB is the change in the voltage applied to the device. It can be seen that the oxide film is destroyed after a certain period of time to keep the current constant. During the measurement, the MOS structure is formed in the same way as the test of the withstand voltage of the oxide film. At the time of measurement, for the measurement method used in the experimental example, a polysilicon electrode of 1 mm2 was used as the electrode. The thickness of the oxide film was 25 nm. Furthermore, the measurement temperature was

486745 、發明說明(1 〇) 125 °C ’外加電流密度是5〇mA/cm2,判定電場是4MV/cm。 在/則疋疋電流T D D B時’正常元件超過1 〇 〇秒而產生瞬間 破壞。可是,對於異常元件時,不產生瞬間破壞,而產生 經時的破壞,所以從丨〇〇秒前緩慢地破壞。為此,通過判 定電場的設定值,其結果,看到在丨〇 〇秒後也產生破壞。 為此’若以定電流TDDB的測定結果作圖,對於正常元件, 在經過一定時間(此時,大致是丨〇〇秒)處標繪點密集,在 其以外,沒有標繪點(參照圖3(A)),在異常元件時,在其 以外經過1 0 0秒離開的場所也打出標繪點(參照圖3 ( b ))。 在熱處理晶圓中,發生這樣的異常,在未摻氮的晶圓熱 處理時是沒有看到的(參照圖3(A)),對於高濃度摻氮的晶 圓進行熱處理時是表現的特有的現象。 另外’ TDDB试驗是半導體元件的可靠性試驗的一種,是 對兀件疲勞的耐性的常規性的指標。而且,TDDB試驗也可 為推測由作為試驗物件的晶圓製造的半導體元件壽命的穩 定性的假設資料。 在該試驗例中,對氫氣熱處理後的各晶圓,測定未研磨 狀態的定電流T D D B和研磨3 a m後的定電流τ D D B。 其結果’付到了下述表中所示的資料。486745, description of the invention (10) 125 ° C 'The applied current density is 50 mA / cm2, and the electric field is determined to be 4 MV / cm. At the current T D D B, the normal element exceeds 1000 seconds and instantaneous damage occurs. However, in the case of an abnormal element, instantaneous damage does not occur, but aging occurs over time. Therefore, it is slowly destroyed from 1000 seconds ago. For this reason, by determining the set value of the electric field, as a result, it was seen that damage also occurred after 1000 seconds. For this reason, if the measurement result of constant current TDDB is plotted, for normal components, the plotted points are dense at a certain time (at this time, approximately 丨 00 seconds), and there are no plotted points outside (refer to the figure) 3 (A)), when a component is abnormal, a plot point is also drawn at a place that has left for 100 seconds (see FIG. 3 (b)). Such anomalies occur in heat-treated wafers, which are not seen during heat-treatment of wafers that are not doped with nitrogen (see Figure 3 (A)). They are unique to heat-treated wafers that are highly doped with nitrogen. phenomenon. In addition, the TDDB test is a type of reliability test of a semiconductor device, and is a general indicator of fatigue resistance of a component. In addition, the TDDB test may be hypothetical data for estimating the stability of the lifetime of a semiconductor device manufactured from a wafer as a test object. In this test example, a constant current T D D B in an unpolished state and a constant current τ D D B after being polished for 3 a m were measured for each wafer after the hydrogen heat treatment. As a result, the information shown in the following table was paid.

89117204.ptd 48674589117204.ptd 486745

89117204.ptd89117204.ptd

第14頁 486745 五、發明說明(12) 表2 估項目 TDDB異常率% 氮濃 (原子/cm3) 0 0 0 0 0 0 0 0 5.00E+13 0 5. 10E + 13 0 6.30E+13 0 1. 13E + 14 0 1·41E+14 0 1. 51E+14 0 1. 64E+14 0 4. 97E+14 18. 9 6.48E+14 8.11 8. 61E+14 13. 5 1. 12E+15 10. 8 1.26E+15 8. 11 在此,表1表示未研磨狀態的氮濃度和定電流TDDB異常 率的關係、表2表示研磨3 //m後的氮濃度和定電流TDDB異 常率的關係。如這些表所示,定電流TDDB顯示正常值的是 氮濃度在4 X 1014原子/cm3以下的場合。因此,通過在上述Page 14 486745 V. Description of the invention (12) Table 2 Estimated project TDDB abnormal rate% Nitrogen concentration (atoms / cm3) 0 0 0 0 0 0 0 0 5.00E + 13 0 5. 10E + 13 0 6.30E + 13 0 1. 13E + 14 0 1.41E + 14 0 1. 51E + 14 0 1. 64E + 14 0 4. 97E + 14 18. 9 6.48E + 14 8.11 8. 61E + 14 13. 5 1. 12E + 15 10. 8 1.26E + 15 8. 11 Here, Table 1 shows the relationship between the nitrogen concentration in the unpolished state and the constant current TDDB abnormal rate, and Table 2 shows the nitrogen concentration after grinding 3 // m and the constant current TDDB abnormal rate. relationship. As shown in these tables, the constant current TDDB shows a normal value when the nitrogen concentration is 4 X 1014 atoms / cm3 or less. So through the above

89117204.ptd 第15頁 486745 五、發明說明(13) 的氧化膜耐壓條件加上該必要條件,摻氮的結晶最適宜條 件是氮濃度在5 X 1 013原子/cm3〜4 X 1 〇i4原子/cm3的範圍内 的熱處理用石夕晶圓’特別是,氮濃度在1 X 1 〇 μ原子/ c W〜 4 X 1 Ο14原子/cm3的範圍内的熱處理用石夕晶圓。 &lt;氬氣退火晶圓之評估&gt; [TZDB(Time Zero Dielectric Breakdown)試驗] 用TZDB試驗測定閘極氧化膜耐壓(G〇I)是用與氫氣退火 時相同條件下進行的。只是在氬氣退火時也測定研磨6 後的閘極氧化膜耐壓。其結果如圖4所示。在圖4中,圖 4(A)是分別重復表示未研磨場合、研磨3 “η後的場合、研 磨6//m後的場合,但圖4(B)是為了與表示氫氣退火的,研 磨3 //m後的圖2容易區分差異,特別是只載有研磨3//m後 的場合的資料。 如圖4表明,摻氮的晶圓,比未摻氮的晶圓,ΤΖ])Β試驗 的閘極氧化膜耐壓特性優良,另外.,沒有觀察到氫氣退火 後的,研磨3 // m後的晶圓那樣的明確上限。因此,對於摻 ^的晶圓,氬氣退火的,不存在^讪試驗上氮濃度的上限 f者比作為氫氣退火時上限值的氮濃度1 χ丨〇15原子“Μ還 靠上處存在上限值。進而,如圖4(A)表明,氬氣退火的, 研磨3 後的、研磨6 //m後的,在TZDB試驗的結果上,沒 有多大不同。 [定電流TDDB(Time Dependent Dielectric Breakd〇wn)試 驗] 定電流TDDB試驗也與TZDB試驗時相同地,用與氫氣退火89117204.ptd Page 15 486745 V. Description of the invention (13) The oxide film withstand voltage conditions plus this necessary condition, the most suitable condition for nitrogen-doped crystals is a nitrogen concentration of 5 X 1 013 atoms / cm3 ~ 4 X 1 〇i4 The Shi Xi wafer for heat treatment in the range of atoms / cm3 'is particularly a Shi Xi wafer for heat treatment having a nitrogen concentration in the range of 1 X 10 μatoms / c W to 4 X 1014 atoms / cm3. &lt; Evaluation of argon-annealed wafer &gt; [TZDB (Time Zero Dielectric Breakdown) Test] The gate oxide film withstand voltage (GOi) was measured by the TZDB test under the same conditions as in the case of hydrogen annealing. The breakdown voltage of the gate oxide film after polishing 6 was also measured only during argon annealing. The results are shown in Fig. 4. In FIG. 4, FIG. 4 (A) shows the unpolished case, the case after grinding 3 ”η, and the case after grinding 6 // m, respectively. However, FIG. 4 (B) is for the purpose of annealing with hydrogen, and polishing. Figure 2 after 3 // m is easy to distinguish the difference, especially only the data of the occasion after grinding 3 // m. As shown in Figure 4, nitrogen-doped wafers are better than non-nitrogen-doped wafers, TZ]) The gate oxide film in the BT test has excellent withstand voltage characteristics. In addition, no clear upper limit is obtained for wafers polished after 3 // m after hydrogen annealing. Therefore, for doped wafers, argon annealed There is no upper limit of the nitrogen concentration on the test, and there is an upper limit than the nitrogen concentration 1 × 1515 atom “M” which is the upper limit during hydrogen annealing. Furthermore, as shown in Fig. 4 (A), the results of the TZDB test were not much different between those annealed with argon and those after grinding 3 and 6 // m after grinding. [Constant current TDDB (Time Dependent Dielectric Breakdwn) test] The constant current TDDB test is also annealed with hydrogen as in the TZDB test.

486745 五、發明說明(14) 時相同條件進行。其結果如圖5〜圖7所示。 圖5及圖6顯示從完全不摻氮的狀態(圖5(A)),提高氮濃 度到5 X 1013 原子/cm3(圖 5(B))、1 X 1〇14 原子/cm3 (圖 5(C))、1· 13 X 1014 原子/cm3 (圖 5(D))、1· 64 X 10“ 原子 /cm3(圖 6(A))、5· 82 X 1014 原子/cm3 (圖 6(B))、1· 37 X 1015 原子/cm3 (圖6(C))時的定電流TDDB試驗的結果。 從這些圖表明,到氮濃度1· 64 X 1014原子/cm3 (圖6(A)) 時’標繪點的密集狀態沒有變化,但氮濃度5· 82 X 1014原 子/cm3 (圖6(B))及1· 37 X 1015原子/cm3 (圖6(C))時,在異常 的元件上有特有的標繪點偏移,但可在丨〇 〇秒更後面處看 到。 另外,在(圖6(D))中將以5· 82 X 1014原子/cm3及1. 37 X 1 015原子/cm3的濃度摻氮的晶圓表示分別重複表示氫氣退 火及氬氣退火的晶圓。另外,圖7表示重複表示圖5(A)〜 圖6(C)。 從該結果可認為對於摻氮晶圓進行氬氣退火的,在進行 定電流TDDB試驗的結果上,在氮濃度丨· 64 χ 1 原子 /cm3 (圖6(A))〜5· 82 X ΙΟ&quot;原子/cm3(圖6(bd之間存在上 限’它在與對於氫氣退火的晶圓在進行定電流TDDB試驗的 結果上作為上限的4x 1〇h原子/cm3(圖6(D))的值在近處存 在。 由此表明’對於將摻氮晶圓進行氫氣退火及氬氣退火 的,在TZDB試驗的結果上,可看到大的差異,但在定電 TDDB試驗的結果上,幾乎看不到大差異。486745 V. Invention description (14) is carried out under the same conditions. The results are shown in Figs. 5 to 7. Figures 5 and 6 show that from the state where nitrogen is not added at all (Figure 5 (A)), the nitrogen concentration is increased to 5 X 1013 atoms / cm3 (Figure 5 (B)), 1 X 1014 atoms / cm3 (Figure 5 (C)), 1.13 X 1014 atoms / cm3 (Figure 5 (D)), 1.64 X 10 "atoms / cm3 (Figure 6 (A)), 5.82 X 1014 atoms / cm3 (Figure 6 ( B)), constant current TDDB test results at 1.37 X 1015 atoms / cm3 (Figure 6 (C)). From these figures, it is shown that the nitrogen concentration is 1.64 X 1014 atoms / cm3 (Figure 6 (A) ), There is no change in the dense state of the plotted points, but the nitrogen concentration is 5.82 X 1014 atoms / cm3 (Fig. 6 (B)) and 1.37 X 1015 atoms / cm3 (Fig. 6 (C)). There is a unique plot point offset on the element, but it can be seen later in 丨 00 seconds. In addition, (Fig. 6 (D)) will be 5.82 X 1014 atoms / cm3 and 1. 37 A wafer doped with nitrogen at a concentration of X 1 015 atoms / cm3 represents a wafer that repeatedly shows hydrogen annealing and argon annealing. In addition, FIG. 7 shows that FIG. 5 (A) to FIG. 6 (C) are repeatedly shown. From this result, It is thought that for nitrogen-doped wafers subjected to argon annealing, based on the results of the constant current TDDB test, the nitrogen concentration 丨 · 64 χ 1 atoms / cm3 (Figure 6 (A)) ~ 5.82 X ΙΟ &quot; atoms / cm3 (Fig. 6 (there is an upper limit between 'bd', it is 4x10h atoms which is the upper limit on the results of the constant current TDDB test with a wafer annealed for hydrogen gas) The value of / cm3 (Fig. 6 (D)) exists nearby. This shows that 'for hydrogen annealing and argon annealing of nitrogen-doped wafers, a large difference can be seen in the results of the TZDB test, but In the results of the fixed power TDDB test, almost no big difference was seen.

\\326\2d-\89-ll\89117204.ptd 第17頁 作為半導體元件用的梦晶圓(半導體 ^),必須多考慮TDDB試驗的結果,所以表日=^夕晶 火及氫氣退火的物件的非氧化性熱處理的 曰’圓虱軋退\\ 326 \ 2d- \ 89-ll \ 89117204.ptd Page 17 As a dream wafer for semiconductor devices (semiconductor ^), the results of the TDDB test must be taken into consideration, so the surface date = ^ evening crystal fire and hydrogen annealing Non-oxidative heat treatment of objects

一個都必須是以定電流TDDB試驗你:的α,任何 ,原子/-以下的濃度含氮的。的、、,口果上作為上限的4 X Β_:』Γ;;ΐ時的上拉單晶中的氮濃度的變化 圖。在圖8中’检抽表不以作為置曰 投入她量作為1蚌的m仆玄乍為 塊材料的聚矽材料的 扠 :為犄的固化率,縱軸表示氮濃度。圖中,“、 表示初始濃度2XW時的曲線、(b)表示初始濃度=) 時的曲線、(c)表不初始濃度5 χ1〇13時的曲 從該圖8表明,隨著氮的偏析 妆的 尾側(在圖的上方表示對岸的矽日嫉&amp;极f拉的矽日日棒的 高。 曰棒的形狀)的氮濃度增 如以上說明,本發明可提供在換氮的石夕晶圓巾, 品顯示了優良的晶圓特性。 〜產One must test you with a constant current TDDB: alpha, any, nitrogen / concentration below. Figure 4 shows the change in the nitrogen concentration in the pull-up single crystal when the upper limit is 4 X Β_: "Γ ;; In Fig. 8, the 'sampling table' is not a polysilicon fork made of a piece of material which is put into a volume of 1 mussel as a mussel. Fork: is the solidification rate of the tritium, and the vertical axis represents the nitrogen concentration. In the figure, "" indicates the curve at the initial concentration of 2XW, (b) indicates the curve at the initial concentration =), and (c) indicates the curve at the initial concentration of 5 x 1013. From this figure, it is shown that the segregation with nitrogen As the above description shows, the nitrogen concentration of the tail side of the makeup (the upper part of the figure shows the high value of the silicon rod on the other side of the silicon rod &amp; pole rod). As described above, the present invention can provide Even wafer towels show excellent wafer characteristics.

486745486745

图1 ”、、員示添加氮濃度和氫氣退火後的氧化膜壓 的關係圖。 卩千 ,2顯示添加氮濃度和氫氣退火後研磨除去表層3 // m時 的氧化膜耐壓良品率的關係圖。 —圖3顯不氫氣熱處理後的矽晶圓的定電流TDDB試驗的測 疋結果圖。其中,圖3 (A)顯示未摻氮的正常元件的定電流 T^DB試驗的測定結果圖。圖3(B)顯示高濃度摻氮的表現異 常的元件的定電流TDDB試驗的測定結果圖。 圖4顯示氬氣退火的摻氮晶圓的TZDB試驗的閘極氧化膜 耐壓(GOI)的測定結果圖。在圖4中,圖4(A)是分別重複表 示未研磨時、研磨後、研磨6//m後的圖,圖4(B)為了 與表示氫氣退火後研磨3 # m後的圖2容易區分差異,特別 是只載有研磨3 // m後時的資料。 圖5顯示氬氣退火的摻氮晶圓的定電流TDDB試驗的結果 圖。特別是,圖5(A)〜圖5(D)顯示從完全不摻氮的狀態 (圖5(A)) ’提高氮濃度到5χ 1〇13原子/ cm3 (圖5(B))、ΐχ 1014 原子/cm3 (圖 5C))、1· 13 X 1014 原子/cm3 (圖 5(D))時的定 電流TDDB試驗的結果。 圖6顯示氬氣退火的摻氮晶圓的定電流TDDB試驗的結果 圖。特別是,圖6(A)〜圖6(C)顯示提高氮濃度到1. 64 X 1014 原子/cm3 (圖 6(A))、5· 82 X 1014 原子/cm3(圖 6(B))、 1· 37 X 1015原子/cm3 (圖6(C))時的定電流TDDB試驗的結果 圖。、16XD)顯示將以5. 82 X 1014原子/cm3及1·〜37 X 1〇ΐ5原子 /cm3的濃度的摻氮晶圓,分別重複表示氫氣退火和氬氣退Figure 1 ”shows the relationship between the nitrogen concentration and the oxide film pressure after hydrogen annealing. 卩 1000, 2 shows the nitrogen film concentration and the yield rate of the oxide film when the surface layer is polished and removed after hydrogen annealing 3 // m. Relationship graph. —Figure 3 shows the measurement results of the constant current TDDB test of the silicon wafer without hydrogen heat treatment. Among them, FIG. 3 (A) shows the measurement result of the constant current T ^ DB test of a normal element not doped with nitrogen. Fig. 3 (B) shows a measurement result of a constant current TDDB test of a high-concentration nitrogen-doped element exhibiting abnormality. Fig. 4 shows a gate oxide film withstand voltage (GOI) of the TZDB test of an argon-annealed nitrogen-doped wafer. Figure 4 shows the results of the measurement. In Figure 4, Figure 4 (A) is a graph showing the unpolished, polished, and polished 6 // m, respectively. Figure 4 (B) shows the polished 3 # after hydrogen annealing. Figure 2 after m is easy to distinguish the differences, especially the data only after grinding 3 // m. Figure 5 shows the results of constant current TDDB test of argon-annealed nitrogen-doped wafers. In particular, Figure 5 ( (A) ~ Figure 5 (D) shows a state where nitrogen is not doped at all (Figure 5 (A)) 'Increase the nitrogen concentration to 5 x 1013 atoms / cm3 (Figure 5 (B)), ΐχ 1014 atoms / cm3 (Fig. 5C)), 1 · 13 X 1014 atoms / cm3 (Fig. 5 (D)), the results of constant current TDDB test. Fig. 6 shows nitrogen doped with argon annealing. A graph of the results of the constant current TDDB test of the wafer. In particular, Figures 6 (A) to 6 (C) show that the nitrogen concentration was increased to 1.64 X 1014 atoms / cm3 (Figure 6 (A)), 5.82 X 1014 atoms / cm3 (Fig. 6 (B)), 1.37 X 1015 atoms / cm3 (Fig. 6 (C)) constant current TDDB test results. 16XD) shows that 5.82 X 1014 atoms / Nitrogen-doped wafers with a concentration of cm3 and 1 · ~ 37 X 105 atomic atoms / cm3, repeating hydrogen annealing and argon degassing, respectively.

\\326\2d-\89-ll\89117204.ptd 第19頁 486745\\ 326 \ 2d- \ 89-ll \ 89117204.ptd Page 19 486745

\\326\2d-\89-ll\89117204.ptd 第20頁\\ 326 \ 2d- \ 89-ll \ 89117204.ptd page 20

Claims (1)

4ρ?ΐΜ 六、甲請荨无nrii 89117204 土 月 曰 修正 «1. 1, 2 5 修正本 1 · 一種矽晶圓,其係摻有氮且施予非氧化性熱處 其特徵為·· 心!有 上j氮濃度在5 X 1013原子/Cm3〜;[χ 1〇15原子/cn]3範圍内。 其特i ^晶圓,其係摻有氮且料非氧化性熱處理者, 上述氮濃度在5x 1〇13原子/cm3〜4χ 1〇14原子/cm3範 献^申請專利範圍第1或2項之矽晶圓,其中上述非氧化 性”、、處理係為氫熱處理或是氬氣退火處理。 4· 一種矽晶圓,其特徵為: y =實施TDDB試驗法或了議試驗法時所製作 几件令命而調整氮的摻雜量。 版。又的 ^ 一種石夕晶圓之評價方法,其特徵為: 藉由測量在經摻氮的熱處理矽晶 τ_式驗法時所製作之假設的= = 驗法或 的石夕:曰曰j是否可作為半導體元件製造用來:用亥亂 ,·”請專利範圍第5項之石夕晶圓之評價方Π吏用。 =i ί=實施TDDB試驗法或tzdb試驗法時所製作之伊 δ又的疋件哥命之方法係為TDDBh驗法。 叮衣作之假 所7製:Π;:;;:!:牛i係包含有:於利用蔡氏法 晶矽施以非氧化性熱處;之“ :雜氮的單 熱處理之單晶石夕⑪予上述非氧化性 平日日y衣k矽晶圓者,其特徵為: ㈣原偏〜1χ1〇15原子/cm3範圍内。4ρ? Ϊ́Μ 6. Please ask for nrii 89117204 Tomotsuki correction «1. 1, 2 5 Revision 1 · A silicon wafer, which is doped with nitrogen and treated with non-oxidizing heat. Its characteristics are ... !! The nitrogen concentration is in the range of 5 X 1013 atoms / Cm3 ~; [χ 1015 atoms / cn] 3. Its special wafer, which is doped with nitrogen and non-oxidative heat treatment, has a nitrogen concentration of 5 x 1013 atoms / cm3 to 4 x 1014 atoms / cm3. The scope of patent application is 1 or 2 Silicon wafer, in which the above-mentioned non-oxidizing, and the treatment system is hydrogen heat treatment or argon annealing treatment. 4. A silicon wafer characterized by: y = produced when the TDDB test method or the test method is implemented Several orders are ordered to adjust the amount of nitrogen doped. Edition. Another ^ An evaluation method of Shi Xi wafers, characterized by: Measured by measuring the nitrogen-doped heat-treated silicon crystal τ_ test method Hypothetical = = Verification method or Shi Xi: whether or not j can be used for the manufacture of semiconductor components: using Hainan, "" please use the appraisal party of Shi Xi wafers in the scope of patent No. 5. = i ί = The method of δ δ and the elder brother's life made when implementing the TDDB test method or the tzdb test method is the TDBdh test method. Ding Yi Zuo Zhi Zhi 7 system: Π;: ;;:!: The bovine i system includes: applying non-oxidizing heat to the crystal using the Chua's method; ": single-crystal spar with heterothermal treatment of nitrogen. Those who donate the above-mentioned non-oxidizing weekday silicon wafers are characterized by: Sugawara is within a range of ~ 1 × 1015 atoms / cm3.
TW89117204A 1999-08-27 2000-08-25 Silicon wafer and method of valuating silicon wafer and method of manufacturing silicon wafer TW486745B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP24118599A JP3479001B2 (en) 1999-08-27 1999-08-27 Silicon wafer manufacturing method
JP24118699 1999-08-27

Publications (1)

Publication Number Publication Date
TW486745B true TW486745B (en) 2002-05-11

Family

ID=26535136

Family Applications (1)

Application Number Title Priority Date Filing Date
TW89117204A TW486745B (en) 1999-08-27 2000-08-25 Silicon wafer and method of valuating silicon wafer and method of manufacturing silicon wafer

Country Status (1)

Country Link
TW (1) TW486745B (en)

Similar Documents

Publication Publication Date Title
US7344689B2 (en) Silicon wafer for IGBT and method for producing same
TW546422B (en) Method of producing silicon wafer and silicon wafer
TWI539041B (en) Semiconductor wafer composed of monocrystalline silicon and method for producing it
JP5439305B2 (en) Silicon substrate manufacturing method and silicon substrate
KR20150118037A (en) Method for heat treatment of silicon wafer and silicon wafer
US7875116B2 (en) Silicon single crystal producing method, annealed wafer, and method of producing annealed wafer
TWI534310B (en) Silicon substrate manufacturing method and silicon substrate
JP2019004173A (en) Production of high precipitate density wafers by activation of inactive oxygen precipitate nuclei by heat treatment
US20100052103A1 (en) Silicon wafer and method for producing the same
CN103237930B (en) Manufacture the method for annealed wafer
TW201000690A (en) Silicon single crystal wafer, method for fabricating silicon single crystal or method for fabricating silicon single crystal wafer, and semiconductor device
TW486745B (en) Silicon wafer and method of valuating silicon wafer and method of manufacturing silicon wafer
KR100720659B1 (en) Silicon wafer and method for manufacture thereof, and method for evaluation of silicon wafer
JP2000026196A (en) Silicon semiconductor substrate and its production
JP3479001B2 (en) Silicon wafer manufacturing method
US8460463B2 (en) Silicon wafer and method for producing the same
TW200406848A (en) Method of eliminating boron contamination of annealed wafer
JP4951927B2 (en) Method for selecting silicon wafer and method for manufacturing annealed wafer
JP2001274165A (en) Silicon semiconductor substrate
JP2820110B2 (en) Semiconductor manufacturing method
JP2000195868A (en) Silicon wafer and heat treatment thereof
JP5434239B2 (en) Silicon wafer manufacturing method
JP2002134514A (en) Silicon wafer and its manufacturing method

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MK4A Expiration of patent term of an invention patent