TW200908120A - Manufacturing method of semiconductor device - Google Patents

Manufacturing method of semiconductor device Download PDF

Info

Publication number
TW200908120A
TW200908120A TW97113542A TW97113542A TW200908120A TW 200908120 A TW200908120 A TW 200908120A TW 97113542 A TW97113542 A TW 97113542A TW 97113542 A TW97113542 A TW 97113542A TW 200908120 A TW200908120 A TW 200908120A
Authority
TW
Taiwan
Prior art keywords
polishing
film
polished
interlayer insulating
insulating film
Prior art date
Application number
TW97113542A
Other languages
Chinese (zh)
Inventor
Noritaka Kamikubo
Original Assignee
Sharp 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
Application filed by Sharp Kk filed Critical Sharp Kk
Publication of TW200908120A publication Critical patent/TW200908120A/en

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/04Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer
    • H01L21/18Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer the devices having semiconductor bodies comprising elements of Group IV of the Periodic Table or AIIIBV compounds with or without impurities, e.g. doping materials
    • H01L21/30Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26
    • H01L21/31Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26 to form insulating layers thereon, e.g. for masking or by using photolithographic techniques; After treatment of these layers; Selection of materials for these layers
    • H01L21/3105After-treatment
    • H01L21/31051Planarisation of the insulating layers
    • H01L21/31053Planarisation of the insulating layers involving a dielectric removal step
    • CCHEMISTRY; METALLURGY
    • C09DYES; PAINTS; POLISHES; NATURAL RESINS; ADHESIVES; COMPOSITIONS NOT OTHERWISE PROVIDED FOR; APPLICATIONS OF MATERIALS NOT OTHERWISE PROVIDED FOR
    • C09GPOLISHING COMPOSITIONS; SKI WAXES
    • C09G1/00Polishing compositions
    • C09G1/02Polishing compositions containing abrasives or grinding agents

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Physics & Mathematics (AREA)
  • Organic Chemistry (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Mechanical Treatment Of Semiconductor (AREA)
  • Finish Polishing, Edge Sharpening, And Grinding By Specific Grinding Devices (AREA)
  • Internal Circuitry In Semiconductor Integrated Circuit Devices (AREA)

Abstract

To provide a planarization method which suppresses defects on a surface of a polished film and can control a thickness of the polished film, and to provide a manufacturing method of a semiconductor device using such the planarization method.An interlayer insulating film 3 is formed on a semiconductor substrate 1 formed with a semiconductor element 2. When this takes place, a projected part 4 having a higher position than a periphery and a non-projected part 5 having a lower position than the projected part 4 exist on a surface of the formed interlayer insulating film 3. For such the interlayer insulating film 3, first, a first polishing processing is conducted by use of abrasive grains having non-Prestonian characteristics to planarize the projected part 4. Thereafter, a second polishing processing in which a polishing pressure is set to 1.5 times or more is executed on the surface of the interlayer insulating film 3.; Thus, the number of defects remaining on the surface of the interlayer insulating film 3 after polishing can be suppressed, and further a control of a film thickness of the interlayer insulating film 3 which is desired to remain is facilitated.

Description

200908120 九、發明說明: 【發明所屬之技術領域】 本發明係半導體裝置之製造方法,尤其係關於在成膜後 用以使表面平坦化之平坦化方法。 【先前技術】BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of fabricating a semiconductor device, and more particularly to a planarization method for planarizing a surface after film formation. [Prior Art]

伴隨半導體積體電路裝置之高積體化,為了於其製造過 程中,高精度形成微細的圖案,而使基板表面均勻地平坦 化之平坦化技術變得重要。作為如此平坦化技術,先前, 廣泛應用使用研磨液(漿料)向研磨布(pad)按壓而對基板進 行摩擦研磨之CMP(Chemical Mechanical P〇nshing:化學 機械研磨)法。 當於利用該CMP法進行平坦化時,要求特別高的平坦性 之情形時,例如,當利用 STI(Shallow Trench Is〇laU〇n, 淺溝槽隔離)法而形成元件分離區域時,將利用該方法而 形成之溝内的埋入絕緣膜之剩餘部分研磨除去之情形時, 如下述專利文獻丨所記載般,廣泛應用以氧化鈽(J稱為 「二氧化鈽」)為研磨粒之漿料。 此係根據如下者,藉由使用以氧化鈽為研磨粒、以適當 的有機化合物為添加劑之聚料,而與先前通常應用之使: 有以氧化石夕(二氧化石夕)為研磨粒之漿料之情形比較,可獲 得更高的氧切膜之研磨速度,及與用作研磨擋止膜之氮 化矽膜相對應之更高的研磨速度選 ^ 於:磨壓力較特定值低之情形時研磨速度變二二= 普雷斯頓(non-Prestonian)特性,因脐认 U此於研磨擋止臈露出之 130439.doc 200908120 階段,可抑制過度地研磨元件分離區域之氧化石夕膜。藉 此,,可實現圖案依存性較小'且具有高平坦性之研磨。 進而’近年來’如下述專利文獻2所示,提供了一種研 磨壓力(研磨時之按壓壓力)低於固定值之情形時之研磨速 度為每分鐘20〜50 nm左右之、非普雷斯頓CP—— 特ί·生較強之研磨材。當使用如此研磨材對具有凸部之形狀 之被研磨膜表面進行研磨之情形時,於凸部存在之狀況 、每刀鐘100〜1000 nm左右之研磨速度進行研磨,相 料此凸部經平坦化而被研磨膜表面大致平坦之階 段,研磨速度急遽下降至每分鐘5〇 nm&右以下,與通常 應用之研磨速度相比研磨幾乎不進行。 亦即,藉由使用如此非普雷斯頓(n〇n_Prest〇nian)特性較 強之研磨粒(研磨材),而如具有於半導體元件或金屬配線 形成後所成膜之凸部的層間絕緣膜等般,於材料與被研磨 膜不同之研磨擋止膜不存在之情形時,亦於凸部經平坦化 之階段研磨速度自動地急遽下降(幾乎不進行,自動停 機)’因此即使對如此層間絕緣膜亦可實現具有高平坦性 之研磨處理。 [專利文獻1]日本專利特開2〇〇131〇256號公報 [專利文獻2]日本專利特開2〇〇6_279〇5〇號公報 【發明内容】 然而’當使用如上所述之非普雷斯頓(n〇n_Prest〇nian)特 性較強之研磨粒,對具有凸部之被研磨膜執行平坦化處理 (研磨處理)之情形時,若藉由使凸部平坦化而使被研磨膜 130439.doc 200908120 表面大致平坦化,則自動地研磨變得幾乎不進行。因此, 產生如下問題’即使於被研磨膜表面所產生之刮痕等缺陷 存在,該缺陷亦不會被除去而是直接殘留,因此平坦化處 理後之被研磨膜表面之缺陷密度變得非常大。又,產生如 下問題’凸部之平坦化後研磨速度非常慢,因此難以根據 預先測定之CMP處理之前階段之成膜量,控制每個處理批 次之CMP處理時之研磨量。 本發明係鑒於上述問題點,且目的在於提供於以非普雷 斯頓(non-Prestonian)特性較強之材料為研磨粒而使經成膜 之被研磨膜之表面平坦化時’抑制存在於該被研磨膜表面 的缺陷,並且可控制被研磨膜厚之平坦化方法,並提供一 種使用該平坦化方法之半導體裝置之製造方法。 用以達成上述目的之本發明之半導體裝置之製造方法的 第1特徵在於包括:成膜步驟’其係於半導體基板上形成 由絕緣膜或者導電膜所構成之被研磨膜;及平坦化步驟, 其係於上述成膜步驟結束後,使上述被研磨膜之成膜表面 平坦化;上述平坦化步驟包括:第1研磨處理,其係使用 具有非普雷斯頓(non-Prestonian)特性之研磨粒對上述被研 磨膜之表面進行研磨處理;及第2研磨處理,其係結束上 述第1研磨處理後,於與上述第1研磨處理相比較為1.5倍 以上之研磨壓力之下,對上述被研磨膜之表面進行研磨處 理’上述第1研磨處理係於使上述被研磨膜之成膜表面變 化為第1表面狀態之階段結束,該第1表面狀態係至少垂直 於上述半導體基板面的方向之高度或者深度為100 nm以上 130439.doc 200908120 之凸部或者凹部不存在者。 根據本發明之半導體裝置之製造方法之上述第〗特徵: 藉由研磨壓力相對小的第1研磨處理,而使存在於被研磨 膜表面之凸部或者凹部平坦化,並且藉由研磨壓力相對大 的第2研磨處理,而於第丨研磨處理後對存在被研磨臈表面 產生缺陷之區域進行研磨,藉此可削減損害研磨後之表面 的缺數量。藉此,不會過剩地研磨經成膜之被研磨膜, 可對該被研磨膜表面進行平坦化處理,並且可較先前大幅 度減少存在於研磨後之被研磨膜内之缺陷數量。 進而,於第1研磨處理之後所進行之第2研磨處理,與第 1研磨處理相比較研磨壓力較大,因此可對第丨表面狀態之 被研磨膜表面以可監視之速度而進行研磨。藉此,可進行 控制以使第2研磨處理於研磨了預先規定之膜厚之時間點 、、’Q束,並且可容易地進行需殘留之被研磨膜之膜厚的調 整。 又,本發明之半導體裝置之製造方法除了上述第丨特徵 以外,其第2特徵在於:於上述第丨研磨處理中,對呈上述 第1表面狀態之成膜表面的研磨速度為剛開始後之研磨速 度之1 /4以下。 根據本發明之半導體裝置之製造方法之上述第2特徵, 藉由第1研磨處理而對凸部或者凹部實施平坦化,於被研 磨膜表面成為第i表面狀態之階段,研磨速度大幅度減 】,因此可各易地識別第1研磨處理之結束時序。又,與 成為第1表面狀態之被研磨膜表面相對應之研磨速度十分 130439.doc 200908120 k ’因此不會於對凸部或者凹部之平坦化處理結束後,過 剩地研磨被研磨膜。 又,本發明之半導體裝置之製造方法除了上述第1或者 第2特徵以外,其第3特徵在於:上述第2研磨處理係於可 對呈上述第1表面狀態之成膜表面進行每分鐘2〇〇 nm以上 之研磨的研磨條件下進行。 根據本發明之半導體裝置之製造方法之上述第3特徵, 可一面與第1研磨處理相比較使研磨後所殘留之缺陷數量 減夕,一面對包含第丨研磨處理後存在於被研磨膜表面之 缺陷的成膜區域進行研磨。又,可對第丨表面狀態之被研 磨膜表面以可監視之研磨速度而進行研磨,因此可使研磨 膜厚之控制谷易化。藉此,可藉由進行控制以僅對較多包 3第1研磨處理後所產生之缺陷之成膜區域進行研磨後, 使第2研磨處理結束,而不過剩地研磨,抑制殘留於研磨 後之被研磨膜的缺陷數量。 又,本發明之半導體裝置之製造方法除了上述第r〜第3 特徵中任一特徵以外,其第4特徵在於:上述第丨研磨處理 係於了對呈上述第1表面狀態之成膜表面進行每分鐘5〇 nm 以下之研磨的研磨條件下進行。 根據本發明之半導體裝置之製造方法之上述第4特徵, 與成為第1表面狀態之被研磨膜表面相對應的研磨速度十 分慢,因此對凸部或者凹部之平坦化處理結束後,不會過 剩地研磨被研磨膜。藉此,可不會由第丨研磨處理過剩地 研磨而轉移至第2研磨處理。 130439.doc 200908120 又,本發明之半導體裝置之製造方法除了上述第1〜第4 特徵中任一特徵以外,其第5特徵在於:上述第2研磨處理 係對上述被研磨膜實施膜厚3〇 nm以上研磨後結束。 根據本發明之半導體裝置之製造方法之上述第5特徵, 可將殘留於第2研磨處理後之被研磨臈的缺陷數量抑制於 不影響其後之製造步驟之範圍内。 又,本發明之半導體裝置之製造方法除了上述第1〜第$With the high integration of the semiconductor integrated circuit device, it is important to form a fine pattern with high precision during the manufacturing process, and to flatten the surface of the substrate uniformly. As such a flattening technique, a CMP (Chemical Mechanical Grinding) method in which a polishing liquid (slurry) is pressed against a polishing pad and rubbed on a substrate is widely used. When flattening by the CMP method is required, when a particularly high flatness is required, for example, when an element separation region is formed by STI (Shallow Trench Isolating) method, it will be utilized. When the remaining portion of the buried insulating film in the trench formed by this method is polished and removed, as described in the following patent document, cerium oxide (J is referred to as "cerium oxide") is widely used as a slurry of abrasive grains. material. This is based on the following, by using cerium oxide as the abrasive particles and a suitable organic compound as an additive, and the conventionally used ones are: using oxidized stone (cerium dioxide) as the abrasive grain In the case of the slurry, a higher grinding speed of the oxygen cut film can be obtained, and a higher grinding speed corresponding to the tantalum nitride film used as the polishing stop film can be selected: the grinding pressure is lower than a specific value. In the case of a grinding speed of two or two = non-Prestonian characteristics, because the umbilical recognition U is exposed to the grinding stop 之 130439.doc 200908120 stage, can inhibit excessively grinding the separation zone of the element of the oxidized stone . As a result, it is possible to achieve a polishing with a small pattern dependency and high flatness. Further, 'in recent years', as shown in the following Patent Document 2, provides a non-prestoning rate at a polishing rate of about 20 to 50 nm per minute when the polishing pressure (pressing pressure at the time of polishing) is lower than a fixed value. CP - a strong abrasive material. When the surface of the film to be polished having the shape of the convex portion is polished by using such a polishing material, the polishing is performed at a polishing speed of about 100 to 1000 nm per knife hour in the presence of the convex portion, and the convex portion is flattened. At the stage where the surface of the polishing film is substantially flat, the polishing rate is rapidly lowered to 5 〇 nm & right and below, and the polishing is hardly performed as compared with the polishing rate which is usually applied. That is, by using such non-Preston (n〇n_Prest〇nian) abrasive particles (abrasive material), such as interlayer insulation having a convex portion formed after formation of a semiconductor element or metal wiring In the case of a film or the like, when the material does not exist in the polishing film different from the film to be polished, the polishing speed is automatically dropped sharply during the flattening of the convex portion (almost no, automatic shutdown). The interlayer insulating film can also realize a polishing process with high flatness. [Patent Document 1] Japanese Patent Laid-Open Publication No. Hei. No. Hei. No. Hei. No. 2, No. Hei. No. 2, No. 2, pp. When a polishing grain having a strong characteristic is used, and a flattening process (polishing process) is performed on a film to be polished having a convex portion, the film to be polished 130439 is obtained by flattening the convex portion. .doc 200908120 When the surface is roughly flattened, the automatic grinding becomes almost impossible. Therefore, there is a problem that even if a defect such as a scratch generated on the surface of the film to be polished exists, the defect does not be removed but remains directly, so that the defect density of the surface of the film to be polished after the planarization treatment becomes very large. . Further, the following problem occurs. The polishing rate after the flattening of the convex portion is very slow, so that it is difficult to control the amount of polishing in the CMP treatment for each treatment batch based on the film formation amount in the stage before the CMP treatment which is measured in advance. The present invention has been made in view of the above problems, and an object thereof is to provide a method for suppressing the surface of a film to be polished by using a material having a non-Prestonian property as a polishing particle. The method of planarizing the surface of the film to be polished and controlling the thickness of the film to be polished, and providing a method of manufacturing a semiconductor device using the planarization method. A first feature of the method for fabricating a semiconductor device of the present invention for achieving the above object includes a film forming step of forming a film to be polished composed of an insulating film or a conductive film on a semiconductor substrate, and a planarization step. After the film forming step is completed, the film forming surface of the film to be polished is planarized; the planarizing step includes: a first polishing process using a non-Prestonian property. The grain is subjected to a polishing treatment on the surface of the film to be polished; and a second polishing process is performed after the first polishing process is completed, and the film is subjected to a polishing pressure of 1.5 times or more as compared with the first polishing process. The surface of the polishing film is subjected to a polishing treatment. The first polishing treatment is completed at a stage where the film formation surface of the film to be polished is changed to a first surface state, and the first surface state is at least perpendicular to a direction of the semiconductor substrate surface. The height or depth is 100 nm or more. 130439.doc 200908120 The protrusion or recess does not exist. According to the above feature of the method of manufacturing a semiconductor device of the present invention, the first polishing process having a relatively small polishing pressure flattens the convex portion or the concave portion existing on the surface of the film to be polished, and the polishing pressure is relatively large. In the second polishing treatment, after the second polishing treatment, the region where the surface of the polishing crucible is defective is polished, whereby the number of defects on the surface after the polishing can be reduced. Thereby, the film-formed film to be polished is not excessively polished, and the surface of the film to be polished can be planarized, and the number of defects existing in the film to be polished after polishing can be greatly reduced as compared with the prior art. Further, since the second polishing treatment performed after the first polishing treatment has a larger polishing pressure than the first polishing treatment, the surface of the surface to be polished in the second surface state can be polished at a monitorable speed. Thereby, it is possible to control so that the second polishing treatment is performed at a time point when the predetermined film thickness is polished, and the Q beam, and the film thickness of the film to be polished which needs to be left can be easily adjusted. Further, in addition to the above-described second feature, the manufacturing method of the semiconductor device of the present invention is characterized in that, in the second polishing process, the polishing rate of the film formation surface in the first surface state is just after the start. The grinding speed is less than 1 / 4. According to the second aspect of the method for manufacturing a semiconductor device of the present invention, the convex portion or the concave portion is planarized by the first polishing treatment, and the polishing speed is greatly reduced when the surface of the polished film is in the i-th surface state. Therefore, the end timing of the first polishing process can be easily identified. Further, the polishing rate corresponding to the surface of the film to be polished in the first surface state is 130,130.doc 200908120 k', so that the film to be polished is not excessively polished after the flattening treatment for the convex portion or the concave portion is completed. Further, in addition to the first or second feature described above, the third method of manufacturing the semiconductor device of the present invention is characterized in that the second polishing process is performed on the film formation surface in the first surface state by 2 Å per minute. The polishing is carried out under grinding conditions of 〇nm or more. According to the third feature of the method of manufacturing a semiconductor device of the present invention, the number of defects remaining after polishing can be reduced as compared with the first polishing process, and the surface is present on the surface of the film to be polished after the second polishing process is included. The film formation area of the defect is ground. Further, since the surface of the surface to be polished of the surface of the second surface can be polished at a monitorable polishing rate, the control of the thickness of the polishing film can be facilitated. Thereby, it is possible to control only the film formation region of the defect generated after the first polishing process by the plurality of packets 3, and then the second polishing process is completed without excessively polishing, thereby suppressing the remaining after polishing. The number of defects in the film being polished. Further, in addition to any one of the above-described rth to third features, the fourth aspect of the present invention is characterized in that the second polishing process is performed on a film formation surface in the first surface state. It is carried out under grinding conditions of 5 μm or less per minute. According to the fourth feature of the method of manufacturing a semiconductor device of the present invention, the polishing rate corresponding to the surface of the film to be polished in the first surface state is extremely slow. Therefore, after the planarization of the convex portion or the concave portion is completed, there is no excess. The ground film is ground. Thereby, it is possible to shift to the second polishing treatment without excessive polishing by the second polishing treatment. In addition to any one of the first to fourth features of the present invention, the second polishing method is configured to apply a film thickness to the film to be polished. Finished after grinding above nm. According to the fifth feature of the method for fabricating a semiconductor device of the present invention, the number of defects remaining in the polishing pad remaining after the second polishing process can be suppressed within a range that does not affect the subsequent manufacturing steps. Moreover, the manufacturing method of the semiconductor device of the present invention is not limited to the above first to the first

特徵中任一特徵以外’其第6特徵在於:上述被研磨膜為 由HDP(Hlgh-Density Piasma :高密度電黎)法所成膜之氧 化矽膜。 根據本發明之半導體裝置之製造方法之上述第6特徵, ;第研磨處理中,可使對成為第1表面狀態之被研磨膜的 研磨迷度十分慢’藉此可抑制過度地研磨被研磨膜。 根據本發明之構成,可削減存在於平坦化後之被研磨膜 表面之缺陷數置’並且可使平坦化後之被研磨膜之殘留膜 厚控制容易。 【實施方式】 以下,參照以下圖丨〜圖3之各圖,就本發明之半導體裝 a &方去(以下,適當稱為「本發明方法」)之實施形 態加以說明。 f 1係模式性地表示本發明方法之各步驟中之概略剖面 構&圖’針對每個步驟分為圖1⑷〜⑷而進行圖示。又, 圖2係本發明方法之製造步驟之流程圖,以下之文中之各 /驟表不圖2所示的流程圖之各步驟。再者,圖1所示之概 130439.doc -10- 200908120 略剖面構造圖僅係模式性地圖示者,實際之構造之尺寸之 比例尺與圖示之比例尺未必一致。 首先,如圖1(a)所示,於半導體基w上形成半導體元件 或者金屬配線層(以下,簡單記載為「半導體元件2」卜並 於其上表面堆積層間絕緣膜3(步驟叫。作為層間絕緣膜 3 ’利用猜(mgh-Density Plasma:高密度電漿)法於成 膜溫度為200〜70〇m、壓力為〇〇1〜1〇 Pa左右之條件下 以膜厚為100〜2000 nm左右,堆積電浆氧化石夕 膜)。再者,堆積層間絕緣臈3之膜厚至少大於上述半導體 元件2之高度。 藉由該步制之層間絕緣媒3之堆積,而如圖】⑷所干’ 於形成有半導體元件2之區域之上部、與除此以外之區域 的上部之間’於表面產生固定之凹凸。以下,將該等凹凸 部分別稱為「凸部4」、「非凸部5」。 再者’所謂此處所假設之凹凸部,係指於層間絕緣膜之 Ο 成膜表2,於垂直於半導體基板丨之基板面的方向上 nm以上向度位置不同之區域。亦即’上述凸部4最上表面 之高度位置,較鄰接之非凸部5之最上表面之高度位置高 100 nm以上。 其次’如® 1(b)及⑷所示,利用CMp法,對基板面進行 使用有包含氧化鈽作為研磨粒之研磨材之研磨處理(以 下’稱為「第i研磨處理」)(步驟#2)。藉此,進行形成於 層間絕緣臈3之上表面之凸部4之平坦化。再者,圖i⑻表 示執行第1研磨處理中途之剖面圖’圖1⑷表示^研磨處 130439.doc 200908120 理之結束時之剖面圖。如圖1⑷所示,於第】研磨處理結束 之時間點,層間絕緣膜3夕矣& 、认 啄膘^之表面成為凸部4不存在之狀態 (以下,適當稱為「第1表面狀態」)。 該步職之第丨研磨處理㈣巾,如衫所進行之利用 包含氧化料為研磨粒之研磨材而進行之⑽法般,對凸 部4維持較高的研磨速度’於第1表面狀態⑷⑷之狀態) 下,於研磨速度自動地下降而產生自動停機之條件下進行 研磨處理。In addition to any of the features, the sixth feature is that the film to be polished is a ruthenium oxide film formed by an HDP (Hlgh-Density Piasma) method. According to the sixth aspect of the method of manufacturing a semiconductor device of the present invention, in the polishing process, the polishing unevenness of the film to be polished in the first surface state can be made very slow, thereby suppressing excessive polishing of the film to be polished. . According to the configuration of the present invention, the number of defects existing on the surface of the film to be polished after planarization can be reduced, and the residual film thickness of the film to be polished after planarization can be easily controlled. [Embodiment] Hereinafter, an embodiment of the semiconductor device of the present invention (hereinafter, appropriately referred to as "the method of the present invention") will be described with reference to the drawings of the following drawings to Fig. 3. The f 1 system schematically shows that the schematic cross-sections of the respective steps of the method of the present invention are shown in Fig. 1 (4) to (4) for each step. Further, Fig. 2 is a flow chart showing the manufacturing steps of the method of the present invention, and the respective steps in the following tables are not the steps of the flowchart shown in Fig. 2. Further, the schematic cross-sectional structural diagram shown in Fig. 1 is only schematically illustrated, and the scale of the actual structure is not necessarily the same as the scale of the figure. First, as shown in FIG. 1(a), a semiconductor element or a metal wiring layer is formed on the semiconductor substrate w (hereinafter, simply described as "semiconductor element 2"), and an interlayer insulating film 3 is deposited on the upper surface thereof (step is called. The interlayer insulating film 3' is made by using the mgh-Density Plasma method at a film forming temperature of 200 to 70 〇m and a pressure of about 1 to 1 〇Pa to a film thickness of 100 to 2000. The thickness of the interlayer insulating spacer 3 is at least larger than the height of the semiconductor element 2. The stacking of the interlayer insulating medium 3 by the step is as shown in the figure (4). The surface of the upper portion of the region in which the semiconductor element 2 is formed and the upper portion of the region in which the semiconductor element 2 is formed are fixed and uneven on the surface. Hereinafter, the uneven portions are referred to as "protrusion 4" and "non-convex". Further, the term "concave-convex portion as used herein refers to a region in which the interlayer film 2 of the interlayer insulating film is different in the direction of the nm or more in the direction perpendicular to the substrate surface of the semiconductor substrate". That is, 'the uppermost surface of the above convex portion 4 The height position is higher than the height position of the uppermost surface of the adjacent non-protrusion portion 5 by 100 nm or more. Next, as shown in Fig. 1 (b) and (4), the substrate surface is used by the CMp method to include cerium oxide as the abrasive grain. The polishing process of the abrasive (hereinafter referred to as "the i-th polishing process") (step #2), thereby flattening the convex portion 4 formed on the upper surface of the interlayer insulating crucible 3. Further, Fig. i(8) FIG. 1(4) shows a cross-sectional view at the end of the polishing station 130439.doc 200908120. As shown in Fig. 1 (4), at the time point when the polishing process is finished, the interlayer insulating film 3 The surface of the 矣& and the 啄膘^ is in a state in which the convex portion 4 does not exist (hereinafter, it is appropriately referred to as "the first surface state"). The 丨 丨 丨 丨 四 四 四 四 四 四 四 四In the case where the oxidized material is a polishing material of the abrasive grains, the high polishing rate is maintained in the first surface state (4) (4) of the convex portion 4, and the polishing rate is automatically lowered to cause automatic shutdown. Grinding is performed.

作為本步驟#1中之更具體的研磨條件,例如,每分鐘滴 下包含旭确子股份有限公司製之氧化㈣磨粒之研磨材 CES-333-2.0約200 d,使研磨時之按壓壓力(以下,簡稱 為「研磨壓力」)約為3 psi(碑每平方英对,大概2i kpa)左 右,使基板(頭)之旋轉速度約為12〇啊,使研磨布 ⑽㈣之旋轉速度約為13〇 rpm,㈣行研磨處王里。旧係 表示使半導體元件2之厚度約為i 8〇 nm,對利用法形 成膜厚約為H)00 nm左右之p_Si〇膜作為層間絕緣臈3之樣 时,於上述研磨條件之下執行第丨研磨處理之情形時的, 凸部4及非凸部5之距半導體基板以上表面之高度位置之 變化冋度位置之測定’係藉由利用分光橢圓偏光法或分 光干涉反射率法所進行之層間絕緣膜3之光學性膜厚測定 法,與利用原子力顯微鏡法所進行之層間絕緣膜3表面之 階差測定法而算出纟,且表示了直徑約為200 mm之半導 體基板(晶圓)上之面内之不同的9個點之測定值之平均。再 者’各點之帛差棒係表示上$面内9個點之測定值之上限 130439.doc •12- 200908120 王卜限的不均者。 此處,實際上,於實施了 10秒以上之第丨研磨處理之階 段,層間絕緣膜3之表面狀態成為上述第1表面狀態,已經 係凸部4不存在之狀態,以下說明中,& 了方便說明於第丄 研磨處理開始前所存在之凸部4與非凸部5之高度位置,藉 由第1研磨處理之執行而如何變化,而將 始前形成有凸牧區域記載為「凸部4」,將第i研磨處= Ο 之開始前形成有非凸部5之區域記載為「非凸部5」。 於圖3中,以塗黑之四方形而繪製之點表示凸部4之高度 位置,且以實線】!表示變化之狀態。另一方面,以空心之 圈繪製之點表不非凸部5之高度位置且以虛線12表示 變=態。又’於各繪製位置上下方向所示之箭頭,表 不對複數個樣品於相同條件下勃 間m☆番从 ’、牛下執仃第1研磨處理時之樣品 二:L均,實線之箭頭表示凸部4之高度位置 的不:,虛線之箭頭表示非凸部5之高度位置的不均。 二:Γ研磨處理開始時(研磨時間。秒),凸部4之 同度位置與非凸部5之高度 件2之厚产,且^ 差,大致等於半導體元 午之厚度,且圖3之圖表中表示了約為18〇⑽。 其次’藉由執行第,磨處理約1()秒 對凸部4實施較多之 〃非凸部5相比 漸接近非凸部5之上表磨面:藉:二之上表面位置逐 上表面位置變得大致相等以後,即使進/第4f非凸部5之 個區域之上表面位置亦幾乎研磨處理兩 行。亦即,認為產生 以清况表示研磨不進 了以’且可謂之表示對凸 130439.doc 200908120 部4之平坦化結束(亦即圖1(c)之狀態),層間絕緣膜3之表 面成為上述第1表面狀態。 再者’第1表面狀態(研磨時間經過丨〇秒後)之層間絕緣 膜3之表面的研磨速度大概為每分鐘23 nrn,且與於開始第 1研磨處理之後的約1 〇秒鐘凸部4被研磨了約丨8〇 nm相比, 可謂係十分慢的研磨速度。 如上所述,該第1研磨處理利用氧化鈽研磨粒作為研磨 材,且具有當研磨壓力低於固定值之情形時研磨速度變 慢’所謂非普雷斯頓(non_prestonian)特性。而且,藉由利 用如此性質,而具有即使於平坦化結束後亦不會過剩地研 磨之特徵。因此,為了利用該特徵,較好的是,使存在凸 部4之狀況下之研磨速度,與對凸部*之平坦化結束後之研 磨速度相比較十分大,藉由於如此條件下進行第丨研磨處 理’可不會過剩地研磨層間絕緣膜3而進行表面之平坦 化。 亦即,第1研磨處理具有於高於特定臨限值之研磨壓力 之下研磨速度快,而於低於該臨限值之研磨壓力之下研磨 速度十分慢的特徵’因此為了充分發揮該特徵,必須調整 研磨條件’以使對存在凸部4之成膜表面之研磨壓力高於 上述臨限值,相反,對不存在凸部4之第丨表面狀態之成膜 表面之研磨壓力低於上述臨限值。具體而言,較好的是, 與平坦化處理結束後之第1表面狀態之層間絕緣膜3之表面 相對應的研磨速度為每分鐘50 nm以下,且於平坦化處理 前之凸部4存在之狀況下以其4倍以上之研磨速度對層間絕 130439.doc • 14 - 200908120 緣膜3之表面進行研磨處理。 然而,與對凸部4之平坦化處理結束後之層間絕緣膜3相 對應的研磨速度,除了根據用作研磨粒之材料而變化外, 亦根據作為被研磨對象之層間絕緣膜3之膜種而變化。例 如,當係藉由 PE-CVD(plasma enhanced chemical vapor deposition ’電漿輔助化學氣相沈積)法而形成之P_TE〇s膜 之情形時,平坦化處理後之研磨速度為每分鐘31 nm左 右,且較之利用上述HDP法而成膜之P-SiO膜所表示之每 刀鐘23 nm之研磨速度尚速’但是係每分鐘$〇 nm以下之十 分慢的研磨速度,因此可謂之藉由將該材料用作層間絕緣 膜3,而不會過剩地研磨可實現對凸部4之平坦化。另一方 面’於藉由熱CVD法而成膜之摻雜有b及p之bpsg膜中, 平坦化處理後之研磨速度為每分鐘450 nm以上,因此可謂 之不適合作為第1研磨處理之被研磨對象。 如此’於第1研磨處理中’當相對於凸部4存在之狀況下 之研磨速度而言,對凸部4之平坦化結束後之研磨速度十 分慢之情形時,例如,如通常所應用般,可藉由連續性地 監視基板(頭)旋轉扭矩或者研磨布(platen)旋轉扭矩之經時 變化,而得知平坦化已結束之情況。如此於確認了平坦化 之結束之時間點,可藉由結束第丨研磨處理,不會過剩地 研磨而執行對凸部4之平坦化。 然而,平坦化結束後之第丨研磨處理之研磨速度十分 慢,因此如圖1(c)所示,係於該第1研磨處理結束後之層間 絕緣膜3之上表面於研磨時所產生㈣痕等缺陷續留之狀 130439.doc 200908120As a more specific polishing condition in the step #1, for example, the abrasive material CES-333-2.0 containing the oxidized (four) abrasive grains manufactured by Asahi Co., Ltd. is dropped every minute for about 200 d, so that the pressing pressure at the time of grinding ( Hereinafter, simply referred to as "grinding pressure") is about 3 psi (the mark per square inch, about 2 i kpa), so that the rotation speed of the substrate (head) is about 12 〇, and the rotation speed of the polishing cloth (10) (four) is about 13 〇 rpm, (four) line grinding at the king. In the case of the conventional system, when the thickness of the semiconductor element 2 is about i 8 〇 nm and the p_Si 〇 film having a film thickness of about H) 00 nm is formed as the interlayer insulating 臈 3, the above-mentioned polishing conditions are performed. In the case of the rubbing treatment, the measurement of the change in the height position of the convex portion 4 and the non-convex portion 5 from the upper surface of the semiconductor substrate is performed by using the spectroscopic ellipsometry or the spectroscopic interference reflectance method. The optical film thickness measurement method of the interlayer insulating film 3 is performed on the surface difference measurement method of the surface of the interlayer insulating film 3 by atomic force microscopy, and the semiconductor substrate (wafer) having a diameter of about 200 mm is shown. The average of the measured values of the nine different points in the plane. Furthermore, the difference bar of each point indicates the upper limit of the measured value of the nine points in the upper surface 130439.doc •12- 200908120 The unevenness of the king's limit. Here, in actuality, in the stage of the second polishing treatment for 10 seconds or more, the surface state of the interlayer insulating film 3 is in the first surface state, and the convex portion 4 is not present, and in the following description, & It is convenient to explain how the height positions of the convex portion 4 and the non-convex portion 5 existing before the start of the second polishing process are changed by the execution of the first polishing process, and the convex region formed before the start is described as "convex" In the portion 4", a region in which the non-protrusion portion 5 is formed before the start of the i-th polishing portion = 记载 is referred to as "non-convex portion 5". In Fig. 3, the dots drawn in the black square shape indicate the height position of the convex portion 4, and are solid lines]! Indicates the state of change. On the other hand, the dot table drawn by the circle of the hollow is not the height position of the convex portion 5 and the variable = state indicated by the broken line 12. In addition, the arrow shown in the up and down direction of each drawing position indicates that the plurality of samples are in the same condition under the same conditions. The sample is in the first grinding process: L, the arrow of the solid line The height of the convex portion 4 is not indicated: the dotted arrow indicates the unevenness of the height position of the non-protrusion portion 5. 2: At the beginning of the Γ grinding process (grinding time. sec), the same position of the convex portion 4 and the height member 2 of the non-protrusion portion 5 are thick, and the difference is substantially equal to the thickness of the semiconductor element noon, and FIG. 3 The chart shows approximately 18 〇 (10). Secondly, by performing the first, the grinding process is performed for about 1 () seconds, and the convex portion 4 is applied more than the non-protrusion portion 5 as compared with the surface of the non-protrusion portion 5: After the surface positions become substantially equal, even if the surface position of the area of the in/out 4f non-convex portion 5 is almost polished, two lines are processed. In other words, it is considered that the polishing is not indicated by the condition of the condition, and it can be said that the flattening of the portion of the convex portion 130439.doc 200908120 is completed (that is, the state of FIG. 1(c)), and the surface of the interlayer insulating film 3 becomes The first surface state described above. Further, the polishing rate of the surface of the interlayer insulating film 3 in the first surface state (after the polishing time elapses after the leap second) is approximately 23 nrn per minute, and approximately 1 〇 second after the start of the first polishing treatment. 4 is compared to a grinding speed of about 8 〇 nm, which is a very slow grinding speed. As described above, the first polishing treatment uses cerium oxide abrasive grains as the abrasive material, and has a non-prestoning characteristic when the polishing pressure is lower than a fixed value. Moreover, by utilizing such a property, it is characterized in that it is not excessively ground even after the flattening is completed. Therefore, in order to utilize this feature, it is preferable that the polishing rate in the case where the convex portion 4 is present is extremely large as compared with the polishing speed after the flattening of the convex portion * is completed, whereby the third step is performed under such conditions. In the polishing treatment, the interlayer insulating film 3 is not excessively polished to planarize the surface. That is, the first grinding process has a feature that the grinding speed is faster than the grinding pressure above a certain threshold, and the grinding speed is very slow below the grinding pressure of the threshold. Therefore, in order to fully utilize the feature The grinding condition must be adjusted so that the polishing pressure on the film forming surface where the convex portion 4 is present is higher than the above-mentioned threshold value, and conversely, the polishing pressure on the film forming surface in which the second surface state of the convex portion 4 is not present is lower than the above Threshold. Specifically, it is preferable that the polishing rate corresponding to the surface of the interlayer insulating film 3 in the first surface state after the planarization process is 50 nm or less per minute, and the convex portion 4 before the planarization treatment exists. In the case of the condition, the surface of the interlayer film 130439.doc • 14 - 200908120 is polished at a polishing rate of 4 times or more. However, the polishing rate corresponding to the interlayer insulating film 3 after the flattening treatment of the convex portion 4 is changed according to the material used as the abrasive grains, and also depends on the film type of the interlayer insulating film 3 to be polished. And change. For example, in the case of a P_TE〇s film formed by a plasma-assisted chemical vapor deposition (PE-CVD) method, the polishing rate after the planarization treatment is about 31 nm per minute. And the polishing rate of 23 nm per knife ring represented by the P-SiO film formed by the above HDP method is still 'but is a very slow polishing speed of less than $〇nm per minute, so it can be said that This material is used as the interlayer insulating film 3, and flattening of the convex portion 4 can be achieved without excessive polishing. On the other hand, in the bpsg film doped with b and p formed by the thermal CVD method, since the polishing rate after the planarization treatment is 450 nm or more per minute, it is not suitable as the first polishing treatment. Grind the object. When the polishing speed in the case where the convex portion 4 is present in the first polishing process is such that the polishing speed after the flattening of the convex portion 4 is completed is very slow, for example, as usual The flattening can be completed by continuously monitoring the substrate (head) rotation torque or the platen rotation torque over time. When the end of the flattening is confirmed as described above, the flattening of the convex portion 4 can be performed without excessive polishing by terminating the second polishing process. However, since the polishing rate of the second polishing treatment after the completion of the planarization is very slow, as shown in FIG. 1(c), the surface of the interlayer insulating film 3 after the completion of the first polishing treatment is generated during polishing (4). Traces and other defects remain unchanged 130439.doc 200908120

此處,與步驟#2之第1研磨處理相比,使研磨壓力為15 倍以上而對層間絕緣膜3重新進行研磨處理(以下,稱為 「第2研磨處理」,步驟#3)。藉由該第2研磨處理,而對層 間絕緣膜3之上表面進行研磨,從而除去殘留於上表面之 缺陷6。於該第2研磨處理中,無需自第丨研磨處理改變研 磨材及研磨布(P丨aten),亦可接著第1研磨處理而連續性地Here, the interlayer insulating film 3 is subjected to a polishing treatment (hereinafter referred to as "second polishing treatment", step #3), in which the polishing pressure is 15 times or more as compared with the first polishing treatment in the step #2. By the second polishing treatment, the upper surface of the interlayer insulating film 3 is polished to remove the defects 6 remaining on the upper surface. In the second polishing treatment, it is not necessary to change the polishing material and the polishing cloth from the second polishing treatment, and the first polishing treatment may be continued continuously.

進行。例如’藉由利用約6 psi(大概41 kPa)之研磨壓力進 行研磨約40秒鐘,而除去約丨00 nm左右之膜厚之層間絕緣 膜3。下述表1係表示與第2研磨處理中之層間絕緣膜3之研 磨除去量相對應的研磨後之層間絕緣膜3之表面所存在的 缺陷數量之表。 [表1]get on. For example, the interlayer insulating film 3 having a film thickness of about 00 nm is removed by grinding with a polishing pressure of about 6 psi (about 41 kPa) for about 40 seconds. Table 1 below shows a table showing the number of defects existing on the surface of the interlayer insulating film 3 after polishing corresponding to the amount of polishing removal of the interlayer insulating film 3 in the second polishing treatment. [Table 1]

第2研磨處理之研磨量「nm! 0(僅第1研麻盘 缺陷數量(個/晶11、>100 nm)The amount of polishing in the second polishing treatment "nm! 0 (only the number of defects in the first grinding disc (number / crystal 11, > 100 nm)

Π3(僅第2研磨處理) 53 再者,上述表1所示之缺陷數量,藉由實施了適當的清 洗後,表示缺陷之大小(自上表面觀察缺陷區域時之平面 形狀之外切長方體之長邊與短邊的平均值)約為⑽⑽以 上之缺陷之、直徑約為2⑻随之每丨牧半導體基板(晶圓) 的數1而表示。以下’將作為缺陷數量而計數之範圍内之 缺陷記載為「缺陷6」。 樣品S1係對僅執行第1研磨處理、而不執行第2研磨處理 130439.doc 200908120 之情形時之缺陷數量進行測定者。又,樣品㈣係對相 =條件下執行第1研磨處理後,分別使第2研磨處理之研磨 置變化時之缺陷數量進行測定者。又,樣品S5係對不進行 第1研磨處理’僅進行使研磨藶力變大之以研磨處理並進 仃凸部4之平坦化處理時之缺陷數量進行測定者。再者, 各樣品S2〜S5之第2研磨處理之研磨量分別如下,樣品叫 28nm’樣品S3為57nm,樣品以為85胆,樣品⑽⑴⑽。Π3 (only the second polishing treatment) 53 Further, the number of defects shown in the above Table 1 indicates the size of the defect by performing appropriate cleaning (the rectangular shape is cut out from the planar shape when the defect region is observed from the upper surface) The average of the long side and the short side is about (10) (10) or more, and the diameter is about 2 (8), which is represented by the number 1 of the semiconductor substrate (wafer). Hereinafter, the defect in the range counted as the number of defects is described as "defect 6". The sample S1 is a measure of the number of defects when only the first polishing process is performed and the second polishing process 130439.doc 200908120 is not performed. Further, in the sample (four), the number of defects in the case where the polishing treatment of the second polishing treatment is changed after the first polishing treatment is performed under the phase = condition is measured. In addition, the sample S5 is not subjected to the first polishing treatment, and only the number of defects in the case where the polishing process is increased to increase the polishing force and the flattening process of the convex portion 4 is performed is measured. Further, the polishing amounts of the second polishing treatments of the respective samples S2 to S5 were as follows, the sample was called 28 nm', the sample S3 was 57 nm, and the sample was 85 biliary, and the sample was (10) (1) (10).

侍知樣品S1之情形時,於執行了第丨研磨處理直後之層 間絕緣膜3之表面上存在超過3咖個非常多之缺陷6。因 此,當藉由僅執行第1研磨處理而結束CMp步驟,進行其 ,之:驟之情形時,因殘留於層間絕緣膜3之上表面的較 多缺陷6 ’故而例如於配線用金屬膜堆積時進人缺陷6中之 金屬材料,產生在配線圖案形成用之蝕刻步驟中未正常蝕 刻之不良’或者’產生在光微影㈣巾缺陷6上部之區域 之圖案消失或殘留多餘圖案等不I從而可能會產生配線 或通孔無法形成所需之形狀等各種障礙。 另—方面,根據樣品S2之結果,執行第i研磨處理後, 藉由第2研磨處理而除去膜厚28 nm之層間絕緣膜3,藉此 每1枚晶圓之缺陷數量減少至大概3〇〇。據此,認為雖然由 第1研磨處理所產生之缺陷數量非常多,但是位於距第上研 磨處理結束後之層間絕緣膜3之上表面位置深度為3〇福 下之位置的缺陷數量為大半,藉由第2研磨處理而除去川 右之層間絕緣膜3,藉此有效地除去缺陷6。又,據此 可謂之’雖然藉由第i研磨處理而產生之缺陷6當然根據研 I30439.doc -17- 200908120 磨裝置之狀態而變動,但是與一般性的研磨粒之大小相比 係較小的缺陷,且難以僅由通常之裝置管理方法而抑制缺 陷本身。 又,根據樣品S4之結果,若藉由第2研磨處理而使層間 絕緣膜3之除去量增加至85 nm左右,則可使每丨枚晶圓之 ' 缺陷數量減少至大概100個以下,如樣品S5般,可抑制為 ' 與僅由第2研磨處理進行CMP步驟之情形時幾乎相同程度 之缺陷數量。 Γϊ 再者,如樣品85般僅執行第2研磨處理之情形時,如上 述表1所不雖然最能夠減少缺陷數量,但是與第丨研磨處理 相比因研磨壓力較大而成為凸部4不存在之第丨表面狀態 後’大幅度ΙΜ亍研磨處自,結果可能導致過剩地執行研磨 處理。亦即,如本發明方法般,藉由第【研磨處理而進行 平坦化處理’藉此一面抑制研磨膜厚為最小限一面使成膜 表面為第1表面狀態之後,藉由第2研磨處理而執行使缺陷 ◎ 》量減少所必須之最小限之研磨處理,藉此可同時實現減 研磨處理後所殘留之缺陷數量與抑制研磨膜厚。 此處,根據上述表1之結果,可知較理想的是藉由第2研 磨處理而研磨層間絕緣膜3之研磨量為30 nm左右以上’ 又,更理想的是為8〇 nm以上。 又,第2研磨處理較第1研磨處理之研磨>1力大,因此與 對具有第1表面狀態之層間絕緣膜3所進行之第}研磨處理 相比研磨速度快。因此,可藉由-面利用-般性的光學性 方法等監視層間絕緣膜3之膜厚_面進行第2研磨處理,而 130439.doc •18· 200908120 容易地控制需殘留之層間絕緣膜3之臈厚,藉此,可使層 間絕緣膜3殘留所需之膜厚而結束CMP步驟。因此,可抑 制層間絕緣膜3之形成步驟之不均、或CMP裝置之研磨速 度之不均。 結束步驟#3之第2研磨處理之後’進行配線步驟、層間 絕緣膜堆積步驟專特定步驟。藉此,可使殘留於層間絕緣 膜之表面之缺陷數量減少,並且可容易地控制需殘留之層 間絕緣膜3之膜厚。 以上,根據本發明方法,分別使研磨條件不同而執行為 了使存在於被研磨膜表面之凸部平坦化所進行之第1研磨 處理,與為了使損害表面之缺陷數量減少所進行之第2研 磨處理,藉此不會過剩地研磨經成膜之被研磨膜,可對該 被研磨膜表面進行平坦化處理,並且可使殘留於研磨後之 表面之缺陷量較先前大幅度減少。進而,之後所進行之第 2研磨處理與第1研磨處理相比研磨壓力較大,因此可對第 1表面狀態之被研磨膜表面以可監視之速度進行研磨。藉 此,可進行控制以使第2研磨處理於研磨了預先規定之膜 厚之時間點結束,並且可容易地調整需殘留之被研磨膜之 膜厚。 再者,上述實施形態中,列舉對層間絕緣膜進行平坦化 處理之情形為例而進行說明,但是作為研磨對象之被研磨 膜不限定於絕緣膜,亦 J J為導電膜。又,於圖1中進行 「凸部」「非&都,夕主、土 」 表達,此係成膜表面所形成之凹凸 區域之稱呼上之一態媒 7 ,右以高度位置較高之區域為基準 130439.doc -19- 200908120 則亦可記載為「凹部」「非凹部 4_ . ^ 」"吓尸坏明弋義為凸部 4不存在之平面狀態之上述「第i表面狀態」係指, 係凹部不存在之平面狀態,若將該等總稱,則係於:膜表 不〇有如下區域的表面狀態,該區域於垂直於半導體美 板1之基板面之方向上高度位置或者深度位置變化1〇〇 以上。 nm 【圖式簡單說明】 ΟIn the case of the sample S1, there are more than 3 very large defects 6 on the surface of the interlayer insulating film 3 after the second polishing treatment. Therefore, when only the first polishing process is performed, the CMp step is terminated, and if it is a step, the metal film is deposited for wiring, for example, due to a large number of defects 6 remaining on the upper surface of the interlayer insulating film 3. When the metal material in the defect 6 is formed, the defect which is not normally etched in the etching step for forming the wiring pattern is generated, or the pattern which is generated in the upper portion of the photolithography (four) film defect 6 disappears or the residual pattern remains. As a result, various obstacles such as wiring or through holes that cannot form a desired shape may occur. On the other hand, after the ith polishing process is performed as a result of the sample S2, the interlayer insulating film 3 having a film thickness of 28 nm is removed by the second polishing process, whereby the number of defects per wafer is reduced to about 3 Å. Hey. Accordingly, it is considered that although the number of defects generated by the first polishing process is extremely large, the number of defects located at a position on the upper surface of the interlayer insulating film 3 after the end of the upper polishing process is 3 〇, is half, The interlayer insulating film 3 is removed by the second polishing treatment, whereby the defect 6 is effectively removed. Further, it can be said that the defect 6 generated by the i-th grinding process is of course changed according to the state of the grinding device of I30439.doc -17-200908120, but is smaller than the size of the general abrasive grain. Defects, and it is difficult to suppress the defect itself by only the usual device management method. Further, according to the result of the sample S4, if the amount of removal of the interlayer insulating film 3 is increased to about 85 nm by the second polishing treatment, the number of defects per wafer can be reduced to about 100 or less. As in the case of the sample S5, it is possible to suppress the number of defects to be almost the same as in the case where the CMP step is performed only by the second polishing treatment. Further, when only the second polishing treatment is performed as in the case of the sample 85, the number of defects can be reduced as much as the above-described Table 1, but the convex portion 4 is not formed by the polishing pressure as compared with the second polishing treatment. After the existence of the third surface state, 'a large amount of grinding is performed, and as a result, the grinding process may be excessively performed. In other words, as in the method of the present invention, the film formation surface is in the first surface state while the film thickness is minimized while the film thickness is minimized, and the second polishing process is performed by the second polishing process. The grinding process is performed to minimize the amount of defects required, thereby achieving the number of defects remaining after the grinding reduction process and suppressing the thickness of the polishing film. In the above, it is preferable that the polishing amount of the interlayer insulating film 3 is about 30 nm or more by the second polishing treatment, and more preferably 8 Å or more. Further, since the second polishing treatment is larger than the polishing of the first polishing treatment, the polishing rate is faster than the polishing treatment for the interlayer insulating film 3 having the first surface state. Therefore, the second polishing treatment can be performed by monitoring the film thickness _ surface of the interlayer insulating film 3 by an optical method such as a general surface, and 130439.doc • 18· 200908120 can easily control the interlayer insulating film 3 to be left. The thickness of the layer is such that the interlayer insulating film 3 can be left to have a desired film thickness to terminate the CMP step. Therefore, unevenness in the formation steps of the interlayer insulating film 3 or unevenness in the polishing speed of the CMP apparatus can be suppressed. After the completion of the second polishing treatment in the step #3, the wiring step and the interlayer insulating film deposition step are specifically described. Thereby, the number of defects remaining on the surface of the interlayer insulating film can be reduced, and the film thickness of the interlayer insulating film 3 to be left can be easily controlled. As described above, according to the method of the present invention, the first polishing treatment for flattening the convex portion existing on the surface of the film to be polished and the second polishing for reducing the number of defects on the damaged surface are performed under different polishing conditions. By this, the film to be polished is not excessively polished, and the surface of the film to be polished can be planarized, and the amount of defects remaining on the surface after polishing can be greatly reduced as compared with the prior art. Further, since the second polishing treatment performed later has a larger polishing pressure than the first polishing treatment, the surface of the surface to be polished in the first surface state can be polished at a monitorable speed. Thereby, it is possible to control so that the second polishing treatment is completed at the time point when the predetermined film thickness is polished, and the film thickness of the film to be polished which can be left can be easily adjusted. In the above-described embodiment, the case where the interlayer insulating film is planarized is described as an example. However, the film to be polished is not limited to the insulating film, and J J is a conductive film. In addition, in Fig. 1, the expressions "protrusion", "non-&, primordial, earth" are expressed, and the concave-convex region formed on the film-forming surface is referred to as a medium 7 and the right height is higher. The area is the reference 130439.doc -19- 200908120, and it can also be described as "concave portion" and "non-recessed portion 4_. ^". "Frightening corpse is the above-mentioned "i-th surface state" in a plane state in which the convex portion 4 does not exist. Refers to the planar state in which the concave portion does not exist. If the general term is used, it is based on the fact that the film surface does not have a surface state in the direction of the substrate surface perpendicular to the semiconductor board 1 or The depth position changes by more than 1〇〇. Nm [Simple diagram description] Ο

圖丨“)〜(d)係本發明之半導體裝置之製造方法之各步驟 中的概略剖面構造圖。 圖2係表示本發明之半導體裝置之製造方法之步驟順序 之流程圖。 圖3係表示第1研磨處理之研磨時間特性之圖表。 【主要元件符號說明】 1 半導體基板 2 半導體元件 3 層間絕緣膜 4 凸部 5 非凸部 130439.doc -20-FIG. 2 is a schematic cross-sectional structural view showing the steps of the method for fabricating the semiconductor device of the present invention. FIG. 2 is a flow chart showing the sequence of steps of the method for fabricating the semiconductor device of the present invention. Graph of polishing time characteristics of the first polishing treatment. [Description of main component symbols] 1 Semiconductor substrate 2 Semiconductor component 3 Interlayer insulating film 4 Projection portion 5 Non-convex portion 130439.doc -20-

Claims (1)

200908120 十、申請專利範圍: 1. 一種半導體裝置之製造方法,其特徵在於包括: 成膜步驟,其係於半導體基板上形成由絕緣臈或者導 電膜所構成之被研磨膜;及 平坦化步驟,其係於上述成膜步驟結束後,將上述被 研磨臈之成膜表面平坦化; 上述平坦化步驟包括: 第1研磨處理’其係使用具有非普雷斯頓(ηοη· Prest〇nian)特性之研磨粒對上述被研磨膜之表面進行研 磨處理;及 第2研磨處理’其係結束上述第1研磨處理後,於與上 述第1研磨處理相比較為1 · 5倍以上之研磨墨力之下,對 上述被研磨膜之表面進行研磨處理; 上述第1研磨處理係於使上述被研磨膜之成膜表面變 化為第1表面狀態之階段結束, 上述第1表面狀態係至少垂直於上述半導體基板面的 方向之高度或者深度為100 nm以上之凸部或者凹部不存 在者。 2. 如請求項1之半導體裝置之製造方法,其中於上述第1研 磨處理中, 對呈上述弟1表面狀態之成膜表面的研磨速度為剛開 始後之研磨速度之1 /4以下。 3. 如請求項1或2之半導體裝置之製造方法,其中上述第2 研磨處理係於可對呈上述第丨表面狀態之成膜表面進行 130439.doc 200908120 每分鐘200 nm以上之研磨的研磨條件下進行。 4. 如請求項1或2之半導體裝置之製造方法,其中上述第i 研磨處理係於可對呈上述第1表面狀態之成膜表面進行 每分鐘50 nm以下之研磨的研磨條件下進行。 5. 如請求項1或2之半導體裝置之製造方法,其中上述第2 研磨處理係對上述被研磨膜實施膜厚3 〇 nm以上研磨後 結束。 6. 如請求項丨或2之半導體裝置之製造方法,其中上述被研 磨膜為由HDP法所成膜之氧化矽膜。 130439.doc200908120 X. Patent application scope: 1. A method for manufacturing a semiconductor device, comprising: a film forming step of forming a film to be polished comprising an insulating germanium or a conductive film on a semiconductor substrate; and a planarizing step, After the film forming step is completed, the surface of the film to be polished is flattened; the planarization step includes: the first polishing process is performed using a non-prestoning characteristic The abrasive grains are subjected to a polishing treatment on the surface of the film to be polished; and the second polishing treatment is performed after the first polishing treatment is completed, and the polishing ink force is 1.5 times or more higher than the first polishing treatment. And polishing the surface of the film to be polished; the first polishing process is completed at a stage of changing a film formation surface of the film to be polished to a first surface state, wherein the first surface state is at least perpendicular to the semiconductor A protrusion or a recess having a height or depth of 100 nm or more in the direction of the substrate surface does not exist. 2. The method of manufacturing a semiconductor device according to claim 1, wherein in the first polishing treatment, the polishing rate of the film formation surface in the surface state of the first surface is 1/4 or less of the polishing rate immediately after the start. 3. The method of manufacturing a semiconductor device according to claim 1 or 2, wherein the second polishing treatment is performed on a polishing condition in which the film formation surface in the surface state of the second surface is subjected to grinding at 130439.doc 200908120 per minute by 200 nm or more. Go on. 4. The method of manufacturing a semiconductor device according to claim 1 or 2, wherein the i-th polishing treatment is performed under polishing conditions in which the film formation surface in the first surface state is polished to 50 nm or less per minute. 5. The method of manufacturing a semiconductor device according to claim 1 or 2, wherein the second polishing treatment is performed after the film is polished to a thickness of 3 〇 nm or more. 6. The method of manufacturing a semiconductor device according to claim 2, wherein the film to be polished is a ruthenium oxide film formed by the HDP method. 130439.doc
TW97113542A 2007-04-27 2008-04-14 Manufacturing method of semiconductor device TW200908120A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2007118407A JP2008277495A (en) 2007-04-27 2007-04-27 Manufacturing method of semiconductor device

Publications (1)

Publication Number Publication Date
TW200908120A true TW200908120A (en) 2009-02-16

Family

ID=39943357

Family Applications (1)

Application Number Title Priority Date Filing Date
TW97113542A TW200908120A (en) 2007-04-27 2008-04-14 Manufacturing method of semiconductor device

Country Status (3)

Country Link
JP (1) JP2008277495A (en)
TW (1) TW200908120A (en)
WO (1) WO2008136240A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI820399B (en) * 2021-02-26 2023-11-01 國立臺灣科技大學 Wafer processing method and wafer processing system

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013031909A (en) * 2011-08-03 2013-02-14 Seiko Instruments Inc Method for polishing glass substrate, method for manufacturing package, piezoelectric vibrator, oscillator, electronic apparatus and radio-controlled timepiece

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002359216A (en) * 2001-05-30 2002-12-13 Mitsubishi Electric Corp Polishing method using ceria slurry, and method of manufacturing semiconductor device
JP2004165424A (en) * 2002-11-13 2004-06-10 Ekc Technology Inc Polishing agent composition and polishing method using the same
JP2006032846A (en) * 2004-07-21 2006-02-02 Sharp Corp Method for manufacturing semiconductor device using cmp

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI820399B (en) * 2021-02-26 2023-11-01 國立臺灣科技大學 Wafer processing method and wafer processing system

Also Published As

Publication number Publication date
JP2008277495A (en) 2008-11-13
WO2008136240A1 (en) 2008-11-13

Similar Documents

Publication Publication Date Title
KR100734305B1 (en) Fabrication method of semiconductor device having dishing-free planarized layer and semiconductor device fabricated using the same
US8518296B2 (en) Slurries and methods for polishing phase change materials
JP2000301454A5 (en)
US20130078784A1 (en) Cmp slurry and method for manufacturing semiconductor device
US20200118827A1 (en) Hard Mask Removal Method
JP5261065B2 (en) Manufacturing method of semiconductor device
KR101752986B1 (en) METHOD FOR PRODUCING SiC SUBSTRATE
JP2008226935A (en) Manufacturing method of semiconductor device
JP2004363191A (en) Chemical mechanical polishing slurry for organic film, method of chemically/mechanically polishing organic film, and method of manufacturing semiconductor device
JP4374038B2 (en) Substrate processing method
JP2005203394A (en) Manufacturing method of semiconductor device
US6899612B2 (en) Polishing pad apparatus and methods
US7491621B2 (en) Method of forming isolation structures in a semiconductor manufacturing process
TW200908120A (en) Manufacturing method of semiconductor device
JP2008021704A (en) Method of manufacturing semiconductor device
US20070269908A1 (en) Method for in-line controlling hybrid chemical mechanical polishing process
JP3161425B2 (en) Method of forming STI
US7186655B2 (en) Method for manufacturing semiconductor device
JP4546071B2 (en) Manufacturing method of semiconductor device
US7109117B2 (en) Method for chemical mechanical polishing of a shallow trench isolation structure
US6897121B2 (en) Method of removing HDP oxide deposition
JP2007019428A (en) Method of manufacturing semiconductor device
KR20080101454A (en) Method for isolation of semiconductor device
US20090098734A1 (en) Method of forming shallow trench isolation structure and method of polishing semiconductor structure
KR101581469B1 (en) Method of polishing substrate