TW423079B - A process for manufacturing ic-components to be used at radio frequencies - Google Patents

A process for manufacturing ic-components to be used at radio frequencies Download PDF

Info

Publication number
TW423079B
TW423079B TW087107482A TW87107482A TW423079B TW 423079 B TW423079 B TW 423079B TW 087107482 A TW087107482 A TW 087107482A TW 87107482 A TW87107482 A TW 87107482A TW 423079 B TW423079 B TW 423079B
Authority
TW
Taiwan
Prior art keywords
layer
region
doped
substrate
electrically insulating
Prior art date
Application number
TW087107482A
Other languages
Chinese (zh)
Inventor
Hans Norstroem
Stefan Nygren
Ola Tylstedt
Original Assignee
Ericsson Telefon Ab L M
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 Ericsson Telefon Ab L M filed Critical Ericsson Telefon Ab L M
Application granted granted Critical
Publication of TW423079B publication Critical patent/TW423079B/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
    • H01L21/02104Forming layers
    • H01L21/02107Forming insulating materials on a substrate
    • H01L21/02109Forming insulating materials on a substrate characterised by the type of layer, e.g. type of material, porous/non-porous, pre-cursors, mixtures or laminates
    • H01L21/02112Forming insulating materials on a substrate characterised by the type of layer, e.g. type of material, porous/non-porous, pre-cursors, mixtures or laminates characterised by the material of the layer
    • 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
    • H01L21/302Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26 to change their surface-physical characteristics or shape, e.g. etching, polishing, cutting
    • H01L21/306Chemical or electrical treatment, e.g. electrolytic etching
    • H01L21/3065Plasma etching; Reactive-ion etching
    • 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
    • H01L21/70Manufacture or treatment of devices consisting of a plurality of solid state components formed in or on a common substrate or of parts thereof; Manufacture of integrated circuit devices or of parts thereof
    • H01L21/77Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate
    • H01L21/78Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices
    • H01L21/82Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components
    • H01L21/822Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components the substrate being a semiconductor, using silicon technology
    • H01L21/8222Bipolar technology
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/70Bipolar devices
    • H01L29/72Transistor-type devices, i.e. able to continuously respond to applied control signals
    • H01L29/73Bipolar junction transistors
    • H01L29/7302Bipolar junction transistors structurally associated with other devices
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/86Types of semiconductor device ; Multistep manufacturing processes therefor controllable only by variation of the electric current supplied, or only the electric potential applied, to one or more of the electrodes carrying the current to be rectified, amplified, oscillated or switched
    • H01L29/92Capacitors having potential barriers

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Physics & Mathematics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Manufacturing & Machinery (AREA)
  • Ceramic Engineering (AREA)
  • Plasma & Fusion (AREA)
  • Bipolar Integrated Circuits (AREA)
  • Bipolar Transistors (AREA)

Abstract

A bipolar transistor of type NPN has an active region at the surface of the component, which is surrounded, as seen along the surface of the component, in the conventional way by thick field oxide areas (18). The active region is partly covered by an electrically isolating surface layer, preferably comprising a nitride layer (34). A base region in the active region is defined by a well-defined opening, that is lithographically produced, in the electrically isolating surface layer. For a bipolar lateral transistor of type PNP, which instead has emitter and collector regions surrounded by such thick field oxide areas, the emitter and collector regions can in the corresponding way be defined by a lithographically defined opening in an electrically isolating surface layer. Owing to the well defined openings the base-collector capacitance and the emitter-collector capacitance respectively can be reduced in these cases, what results in better high frequency characteristics of the transistors. A region of the silicon nitride layer (34) is at the same time used as an efficient dielectric in a simultaneously manufactured capacitor. The NPN-transistor can be provided with a thin side string made of nitride for isolation between an emitter connection and a base connection. In the same semiconductor plate special, deep and shallow substrate connecting terminals can be provided for electrically isolating component areas. Trenches (22) can be used for electrical isolation of component areas and they can at their sidewalls have a laminate of at the bottom oxide (23) and thereon nitride (25) in order to facilitate planarizing etching and in order to act as a diffusion barrier.

Description

r 4 2 3 Ο 7 9 A7 B7 五、發明説明( 經濟部中央標準局負工消費合作社印裝 發明領媸 本發明係和製造適合射頻範圍内之信號、且係利 用以矽為基本的雙極性技術製造的IC元件有關,特 別是適合同時在一矽基質上製造垂直NpN電晶體、 電容和橫向PNP電晶體,和適合用以產生深基質接 點。 發明背景 迄至^前,要使用雙極性電晶體(Bip-電晶體)產 生具有高封裝密度的快速雙極性電路是可能的a,其 製造方式使得他們具有兩層g乙醯胺,因而為俗稱 的雙多晶矽型電晶體,其使用自行對準或自行對準 技術聯合圍繞在電晶體四週之溝槽所提供的電子絕 緣(所稱之溝槽絕緣)。這類,,雙多晶矽型,,的習知溝 槽絕緣雙極性電晶體的剖面示意圖如圖丨所示,它 是使用自行對準技術製造而成。 匕 在已知製造雙多晶矽型雙極性電晶體的製程中, 先令第一個沉積的聚乙醯胺層形成一基本連接。若 此電^體是NPN型,則此多晶矽層被強烈摻雜成p 型。最後沉積的聚乙醯胺層(它被強烈摻雜成N型並 且構成一射極電極)實體上是與構成基極連接的第 一沉積聚乙醯胺層分開一下方的電絕緣層和絕緣側 條’亦稱為”間隔物,,。在製造雙極性電晶體時使用 自行對準方法的優點是可使基極阻抗和基極與集極 之間的電容降低。此外’以溝槽提供絕緣可澈底地 本纸張尺度適用中國國家標準(CNS ) Α4規格(2丨0Χ297公釐) _ (請先聞讀背面之注意事項再填寫本頁) 丁 *-·° Γ4 2 3 Ο 7 9 五、發明説明(2 ) 地降低集極和基極之間的電容。因此,電路之效能 可以大為改良。 本文後將由參考圖2至6詳細說明一經常使用之已 知的製造一NPN型的溝槽絕緣之雙極性電晶體之方 法。作為基本材料的p型單晶矽基質1〇1之表面被設 置於一矽結晶結構之一(1〇〇)平面上,請見圖2 ^底 部擴散102(亦稱為,’屏障層,,,它可以由例如砷或銻 的離子植入層組成)以微影蝕刻術界定而成,接下 來將厚度約為2" m的蟲晶碎層1〇3施於該平面1〇1 上。然後使用微影触刻術聯合離子植入的方式在平 面101上界定出N-和P_區域。N_區域1〇4是以磷離子 植入的方式做成,它被直接放置在N+型的底擴散層 102上方。另—個區域1〇5係設置於〜區域ι〇4之間 ,它是P摻雜的且是以例如硼離子植入的方式製造 ,見圖2所示。 接著以傳統的LOCOS方法(’’LOCal碎氧化,,)界定 主動區’見 J.A_ Appel 等人在 Philips Researchr 4 2 3 Ο 7 9 A7 B7 V. Description of the invention (Printed invention collar printed by the Consumers' Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs) The present invention relates to the manufacture of signals suitable for the radio frequency range, and uses silicon-based bipolarity. IC components manufactured by technology are particularly suitable for simultaneously manufacturing vertical NpN transistors, capacitors, and lateral PNP transistors on a silicon substrate, and are suitable for creating deep substrate contacts. BACKGROUND OF THE INVENTION Until now, bipolarity was used Transistors (Bip-transistors) are possible to produce fast bipolar circuits with high packaging density. A is manufactured in such a way that they have two layers of g-acetamide, so they are commonly known as dual polycrystalline silicon transistors. The quasi-or self-alignment technology combined with the electrical insulation provided by the trenches surrounding the transistor (so-called trench insulation). The profile of the conventional trench-insulated bipolar transistor of this type, a dual polysilicon type, The schematic diagram is shown in Figure 丨, which is manufactured using self-alignment technology. In a known process for manufacturing a bi-polycrystalline silicon bipolar transistor, the first polymer The acetamide layer forms a basic connection. If the electrical body is of the NPN type, the polycrystalline silicon layer is strongly doped to the p type. The finally deposited polyacetamide layer (which is strongly doped to the N type and constitutes a Emitter electrode) is physically separated from the first deposited polyethylene layer constituting the base by an electrically insulating layer and insulating side strips, which are also referred to as "spacers." It is used in the manufacture of bipolar transistors. The advantage of the self-aligning method is that the base impedance and the capacitance between the base and the collector can be reduced. In addition, the insulation provided by the trench can be used. The paper size is applicable to the Chinese National Standard (CNS) Α4 specification (2 丨0 × 297mm) _ (Please read the notes on the back before filling in this page) Ding *-· ° Γ4 2 3 〇 7 9 V. Description of the invention (2) The capacitance between the collector and the base is reduced. Therefore The performance of the circuit can be greatly improved. A conventionally known method for manufacturing an NPN-type trench-insulated bipolar transistor will be described in detail later with reference to FIGS. 2 to 6. As a basic material, a p-type single crystal The surface of the silicon substrate 101 is set on a silicon crystal structure On one (100) plane, see Figure 2 ^ bottom diffusion 102 (also known as a 'barrier layer,' which can consist of an ion implanted layer such as arsenic or antimony) defined by lithography Then, a fragment of worm crystal 10 with a thickness of about 2 m is applied on the plane 101. Then, lithography and ion implantation are used to define N-and plane 101. P_area. N_area 104 is made by phosphorus ion implantation, and it is placed directly above the N + -type bottom diffusion layer 102. Another area 105 is located in ~ area ι〇4 In between, it is P-doped and is manufactured by, for example, boron ion implantation, as shown in Figure 2. Then the traditional LOCOS method ("LOCal fragmentation oxidation,") is used to define the active region. See J.A_ Appel et al. At Philips Research

Report,Vol. 25, 1970 第 118-132 頁中之,,矽之區域 經濟部中央橾準局負工消费合作社印製 (請先閱讀背面之注意事項再填转本頁) 式氧化和其在半導體技術中之施加” ◊然後首先施 加一適當材料的絕緣罩1 〇6 ’見圖3,再做微影蝕刻 術佈圖。接下來,在罩幕1〇6的孔隙内熱成長矽1〇7 ’使得電晶體的基極區域1 和集極區域1 〇9得被保 留下來’並形成於罩幕覆蓋表面的區域内。在經此 方式界定出主動區以後(以氧化層區域丨〇9,分開), 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公董) 1423079 五、發明説明(3 ) 經濟部中央標準局員工消費合作社印製 再以微影蝕刻術方式界定絕緣溝槽丨〗〇,然後使用 同位性乾触刻將氧化矽材料1 〇7和基質材料1 〇丨蝕刻 掉1直到溝槽11 〇取得所需深度為止(約5至丨〇 v m) ’並且向下延伸進入非有效的p基質ιοί中。 溝槽11 〇的壁被熱氧化,以致於可獲得一薄的、 電絕緣層,然後,該溝槽被充填以絕緣的或半絕緣 的材料1 1 1,例如氧化矽或多晶矽,亦稱為多晶矽 型或聚乙醯胺。該填充材料然後以乾蝕刻方式蝕刻 ,,直到得到一平坦表面為止。接著使此平面表面 氧化,特別是在溝槽已填充有聚乙醯胺的情況下, ,溝槽的開口内的矽材料氧化,以在開口表面上獲 得一絕緣層。若溝槽在開始時就已經只充填有氧時 ’則不需要額外的氧化步驟。其結果如圖3所示。 可發現在圖3中,基極區域1〇8的延展是使用如上之 LOCOS方法界定而得的。本方法之缺點將於下文說 明一製造電晶體之修正製程時討論β Β 在形成溝槽1 〇以後,以微影蝕刻術界定一集極插 口 η 2,見圖4所示,亦即,在集極區域丨〇9内的元 件平面表面和底擴散面之間界定一低阻抗的連接。 之後以離予植入的方式將一掺雜子(通常是磷)摻 雜至該微影蝕刻界定的開口中。 接下來的製程敘述將以上述之具一自行對準的基 射極接面的雙多晶矽型ΝΡΝ電晶體說明’因為此類 型的元件通常會聯合以溝槽取得的電絕緣。 本纸張尺度適用中關家標準(CNS)从胁(21Q><297公羡) (請先閱讀背面之注意事項再填寫本頁} 訂 14 23 0 Α7 Β7 經濟部中央榡準局—工消費合作社印製 五、發明説明(4 ) 在前述之主動區108、1〇9界定和集極插口 112形 成以後,沉積一薄的聚乙醯胺層113,其厚度為幾 百,如圖4所示。然後以離子植入硼的方式使該 聚乙醯胺113摻雜成P+型,然後在此聚乙醯胺層上 方經CVD(“化學汽相蒸鍍”)沉積一薄的氧化層114。 在此製程後,該摻雜有硼的P +型聚乙醯胺層將形成 一俗稱的本質(intrinsic)之基座或是基本連接或終 端。以CVD產生的氧化層1 14以及位於其下方的聚 乙酿胺層1 1 3被以微影姑刻方式佈圖,以界定出一 位於基極區域1 〇 8内的射極開^口 1 1 5。接著以如電漿 蚀刻之乾蝕刻步驟將這兩層的那些未被微影蝕刻罩 幕覆蓋的部份移除掉。在完成射極開口 115之佈圖 以後,成長一薄的熱氧化層丨丨6,以保護該射極開 口内的表面,然後以一離子植入硼的方式形成一俗 稱之本質的基極11 7。因此,此本質的基極〗丨7會被 精密、地放在射極開口 1 1 5之内和下方處。 為使欲產生的射極與該本質的基質分開,沿著射 极開口 11 5的側邊开;#成”間隔物”或側條1 1 8,如圖5 ,示。其形成方式使得首先在平面上以Cvd沉積一 氧化層,然後使用一各向異性的乾蝕刻步驟將該平 面的平坦表面部份上的此氧化層姓刻掉,因而可伴 隨著當佈圖以產生射極開口 u 5之步驟形成CVD氧 化物的側條或間隔物n8。形.成此間隔物i 18之後, 况積一其厚度在幾百nm左右的薄聚乙醯胺層119於 ---------------—1T------.戒 (請先閑讀背面之注意事項再填寫本頁) -7-Report, Vol. 25, 1970, pages 118-132, printed by the Consumer Electronics Co., Ltd. of the Central Economic and Technical Bureau of the Ministry of Regional Economics (please read the precautions on the back before filling this page) "Application in semiconductor technology" ◊ Then first apply an insulating cover 1 0 6 'of appropriate material, see Figure 3, and then make a photolithographic layout. Next, silicon 1 is thermally grown in the pores of the cover 1 06. 7 'Make the base region 1 and collector region 1 of the transistor be retained' and form in the area covered by the mask. After the active region is defined in this way (with the oxide layer region) 09 , Separate), this paper size applies Chinese National Standard (CNS) A4 specification (210X297 public director) 1423079 V. Description of the invention (3) Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs and then defining the insulation trench by lithography丨 〖〇, and then use isotopic dry contact etching to etch the silicon oxide material 1 〇07 and the matrix material 1 〇 丨 until the trench 11 〇 to achieve the required depth (about 5 to 丨 vm) 'and extend downward Into non-effective p-matrix The walls of the trench 110 are thermally oxidized so that a thin, electrically insulating layer is obtained, and the trench is then filled with an insulating or semi-insulating material, such as silicon oxide or polycrystalline silicon, or It is called polycrystalline silicon or polyvinylamine. The filling material is then etched by dry etching until a flat surface is obtained. This flat surface is then oxidized, especially if the trench is filled with polyethylene. The silicon material in the opening of the trench is oxidized to obtain an insulating layer on the surface of the opening. If the trench is only filled with oxygen at the beginning, no additional oxidation step is required. The result is shown in Figure 3. It can be found that in FIG. 3, the extension of the base region 108 is defined by using the LOCOS method as described above. The disadvantages of this method will be discussed below when a modified process of manufacturing a transistor is discussed. Β Β forms a trench. After slot 10, a collector socket η 2 is defined by lithography, as shown in FIG. 4, that is, a low-impedance boundary is defined between the planar surface of the element in the collector region and the bottom diffusion surface. Connect after The implantation method is doped with a dopant (usually phosphorus) into the opening defined by the lithographic etch. The following process description will be based on the above-mentioned dual polycrystalline silicon type with a self-aligned base-emitter junction. NPN transistor description 'Because this type of component is usually combined with the electrical insulation obtained by the groove. This paper standard applies the Zhongguan standard (CNS) Congxie (21Q > < 297 public envy) (Please read the back of the first Matters needing attention to fill out this page again} Order 14 23 0 Α7 Β7 Printed by the Central Bureau of Standards of the Ministry of Economic Affairs-Industrial and Consumer Cooperatives V. Description of the invention (4) After the aforementioned active zones 108 and 109 were defined and the collector socket 112 was formed A thin polyethyleneamine layer 113 is deposited with a thickness of several hundred, as shown in FIG. 4. Then, the polyethylenimine 113 is doped with P + by ion implantation of boron, and a thin oxide layer 114 is deposited on the polyethylenimide layer by CVD ("chemical vapor deposition"). After this process, the boron-doped P + -type polyethyleneamine layer will form a commonly known intrinsic pedestal or a basic connection or terminal. The oxide layer 1 14 generated by CVD and the polyethylenamine layer 1 1 3 located below it are patterned in a lithographic manner to define an emitter opening located in the base region 108. 1 5. The parts of the two layers that are not covered by the lithographic etching mask are then removed by a dry etching step such as plasma etching. After the layout of the emitter opening 115 is completed, a thin thermal oxide layer 6 is grown to protect the surface inside the emitter opening, and then an ion-implanted boron is used to form a commonly known essence base 11 7. Therefore, the base of this essence is placed precisely within and below the emitter opening 1 1 5. In order to separate the emitter to be generated from the essence of the matrix, open along the side of the emitter opening 115; # 成 "spacers" or side bars 1 1 8 as shown in Figure 5. It is formed in such a manner that an oxide layer is first deposited on the plane by Cvd, and then an anisotropic dry etching step is used to engrav the surname on the flat surface portion of the plane, which can be accompanied by the layout The step of generating the emitter opening u 5 forms a side strip or spacer n8 of the CVD oxide. After forming the spacer i 18, a thin polyacetamide layer 119 having a thickness of about several hundred nm is formed at ---------------- 1T ---- -. Caution (please read the precautions on the back before filling out this page) -7-

r 4 2 3 ο 7 9五、發明説明(5 ) A7 B7 經濟部中央橾準局頁工消費合作社印製 該平面表面上。此層被植入砷,以變成N+型,且將 在做退火處理後形成電晶體的射極電極12〇。在佈 圖和蝕刻該層1 1 9以產生射極電極後,其結構如圖5 所示者。通常吾人會令上方的聚乙醯胺層119(形成 射極電極)仍保持於射極區域1 〇 9的上部,見圖3所 示’以及集極插口 112的上部,在此處該集極插口 112作為一集極終端121。 接著電路被鈍化一例如氧化矽的層丨2 2,如圖6所 示,其中電晶體的基極、射極和集極之接觸孔123 、124、125皆以微影蝕刻術$定,在蝕刻此接觸孔 之後,電路被濺鍍以一金屬層丨26,例如鋁,它會 穿透該接觸孔123、124 ' 125 ’且將形成與外部世 界接觸的電接點。然後以微影蝕刻方式和蝕刻界定 導電層丨26,以產生外部接點丨2?、U8、129,其最 終的結構如圖6所示’與圖丨作—比較。圖1所示為 最後元件的較佳圖形,縱使在某些例子中,層之厚 度被跨大了些亦然。 如前所述,基極區域1〇8是經由L〇c〇s*法界定 而得,見圖3所示。然後最好使用一雙層結構,此 結構是由一直接設於單晶矽上方的二氧化矽以及可 在,以熱成長一所稱之場氧化物1 〇7時作為一區域 性氧化罩幕的氮化矽組成。當做場區氧化時,某此 氧的橫向擴散將會沿著單晶矽和二氧化矽之間的^ 界層產纟,然後’亦會有某些氧成長會發生在氮化 (紙張尺度辟(叫Α4· < 2敝297公釐 (請先閱讀背面之注意事項再填寫本頁)r 4 2 3 ο 7 9 V. Description of the invention (5) A7 B7 Printed on the flat surface by the Pager Consumers Cooperative of the Central Bureau of Quasi-Ministry of Economic Affairs. This layer is implanted with arsenic to become N + type, and an emitter electrode 120 of a transistor will be formed after annealing treatment. After the layer 1 1 9 is laid out and etched to produce an emitter electrode, its structure is as shown in FIG. 5. Usually we will keep the upper polyethylene layer 119 (forming the emitter electrode) on the upper part of the emitter region 10, as shown in Figure 3 'and the upper part of the collector socket 112, where the collector The socket 112 serves as a collector terminal 121. Then the circuit is passivated by a layer such as silicon oxide. As shown in FIG. 6, the contact holes 123, 124, and 125 of the base, emitter, and collector of the transistor are all determined by lithography. After the contact hole is etched, the circuit is sputtered with a metal layer 26, such as aluminum, which will penetrate the contact holes 123, 124 '125' and will form an electrical contact with the outside world. The lithographic etching method and etching are then used to define the conductive layer 丨 26 to generate external contacts 丨 2 ?, U8, 129. The final structure is shown in Figure 6 'and compared with Figure 丨. Figure 1 shows a better picture of the final component, even if the thickness of the layer has been increased in some cases. As mentioned above, the base region 108 is defined by the Loccos * method, as shown in FIG. 3. Then it is best to use a two-layer structure. This structure is composed of a silicon dioxide directly above the single crystal silicon and can be used as a regional oxide mask when thermally growing a so-called field oxide 107. Composition of silicon nitride. When the field is oxidized, the lateral diffusion of this oxygen will be generated along the ^ boundary layer between single crystal silicon and silicon dioxide, and then some oxygen growth will also occur in the nitride (paper scale development (Called Α4 · < 2 敝 297mm (please read the precautions on the back before filling this page)

、1T, 1T

In· I— I: Γ - -I . -· I. , —I- n 1423079 ^五、發明説明(6 ) 經濟部中央標準局員工消费合作社印製 層的邊緣上’見圖3之130。此氧化物13〇即廣稱的” 烏嘴’’。因此基極區域的延伸只能到達微影姓刻界 定的氮化·氧化-罩幕結構所界定的程度。可說該區 域的正確性是由完成製造後所剩下的”烏嘴,,所界定 的。為能補償因為產生這些,,烏嘴”而缺乏的正確性 以及製程變化,該基極區域1 08不需要很大。因而 可在基極和集極之間獲得不需要的大電容。 此外’當在N -區域1 〇 4内產生場氧化物1 〇 7時,將 有摻雜濃度13 1(俗稱,’摻雜堆積,,(d〇pant pUe up))發 生在該場氧化物1〇7和單晶矽基質ι〇1的表面之邊界 内’如圖3所示。當P +型的聚乙醯胺丨13 (其形成本 質的基極)與側條1 18外的基極區域1〇8接觸時,它 會使得完成的NPN電晶體中之基極與集極之間的電 谷增加’見圖3和4所示。 發明概要 本發明之一目的在於解決上述之問題,因而可提 供特別是在雙極性電晶體中的半導體元件,其具有 較尚的效能,特別是一雙極性NPN電晶體在其基極 和集極之間之電容值降低,和一雙極性PNP電晶餅 在其射極和集極之間之電容值降低p 本發明之另一目的在於提供上述問題的解決方法 ’其中一基質電容(亦即一位於基質之表面上的被 動性電容元件)在當—NPN電晶體的基質和集極之 間的電容降低之同時形成。 本紙張尺度· (請先閲讀背面之注意ΐ項再填寫本頁) -ί,. 訂 14 2 -(: 7 '' at 14 2 -(: 7 '' at 經濟部中央標準局員工消費合作社印製 五、發明説明(7 ) 本發明之另一目的在於提供用以電絶綠在一基質 表面上的元件區域之中空的和深的基質接點。 本發明之又一目的在於提供一電晶體,通常是指 一雙極性裝置,其以一溝槽或多個構槽隔離,其中 該溝槽可以一有效的方式產生。 為產生一 NPN電晶體,引入一二氧化矽和氮化矽 之疊置層’它是被設置於此Npn電晶體的有效區域 (集極區域)之上端。此疊層是以微影蝕刻方式佈圖 ,以致於該電晶體的基極區域可由疊置層内的開口 界定。在此同時,可使用微穸蝕刻術界定的開口產 生一橫向的PNP電晶體,以產生此電晶體的射極和 集極。 亦有可能無須使用任何額外的罩幕步驟形成一使 用氮矽層作為一介電層的基質電容,在此同時,在 此雙極性NPN電晶體的基極和射極之間的電容值可 降低。 一半導體元件(可以是NPN型的雙極性電晶體)具 有一位於元件表面的主動區,該主動區是以一習知 的方式由厚的場氧化物區域包園,如可沿元件表面 看,者。此主動區係部份由一電絕緣表面層所涵蓋 ,最,包括一氮化層。在此主動區内的基極區是由 一在該電絕緣表面層内微影蝕刻產生的已界定開口 決疋。右此時之半導體元件是一 pNp型的雙極性電 晶體時代之纟,是射㈣集極區域在元件表 __-10- 本紙狀度適^關家標準(A4規格(2丨&297公董-— (請先閲讀背面之注意事項再填寫本頁)In · I— I: Γ--I.-· I., —I- n 1423079 ^ V. Description of the invention (6) On the edge of the printed layer of the staff consumer cooperative of the Central Standards Bureau of the Ministry of Economic Affairs, see Figure 3130. This oxide 13 is widely known as the "black mouth". Therefore, the extension of the base region can only reach the extent defined by the nitriding and oxidation-mask structure defined by the lithographic name. It can be said that the region is correct It is defined by the "Uzui," which is left after the manufacturing is completed. In order to compensate for the lack of correctness and process variations due to the generation of these, the base region does not need to be very large. Therefore, an unnecessary large capacitance can be obtained between the base and the collector. In addition, ' When a field oxide 1 07 is generated in the N-region 1 0 4, a doping concentration 13 1 (commonly known as 'dopant pUe up') occurs at the field oxide 1. 7 and the boundary of the surface of the single crystal silicon substrate ι〇1 'are shown in Figure 3. When the P + type polyacetamide 丨 13 (which forms the essential base) and the base region outside the side strip 1 18 When contacting 108, it will increase the valley between the base and the collector in the completed NPN transistor, as shown in Figures 3 and 4. SUMMARY OF THE INVENTION An object of the present invention is to solve the above problems, and therefore Can provide semiconductor elements, especially in bipolar transistors, which have relatively high performance, especially a bipolar NPN transistor with reduced capacitance between its base and collector, and a bipolar PNP transistor A reduction in the capacitance of a cake between its emitter and collector p Another object of the present invention is to provide the above problem Solution 'One of the substrate capacitors (that is, a passive capacitor element on the surface of the substrate) is formed at the same time when the capacitance between the substrate and the collector of the NPN transistor is reduced. This paper size · (Please read first Note the items on the back, please fill out this page again) -ί ,. Order 14 2-(: 7 '' at 14 2-(: 7 ``) Printed by the Staff Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs V. Invention Description (7) Another object of the present invention is to provide hollow and deep substrate contacts for electrically insulating the element area on a substrate surface. Another object of the present invention is to provide a transistor, generally referred to as a bipolar device. It is isolated by a trench or a plurality of trenches, wherein the trenches can be produced in an efficient manner. To generate an NPN transistor, a superimposed layer of silicon dioxide and silicon nitride is introduced. The upper end of the active area (collector area) of this Npn transistor. This stack is patterned by lithographic etching so that the base area of the transistor can be defined by the openings in the stacked layer. At the same time, the Apertures defined using micro-etching A lateral PNP transistor to generate the emitter and collector of the transistor. It is also possible to form a matrix capacitor using a silicon nitride layer as a dielectric layer without using any additional masking steps. The capacitance value between the base and the emitter of the bipolar NPN transistor can be reduced. A semiconductor device (which can be an NPN type bipolar transistor) has an active area on the surface of the device. The known method consists of a thick field oxide region. If you can see along the surface of the element, this active area is partially covered by an electrically insulating surface layer, and most of which includes a nitride layer. In this active area The base region is defined by a defined opening in a photolithographic etch in the electrically insulating surface layer. Right the semiconductor device at this time is a pNp-type bipolar transistor era, is the emitter collector region in the component table __- 10- The paper degree is appropriate Public Manager-— (Please read the notes on the back before filling this page)

-1T A7 B7 五、發明説明(8 經濟部中央標準局舅工消費合作社印袈 面上被該厚的場氧化物區域包圍,如沿元件表面看 出者.射極區域和/或一集極區域也可以一對應 的^式由一電絕緣表面層内的微影蝕刻界定的開口 決定。利用這兩例子中的微影蝕刻界定,該電絕緣 表面f將延伸超出和至該圍繞的場氧化物區域下方 使4于在基極區域之間和射極或集極之間以及最接 此I域的i每乳化物區各存在一條狀的電絕表面層 該電絕緣表面層有利地包括一位於上部的氮化矽 層和位於下方的二氧化矽層多疊層。該氮矽層最好 作為一同時產生的電容内之有效介電。 溝槽可用以絕緣元件,且可以傳統的方式經蝕刻 而成。接下來將一疊層加在該溝槽的壁上,此層包 含在底部的熱成長二氧化矽層和其上之以沉積方式 形成的薄氮化矽層。最後該溝槽剩下的主要部份會 因為施加一電絕緣或是半絕緣層,如一二氧化矽層 或是一未摻雜矽層(例如一單晶矽)於平面表面(例二 以適當的沉積方式)上的方式而被充填β然後該氮 化矽層在後來的平坦化蝕刻中作為一蝕刻終止層, 以使層平坦化,且該溝槽的主要部份被充填。此外 ,若當充填該溝槽時使用的材料具有雜質時,該氮 矽層會防止他們擴散進入基質材料中’如此的擴散 亦會降低溝槽的電絕緣功能。_ 圖式之簡述 -11 Μ-氏張尺度適用中國國家標準(CNS ) Α4規格(210Χ297公楚) (谇先閱讀背面之注意事項再填寫本頁) 訂 ;423079 A7 B7 經濟部中央樣準局員工消費合作·杜印製 五、發明説明(9 ) 現將以下列之非限制式實施例和附圖之參考描述 本發明,其中: 胃' 圖1所示為一具有因溝槽所提供的電絕緣力之雙 極性自行對準之雙多晶矽電晶體的剖面圖; 圖2所示為在形成一底擴散層和一磊晶表面層以 後’為產生如圖丨所示的電晶體之開始材科的剖面 圖; 圖3所示為類似於圖2、但在界定了主動區和以溝 槽絕緣以後的剖面圖; 圖4所示為類似於圖2、但在界定了射極開口和— 本質基極以後的剖面圖; 圖5所示為類似於圖2、但在界定了間隔物和射極 及集極以後的剖面圖; 圖6所示為類似於圖2、但在界定了第一金屬層以 後的剖面圖; 圖7所示為一碎層和其上配置之層的剖面圖,其 主要係用以產生一具有好的高頻特性的NPN電晶體 ,但亦為了產生一電容和一橫向的PNP電晶體,其 剖面圖顯示在形成底部擴散之前的平面; 圖8所示為類似於圖7之剖面圖,其顯示在形成底 部擴散以後的平面之狀態; 圖9所示為類似於圖8之剖面圖,但顯示在形成底 部擴散以後的剖面圖; 圖丨0所示為類似於圖9之剖面圖,其顯示在產生 -12 - 本紙張尺度迺用中國國家標準(CNS ) A4规格(210X297公釐) ssi· --------- Γ·_ * r i fur ^ ^^^^1 —fltp T1« (請先w讀背面之注意事項再填寫本頁) r Λ 2 3 0 a? _____Β7 五、發明説明(10 ) 一額外的P摻雜時之平面狀態: 圖1 1所示為類似於圖1 〇之剖面圖,顯示在平面表 面上沉積了磊晶矽以後之剖面圖; 圖12所示為類似於圖η之剖面圖,其顯示當選擇 性形成N-區域時之平面狀態; 圖13所示為類似於圖1 2之剖面圖,其顯示在選擇 式氧化該N-區域和形成自行對準的p_區域後之剖面 圖; 圖1 4所示為類似於圖1 3之剖面圖,其顯示在界定 元件區域以後之剖面圖’其中指示出不同的元件區 域; ' 圖1 5 a和1 5 b所示為類似於圖1 4之剖面圖,但在場 區氧化以後’其分別顯示一 NPN電晶體的區域和一 電容以及用於一橫向PNP電晶體的區域; 圖16所示為類似於圖15之剖面圖,但在界定了溝 槽以後的剖面圖; 圖1 7所示為類似於圖1 6之剖面圖,但已去除了一 硬的罩幕和屏障層,且氧化了溝槽内壁; 經濟部中央標準局員工消費合作社印裝 (婧先閱讀背面之注意事項再填寫本頁) 圖1 8所示為類似於圖1 7之剖面圖,其顯示在以一 聚乙醯胺層充填溝槽之後的剖面圖; 圖19所示為類似於圖18之剖面圖,其顯示在氧化 該溝槽之開口内的聚乙醯胺後之剖面圖; 圖2 0a和20b所示為類似於圖19之剖面圖,其顯示 當分別在用於NPN電晶體和一電容的區域上以及用 -13 - 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐} 2 3 0 7 9 A7 B7 五、發明説明(u 於一檢向PNP電晶體的—區域上形成集極時之平 之狀態; 圖2 la和2 lb所示為類似於圖2〇a和20b,分別顯示 在沉積一氮矽層和界定一射_基區域後之剖面圖; 圖22所示為類似於圖2 1之剖面圖,其顯示在界定 一基極區域和沉積鱗碎後之剖面圖; 圖23a和3b所示為類似於圖22之剖面圖,其分別 顯示在界定一射-基極區域、一上電容板和ϋ連 接後之剖面圖,顯示用於一 ΝΡΝ電晶體和—電容之 區域和用於一橫向ΡΝΡ電晶體之區域:-1T A7 B7 V. Description of the invention (8 The printed surface of the Central Government Bureau of Standards and Technology Cooperatives of the Ministry of Economic Affairs is surrounded by the thick field oxide region, as seen along the element surface. The emitter region and / or a collector The area can also be determined by a corresponding ^ formula defined by the opening defined by the lithographic etching in an electrically insulating surface layer. Using the lithographic etching defined in these two examples, the electrically insulating surface f will extend beyond and to the surrounding field oxidation A strip of electrically insulating surface layer exists between the base region and the emitter or collector and between each of the emulsion regions closest to this I region under the object region. The electrically insulating surface layer advantageously includes a The upper silicon nitride layer and the lower silicon dioxide layer are stacked. The silicon nitride layer is preferably used as an effective dielectric in a capacitor that is generated at the same time. The trench can be used as an insulating element and can be passed in a conventional manner. It is etched. Next, a stack is added to the wall of the trench. This layer includes a thermally grown silicon dioxide layer at the bottom and a thin silicon nitride layer formed thereon by deposition. Finally, the trench The remaining major part will An electrically insulating or semi-insulating layer, such as a silicon dioxide layer or an undoped silicon layer (such as a single crystal silicon), is filled with β on a flat surface (eg, in a proper deposition method) and then the The silicon nitride layer is used as an etch stop layer in the subsequent planarization etching to flatten the layer and the main part of the trench is filled. In addition, if the material used when filling the trench has impurities, The nitrogen-silicon layer will prevent them from diffusing into the matrix material. 'Such diffusion will also reduce the electrical insulation function of the trenches. _ Brief description of the drawings -11 M-scale scales are applicable to Chinese National Standard (CNS) A4 specifications (210 × 297 Gongchu) (谇 Please read the precautions on the back before filling this page) Order; 423079 A7 B7 Consumption Cooperation by Staff of the Central Bureau of the Ministry of Economic Affairs · Du printed 5. Description of Invention (9) Will now be implemented in the following non-limiting manner The invention is described with reference to examples and drawings, in which: stomach 'FIG. 1 is a cross-sectional view of a bi-polycrystalline silicon transistor having bipolar self-alignment due to the electrical insulating force provided by the trench; FIG. 2 is a view In forming a bottom diffusion layer and After the epitaxial surface layer is a cross-sectional view of the starting material family of the transistor shown in FIG. 丨; FIG. 3 is a cross-sectional view similar to FIG. 2 but after the active area is defined and the trench is insulated; Figure 4 is a cross-sectional view similar to Figure 2 but with the emitter opening and the essential base defined; Figure 5 is similar to Figure 2 with the spacer and the emitter and collector defined 6 is a cross-sectional view similar to FIG. 2 but after the first metal layer is defined; FIG. 7 is a cross-sectional view of a broken layer and a layer disposed thereon, which are mainly used for Generate a NPN transistor with good high frequency characteristics, but also to generate a capacitor and a lateral PNP transistor, the cross-sectional view shows the plane before the bottom diffusion is formed; Figure 8 shows a cross section similar to Figure 7 FIG. 9 shows a state of the plane after the bottom diffusion is formed; FIG. 9 shows a cross-sectional view similar to FIG. 8, but shows a cross-sectional view after the bottom diffusion is formed; and FIG. 0 shows a cross-section similar to FIG. 9. Figure, which shows the -12-This paper scale uses China Home Standard (CNS) A4 Specification (210X297mm) ssi · --------- Γ · _ * ri fur ^ ^^^^ 1 —fltp T1 «(Please read the precautions on the back before filling (This page) r Λ 2 3 0 a? _____ Β7 V. Description of the invention (10) An additional P-doped planar state: Figure 11 shows a cross-sectional view similar to Figure 10, showing deposition on a planar surface A cross-sectional view after epitaxial silicon is shown; FIG. 12 is a cross-sectional view similar to FIG. Η, which shows a planar state when an N-region is selectively formed; FIG. 13 is a cross-sectional view similar to FIG. It shows a cross-sectional view after selective oxidation of the N-region and formation of a self-aligned p_ region; FIG. 14 shows a cross-sectional view similar to FIG. 13, which shows a cross-sectional view after the element area is defined. The different element regions are indicated; 'Figures 15a and 15b show cross-sections similar to Figure 14 but after the field region has been oxidized', which respectively show a region of an NPN transistor and a capacitor and the In a region of a lateral PNP transistor; FIG. 16 is a cross-sectional view similar to FIG. 15 but with a trench defined; FIG. 1 7 is a cross-sectional view similar to FIG. 16 but with a hard cover and barrier layer removed and the inner wall of the trench oxidized; printed by the Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs (Jing first read the note on the back) Please fill in this page again) Figure 18 shows a cross-sectional view similar to Figure 17, which shows a cross-sectional view after filling the trench with a layer of polyethylene; Figure 19 shows a cross-section similar to Figure 18 Figure 20 shows a cross-sectional view after oxidizing the polyethylene in the opening of the trench; Figures 20a and 20b are cross-sectional views similar to Figure 19, showing when used for NPN transistors and a Capacitor area and use -13-This paper size applies Chinese National Standard (CNS) A4 specification (210X297 mm) 2 3 0 7 9 A7 B7 V. Description of the invention (u on the area of a PNP transistor) The flat state when the collector is formed; Figures 2a and 2lb are similar to Figures 20a and 20b, showing cross-sectional views after depositing a silicon-silicon layer and defining an emitter region; Shown as a cross-sectional view similar to FIG. 21, showing a cross-sectional view after delimiting a base region and sedimentary scales 23a and 3b are cross-sectional views similar to FIG. 22, which respectively show cross-sectional views after defining an emitter-base region, an upper capacitor plate, and a puppet, and are used for an NPN transistor and a capacitor. And the area for a lateral PN transistor:

圖24所示為類似於圖23a之剖面圖,其顯示舍做 一基質植入時之平面之狀態; W 圖25a所示為類似於圖24之剖面圖,其顯示序在 基質連接終端和射極連接終端之間形成用以^ &的 間隔物時之平面之狀態; ’ 圖25b所示為圖24在形成一變化設計的 之剖面圖的一部份; 圖25c所示為根據圖25a之射極結構以〜 ^ 览子顯微 鏡攝取的圖像; 短濟部中央標準局員工消費合作社印裝 電子顯微 圖25d所示為根據圖25b之射極結構 鏡攝取之圖像; 其顯示當形 聚乙醯胺層 (請先聞讀背面之注意事項再填寫本頁:> 圖26a所示為類似於圖25a之剖面圖 成一射極時之平面之狀態,其中顯示 在蚀刻之前和之後的圖形和顯示用於〜而晶體 -14- 本紙張尺度適用中國國家標準(CNS ) A4規格(21〇Χ297公釐) M濟部中央標準局員工消費合作·社印製 『4 230 7 9 a? ------ B7 五、發明説明(12 ) 和一電容的區域; 圖2 6 b b所示為結構之一部份的剖面圖,其部份出 現在圖26a中’它並顯示低阻抗和高阻抗電阻器的 產生; 圖2 6 c所示為類似於圖2 6 a之剖面圖,其顯示用於 一橫向PNP電晶體的區域; 圖27a和27b所示分別為類似於圖26a和26b之剖面 圖,其顯示在蝕刻施加於一被摻雜成P+的聚乙醯胺 層之上方的氧化層以後之剖面圖; 圖28&和2813所示分別為類似於圖27&和2715之剖面 圖,其顯示在射極和基極區域驅入(drive_in)和蝕刻 以產生額外的間隔物以後之剖面圖; 圖29所示為一製得的NPN電晶體之摻雜外觀圖, 該外觀係經由SIMS攝得; 圖30所示為當蝕刻圖28a和28b中所示之間隔物時 ,由電阻上方之保護罩幕層上視之之圖示; 圖3 1所示為圖2 7 a的剖面圖部份,其只顯示產生 ό-j ^ 0¾ fa 圖32a和32b所示分別類似於圖2.8a和28b之剖面圖 ,其顯示在沉積鈦和矽化和化學移除該鈦和氮化鈦 以後的剖面圖; 圖3 3所杀分別類似於圖3 2 a之剖面圖,其顯示在 蝕刻用於一基質的電連接之深接點孔後之剖面圖; 圖34a和34b所示分別類似於圖33之剖面圖,其顯 15 - 本紙張尺度適用中國國家標準(CNS ) A4規格(2丨OX2?7公釐) ---------X------訂------点 (請先閲讀背面之注意事項再填寫本頁) Γ4 2 3 Ο 7 9 Α7 Β7 五、發明説明(13 經濟部中央標準局員工消費合作社印製 示在蝕刻了所有接點孔後之剖面圖; 圖35a所示為最後製得的電路元 包含兩種電晶體、兩種電容和—電阻元|牛..,其 圖圖35b所示係類似於圖35a之剖面圖,i 實質相同的最後製得的電路元件 ^中顯不 溝槽分開; 仁匕們彼此以一 圖3 61示為以一電子顯微鏡攝得的最終產品 圖,其中深基質接點被充填以鎢,可看電路 胺電阻和一以溝槽絕緣的NPN電晶體;和,乙醯 。圖37所示為上視®,其說明不同的元件如何” 較佳實施例之推诚 4考示’以下將敘述具有高效能的不同 4件的產生万式’這些元件皆是同時在 基質上產生的。這些圖中某些(顯示—基 圖)是非常示意的,然而其他的圖式則以昭片= 像顯示1然是最佳的最後結構。亦可觀察到二某: 元件而言1下文中之描述是具有限定接雜型式: 材料產生和建構的’但這些對應的元件也可以由相 反挣雜的型式疋材料產生’亦即以有限的ρ摻雜第 -材料和有限的Ν摻雜的第二材料做成的元件在某 些例子中也可以以對應的N摻雜之第一材料和對應 的Ρ摻雜之第二材料做成。 " 圖7係顯示一 Ρ型的矽平面丨在形成一底部擴散或 16 本纸張足度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閱讀背面之注意事項再填寫本頁} 乂 丁 --=° 旅 2 3 Ο 7 9 五、發明説明(14 Α7 Β7 經濟部中央標準局男工消費合作社印製 埋入式Ν型層之前的剖面圖。此矽平面1可以由一 同質性、非常弱摻雜的Ρ型平面組成,其基本上之 阻抗為10-20歐姆-公分’它也可以以ρ型或是所稱 义epi平面敘述’其中基質”丨,,是由基本上為幾十微 歐姆-公分的阻抗之P型(P+)高摻雜平面組成,其上 已成長有P型的磊晶弱摻雜層(P__)。該磊晶成長的 P--型基本上有,其阻抗值為歐姆· 公分。一類似於前述之最後提到之例子的開始材料 被使用於雙極性結構中,它被揭示於v delta T〇rre 等人在”MOSAlC V - A Very High Perf〇rmance Technology” ’ BCTM1991 第 21_24 頁中所揭示之文 獻中。根據此文獻,使用一高度摻雜的p+型基質, 且其上為磊晶本質層,亦即無適當摻雜的層。然後 在此本質層的表面上形成類似於將於下文描述的結 構,如底部擴散等等。 ° 使用某一已知的方法將一相當厚的二氧化矽保護 層2加在咬平面⑼表面上,例如以熱氧化法。此氧 化層2的厚度被選擇為最好约為〇·8…以施加和 佈圖一光阻層3的方式對此氧化層2做微影蝕刻佈圖 接:來將這些部份内未為該光阻層3保護的氧 化物溶解或蚀刻掉,如_ 7 & + ^ 干如園/所不。此氧化物的移除 可以已知的濕化學或數仆堅士 、,予為乾化子万法進行,然後某—已 知的方式將孩光阻層3去除。 -基本上厚度為幾百埃的薄保護氧化層4被熱成 (請先閱讀背面之注意事項再填荇本頁) *1Τ 17- 私紙張尺度適用中國國家標準()Α4^72Ϊ〇'χ 297;^ r423〇 7 9 五、發明説明(15 ) 經濟部中央標车局貝工消費合作社印褽 長於平面^的表面上,見圖8所示,此保護層特別被 $置在先則施加的二氧化矽層2之相當厚的剩餘部 份之間的區域上,因為如此小的厚度在該部份内還 不致被/主意到。接本來產生一 N+型的底部擴散層 或一稱為” N+埋入層”,其方式是首先做一離子植入 ,如圖箭頭所示。在此離子植入步驟中使用砷 ,,植入之能量約為50keV和其摻雜濃度約為3 χ 1〇離子/平方公分。該厚氧化層2的剩下部份在植 入期間將作為一罩幕,接著在植入時之能量必須被 採用,以致於只有那些撞擊,氧化層4的離子可以 穿透矽基質,如圖8中之4a處的十字所指示者而 其他的離子則會受到該厚氧化層之阻隔。完成離子 植入以後,即以尚溫(基本上在1 1 〇 〇 I時進行3 〇分 鐘以内)進行退火,以驅入植入之摻雜子(亦即砷原 子)’以產生底部擴散層5,見圖8所示。做完此驅 ^步驟以後,該N+型的底部擴散區域5(“N+埋入層 ”)的最後深度約為丨.5jCZm,見圖9所示。在此驅又 操作時,同時切表面上有進—步的氧化,該表面 上的矽原子被消耗,使得該薄二氧化矽保護層斗之 厚度增加到2〇〇nm,藉此而獲得一有較 6產生。因切原子消耗之故,在該 <被先前施加的厚氧化層2覆蓋之區域以及現在辑 得較厚且仍為一相當薄的氧化層覆蓋的區域之間= 表面上會有一階梯或一罩蔽,而這些階梯在後續的 -----____ ,18 - 本紙 標準(CNS) Α4· (21()><297公 (請先閱讀背面之注意事項再填寫本莧) 訂 g:d23〇79 經濟部中央標準局貝工消费合作社印製 A7 B7 五、發明説明(16 ) 製程中被用作一對齊罩幕。 自然地也可使用其他的N型摻雜植入,以產生戈 底部擴散層5’如銻。但是,銻之使用需要在較; 的驅入溫度下退火,基本上是在約125crc下進行半 個小時。 在用以產生全部為氧化物的N+型底部擴散5之驅 入操作自平面表面離開以後,最好是以一濕化學的 万式,接下來可看見在單晶矽平面之表面上之前述 階梯形狀,如圖10所示。接下來在平面表面上產生 一約為30-4Onm厚的薄保護氧化層7,其產生方式最 好是以熱成長。然後以約丨0keV之能量和約為4 χ 1〇12離子/平方公分的摻雜濃度在整個平面表面上植 入最好是硼的離子,而在該Ν+型的底部擴散區域5 之間的區域内產生額外的ρ_型摻雜,如圖1〇之箭頭 所示。該植入能量和這些摻雜濃度可以調整,以使 這些接著被植入該已摻雜有坤的Ν+型底擴散區域5 内的硼原子可完全地被這些區域内的摻雜封閉和補 償,藉此’該底擴散5會繼續為Ν+型,其摻雜原子 之内容之減少相當地少或是可予忽視。 值得觀察吾人可以不需使用上述之硼植入即可管 理和獲得良好功能的元件,其方式是增加原本非常 低摻雜的開始材料中的接雜程度,使其接近於本質 的’成為Ρ -型。但是,本例中在最後冗成的元件中 ’由Ν+型的底部擴散區域5中分佈的電容將較高。 -19- 本纸伕尺度適用中國國家橾準(CNS ) Α4規格(210X 297公釐) (請先閱讀背面之注意事項再填寫本頁)FIG. 24 is a cross-sectional view similar to FIG. 23a, which shows the state of the plane when a matrix is implanted; W FIG. 25a is a cross-sectional view similar to FIG. Figure 25b shows a part of the cross-sectional view of FIG. 24 in forming a changed design; FIG. 25c shows a part according to FIG. 25a The image of the emitter structure was taken with a ~ ^ microscope; the printed electron micrograph of the Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs printed in Figure 25d is an image taken according to the emitter structure mirror of Figure 25b; Polyacetamide layer (please read the notes on the back before filling out this page:> Figure 26a shows the state of the plane when the cross-sectional view similar to that of Figure 25a forms an emitter, which shows before and after etching The graphics and display are used for ~ and crystal-14- This paper size is applicable to Chinese National Standard (CNS) A4 specification (21 × 297 mm) M Printed by the Ministry of Economic Affairs Central Standards Bureau Consumer Consumption Cooperative · 4 230 7 9 a ? ------ B7 V. Description of the invention (12) and Yidian Figure 2 6 bb shows a cross-sectional view of a part of the structure, a part of which appears in Figure 26a 'It also shows the production of low and high resistance resistors; Figure 2 6 c shows a similar Fig. 26a is a cross-sectional view showing a region for a lateral PNP transistor; Figs. 27a and 27b are cross-sectional views similar to Figs. 26a and 26b, respectively, showing the etching applied to a doped to P + Sections after the oxide layer above the polyacetamide layer; Figures 28 & and 2813 show cross-sections similar to Figures 27 & and 2715, respectively, showing drive_in in the emitter and base regions A cross-sectional view after etching and etching to generate additional spacers; FIG. 29 shows a doped appearance of a prepared NPN transistor, the appearance is taken by SIMS; FIG. 30 shows when etching FIG. 28a and 28b The spacer shown in the figure is viewed from the top of the protective cover layer above the resistor. Figure 31 shows the cross-sectional view of Figure 2 7a, which only shows the ό-j ^ 0¾ fa figure. Figures 32a and 32b show cross-sections similar to Figures 2.8a and 28b, respectively, showing the Sectional views after the titanium and titanium nitride; FIG. 33 is a cross-sectional view similar to that of FIG. 3a, showing a cross-sectional view after etching a deep contact hole for electrical connection of a substrate; FIG. 34a Figures 34 and 34b are similar to the cross-sectional views shown in Figure 33, respectively. 15-This paper size applies to the Chinese National Standard (CNS) A4 specification (2 丨 OX2? 7mm) --------- X-- ---- Order ------ points (please read the notes on the back before filling this page) Γ4 2 3 〇 7 9 Α7 Β7 V. Description of the invention (13 Printed by the Staff Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs Cross-section view after all contact holes have been etched; Figure 35a shows that the final circuit element contains two transistors, two capacitors, and a -resistor element | new .. Its figure is similar to that shown in Figure 35b The cross-sectional view of FIG. 35a shows that the substantially-reduced circuit elements ^ are substantially the same without separation of the trenches; the final products are taken by an electron microscope as shown in FIG. 36 to each other, in which the deep matrix is connected The dots are filled with tungsten to see the circuit amine resistance and a trench-insulated NPN transistor; and, acetamidine. Figure 37 shows Topview®, which explains how different components are. "The best example of the 4th study" The following will describe the production of different 4 pieces with high performance "These components are all on the substrate at the same time Produced. Some of these diagrams (display-base diagram) are very schematic, but others are shown in the film = image 1 is the best final structure. You can also observe the two: 1 The description below has a defined doping pattern: the material is produced and constructed 'but these corresponding elements can also be produced from the oppositely doped type' material, that is, the first material is doped with a limited ρ and the finite N In some examples, a component made of a doped second material may also be made of a corresponding N-doped first material and a corresponding P-doped second material. &Quot; Figure 7 shows a P-type Silicon plane 丨 spreading at the bottom or 16 papers are fully compatible with Chinese National Standard (CNS) A4 specifications (210X297 mm) (Please read the precautions on the back before filling this page} 乂 丁-= ° 旅 2 3 Ο 7 9 V. Description of the invention (14 Α7 Β7 Sectional drawing of the male bureau of the Standards Consumer Cooperative before printing the buried N-type layer. This silicon plane 1 can be composed of a homogeneous, very weakly doped P-type plane, and its basic impedance is 10-20 ohms- Cm 'It can also be described in terms of the ρ-type or the epi plane, where the matrix', which is composed of a P-type (P +) highly doped plane with an impedance of tens of micro-ohm-cm, above which A P-type epitaxial weakly doped layer (P__) has been grown. The P-type of this epitaxial growth basically has an impedance value of ohm · cm. A starting material similar to the last-mentioned example mentioned above Used in bipolar structures, it is disclosed in the literature disclosed by v delta Torre et al. In "MOSAlC V-A Very High Perfance Technology" 'BCTM1991 pages 21_24. According to this document, a A highly doped p + -type substrate with an epitaxial intrinsic layer, that is, a layer without proper doping, and then a structure similar to that described below, such as bottom diffusion, is formed on the surface of this intrinsic layer. ° Using a known method The silicon dioxide protective layer 2 is added on the surface of the bite plane, for example, by a thermal oxidation method. The thickness of this oxide layer 2 is preferably selected to be about 0.8... By applying and patterning a photoresist layer 3 This oxide layer 2 is lithographically etched and patterned: to dissolve or etch away the oxides in these parts that are not protected by the photoresist layer 3, such as _ 7 & + ^ as dry as the garden / not. This oxidation Removal of the material can be carried out by known wet chemistry or by several methods, and then by desiccant method, and then-a known method to remove the photoresist layer 3.-Basically a thickness of several hundred angstroms The thin protective oxide layer 4 is thermally formed (please read the precautions on the back before filling this page) * 1Τ 17- The standard of the private paper applies the Chinese national standard () A4 ^ 72Ϊ〇'χ 297; ^ r423〇7 9 V. Description of the invention (15) The seal of the Beige Consumer Cooperative of the Central Bureau of Standard Vehicles of the Ministry of Economic Affairs is longer than the surface of a flat surface, as shown in Fig. 8. This protective layer is particularly thick, and the silicon dioxide layer 2 applied in advance is quite thick. In the area between the remaining parts of the part, because such a small thickness is not yet noticed / attended in that part. An N + -type bottom diffusion layer or an “N + buried layer” is originally generated. The method is to first do an ion implantation, as shown by the arrow. Arsenic is used in this ion implantation step. The implantation energy is about 50 keV and its doping concentration is about 3 x 10 ions / cm 2. The remainder of the thick oxide layer 2 will serve as a veil during implantation, and then the energy during implantation must be used so that only those impacting, the ions of the oxide layer 4 can penetrate the silicon substrate, as shown in the figure. The cross at 4a of 8 indicates that the other ions are blocked by the thick oxide layer. After the ion implantation is completed, annealing is performed at a still temperature (basically within 30 minutes at 11000) to drive the implanted dopants (that is, arsenic atoms) to generate a bottom diffusion layer 5, as shown in Figure 8. After completing this driving step, the final depth of the N + -type bottom diffusion region 5 ("N + buried layer") is about .5jCZm, as shown in FIG. During the operation of this driver, there is further oxidation on the cut surface at the same time, the silicon atoms on the surface are consumed, so that the thickness of the thin silicon dioxide protective layer bucket is increased to 2000 nm, thereby obtaining a There are more than 6 produced. Due to the consumption of tangential atoms, between the area covered by the previously applied thick oxide layer 2 and the area now thicker and still covered by a relatively thin oxide layer = there will be a step or Cover, and these steps in the subsequent -----____, 18-Standard of this paper (CNS) Α4 · (21 () > < 297g (Please read the precautions on the back before filling in this card) Order g : D23〇79 Printed by A7 B7, Shellfish Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs 5. Description of the Invention (16) The process is used as an alignment mask. Naturally, other N-type doping implants can also be used to produce The bottom diffusion layer 5 'is like antimony. However, the use of antimony needs to be annealed at a relatively low drive temperature, which is basically performed at about 125 crc for half an hour. It is used to generate all N + type bottom diffusions that are oxides. After the drive-in operation of 5 leaves from the flat surface, it is best to use a wet chemical method, and then the aforementioned step shape on the surface of the single crystal silicon plane can be seen, as shown in Fig. 10. Next on the flat surface A thin protective oxide layer 7 having a thickness of about 30-4 Onm is generated on the The growth method is best to grow by heat. Then, with the energy of about 0 keV and the doping concentration of about 4 x 1012 ions / cm 2, the ions, preferably boron, are implanted on the entire planar surface, and Additional p-type doping is generated in the region between the + -type bottom diffusion regions 5 as shown by the arrow in Figure 10. The implantation energy and the doping concentration can be adjusted so that these are then implanted into the The boron atoms in the N + -type bottom diffusion regions 5 which have been doped with Kun can be completely blocked and compensated by the doping in these regions, thereby 'the bottom diffusion 5 will continue to be N + type, and The reduction of the content is relatively small or can be ignored. It is worth observing that we can manage and obtain good functioning components without using the above boron implants, by increasing the doping in the original material with very low doping To the extent that it is close to the essence, it becomes a P-type. However, in the last redundant element in this example, the capacitance distributed in the bottom diffusion region 5 of the N + type will be higher. -19- Paper Standards apply to China National Standards (CNS) Α4 specifications (210X 297 PCT) (Please read the notes and then fill in the back of this page)

鲤濟部中央標準局員工消費合作杜印製 ^ 4 2 3 0 7 9 ' a? -----------B7 五、發明説明(Π) 了般用以產生N+型底部擴散和某些p胃型的中間區域 亦已揭示於Havemann的美國專利第5,374,845號中 ’見該專利之圖2和3之敘述。 。做完了上逑之較佳硼離子植入後,當在底部擴散 區,5 I間取得一弱p摻雜(即p_型)的區域時,所有 的氧化層再度被自平面表面上去除,最好是以濕化 學方式’使得階梯形狀再度出現,使用一某一已知 的方法在基質表面上成長一磊晶矽層9,見圖丨丨所 示此,々為m厚之县晶珍層9最好未摻雜(即” 本質的矽”)。若需要,此層$可在該磊晶成長中已 被摻,成N-型式,本例中其摻雜的内容基本上是ι X 10 /cm3之大小。在前述之Havemann之美國專利 案中’對應的磊晶層被非常輕微地摻雜,其阻抗大 於10歐姆-公分,但是也可說是實質為本質的,亦 即未摻雜。但是,一同質性摻雜的磊晶層將使得在 ,所謂的”上下接點”時之基質與表面的後來連接變 得j為困難《在該磊晶成長時,所使用之溫度使得 在前次植入的P-型區域内之接受器原子擴散到基質 1内,且同時形成該磊晶層9 ,以致於可在基質i和 磊晶矽層9之連接處之不存在有該N+型底部擴散層 5的位置上形成P-型的埋入區域8,如圖丨丨所示。再 一次地,該磊晶層9之上外表面上具有階梯狀。 如後文將描述者,該磊晶層9將被選擇性地摻雜 ,以個別地形成N-和P-型的區域(所稱之”N_穴和 -20- 本纸裱尺度適用中國國家標準(CNS ) A4規格(210X297公整 (請先閱讀背面之注意事唄再填蹲本頁」Printed by the Consumer Standards Department of the Central Bureau of Standards of the Ministry of Economic Affairs of the People's Republic of China Du 4 ^ 2 3 0 7 9 'a? And certain p-stomach-shaped intermediate regions have also been disclosed in Havemann, US Patent No. 5,374,845, 'see that patent for Figures 2 and 3. . After the completion of the implantation of the better boron ions, when a weak p-doped (ie p-type) region is obtained between 5 I in the bottom diffusion region, all oxide layers are removed from the planar surface again. It is best to 'wet chemically' make the step shape reappear, and use a known method to grow an epitaxial silicon layer 9 on the surface of the substrate, as shown in Figure 丨 丨. Layer 9 is preferably undoped (ie "essential silicon"). If necessary, this layer can be doped during the epitaxial growth to form an N-type. In this example, the content of the doped layer is basically ι X 10 / cm3. In the aforementioned US patent of Havemann, the corresponding epitaxial layer is very slightly doped, and its impedance is greater than 10 ohm-cm, but it can also be said to be essentially essential, that is, undoped. However, the homogeneously doped epitaxial layer will make the subsequent connection between the substrate and the surface at the so-called "upper and lower contacts" difficult. "When the epitaxial growth, the temperature used makes the former Receptor atoms in the sub-implanted P-type region diffuse into the matrix 1 and form the epitaxial layer 9 at the same time, so that the N + type may not exist at the junction of the matrix i and the epitaxial silicon layer 9 A P-type buried region 8 is formed at the position of the bottom diffusion layer 5, as shown in FIG. Once again, the epitaxial layer 9 has a stepped shape on the outer surface. As will be described later, the epitaxial layer 9 will be selectively doped to individually form N- and P-type regions (known as "N_holes and -20- National Standard (CNS) A4 specification (210X297 metric whole (please read the precautions on the back first, then fill in this page)

五、發明説明(IS ) 經濟部中夾樣準局員工消費合作社印製 穴)。在N-型區域中(直接放置在該N+型底部擴激 層5的上方),將形成雙極性電晶體和電容。在中間 部分上’在芫成製作連接路徑或區域後將存在p—區 域’以作為在表面和基質1上形成的電路或元件之 間的連接。 在成長%該磊晶層9以後’在晶片表面上產生一 薄的屏障氧化層1〇(最好以熱氧化方式形成),見圖 12所示。此氧化層的厚度基本上約為4〇nm。此屏 障氧化層10的上方是一薄的氮矽層丨丨,基本上其厚 度約為1 30nm,它是經由LPC VD方法(”低愿化學汽 相蒸鍍法”)沉積而成。此氮化層丨丨經施加一光阻層 Π ’和對其佈圖的方式微影蝕刻形成,在未受該光 阻層1 1 ’保護的部份上之氮化層被姓刻掉,如圖i 2 所示,且這些部份包括元件的區域,其中現在只剩 下二氧化矽層10而已。該氮矽層1丨最好以適當的乾 蚀刻製程蝕刻掉’該蝕刻過程可以選擇性地只去除 氮化層但留下在氮化層下方的氧化層10,接下來如 圖1 2之箭頭所示,在其上做一離子植入步驟,以在 該為晶矽層9内產生上述之N型或是N區域13(所稱 之’’N穴”)’在其下方的薄的氧化矽層1〇作為該磊晶 層9的表面之保護層。 在一較佳實施例中’此用以形成該N_區域丨3的離 子植入步驟(見圖1 3)最好是_由約45 OkeV能量和約 1.5X1012離子/平方公分植入磷的方式進行。但是 (請先閱讀背面之注意事項再填寫本頁) I l^n t n , -Λ. ▲ -21 - 本纸蒗尺度適用中國國家標準(CNS ) A4規格(210X:297公釐) fA23〇^9 五、發明説明(19 ) A7 B7 n^i *11— mi 經濟部中央榡準局員工消費合作社印製 ,若在該N-區域13中有另一種摻雜形式,則此植入 條件可以修改。 在巧入以後,光阻層〗1 ’被以某已知的方式去除 ,在氮化層11的開口内之薄氧化層1〇會因瓿成 增厚,以致於在成長以後,可得到約450nm厚的氧 化石夕層1 2,見圖1 3所示。用於前述之植入的捧雜子 在本幸X佳例中疋磷,如前所述,在此氧化步驟期間 ,L將擴散至泫羞晶層9中,藉此該N區域13的製 造已部分f成。其中之某些摻雜子的擴散將在接下 來的包括高溫之步驟期間發生。剩下的氮化層u部 份(它在矽平面表面之熱氧化時作為屏障層使用)會 導使氧化物在此熱成長期間只成長在氮化物層丨i未 被去除的區域上,亦即在已經被植入的這些區域中 。氧化以後,將氮化層丨丨完全地自平面上去除最 好以濕化學方法。在其下方的薄氧化層1 〇保留,並 在較厚的氧化矽區域12的邊緣上形成階梯狀,該薄 的氧化層10在接下來用以產生前述之p_型或p_區域 (即所稱的”P穴”)的植入步驟中作為—保護層使用 〇 在此離子植入步驟中的能量可以調整,以使離子 只能夠通過其表面上具有該薄氧化層1〇的區域,且 被其表面具有較厚的氧化層丨2之區域阻隔^經此可 獲得與前述之植入的N-區域13自行對齊的p_區域14 或P-穴。在較佳實施例中,此用以形成p_區域丨4的 22 本纸張尺度適用中國國家標準(CNS ) A4規格(ZIOXM7公楚} (請先閱讀背面之注意事項再填寫本頁) 乂. 訂 lA23〇"^ 9 at 五、發明説明(20 ) 經涛部占-央標準局員工消費合作社印製 離子植入之方式是以約5〇keV之能量和約2χ 1〇"_2 X 1〇14離子/平方公分的方式植入硼。但是,此植入 條件在該Ρ-區域14中希望有另—種摻雜子型式的情 況下亦是可以改變的。植入步驟完成以後,在一高 溫下(基本上在约1000。(:時進行4小時)做退火,以 驅動所植入的摻雜子,以獲得該队和?區域13、14 所需的擴散深度《經過退火後之結構如圖丨3所示。 上述之用以產生N_和p_區域的製程亦已描述於 Havemann的美國專利中。 驅入步騾以後,所有的氧化區域皆以最好是濕化 學触刻去除’亦即,具有薄氧化層1〇的區域、具有 厚氧化層1 2的區域和可能的額外氧化層(此氧化層 可在先前為驅入所植入原子而做之退火時在平面表 面上發現)。移除了該氧化層以後,矽晶片的表面 上出現階梯狀。接著使用已知的L〇c〇S方法界定用 於要被製造的元件之有效開口,因此,首先最好以 熱氧化的方式在平面的整個表面上施加—薄的屏障 層15 ’見圖14所示。在此氧化層15的上部沉積有— 較厚的氧化層16,其厚度基本上是2〇 〇nm,最好以 LPCVD方法形成。該氮化層16是以施加一光阻層17 和對其做佈圖以界定元件區域的方式以微影蝕刻地 佈圖’該氮化層1 6上未為光阻層1 7保護的部份被蝕 刻掉,如圖14所示。此氮化層丨6最好以一適當的乾 姓刻過程去除’此蝕刻過程長可選擇性地只去除氮 -23- 本紙張尺度通用中國國家標準(CNS ) A4規格(2】0>7297公釐) (請先閱讀背面之注意事項再填寫本頁〕 訂 *d23〇f A7 B7 五、發明説明(21 ) 經濟部中央螵辛局負工消費合作社印衷 化層和留下在其下方的薄屏障氧化層】5。 圖14中顯示三個分開的N_區域丨3,其中由左到右 分別形成一橫向的NPN電晶體、一電容和一垂直的 NPN電晶體。接著氮化矽層16大致上覆蓋於要形成 —橫向PNP電晶體之基極連接、集極和射極的區域 上,其中部份的電極連接會被形成於一區域,而另 一區域中則要於電容中形成一電極連接和一介面層 ,以及一用於垂直式NPN電晶體的主動區域和用於 集極連接的區域。 將光阻層之開口内的氮化層1 6轴刻掉後,該光阻 層會以一已知的方式去除’然後在其上面以熱成長 的方式將一約60nm厚的氧化矽層1 8(所稱之場氧化 物)成長於該氮化層1 6的開口内。在較佳實施例中 ’該場氧化物1 8最好是在基本上9 5 0。(:之濕大氣下 成長而成,該氮化層16之出現(其在矽表面的熱氧 化期間作為一屏障層)會使該氧化矽只會成長於氮 化層已被去除的區域内。既然在氮化層丨6的開口内 會有部份矽會在其轉換成二氧化矽時被消耗掉,因 此場氧化物1 8在基質表面内或是該磊晶層9的正表 面上將會部分凹陷,稱為,,半凹陷”,其結果顯示於 圖15a和15b中,其中圖15a顯示其内要產生電容和 NPN電晶體的區域,而圖丨5B則顯示其内要產生橫 向PNP電晶體的區域。在後圖中,亦可以看到該場 乳化物1 8之區域如何在氣化珍層16的邊緣處成長。 -24 - 本紙ft尺度適用中國國家標準(CNS ) A4规格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁) 五、發明説明(22 :) 經濟部中央標準局員工消費合作衽印製 經過場氧化之後,該氮化層1 6和氧化矽層1 5最好 以濕化學方式去除,然後在其上熱成長—其厚度約 為30nm的氧化矽層15,稱為κ〇〇Ι·氧化層。此層只 有可在場氧化物1 8之間的區域内看見◊然後將—多 晶矽或是多晶矽型材料的薄屏障層19(其厚度約為 t〇nm)沉積於平面之表面上,如圖丨6所示。在較佳 貝施例中’該聚乙醯胺層是以LPCVD沉積而得。但 ,,該屏障層19最好是由另一種材料製成,本文係 等效於如微晶矽或非結晶矽之矽種類。在此聚乙醯 胺之屏障層19的上部為二氧化矽層2〇,其基本上是 沉積產生,厚度约250nm。在較佳實施例中,該氧 化層 20 是以 TEOS(tetraethyl-orthosilicte)之熱分解 的LPCVD方法沉積而成。經過此沉積之後,此氧化 層20經一退火過程使其密集化’基本上它是在一濕 的大氣下在800C下持續3小時。該氧化層20也可以 由俗稱之LT0-氧化物(“低溫氧化物”)或是pECVD_ 氧化物(“電漿增進的化學汽相蒸鍍沉積,,)製成,因 為此氧化層的目的只是作為其後之包括蝕刻溝槽之 步驟中的硬罩幕使用,此亦可見於Eklund等人的美 國專利第4,958,213號。但是’此已知製程中是以一 氣化層取代該氧化層20。 接著對該結構做微影姓刻式佈圖,其先施加一光 阻層2 1和在其中做開口,以界定在要製造的個別元 件和/或元件組四週之深的電絕緣溝(稱為溝槽),見 {請先閱讀背面之注意事項再填寫本頁)V. Description of the Invention (IS) Printed by the Consumer Cooperative of the Sample Procurement Bureau in the Ministry of Economic Affairs). In the N-type region (directly above the N + -type bottom diffusion layer 5), a bipolar transistor and a capacitor will be formed. On the middle portion, 'the p-region' will exist after forming the connection path or region as a connection between the surface or the circuit or element formed on the substrate 1. After the growth of the epitaxial layer 9 ′, a thin barrier oxide layer 10 (preferably formed by thermal oxidation) is produced on the wafer surface, as shown in FIG. The thickness of this oxide layer is approximately 40 nm. Above the barrier oxide layer 10 is a thin silicon nitride layer, which has a thickness of approximately 130 nm, and is deposited by the LPC VD method ("low-wish chemical vapor deposition method"). This nitride layer 丨 丨 is formed by applying a photoresist layer Π 'and patterning it by lithography. The nitride layer on the portion not protected by the photoresist layer 1 1' is engraved with a surname. As shown in FIG. I 2, and these parts include the area of the device, only the silicon dioxide layer 10 is left. The nitrogen-silicon layer 1 is preferably etched away by a suitable dry etching process. The etching process can selectively remove only the nitride layer but leave the oxide layer 10 under the nitride layer. Next, the arrow shown in FIG. 12 is shown. As shown, an ion implantation step is performed thereon to generate the above-mentioned N-type or N-region 13 (referred to as "N-hole") in the crystalline silicon layer 9 below the thin oxide. A silicon layer 10 is used as a protective layer on the surface of the epitaxial layer 9. In a preferred embodiment, the ion implantation step (see FIG. 13) used to form the N_region 3 is preferably performed by Approx. 45 OkeV energy and approximately 1.5X1012 ions / cm² of phosphorus implantation. But (Please read the precautions on the back before filling out this page) I l ^ ntn, -Λ. ▲ -21-This paper's dimensions apply China National Standard (CNS) A4 specification (210X: 297 mm) fA23〇 ^ 9 V. Description of the invention (19) A7 B7 n ^ i * 11— mi Printed by the Consumer Cooperatives of the Central Government Bureau of the Ministry of Economic Affairs. There is another form of doping in the N-region 13, and this implantation condition can be modified. After the incorporation, the photoresist layer 1 ′ is used in a known way. After removal, the thin oxide layer 10 in the opening of the nitrided layer 11 will be thickened by the ampoule formation, so that after growth, an approximately 450 nm-thick oxide layer 12 can be obtained, as shown in Fig. 13. The implanted heterozygote in the aforementioned implant is osmium phosphorus in the best example of Ben X. As mentioned earlier, during this oxidation step, L will diffuse into the crystalline layer 9 so that the fabrication of the N region 13 has been completed. Part f. Diffusion of some of these dopants will occur during subsequent steps including high temperatures. The remaining u part of the nitride layer (which is used as a barrier layer during thermal oxidation of the planar surface of silicon) will This causes the oxide to grow only in the areas where the nitride layer 丨 i has not been removed during this thermal growth, that is, in those areas that have been implanted. After oxidation, the nitride layer 丨 丨 is completely from the plane Removal is best by a wet chemical method. A thin oxide layer 10 below it is retained, and a stepped shape is formed on the edge of the thicker silicon oxide region 12, which is then used to generate the aforementioned p _-Type or p_-region (known as "P-hole") as a protective layer 〇 The energy in this ion implantation step can be adjusted so that ions can only pass through the area with the thin oxide layer 10 on its surface, and be blocked by the area with the thick oxide layer 丨 2 on its surface ^ A self-aligned p-region 14 or P-hole can be obtained with the aforementioned implanted N-region 13. In a preferred embodiment, this is used to form a p_region 丨 4 22 This paper size applies Chinese national standards (CNS) A4 specification (ZIOXM7). (Please read the precautions on the back before filling out this page.) 乂. Order lA23〇 " ^ 9 at V. Description of the invention (20) Occupied by the Ministry of Economic Affairs of the Central Bureau of Standards and Communications The cooperative printed ion implantation method implants boron with an energy of about 50 keV and about 2 x 10 " _2 X 1014 ions / cm2. However, this implantation condition can also be changed if another type of dopant is desired in the P-region 14. After the implantation step is completed, annealing is performed at a high temperature (basically at about 1000. (: 4 hours) to drive the implanted dopants to obtain the team and? Regions 13, 14 required Diffusion depth "The structure after annealing is shown in Figure 丨 3. The above-mentioned process for generating N_ and p_ regions has also been described in Havemann's US patent. After driving in, all the oxidized regions are It is best to remove by wet chemical contact, that is, a region with a thin oxide layer 10, a region with a thick oxide layer 12 and possibly additional oxide layers (this oxide layer can be made previously to drive implanted atoms Found on a planar surface during annealing). After the oxide layer is removed, a stepped shape appears on the surface of the silicon wafer. Then, the known loc method is used to define the effective opening for the component to be manufactured, Therefore, it is best to first apply a thin layer of barrier layer 15 'on the entire surface of the plane by thermal oxidation-as shown in Figure 14. Above this oxide layer 15 is deposited-a thicker oxide layer 16, the thickness of which is basically The upper limit is 200 nm, preferably LPCVD The nitride layer 16 is formed by applying a photoresist layer 17 and patterning it to define the element area. The nitride layer 16 is patterned by photolithography. The nitride layer 16 is not a photoresist layer 1 7 The protected part is etched away, as shown in Figure 14. The nitride layer 6 is best removed by an appropriate dry last engraving process. This etching process is long and can selectively remove only nitrogen-23. This paper is universal China National Standard (CNS) A4 Specification (2) 0 > 7297 mm) (Please read the notes on the back before filling out this page] Order * d23〇f A7 B7 V. Description of Invention (21) Central Government Bureau of Economic Affairs, Ministry of Economic Affairs Negative work consumer co-operative layer and thin barrier oxide layer left under it] 5. Figure 14 shows three separate N_regions, where a lateral NPN transistor is formed from left to right, A capacitor and a vertical NPN transistor. Then the silicon nitride layer 16 covers approximately the area of the base connection, collector and emitter of the lateral PNP transistor to be formed, and some of the electrode connections will be formed In one area, an electrode connection and a dielectric are formed in the capacitor in another area. Layer, and an active area for the vertical NPN transistor and an area for collector connection. After the nitride layer 16 in the opening of the photoresist layer is etched off, the photoresist layer will be known as a The method is to remove the silicon oxide layer 18 (the so-called field oxide) having a thickness of about 60 nm in the opening of the nitride layer 16 by thermal growth on it. In a preferred embodiment, 'The field oxide 18 is preferably grown under a substantially humid atmosphere. (: The appearance of the nitride layer 16 (which acts as a barrier layer during the thermal oxidation of the silicon surface) The silicon oxide will only grow in the area where the nitride layer has been removed. Since some silicon in the opening of the nitride layer 6 will be consumed when it is converted into silicon dioxide, the field oxide 1 8 will be partially recessed in the surface of the substrate or on the front surface of the epitaxial layer 9, called "semi-recessed". The results are shown in Figures 15a and 15b, where Figure 15a shows that capacitance and NPN are to be generated therein. The region of the transistor, and Figure 5B shows the region in which the lateral PNP transistor is to be generated. In the following figure, it can also be seen how the area of the field emulsion 18 grows at the edge of the gasification layer 16. -24-The ft dimension of this paper applies the Chinese National Standard (CNS) A4 specification (210X297 mm) (Please read the precautions on the back before filling this page) 5. Description of the invention (22 :) Staff consumption cooperation of the Central Standards Bureau of the Ministry of Economic Affairs 衽After the field oxidation is printed, the nitride layer 16 and the silicon oxide layer 15 are preferably removed by wet chemistry and then thermally grown thereon—a silicon oxide layer 15 having a thickness of about 30 nm is called κ〇〇〇 I. Oxidation layer. This layer can only be seen in the area between the field oxides 18 and then deposit a thin barrier layer 19 of polycrystalline silicon or polycrystalline silicon type material (its thickness is about 0 nm) on a flat surface, as shown in Figure 丨6 shown. In the preferred embodiment, the polyacetamide layer is obtained by LPCVD deposition. However, the barrier layer 19 is preferably made of another material, and this text is equivalent to a silicon type such as microcrystalline silicon or amorphous silicon. The upper part of the barrier layer 19 of polyethylene is a silicon dioxide layer 20, which is basically produced by deposition and has a thickness of about 250 nm. In a preferred embodiment, the oxide layer 20 is deposited by a thermal decomposition LPCVD method of TEOS (tetraethyl-orthosilicte). After this deposition, the oxide layer 20 is densified by an annealing process. Basically, it is in a humid atmosphere at 800C for 3 hours. The oxide layer 20 can also be made of LT0-oxide ("low temperature oxide") or pECVD_ oxide ("plasma-enhanced chemical vapor deposition,") because the purpose of this oxide layer is only Used as a hard mask in subsequent steps including etching the trench, this can also be seen in US Patent No. 4,958,213 to Eklund et al. However, 'the known process uses a gasification layer to replace the oxide layer 20. Next The structure is lithographically engraved. First, a photoresist layer 21 is applied and an opening is made in it to define a deep electrical insulation trench around the individual component and / or component group to be manufactured (referred to as Groove), see {Please read the notes on the back before filling this page)

'1T -25- 尽紙張尺度適用中國國家標準(CNS > M規格(2I0X297公餐 d23〇79 ^ 五、發明説明(23 ) 圖16所示。在本較佳實施例中,設置有用以產生溝 槽的開口,以致於他們被放置於場氧化物層丨8的區 域上方’並且完全地或部份地重疊指示p_區域或p_ 八14以及N-區域或N-穴1 3之間的連接(這些區域主 要5又置於该羞晶層9内)之正碎表面。 經濟部中央標準局員工消費合作社印裝 (請先閲讀背面之注意事項再填寫本頁} 在未受光阻層21保護的位置上,位於上方的氧化 層20、下方的聚乙酿胺屏障層Η以及底部的場氧化 物層1 8會蝕刻掉,向下到達該磊晶矽層9的表面。 ^蝕刻過程(它以數個步驟組成)(用以去除個別材料 )最好是以乾蝕刻法完成。在蝕刻後,該光阻層2上 以某一已知的方法去除,然後在其上根據佈圖,以 向下蝕刻該磊晶層1 9至該矽基質的方式產生深的溝 槽22 ’該佈圖是由開口界定,此開口之前才剛以微 影姓刻方式在最上面的氧化層2 〇上產生。在此溝槽 蝕刻步驟期間,此氧化層2丨作為一罩幕層_硬罩幕 。在本較佳實施例中,該溝槽22約為1 " m寬和約 6 _ 5 v m深。他們的產生方式使得它們的側壁幾乎垂 直該蟲晶層1 9的表面,且其產生方式使得當該溝槽 向下穿透和被其底部的一弱圓形形式終止時可做的 非常地窄,如圖1 6所示。其輪郭的目的有助於接下 來以聚乙醯胺充填或插入溝槽22,和助於減少基質 1内的機械張力,亦即,為降低該基質的張力以防 止在該深溝槽22内破裂,因此該深溝槽22形成破裂 之指標。這類的溝槽亦已揭示於瑞典專利第 -26- 本紙張尺度適用中國國家標準(CNS ) A4規格(210χ29·7公着) 雾423〇79 A7B7 五、發明説明(24 鲤濟部中央標準局員工消費合作社印製 9701934-3號案中,其名稱為,,„積體電路 元作和其製作方法”。 門们 在一較佳實施例中,根據上述討論,開始材料是 由一俗稱epi-晶片或磊晶晶片组成,亦即,基質包 含-高度摻雜的? +型矽平面’基本上其阻抗值為幾 十微歐姆-公分,其上並已成長有—磊晶之弱摻雜 的P-型碎層。此成長的羞晶層基本上之厚度約為6 # m ’阻抗值約為!〇_2〇歐姆-公分。因而,溝槽22 的深度將使它們可以向下到達p +型的高度摻雜I大 量秒材料,請見上述之由V. delta Torre等人所發表 的引證文獻。如此確使沿溝槽22的較低部份之高電 場臨界,並因而可防止洩漏電流通經該溝槽22的封 套表面。既然該P-區域14和N-區域13由溝槽分開, 且元件要被在遠N -區域内產生,尚可在要被製作的 不同元件之間形成良好的電絕緣。在使用p_型之弱 摻雜的基質中(其基本之阻抗值為10-20歐姆-公分) ,在用以產生溝槽22而做的蝕刻後,進行另一個離 子植入步驟,以增加在溝槽22之較低部份以下的電 %臨界值,亦即,做一俗稱的”溝槽通道終止”。然 後最好以約20keV之能量和5X 1013離子/平方公分 之摻雜濃度在0 °c之”傾斜角”下植入硼原子,見上 述之Eldund等人之美國專利。在此植入步驟中之能 量和摻雜濃度可以隨著產生溝槽22的製程步騾中之 條件而改變。本例中用於一弱摻雜基質的製程已簡 -27- 本纸張尺度適用中國國家標準(CNS 規格(210><297公釐) (請先閱讀背面之注意事項再填寫本頁) 訂 i423〇79 A7 i423〇79 A7 經濟部中央標準局員工消費合作社印製 B7 五、發明説明(25 ) 短地描述於P.C, Hunt等人之文獻中,其名稱為”製 程HE :用於類比和數位施加的高度增進式溝槽絕 緣雙極性技術”,Proceeding of IEEE 1988,Custom and Integrated Circuits Conference,New York,五 月16-19日。 在完成溝槽之蝕刻以後,氧化層20的剩下部份(“ 硬罩幕”)被蝕刻掉。然後在其下方的聚乙醯胺層19 作為一蝕刻終止層。接下來該聚乙醯胺層1 9經由選 擇蝕刻劑或是蝕刻條件而被去除,留下直接在此聚 乙醯胺下方的場氧化物部今1 8,且該二氧化矽層 1 5 b也未受到影響。藉此可保留該場氧化物的厚度 之良好均勻性。此蝕刻步驟最好在完成溝槽22的蝕 刻以後直接在一多室系統中或”叢集系統”中做乾蝕 刻的方式進行。 在蝕刻該溝槽22和去除該二氧化矽硬罩幕20和聚 乙醯胺屏障層1 9以及二氧化矽層1 5 b以後,平面的 表面在濕大氣下約900°C下做熱平坦化處理。然後 在溝槽22的壁上亦被氧化,且在該溝槽的壁上之氧 化層23的最終厚度將可達約3 Onm,如圖1 7所示。 因而在晶片的上表面上獲得屏障氧化層24,此層是 以熱氧化方式與溝槽壁上的氧化層23同時成長,其 厚度約為3Onm,如圖1 7所示,並且形成於平面的 表面上之特別要被構成元件之主動區域(稱為ΚΟΟΙ-氧化物)的區域上。然後於此屏障氧化層24的上方 -28 - 本紙張尺歧财關家鮮(CNS ) Α4規格(210X297公疫) ---------.乂------il------旅 (請先閱讀背面之注意事項再填寫本頁} f4 2 3 Ο 7 9 A7 B7 經濟部中央標準局負工消費合作社印製 五、發明说明(26 ) 沉積一薄的氮化矽層25(最好以LPCVD形成),見圖 18所示《其上方亦以最好是TE〇s.分解法(亦使用 LPCVD)沉積另一個氧化層26。因為使用LpcvD可 獲得均勻的沉積之故,對應於氮化層25的表面層以 及經此獲得的氧化層25亦可沿溝槽22的封套表面和 底邵發現。最後在晶片表面上沉積其厚度約}. 5 V扭 、本負上疋未摻雜的微晶矽或是電子半絕緣型的聚 乙醯胺(亦是使用LPCVD)之厚層27,使得所有的溝 槽22可以完全地充填以此矽層。在本較佳實施例中 使用微晶矽,因為它能給予了良好的充填程度。 沉積了孩微晶矽或聚乙醯胺層27以後,此層可以 乾蝕刻法自平面的所有上部或外部表面上去除,見 圖丨7所示,使得來自此矽層的材料只會保留於溝槽 2、2中二在本較佳實施例中,當在此微晶矽層”正 方的氧化層26被曝露於平面的上部或外部表面部丫八 上時’該姓刻步驟即終止。因而可避免過度飯刻^ 溝槽22之充填材料的微晶矽。在此步驟以後, 發現溝槽2 2並未被完全充填。 完成蝕刻之後,矽晶片在一濕氣中約95〇t d熱广-V然後該碎層27的上表面(充埴該溝 槽22者)被乳化,使得約有〇4#m厚度的二氧化 絕緣層28(稱為”氧化帽”)被形成於溝槽的開口 見圖19所示。本發明之氮化層15作為— 可防止平面的其他部份被進_步氧化。並後^ θ -29 本纸張尺纽财關家辟(6νΓΓα4^(21〇Χ29Τ^ (请先Μ讀背面之注意事項再填寫本頁j -*17 丨·^ r4 23〇79 a? £________B7五、發明説明(27 ) 經濟部中央樣準局員工消費合作社印製 之上表面上之上方之薄氧化層26和此氧化層26下方 的氮化層25接著以乾蝕刻法去除。此乾蝕刻步驟到 達該氮化層2 5正下方的層時即停止,亦即,當場氧 化物區18和屏障氧化層24被曝露出來為土。 該溝槽22亦被充填以電絕緣材料,如氧化矽,因 此,若沒有微晶矽層27,取而代之者為較厚的氧化 珍層26時’可使溝槽22被完全地充填,如使用 SACVD法(”次大氣化學蒸鍍沉積法”)。此氧化矽層 由某一適當的乾蝕刻過程蝕刻掉,直到在該溝槽22 的開口上形成一平坦表面為止。與此平坦化步鄉有 關者’在氧化矽層正下方的氮矽層25亦會被去除( 在平坦或是水平的部份),而此蝕刻步驟中該氮矽 層之作用如同一姑刻終止層。本例中不需要以熱氧 化以形成絕緣層28。該氮矽層25保留於溝槽22中, 且在上述之以微晶ί夕充填時作為一防止該充填材料 中可能的雜子之擴散屏障層。 開始時曾提到已知在一 ΝΡΝ電晶體中,Ν +丑的底 部擴散層係作為一低阻抗的集極電極。為使矽表面 和埋入式底擴散5的集極連接終之間有低的阻抗, 形成一所稱之集極插栓。此插栓是利用在晶片的整 個表面上施加一光阻層3 1和對此層做佈圖的方式微 影姓刻式界定而成,使得用於此光阻層3 1的插栓之 開口區域3 0 ’可在所考慮的區域上形成。在本較佳 實施例中,該底部擴散層5亦作為同時產生的平面 -30- 本紙乐尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閱讀背面之注意事項再填寫本頁) ,1Τ ί·#--- • J . I - In |:4 2 3 0 7 9 A7 B7 五、發明説明(28 ) 經濟部中央標準局員工消費合作社印製 電容的電極之一’因此,在此製程步騾中,亦界定 作為光阻層内之電極的區域3 〇 ’’,在此區域内可 製造此電容’且該區域包含兩個與N +型埋入區域5 連接的分開的連接插栓區域’以降低對此埋入式底 部擴散層5(構成與電容電極之一連接的一部份)的 串聯電阻。該光阻層3 1内的開口之製造方式使得它 們可以覆盍場氧化物層區域之間的整個面積,此亦 表光阻層3 1的剩下部份覆蓋場氧化物層丨8區域之 間的其他的整個表面《此狀況也可說是光阻層内的 開口邊緣始終位於場氧化物層丨8的區域上方。開口 亦I疋供於區域3 0,’ ’上方,以利橫向pNp電晶體的基 極連接之製作,見圖20b所示。 佈圖完光阻層3 1以後,即對光阻層3丨的開口(亦 即用於集插栓的區域30’之開口)、在用於電容電極 之區域30”以及用於基極連接的區域之開口 3〇,,,進 行摻雜,以得到被強烈摻雜成的區域,如圖2〇a 和20b之箭頭所示。此摻雜步驟最好是以約5〇keV之 能量和5 X 1015離子/平方公分的摻雜濃度離子植入 例如磷的方式進行。重要地是在此植入步驟中之能 量之選擇可使因為此植入步驟而引入矽内的缺陷位 置不會向下通過到達該場氧化物層丨8之底部或是下 表面所界定的深度,亦即,該層終止的位準,如平 面表面之向下方向所見者。若缺陷更向下延伸時, 會造成射-基極接面和/或基_集極接面之鄰近區域中 一— -31 - ^張尺 中國國家標準(CNS )从規格(2ΐ〇χ295^ I tn ! rn t— _ _ -.---I i _ _ T _ _ _ _ _ _ ,展 * ·-0'" (請先閱讀背面之注意事項再填寫本頁) A7 A23QT9 ^ 五、發明説明(29 ) 的錯置以及使要製作的NPN電晶體中產生极關的洩 露問題《因此,植入能量和摻雜濃度有時需依先前 為產生場氧化物層1 8所做的場氧化之條件而改變, 特別如前述之瑞典專利申請案第9701934-3號中所 述者。 經過植入步驟以後’該薄的、保護氧化層24自該 植入區域上方去除,最好是以乾蝕刻方式為之,如 圖20a和20b所示。但是,可看出此氧化層24會保留 於光阻層3 1所覆蓋的表面部份上,亦即在雙極性 NPN電晶體中將形成一基極區域36’的位置上的其 他區域之間,如圖2 U所示。然後以一已知的方式 去除光阻層3 1,接著為使植入步驟中所引入的摻雜 子驅入,平面在最好非氧化的環境下(如包含氮或 或)以基本上9 0 〇 C的溫度進行退火約半小時β經過 此退火過程後之最後的Ν+型集極插栓3〇,、集極電 極32足一和其連接32,(亦為ν+型)以及用以連接橫 向ΡΝΡ電晶體内的埋入式基極連接層5之插栓32,,如 圖21a和21b所示。 、冗成上述之退火步驟後,以稀釋的氫氟酸蝕刻該 平面—暫時間’將用於NPN電晶體的集極插栓2 1 上方和在用於PNP電晶體的基極連接的區域3〇,,,内 之可能的薄氧化層去除(該層早已形成於用於電容 的區域30”内和區域3〇,内),蝕刻後,即最好以 LPCVD的方式沉積一薄的氮矽層34於平面上,如圖 __ -32- 本紙張尺度(eNS} Α4· (_21()χ297/'1T -25- Applicable to Chinese national standards (CNS > M specifications (2I0X297 public meal d23〇79) as far as possible paper size) 5. Description of the invention (23) Figure 16. In this preferred embodiment, the setting is useful to generate The openings of the trenches, so that they are placed above the area of the field oxide layer 8 and fully or partially overlap the p_area or p_ 8-14 and the N-area or N-hole 1 3 The broken surface of the connection (mainly 5 of these areas are placed inside the shame layer 9). Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs (please read the precautions on the back before filling this page) In the protected position, the upper oxide layer 20, the lower polyethylene barrier layer Η and the bottom field oxide layer 18 will be etched away and reach the surface of the epitaxial silicon layer 9 downward. ^ Etching process ( It consists of several steps) (to remove individual materials) is best done by dry etching. After etching, the photoresist layer 2 is removed by a known method, and then according to the layout, Deep etching is performed by etching the epitaxial layer 19 down to the silicon substrate. Slot 22 'The layout is defined by an opening, which was just created on the top oxide layer 2 by lithography last before the opening. This oxide layer 2 is used as a mask layer during the trench etching step. _Hard cover. In the preferred embodiment, the grooves 22 are about 1 " m wide and about 6_5 vm deep. They are generated in such a way that their sidewalls are almost perpendicular to the surface of the worm crystal layer 19 And the way it is generated is that when the trench penetrates downwards and is terminated by a weak circular form at the bottom, it can be made very narrow, as shown in Figure 16. Its purpose is to help the next Polyethylenimine fills or inserts the trench 22 and helps reduce the mechanical tension in the substrate 1, that is, in order to reduce the tension of the substrate to prevent cracking in the deep trench 22, the deep trench 22 forms an indicator of cracking This type of groove has also been disclosed in Swedish Patent No. -26- This paper size applies to Chinese National Standard (CNS) A4 specification (210x29 · 7) Fog 423〇79 A7B7 V. Description of the invention Standard Bureau employee consumer cooperative printed 9701934-3, its name For, "Integrated circuit element and its manufacturing method". In a preferred embodiment, according to the above discussion, the starting material is composed of a commonly known epi-wafer or epitaxial wafer, that is, the matrix contains- The highly doped? + -Type silicon plane 'basically has an impedance value of tens of microohm-cm, and a weakly doped P-type fragment layer that has been epitaxially grown thereon. The grown shame layer is basically The thickness is about 6 # m 'The impedance value is about! 0_2 ohm-cm. Therefore, the depth of the trenches 22 will allow them to reach down to the p + -type highly doped I-second material, see The above cited references are published by V. delta Torre et al. This makes the high electric field along the lower portion of the trench 22 critical, and thus prevents leakage current from flowing through the surface of the envelope of the trench 22. Since the P-region 14 and the N-region 13 are separated by trenches, and the components are to be produced in the far N-region, good electrical insulation can still be formed between the different components to be fabricated. In a p-type weakly doped matrix (its basic impedance value is 10-20 ohm-cm), after the etching used to generate the trench 22, another ion implantation step is performed to increase The critical% of electricity below the lower portion of the trench 22, that is, what is commonly known as "trench channel termination". It is then preferable to implant boron atoms at an energy of about 20 keV and a doping concentration of 5X 1013 ions / cm 2 at a "tilt angle" of 0 ° C, as described in the aforementioned U.S. patent to Eldund et al. The energy and doping concentration in this implantation step may be changed according to the conditions in the process step of generating the trench 22. The manufacturing process for a weakly doped matrix in this example has been simplified-27- This paper size applies to Chinese national standards (CNS specifications (210 > < 297 mm) (Please read the precautions on the back before filling this page) Order i423〇79 A7 i423〇79 A7 Printed B7 by the Consumers' Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 5. The description of the invention (25) is briefly described in the literature of PC, Hunt et al. Its name is "Process HE: Used for analogy And Digitally Applied Highly Enhanced Trench Insulation Bipolar Technology ", Proceeding of IEEE 1988, Custom and Integrated Circuits Conference, New York, May 16-19. After the trench etch is completed, the oxide layer 20 remains A portion ("hard mask") is etched away. Then, the polyamine layer 19 under it serves as an etch stop layer. Next, the polyamine layer 19 is selected by using an etchant or an etching condition. It is removed, leaving a field oxide portion directly below the polyacetamide, and the silicon dioxide layer 15 b is not affected. As a result, the thickness uniformity of the field oxide can be maintained. This etching step is best done after the trench The etching of 22 is performed directly in a multi-chamber system or a “cluster system” by dry etching. After etching the trench 22 and removing the silicon dioxide hard mask 20 and the polyethylene barrier layer 19 and After the silicon dioxide layer 15 b, the planar surface is thermally planarized at about 900 ° C in a humid atmosphere. Then, the wall of the trench 22 is also oxidized, and the oxide layer on the wall of the trench The final thickness of 23 will be about 3 Onm, as shown in Figure 17. Therefore, a barrier oxide layer 24 is obtained on the upper surface of the wafer. This layer is grown simultaneously with the oxide layer 23 on the trench wall by thermal oxidation. Its thickness is about 3 nm, as shown in FIG. 17, and is formed on the surface of the plane, in particular, the area of the active region (referred to as KO-oxide) of the component to be constituted. Then on this barrier oxide layer 24 -28-The paper ruler Qi Cai Guan Jiaxian (CNS) Α4 size (210X297 public epidemic) ---------. 乂 ------ il ------ Travel (Please read first Note on the back, please fill out this page again} f4 2 3 Ο 7 9 A7 B7 Printed by the Consumers ’Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs ) A thin silicon nitride layer 25 (preferably formed by LPCVD) is deposited, as shown in FIG. 18 "on top of it, another oxide layer 26 is also deposited, preferably using a TEOS decomposition method (also using LPCVD). Because Because LpcvD can be used to achieve uniform deposition, the surface layer corresponding to the nitride layer 25 and the oxide layer 25 obtained therefrom can also be found along the envelope surface and bottom of the trench 22. Finally, the thickness of the wafer is about}. 5 V twisted, undoped microcrystalline silicon or a thick layer 27 of electronic semi-insulating polyvinylamine (also using LPCVD), so that all The trench 22 can be completely filled with this silicon layer. Microcrystalline silicon is used in the preferred embodiment because it gives a good degree of filling. After the microcrystalline silicon or polyethylenimine layer 27 is deposited, this layer can be removed from all the upper or outer surfaces of the plane by dry etching, as shown in Figure 丨 7, so that the material from this silicon layer will only remain on Trenches 2, 2 and 2 In the preferred embodiment, when the "square-shaped oxide layer 26" of the microcrystalline silicon layer is exposed on the upper part of the plane or on the outer surface part Yaba, the step of engraving is terminated. Therefore, it is possible to avoid excessively engraving the microcrystalline silicon of the filling material of the trench 22. After this step, it is found that the trench 22 is not completely filled. After the etching is completed, the silicon wafer is heated in a humidity of about 95 ° td. Guang-V Then the upper surface of the crushed layer 27 (filling the groove 22) is emulsified, so that a dioxide insulation layer 28 (called "oxide cap") having a thickness of about 0.4 m is formed in the groove. The opening is shown in Fig. 19. The nitride layer 15 of the present invention serves as-it can prevent the other parts of the plane from being further oxidized. ^ Θ -29 〇Χ29Τ ^ (Please read the notes on the back before filling in this page j-* 17 丨 · ^ r4 23〇79 a? £ ________ B7 (27) The thin oxide layer 26 printed on the upper surface and the nitrided layer 25 below the oxide layer 26 printed on the upper surface by the Consumer Cooperative of the Central Procurement Bureau of the Ministry of Economic Affairs are then removed by dry etching. This dry etching step reaches the nitrogen The layer immediately below the formation layer 25 is stopped, that is, the field oxide region 18 and the barrier oxide layer 24 are exposed as soil. The trench 22 is also filled with an electrically insulating material, such as silicon oxide. Therefore, if Without the microcrystalline silicon layer 27, the thicker oxide layer 26 can be used instead to 'completely fill the trench 22, such as using the SACVD method ("sub-atmospheric chemical vapor deposition method"). This silicon oxide layer consists of A suitable dry etching process is etched away until a flat surface is formed in the opening of the trench 22. Those who are involved in this planarization step will also remove the nitrogen silicon layer 25 directly below the silicon oxide layer ( In the flat or horizontal part), the nitrogen-silicon layer functions as the same etch stop layer in this etching step. In this example, thermal oxidation is not required to form the insulating layer 28. The nitrogen-silicon layer 25 remains in the trench In the groove 22, It is used as a diffusion barrier layer to prevent possible impurities in the filling material. At the beginning, it was mentioned that the bottom diffusion layer of N + is known as a low impedance collector electrode in an NPN transistor. The low impedance between the silicon surface and the collector connection of the buried bottom diffusion 5 forms a so-called collector plug. This plug uses a photoresist layer 3 1 on the entire surface of the wafer. It is engraved with the lithographic surname in the way that this layer is laid out, so that the opening area 3 0 ′ of the plug for this photoresist layer 31 can be formed on the area under consideration. In this preferred implementation In the example, the bottom diffusion layer 5 is also used as a flat surface at the same time. -30- The paper scale is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) (please read the precautions on the back before filling this page), 1Τ ί · # --- • J. I-In |: 4 2 3 0 7 9 A7 B7 V. Description of the invention (28) One of the electrodes printed by capacitors in the Consumer Cooperatives of the Central Bureau of Standards of the Ministry of Economic Affairs. Therefore, in this process step In the figure, the region 3 ″ as the electrode in the photoresist layer is also defined. Here, This capacitor can be manufactured within the domain 'and this region contains two separate connection plug regions which are connected to the N + -type buried region 5' to reduce this buried bottom diffusion layer 5 (constituting a capacitor connected to one of the capacitor electrodes) Section) series resistance. The openings in the photoresist layer 31 are manufactured in such a way that they can cover the entire area between the field oxide layer regions. This also means that the remaining part of the photoresist layer 31 covers the field oxide layer. The entire other surface of this interval can be said to be that the edge of the opening in the photoresist layer is always located above the area of the field oxide layer. The opening is also provided above the area 30, '' to facilitate the fabrication of the base connection of the lateral pNp transistor, as shown in Figure 20b. After the layout of the photoresist layer 31, the opening to the photoresist layer 3 (that is, the opening of the region 30 'for the collecting plug), the region 30 "for the capacitor electrode, and the base connection The opening of the region 30, is doped to obtain a strongly doped region, as shown by the arrows in Figures 20a and 20b. This doping step is preferably performed with an energy of about 50keV and 5 X 1015 ions / cm² doping concentration ion implantation such as phosphorus. It is important that the energy selection in this implantation step prevents the defect sites introduced into the silicon due to this implantation step from By reaching the bottom of the field oxide layer or the depth defined by the lower surface, that is, the level at which the layer terminates, as seen in the downward direction of the planar surface. If the defect extends further downward, it will One of the adjacent areas causing the emitter-base junction and / or the base-collector junction — —31 — ^ Zhang Chi Chinese National Standard (CNS) from the specification (2ΐ〇χ295 ^ I tn! Rn t — _ _- .--- I i _ _ T _ _ _ _ _ _, exhibition * · -0 '" (Please read the notes on the back before filling this page) A7 A23QT9 ^ 2. The misplacement of invention description (29) and the extremely leakage problem in the NPN transistor to be produced. Therefore, the implantation energy and doping concentration sometimes need to be based on what was previously done to generate the field oxide layer 18. The conditions of the field oxidation change, especially as described in the aforementioned Swedish Patent Application No. 9701934-3. After the implantation step, 'the thin, protective oxide layer 24 is removed from above the implanted area, preferably This is done by dry etching, as shown in Figures 20a and 20b. However, it can be seen that the oxide layer 24 will remain on the surface portion covered by the photoresist layer 31, that is, in the bipolar NPN transistor. A base region 36 'is formed between other regions, as shown in FIG. 2 U. Then, the photoresist layer 31 is removed in a known manner, and then the dopants introduced in the implantation step are removed. Drive in, the plane is annealed in the best non-oxidizing environment (such as containing nitrogen or OR) at a temperature of substantially 90 ° C for about half an hour. Β The final N + type collector plug after this annealing process 3 〇, the collector electrode 32 feet and its connection 32, (also ν + type) and Connect the plug 32 of the embedded base connection layer 5 in the lateral PN transistor, as shown in Figures 21a and 21b. After the above annealing step is completed, the plane is etched with diluted hydrofluoric acid—temporarily Time 'will remove the possible thin oxide layer above the collector plug 2 1 for the NPN transistor and within the base connection area 30 for the PNP transistor (this layer has already been formed for Capacitor area 30 "and area 30, within), after etching, it is best to deposit a thin silicon nitride layer 34 on the plane by LPCVD, as shown in Figure __32- This paper size (eNS) Α4 (_21 () χ297 /

---------i------1T------^ f請先閲讀背面之注意事項再填舄本頁J 經濟部中央標率局員工消費合作社印製 r4 2 3 Ο 7 9五、發明説明(30 ) A7 B7 經濟部中央標準局員工消費合作社印製 ==所示。'匕氮硬層34的作用是為在製程中取 仔兩項特殊的目的: 1)孩氮矽層34之直接與矽平面之表面的區域(此區 f涵括於電容區域3〇”中,且其中亦欲形成電容電 極30,,,< 一)接觸的部份將作為要被製作的電容之 1電層^此氮矽層具有比二氧化矽為高的介電常數 ,而使用一氮化物介電的電容與具有二氧化碎介電 層足電容比較起來每表面單元具有高的電容。此氮 化層的厚度使得電容的電容約為2 4fF/ # 此對 應於以LPCVD沉積且厚度約多之八爪的氮化層34。 π)該氮矽層34沉積於主動區36,内之剩下的氧化 層24(此處將形成後來要產生的雙極性NpN電晶體) 上的部份會使其絕緣區域的厚度增加,因此對集_ 基極接面而言會有較小的寄生電容。 如圖21a和21b所示,在施加了該氮矽層34以後, 即以微影蝕刻的方式体圖該平面,其方式是先施加 一光阻層35,然後對其做適當的開口,以界定用於 將被產生的NPN電晶體之基極區域36’和用於一集 極的開口 37”、37’,,以及將被產生的橫向pNp電晶 體的射極,和用以界定用以與P_區域内的基質接點 37’連接的開口。該用以界定將被製作的NpN電晶 體的基極區域3 6 ’的開口之位置使得它被設置在無 %氧化物層1 8的區域上,且使得開口的邊緣會被設 置在與場氧化物層18之區域不太小的距離上。用於 33 本紙乐尺度適用中國國家標準(CNS ) M規格(2ΐ〇χ297公釐) (請先閲讀背面之注意事碩再填寫本頁} •乂 -*17 i ^ 1---------- /423079五、發明説明(31 ) A7 經濟部中央標準局員工消費合作社印製 將被製造的橫向PNP電晶體内的集極和射極的開口 以相同的方式設置在無場氧化物層1 8的區域上方。 但是,此開口的邊緣則與場氧化物層丨8的邊緣相鄰 ’如圖2 1 b所示。這些開口亦放置在N-區域1 3上, 因而在N +型的底部擴散層5的上方。相反地,在光 阻層3 5内之用以基質連接的開口被設置在p_區域j 4 的上方,因而在P-型的底部擴散層8的上方。 接下來對違光阻層3 5的開口做蚀刻,最好以乾名虫 刻為之。此乾蚀刻步驟是順序進行的,先使氮化層 34去除,然後將下方的氧化層24蝕刻掉,而當露出 矽表面時即停止蝕刻。本文所描述的製程中所特定 的佈圖步驟對將製作的NPN電晶體而言可減少基極 區域的面積,此面積已由場氧化物層内的開口決定 ,但在本文中是以光阻層3 5的開口邊緣予以決定。 此外,它可避免將在NPN電晶體中產生的基極區域 很接近場氧化物的邊緣,此處因為由N_區域或N穴 13的摻離子之”堆疊”(pile_up),故會存在增加之摻 雜子濃度,如前所述者。用以產生在場氧化物層1 8 之區域間的氮化層34和氧化層24中的開口之佈圖目 的在於降低在將製造的NPN電晶體之集極和基極之 間的电谷值,其事實基礎是可獲得一良好界定的開 二,而氮化層的剩下部份則用以形成要被製造的電 谷之介電層。此外,此相同的佈圖步驟也可用以界 定欲製作的橫向PNP電晶體的射極和集極之間距。 (請先閱讀背面之注意事項再填寫本頁) Λ 訂 丨成 nil t . ___-34- 本紙張尺度適用中國 A7 B7 ^ d23〇79 五 '發明説明(32 ) {請先閲讀背面之注意事項真填寫本寅) 此製程的優點是其中之射極和集極的間距可被良好 地界定,且同時,該射極和集極開口可做的更小, 如此降低了這些電極之間的電容性耦合。此距離也 可以由%氧化物串1 8界定,如圖2 1 b所示。 如上所述之製紅優點為電容的介電層(它是由氮 化層在構成電容的一電極的插栓3 2上形成的)可以 在界定欲製作的NPN電晶體的射_基極區域36,的層 之同時產生,在此同時,由此NpN電晶體的集極和 基極之間的電容值之寄生分配降低,和因為它是以 微影蝕刻方式界定之故,故严在欲被製作的NpN電 晶體内的射-基極區域36’會被良好界定,且在欲被 製作的PNP電晶體内的射極和集極之間距亦可被良 好界定。 經濟部中央標準局貝工消費合作社印製 在完成氮化層34和氧化層24之蝕刻以界定用於欲 被製造的NPN電晶體的基極區域3 6,、用於要被製 造的柄向PNP電晶體的集極窗3 7 ’’、3 7 ’,,以及基質 連接37’以後,以一已知的方法將光阻層35去除。 接下來’在本較佳實施例中,沉積一厚度約為 200nm的薄非晶矽層38於平面的表面上(最好以 LPCVD方式)’如圖22所示。此矽層38在後來的製 程中會構成用以連接要被製作的NPN電晶體之基極 的導電路徑、要被製造之電容的電極以及用以連接 要被製作的橫向PNP電晶體的射極和集極之導電路 徑以及與基質接點之連接’該矽層3 8亦可以微晶矽 35 本紙ft尺度適用中1國家標準(CNS ) A4規格(210X297公釐) r4 2 3 0 7 9 五、發明説明(33 A7 B7 經濟部中央標準局一貝工消費合作社印製 或聚乙驢胺等材料組成。 在後來的離子植入中(如 :::層38被掺雜以成為強接雜的Ρ-型。在_ 約離子植入步驟最好以約50keV的能量禾 1〜,、子7平方公分的掺雜濃度植入BF2的方4 f: ’琢植入能量亦可調整,以使植入的硼原子不 :向下到達該县晶層9的表面上。該摻雜濃度和能 =亦可根據非晶矽層38和其上自然產生的厚度而改 支。此離子植入亦可使用其他的硼元素和/或原子 硼:入該攻層+,但本例中其能量和接雜子必須被 s周整成適當的值。 該非晶矽層38上方沉積有—氧化矽層39,其厚度 基本上約為15〇nm。在本較佳實施例中,此氧化層 39是以PECVD法沉積而得’但纟他如所稱的低溫氧 化法(以某些適合的CVD法,如LT〇)之型式亦可使 用、。在本較佳實施例中,當沉積該氧化層39時之溫 度被維持很低,使得該非晶矽層3 8不可以再結晶化 。於產生i用以連接NPN電晶體之基質之導體時, 在氧化碎保護層39下方使用植入有bf2的非晶矽之 優點(以P E C V D法沉積而得)已如瑞典專利申請案第 9504 1 50-5號所述。 在沉積了氧化矽層39以後’平面之表面被覆以一 光阻層40,此層並被微影蝕刻方式佈圖,界定出一 如40所示的區域’該區域係涵括於電容區域中 36 本紙伕尺度適用中國國家標準(CNS ) Α4規格(210 X 297公釐) (請先聞讀背面之注意事項再填寫本頁) 乂 訂 # A7 B7 _yA 2^0 Ί Q— 五、發明説明(34 ) (请先閲讀背面之注意事項再填寫本頁〕 ,並且用於屬於欲製作的平面電容的上部電極,且 被放置在整個區域30”存在的氮化介電層μ上方, 因此’區域40’被覆以光阻層40。此外,光阻層區 域40覆蓋圍繞用於欲被製作的開始射極基極區域 36’’且它尚覆蓋用於射極的區域37,,,和用於將被 製作的橫向P N P電晶體的集極之區域3 7 ”上,以利 後續的接點連接’見圖23a和23b所示。使用現在施 加且佈圖的光阻層40作為一罩幕的方式,在光阻層 開口中的該氧化矽層39和下方的非晶矽層38會被蝕 刻掉。此姓刻過程最好以在多室系統或是,,叢集,,系 統中的接續乾蝕刻方式進行,此過程將在光阻層4〇 上存有此氮化層的開口中完全地曝露出該氮矽層34 時停止,如在場氧化物之上部和在其中經由埋入式 擴散區域與要被形成的電容之底部電極完成連接的 ^域40”内。其結果如圖233和23b所示。在本較佳 貫施例中,所採用的蝕刻順序可使用於開始之射_ 基極區域36’的開口内之約2〇_4〇11111厚的矽基質在最 後的蝕刻過程中(即俗稱的過度蝕刻步驟)被消耗掉 〇 經濟部中央標準局員工消費合作社印製 .元成上述蝕刻步驟以後,對將形成NpN電晶體的 集極足區域上進行另一個摻入步驟,以減小所稱的 ,’基極加寬”,並因而改良了電晶體的高頻特性,如 M.C.Wilson 於 199〇 年 9 月在 ESSDERC,90 、 Nottingham之文獻”The applicaU〇n 〇f 3 selective -37- 藿A23〇 7 α Α7Β7 經濟部中央標準局員工消費合作社印裳 五、發明説明(35 implanted c〇Uect〇r to an advanced bip〇lar pr〇ces$„--------- i ------ 1T ------ ^ f Please read the notes on the back before filling in this page. J Printed by the Consumer Cooperative of the Central Standards Bureau, Ministry of Economic Affairs, r4 2 3 Ο 7 9 V. Description of the invention (30) A7 B7 Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs == shown. The role of the nitrogen hard layer 34 is to take two special purposes in the manufacturing process: 1) The area of the silicon nitrogen layer 34 directly on the surface of the silicon plane (this area f is included in the capacitor area 30). And, where the capacitor electrode 30 is also to be formed, < a) The contacting portion will be used as the electric layer of the capacitor to be made. This nitrogen silicon layer has a higher dielectric constant than silicon dioxide, and is used. A nitride dielectric capacitor has a high capacitance per surface unit compared to a foot capacitor with a shattered dielectric layer. The thickness of this nitride layer makes the capacitance of the capacitor approximately 2 4fF / # This corresponds to LPCVD deposition and The thickness of the nitride layer 34 is about eight. The silicon nitride layer 34 is deposited on the active region 36, and the remaining oxide layer 24 (here, a bipolar NpN transistor to be generated later) is formed. It will partially increase the thickness of the insulation area, so there will be less parasitic capacitance for the collector-base junction. As shown in Figure 21a and 21b, after the silicon-nitrogen layer 34 is applied, The method of shadow etching is used to illustrate the plane. The method is to apply a photoresist layer 35 first, and then apply it to the surface. To define the base region 36 'for the NPN transistor to be generated and the openings 37 ", 37' for a collector, and the emitter of the lateral pNp transistor to be generated, and To define an opening to connect with the substrate contact 37 'in the P_ region. The position of the opening to define the base region 3 6 ′ of the NpN transistor to be fabricated is such that it is set on the area without the% oxide layer 18, and the edge of the opening is set on the field oxide. The area of the object layer 18 is not too small. For 33 paper music scales applicable to the Chinese National Standard (CNS) M specification (2ΐ〇χ297 mm) (Please read the cautions on the back before filling out this page} • 乂-* 17 i ^ 1 ------ ---- / 423079 V. Description of the invention (31) A7 Printed by the Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs The openings of the collector and emitter in the lateral PNP transistor to be manufactured are set in the same way in field-free oxidation Above the region of object layer 18. However, the edge of this opening is adjacent to the edge of the field oxide layer 8 as shown in Figure 2 1b. These openings are also placed on the N-region 13 and thus on N Above the + -type bottom diffusion layer 5. On the contrary, the opening for matrix connection in the photoresist layer 35 is provided above the p_ region j 4 and thus above the P-type bottom diffusion layer 8 Next, the opening of the photoresist layer 35 is etched, preferably with a dry name. This dry etching step is performed sequentially, first the nitride layer 34 is removed, and then the lower oxide layer 24 is etched. And stop etching when the silicon surface is exposed. The specific layout steps in the process described in this article The NPN transistor can reduce the area of the base region. This area has been determined by the opening in the field oxide layer, but in this article it is determined by the opening edge of the photoresist layer 35. In addition, it can avoid The base region generated in the NPN transistor is very close to the edge of the field oxide. Here, because of the "pile_up" of the doped ions in the N_ region or N hole 13, there will be an increased dopant concentration. As mentioned above, the layout of the openings in the nitride layer 34 and the oxide layer 24 between the regions of the field oxide layer 18 is to reduce the collector and base of the NPN transistor to be manufactured. The electric valley value between them is based on the fact that a well-defined Kai 2 can be obtained, and the rest of the nitride layer is used to form the dielectric layer of the electric valley to be manufactured. In addition, this same cloth The steps in the figure can also be used to define the distance between the emitter and the collector of the lateral PNP transistor to be made. (Please read the precautions on the back before filling this page) Λ Order nil t. ___- 34- This paper size applies China A7 B7 ^ d23〇79 Five 'invention description (32) {Please read the back (Notes should be filled in this manual) The advantage of this process is that the distance between the emitter and collector can be well defined, and at the same time, the emitter and collector openings can be made smaller, thus reducing the Capacitive coupling. This distance can also be defined by the% oxide string 18, as shown in Figure 2 1b. As described above, the advantage of making red is that the dielectric layer of the capacitor (which is formed by the nitride layer on the plug 32 of one electrode constituting the capacitor) can define the emitter_base region of the NPN transistor to be fabricated. At the same time, the layer of 36 ′ is generated at the same time. At the same time, the parasitic distribution of the capacitance value between the collector and the base of the NpN transistor is reduced, and because it is defined by lithographic etching, it is strictly necessary The emitter-base region 36 'in the fabricated NpN transistor will be well defined, and the distance between the emitter and collector in the PNP transistor to be fabricated can also be well defined. The Central Standards Bureau, Shelley Consumer Cooperative, Ministry of Economic Affairs, printed the finished nitride layer 34 and oxide layer 24 to define the base region 3 for the NPN transistor to be manufactured, and the handle to be manufactured. After the collector windows 3 7 ″, 37 ′ of the PNP transistor, and the substrate connection 37 ′, the photoresist layer 35 is removed by a known method. Next, in the present preferred embodiment, a thin amorphous silicon layer 38 with a thickness of about 200 nm is deposited on a flat surface (preferably by LPCVD) as shown in FIG. 22. This silicon layer 38 will form a conductive path for connecting the base of the NPN transistor to be manufactured, an electrode of the capacitor to be manufactured, and an emitter of the lateral PNP transistor to be manufactured in a later process. The conductive path to the collector and the connection to the substrate contacts. The silicon layer 3 8 can also be microcrystalline silicon 35 This paper ft scale applies to the national standard (CNS) A4 specification (210X297 mm) r4 2 3 0 7 9 5 Description of the invention (33 A7 B7 Printed by Polycarbonate and other materials such as a shellfish consumer cooperative of the Central Standards Bureau of the Ministry of Economic Affairs. In the subsequent ion implantation (such as ::: layer 38 is doped to become a strong hybrid P-type. In the ion implantation step, it is best to implant the square of BF2 with an energy of about 50 keV and a doping concentration of 7 square centimeters. F: 'The implantation energy can also be adjusted to The implanted boron atoms do not reach down to the surface of the crystal layer 9 of the county. The doping concentration and energy = can also be changed according to the amorphous silicon layer 38 and the thickness naturally generated thereon. This ion implantation Other boron elements and / or atomic boron can also be used: into the attack layer +, but in this example its energy and doping The atom must be rounded to an appropriate value. A silicon oxide layer 39 is deposited over the amorphous silicon layer 38 and has a thickness of approximately 150 nm. In the preferred embodiment, the oxide layer 39 is PECVD. It can be obtained by the method of deposition, but the type of low-temperature oxidation (such as some suitable CVD method, such as LT), which can be called, can also be used. In the preferred embodiment, when the oxide layer 39 is deposited, The temperature is kept very low, so that the amorphous silicon layer 38 cannot be recrystallized. When a conductor that is used to connect the substrate of the NPN transistor is produced, an amorphous body implanted with bf2 is used under the oxide crush protection layer 39. The advantages of silicon (deposited by PECVD) have been described in Swedish Patent Application No. 9504 1 50-5. After the silicon oxide layer 39 has been deposited, the surface of the 'plane is covered with a photoresist layer 40, and this layer is The layout is lithographically etched to define an area as shown in 40. This area is included in the capacitor area. 36 paper sizes are applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) (please first Read the notes on the back of the page and fill in this page) 乂 定 # A7 B7 _yA 2 ^ 0 Ί Q— V. Description of the invention (34) (Please read the precautions on the back before filling in this page), and it is used for the upper electrode belonging to the planar capacitor to be made, and it is placed in the entire area 30 "of the nitride dielectric layer μ Above, the 'region 40' is covered with a photoresist layer 40. In addition, the photoresist layer region 40 covers the starting emitter base region 36 '' which is to be made and it also covers the region 37 for the emitter, , And the collector region 3 7 ″ for the lateral PNP transistor to be fabricated to facilitate subsequent contact connection ′, as shown in FIGS. 23 a and 23 b. Using the currently applied and patterned photoresist layer 40 as a mask, the silicon oxide layer 39 and the underlying amorphous silicon layer 38 in the opening of the photoresist layer will be etched away. This process of engraving is best performed by continuous dry etching in a multi-chamber system or, cluster, system. This process will be completely exposed in the opening where the nitride layer is stored on the photoresist layer 40. The silicon-nitrogen layer 34 is stopped at times such as above the field oxide and in the region 40 "in which the bottom electrode of the capacitor to be formed is connected via the buried diffusion region. The results are shown in Figures 233 and 23b. In the preferred embodiment, the etch sequence used allows the silicon substrate with a thickness of about 20-4001111 in the opening of the base region 36 ′ to be used in the final etching process ( It is commonly known as the over-etching step.) It is consumed. It is printed by the Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs. After the above-mentioned etching step, another incorporation step is performed on the area where the NpN transistor will be formed. The so-called "widening of the base", and thus improved the high-frequency characteristics of the transistor, such as MCWilson's literature in ESSDERC, 90, Nottingham, September 1990, "The applicaUon 0f 3 selective- 37- 藿 A23〇7 α Α7Β7 Warp Ministry of Consumer Cooperatives Central Bureau of Standards staff printed skirts V. invention is described in (35 implanted c〇Uect〇r to an advanced bip〇lar pr〇ces $ "

所述。在本較佳實滿你丨中,沙松咕3 P AA , , r权住貫施例中此摻雜是以離子植入磷 的万式進行,如圖23a和23b之箭頭所示,且最好以 兩個步驟進行。在第一個步驟期間’以約2〇处以之 能量=約1 X 10>2離子/平方公分的摻雜濃度植入磷 ,在第二個步驟期間,則以約46〇keV的能量和約 1.8X 10離子/平方公分的摻雜濃度植入鱗。這兩 個植入步驟的相互順序是可以改變的。在正確的製 私中’各個植入捧雜濃度和能量之小調整始終可以 用以補償微小的製程變化,例如該磊晶層9的厚度 變化等。接著可觀察到摻雜子係與開始之射-基極 區域36’之開口一致或對齊,且在植入過程中光 阻層40仍保留於平面上,以防止摻雜子(較佳實施 例中為鱗)進入非延伸地方上的磊晶層9中。因此, 在冗成製程步驟後,所增加的集極摻雜將不會存在 於所稱的本質基質下方’亦即,沿區域36,的邊緣 之區域’此處,P+型的非晶矽層38會與該磊晶層9 的表面接觸。經此方式得保持欲製作的NPN電晶體 内的集極和基極之間的低電容性。 完成植入步驟後,以一已知的方式將該光阻層4〇 去除’然後在平面表面上沉積一厚度約為20nm的 薄二氧化矽層42 ’使得它可以特別地覆蓋在開始的 射-基極區域處的開口 3 6 ’,見圖24所示。在本較佳 實施例中’此氧化層42最好在800°C濕環境下以熱 -38 - ‘紙張尺度適用中國囷家標準(CNS > A4現格(210X297公釐 ------1T------# (請先聞讀背面之注意事項再填寫本頁) 經濟部中央標準局員工消費合作杜印製 fA23〇^ 五、發明説明(36 ) 氧化的方式沉積而成=在此氧化步驟中,前面所施 加的氧化層39(如上所述,此層是以PECVD在低溫 下沉積而成)將在一二氧化矽層41被形成於非晶矽 層38的垂直自由側壁或是邊緣表面上的同時被密集 化。在氧化過程中(其本身包含一退火步驟),該非 晶矽層3 8在植入之硼被再分配時會被轉換多晶矽或 是聚乙醯胺(亦即它部份結晶化)。前面的非晶矽層 3 8將於本文後稱為p +型聚乙醯胺層,其結果如圖24 所示。 如上所述’在產生二氧化矽層42以後,最好將硼 植入平面中,如圖24之箭頭所示者,以形成欲製造 的NPN電晶體的本質之基極區域或是有效的基極區 域。在本較佳實施例中,硼是以約丨〇keVi能量和 約7X 1〇13離子/平方公分的方式離子植入的。此最 上方之施加氧化層42的厚度之些微變化皆將使能量 和/或摻雜濃度做對應的調整。該植入只會穿透只 有氧化矽層42直接設置在磊晶層9之上表面的位置 處(亦即在開始的射-基極區域3 6,内)的不同施加的 氧化層、ί夕和氮化層。 完成如上所述的基極植入以後,平面被熱氧化( 最好在800 °C的濕大氣下持續約2小時),而進一步 地降低硼原子的表面濃度。在本較佳實施例中,此 平面上會由LPCVD方式均勻地覆蓋一層約180nm厚 的氮化層44 ’如圖25a所示。在本較佳實施例中’ ____ -39- 本纸張尺度適用中國國家標準(CNS ) A4規格UlOx^T^i-) ---------衣------ΐτ------^ (請先閎讀背面之注意事項再填寫本頁) A7 B7 經濟部中央標率局員工消費合作社印褽 fA23〇^9 五、發明説明(S7 ) 此氮化層可使用一特別的各向異性乾蝕刻過程蝕刻 ,直到以已知的方法在該氮妙層44内有最大階梯的 位置上(如用於欲被製作的NPN電晶體之開始射-基 極區域3 6之開口處)仍保留有該氮石夕側條或,,間隔 物”45為止。在此蝕刻步驟中,不只氮化層44被蝕 刻掉,同時前面所施加之在最後施加的氮化層44正 下方的區域處之氮化層3 4亦被蝕刻掉。當場氧化物 1 8和氧化沙層4 2的區域表面曝露出來時,即停止此 蚀刻過和。在開始之射-基極區域3 6 ’内之開口(它是 在氮化層4 4的此触刻過程中形成的)將構成所稱的 射極開口 36” ^完成製作以後,在所製造NpN電晶 體中之射極將因為該氮化侧條45和聚乙酿胺層的邊 緣表面上的氧化層41而與該為p+型的聚乙醯胺層38 分開。此蝕刻過程中,該矽表面亦曝露於區域4〇,, 中,此區域40”處可形成與欲製作的電容内之底部 電極之連接砂表面亦露出於區域45,中,其中可 形成與欲製造的橫向PNP電晶體之基極連接。 剩下的只在射極開口中為明顯的氧化矽層42則以 一濕化學方法或是乾蝕刻法去除。在本較佳實施例 中’使用兩步驟的乾餘刻法,首先是以在 Ar/CHF^CF4之電漿中之RIE(“反應性離子蝕刻”)移 除氧化層42,然後在Ar/NF3的大氣中以溫和的同位 ^矽蝕刻法移除前一 RIE步驟產生的雜子和輻射損 壞。在Ar/NF3中之蝕刻步驟會自射極開口 36,,内的 -40- 本紙張尺度適财關家剩t ( CNS ) Α4·(加幻97公爱 (诗先閲讀背面之注意事項再填寫本頁)As described. In this preferred embodiment, the dosing of the sasongo 3 P AA,, r in this embodiment is performed in the form of ion implanted phosphorus, as shown by the arrows in Figures 23a and 23b, and This is best done in two steps. During the first step, the phosphorus was implanted at a doping concentration of about 20 × with an energy of about 1 × 10> 2 ions / cm 2. During the second step, the energy was about 46 × keV and about The scale was implanted at a doping concentration of 1.8X 10 ions / cm 2. The mutual order of these two implantation steps can be changed. In the correct manufacturing process, small adjustments of the implantation impurity concentration and energy can always be used to compensate for small process variations, such as the thickness of the epitaxial layer 9. It can then be observed that the dopant system is consistent or aligned with the opening of the initial emitter-base region 36 ', and the photoresist layer 40 remains on the plane during the implantation process to prevent dopants (a preferred embodiment Middle scale) into the epitaxial layer 9 on the non-extended place. Therefore, after the redundant process steps, the added collector doping will not exist under the so-called essential matrix, that is, the region along the edge of the region 36 ′. Here, the P + type amorphous silicon layer 38 will be in contact with the surface of the epitaxial layer 9. In this way, the low capacitance between the collector and the base in the NPN transistor to be produced must be maintained. After the implantation step is completed, the photoresist layer 40 is removed in a known manner, and then a thin silicon dioxide layer 42 ′ having a thickness of about 20 nm is deposited on the planar surface so that it can particularly cover the initial radiation. -The opening 3 6 ′ at the base region, see FIG. 24. In this preferred embodiment, 'this oxide layer 42 is preferably heat-38-' in a wet environment at 800 ° C, and the paper size is in accordance with Chinese standards (CNS > A4) (210X297 mm ----- -1T ------ # (Please read the notes on the reverse side before filling out this page) Consumption Cooperation by Staff of the Central Bureau of Standards, Ministry of Economic Affairs, printed fA23〇 ^ 5. Description of the invention (36) Deposited by oxidation = In this oxidation step, the previously applied oxide layer 39 (as described above, this layer is deposited at low temperature by PECVD) will form a silicon dioxide layer 41 on the vertical free of the amorphous silicon layer 38 Simultaneously densified on the sidewall or edge surface. During the oxidation process (which itself includes an annealing step), the amorphous silicon layer 38 will be converted to polycrystalline silicon or polyethylenimide when the implanted boron is redistributed. (That is, it is partially crystallized). The previous amorphous silicon layer 38 will be referred to as a p + -type polyethyleneamine layer later in this document, and the results are shown in FIG. 24. As described above, 'silicon dioxide is being generated. After layer 42, it is better to implant boron in the plane, as shown by the arrow in FIG. 24, to form the NPN transistor to be manufactured. The mass base region or the effective base region. In the preferred embodiment, boron is ion-implanted in a manner of about kekeVi energy and about 7X 1013 ions / cm 2. This topmost Applying slight changes in the thickness of the oxide layer 42 will make corresponding adjustments in energy and / or doping concentration. The implantation will only penetrate through the silicon oxide layer 42 directly at the position above the epitaxial layer 9 ( That is, the oxide, base, and nitride layers applied differently in the initial emitter-base region 3,6. After completing the base implantation as described above, the plane is thermally oxidized (preferably at 800 °). C in a humid atmosphere for about 2 hours), and further reduce the surface concentration of the boron atom. In the preferred embodiment, a layer of approximately 180 nm thick nitride layer 44 ′ is uniformly covered on this plane by LPCVD. As shown in Figure 25a. In the preferred embodiment, '____ -39- This paper size is applicable to the Chinese National Standard (CNS) A4 specification UlOx ^ T ^ i-) --------- clothing --- --- ΐτ ------ ^ (Please read the notes on the back before filling out this page) A7 B7 Employee Co-operation of Central Standards Bureau, Ministry of Economic Affairs社 印 褽 fA23〇 ^ 9 V. Description of the invention (S7) The nitride layer can be etched using a special anisotropic dry etching process until the position where the largest step is formed in the nitrogen layer 44 by a known method. (Such as the opening of the NPN transistor to be made at the opening of the base region 36) still retains the nitrogen stone stripe, or, spacers "up to 45. In this etching step, not only nitrogen The nitrided layer 44 is etched away, and at the same time, the previously applied nitrided layer 34 at the area immediately below the lastly applied nitrided layer 44 is also etched away. When the area surfaces of the field oxide 18 and the oxide sand layer 4 2 are exposed, the etching is stopped. The opening in the beginning of the emitter-base region 3 6 ′ (which is formed during this etching of the nitride layer 4 4) will constitute the so-called emitter opening 36 ″ The emitter in the NpN transistor will be separated from the p + type polyethyleneamine layer 38 due to the nitrided side strip 45 and the oxide layer 41 on the edge surface of the polyethylenamine layer. During this etching process, the The surface of the silicon is also exposed in the area 40, where the connection sand surface that can form the bottom electrode in the capacitor to be made at this area 40 "is also exposed in the area 45, where the lateral PNP electricity to be produced can be formed. The base of the crystal is connected. The remaining silicon oxide layer 42 which is only visible in the emitter opening is removed by a wet chemical method or a dry etching method. In the preferred embodiment, 'a two-step dry-etching method is used. First, the oxide layer 42 is removed by RIE ("Reactive Ion Etching") in a plasma of Ar / CHF ^ CF4. In the atmosphere of NF3, the mild in-situ ^ silicon etching method is used to remove heterodyne and radiation damage generated in the previous RIE step. The etching step in Ar / NF3 will self-emitter opening 36, within -40- This paper size is suitable for wealth and family (t) (CNS) Α4 · (plus magic 97 public love (read the notes on the back of the poem first) (Fill in this page)

經濟部中央標準局員工消费合作社印製 ¥42307 9 A7 ____ B7 五、發明説明(38 ) 羞晶層9(本質的基極)的自由表面上移除約15〇_2〇〇 埃的矽厚度。既然此蝕刻步驟會影響該本質基極的 外形’它可以要求蝕刻深度依要製造的電晶體之電 流增益因素(Hfe)而改變。 在圖2 5 b所示的變化實施例中(其中使用所稱之,,可 丟棄的間隔物”),該氮咬層4 4可以由某一較薄的氮 化層144(其厚度約為50nm)予以取代。此氮化層144 亦以LPCVD法被均勻地沉積於平面上,在此氮化層 的上方則沉積一約為150nm厚的氧化矽層148。此 氧化矽層148是由PECVD-TEOS-氧化物或是 SACVD-氧化物構成’它以接近均勻的方式覆蓋於 平面上。此氧化層148的特徵在於以一低溫產生, 例如約400°C,因而它不會由stoichiometrically合成 的一氧化碎组成’而是更接近於,’參透性的 ’’(porous)。最後提到的特性被使用於下列的步驟中 〇 上述儿積步驟以後’以各向異性乾蚀刻的方式將 该乳化層148和下方的氮硬層144去除°如本文所述 之例子’此再蝕刻作用作為一具有三個步驟的rie 製程’其中首先以Ar、CHF3和CF4之混合氣體去除 氧化層148’當該氮化層44已露出在平坦的、平行 的表面上(例如在場氧化物區域1 8的上方的氮化表 面上)時触刻停止’而該漆透.性的氧化層1 4 8之側條 則保留於垂直表面上。接著在步驟2時,使用最後 41 - 本紙依尺度適用中國國家標隼(CNS ) A4規格(210X297公釐) H tnn In· n flufi 1^1. (請先閏讀背面之注意事項再填寫本頁) -^4^3-CLZa. 五、發明説明(39) 經濟部中央標準局員工消費合作社印製 Π匕側條作為罩幕將該氮化層144去除,亦即斥 2所有平行表面上者去除。當露出射 牛二Γ化物區域18之表面和氧化矽層42時,此蝕刻 艾驟即停止。步驟3類似於本實施例’以rie將射極 開口 2所剩下的氧化層42去除,然後進行以Ar/NF3 中之和的矽蝕刻,以去除表面雜質和輕射之損害 α 完成了上述之乾蝕刻以後,射極開口將由一以” 参透性,,氧化物/氮化物/氧化物148、144、39組成的 合成側條或間隔物包圍。將#合成的側條放在HF( 氬氟酸)中一短暫時間,可去除位於外側的,,滲透性 氧化物’而不會顯著地踫撞具有熱氧化的區域。 在例如”FSI/Excalibur’’的設備中之HF-蒸汽中可有 效地去除該”滲透性,,氧化物,但是也可以使用一以 HF為基礎的濕化學浴。此触刻過程後之侧條將不 具有三角形的截面積,但是它們的截面將更相似於 ” L ” ’如圖2 5 b所示。 前述之包含一所謂的”可丟棄的間隔物,,之方法的 優點為P +型的聚乙醯胺層38中之開口將不會被窄化 。因此有助於下面的沉積和使摻雜子由N+型聚乙 酿胺4 6向外擴散,亦即,可塵抑所謂的”多-插栓效應”。 圖25c顯示根據圖25a而描.述的實施例而製造的 NPN電晶體之截面底視圖,亦即在射極開口内留有 -42- 本纸張尺度適用中國國家標準(CMS ) A4規格(210X297公釐) (請先閱讀背面之注意事項再填寫本頁) 訂 瘻Α 2·3 Ο 7 9五、發明説明(40 ) 绍濟部中央標準局負工消费合作社印製 氮化物側條。圖2 5 d顯示使用根據上述的’’可丟棄間 隔物”之方法製造之對應結構圖,亦可見於圖25b。 其截面圖像是以傳輸電子顯微鏡XTEM攝得。 在曝露出在射極開口 3 6 ”内的單晶碎表面後,以 LPCVD法在欲被製造的ΝΡΝ電晶體的集極插栓3 1上 和在底邵電容電極32的區域40”内之接點區域上以 及在欲被製造的橫向ΡΝΡ電晶體的基極連接上沉積 一約250nm厚的聚乙醯胺層46,見圖2以和26c所示 。然後以最好是離子植入坤和/或磷的方式摻雜該 聚乙醯胺層46,如圖26a和26d之虛線箭頭所示。可 看出該圖顯示該聚乙醯胺層46是在最後佈圖前(虛 線)和之後(實線)之圖形,其内容將於下文敘述之。 在本較佳實施例中’此離子植入是在下文將描述 之對之前所施加的聚乙醯胺層46做佈圖之步驟前, 以二個挺作步驟進行的。其第一個步騾是以約 5 0keV之能量和約3 X ίο!5離子/平方公分的摻離濃 度全面性地在平面上植入砷,然後以一光阻層48和 對Θ光阻層之開口的方式微影蚀刻式地對該平面佈 圖,留下該平面上有高阻抗電阻(稱為Rhi,下文將 描述)之部份上的光阻層48。利用此光阻層48作為 一罩幕,進行另—次砷離子植入,但此次之離子摻 離濃度約為1·2 X 1〇16離子/平方公分,能量約為 1 50keV,其情況如圖26b所示。應了解在此平面上 的所有郅份,除了高阻抗電阻Rhi的區域以外,皆 (諳先聞讀背面之注意事項再填寫本頁) 訂 .丨咸 ____-43· 本紙張尺度適用中niT冢標準(CNS) M規格(ϋ97 瘳A 2 3 ◦ 7 9 a? _____B7 五、發明説明(W ) 進行了上述之兩次植入。 接下來對平面再做一次微影蝕刻術佈圖,然後界 定出低阻抗電阻RL〇的區域’在最後提到的情況中 ’在此佈圖步驟所施加且未顯示的光阻層將會留在 該平面表面的所有部份上,除了低阻抗電阻Rl〇的 開口以外》使用此光阻層作為一罩幕,以約25keV 之能量和約4 X 1 015離子/平方公分的摻雜濃度植入 場’這些操作方式不顯示於圖中,但是與圖26b比 較。在完成了本文所述的製作步驟後,如前所述的 植入步驟可獲得具有約500气姆/平方的表面阻抗之 高阻抗電阻Rhi以及約為100歐姆/平方的表面阻抗 之低阻抗電阻RL0。為補償其他的製程變化可以對 該植入摻雜濃度和/或能量做小小的調整。 經濟部中央標準局貝工消費合作杜印製Printed by the Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs ¥ 42307 9 A7 ____ B7 V. Description of the invention (38) The thickness of the silicon layer is removed from the free surface of the crystal layer 9 (essential base) by about 15-20 angstroms. . Since this etching step will affect the shape of the essential base ', it may require that the etching depth be changed according to the current gain factor (Hfe) of the transistor to be manufactured. In the modified embodiment shown in FIG. 2b (where the so-called, disposable spacer is used), the nitrogen biting layer 44 may be composed of a thinner nitride layer 144 (its thickness is about 50nm) was replaced. The nitride layer 144 was also uniformly deposited on the plane by the LPCVD method, and a silicon oxide layer 148 with a thickness of about 150 nm was deposited on the nitride layer. The silicon oxide layer 148 was made by PECVD. -TEOS-oxide or SACVD-oxide composition 'It covers the surface in a nearly uniform manner. This oxide layer 148 is characterized by being produced at a low temperature, such as about 400 ° C, so it will not be synthesized by stoichiometrically The oxidative breakdown composition is closer to that of 'porous'. The last-mentioned characteristics are used in the following steps. After the above-mentioned product step, the anisotropic dry etching is used to apply the composition. The emulsified layer 148 and the lower nitrogen hard layer 144 are removed. As the example described herein, "This re-etching is a three-step rie process." Among them, the oxide layer 148 is first removed with a mixed gas of Ar, CHF3 and CF4. The nitride layer 44 has been exposed on a flat surface On a parallel surface (for example, on a nitrided surface above the field oxide region 18), the etch stops, and the side of the paint-permeable oxide layer 1 4 8 remains on the vertical surface. Then In step 2, use the last 41-this paper applies the Chinese National Standard (CNS) A4 size (210X297 mm) according to the size H tnn In · n flufi 1 ^ 1. (Please read the precautions on the back before filling this page )-^ 4 ^ 3-CLZa. V. Description of the Invention (39) The Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs printed the dagger side bar as a cover to remove the nitrided layer 144, that is, to repel 2 on all parallel surfaces. Removal. When the surface of the emitter region 18 and the silicon oxide layer 42 are exposed, this etching step is stopped. Step 3 is similar to this embodiment 'removing the oxide layer 42 remaining in the emitter opening 2 with rie. Then, the silicon etching with the sum of Ar / NF3 is performed to remove the surface impurities and light shot damage. After the dry etching is completed, the emitter opening will be changed by a “permeability”, oxide / nitride / oxidation. Synthetic side bars or spacers composed of objects 148, 144, 39. Placing the #synthesized side strip in HF (argon fluoride acid) for a short time can remove the outer, permeable oxide 'without significantly striking the area with thermal oxidation. The "permeability," oxides are effectively removed from HF-steam in equipment such as "FSI / Excalibur '", but a HF-based wet chemical bath can also be used. The side bars after this engraving process will not have a triangular cross-sectional area, but their cross sections will be more similar to "L" as shown in Figure 2 5b. The aforementioned method, which includes a so-called "disposable spacer", has the advantage that the openings in the P + -type polyethyleneamine layer 38 will not be narrowed. Therefore, it facilitates the following deposition and doping The electrons diffuse outward from the N + -type polyethylenamine 4 6, that is, the so-called "multi-plug effect" can be suppressed. Fig. 25c shows the NPN transistor manufactured according to the embodiment described in Fig. 25a. Bottom view of the cross section, that is, -42 is left in the emitter opening. This paper size is applicable to the Chinese National Standard (CMS) A4 size (210X297 mm) (Please read the precautions on the back before filling this page). 2. · 3 Ο 7 9 V. Description of the invention (40) The side strips of nitride printed by the Consumers' Cooperative of the Central Standards Bureau of Shaoxing Ministry. Figure 2 5d shows the use of the method according to the above-mentioned "disposable spacer" method. The corresponding structure diagram can also be seen in Figure 25b. The cross-sectional image was taken with a transmission electron microscope XTEM. After being exposed on the broken surface of the single crystal in the emitter opening 36 ", the LPCVD method is used on the collector plug 31 of the NPN transistor to be manufactured and in the region 40" of the bottom capacitor electrode 32. A 250 nm thick polyethylene layer 46 is deposited on the contact area and on the base connection of the lateral PN transistor to be manufactured, as shown in FIGS. 2 and 26c. The polyvinylamine layer 46 is then doped in a manner that is preferably ion-implanted and / or phosphorous, as shown by the dashed arrows in Figs. 26a and 26d. It can be seen that the figure shows that the polyamide layer 46 is a figure before (dotted line) and after (solid line) the final layout, and its content will be described later. In this preferred embodiment, the ion implantation is performed in two steps before the step of patterning the previously applied polyvinylamine layer 46 as described below. The first step is to comprehensively implant arsenic on a plane with an energy of about 50 KeV and a doping concentration of about 3 X ί! 5 ions / cm 2, and then use a photoresist layer 48 and a photoresist to Θ The plane of the layer is lithographically patterned on the plane, leaving a photoresist layer 48 on a portion of the plane having a high resistance (referred to as Rhi, described later). Using this photoresist layer 48 as a mask, another arsenic ion implantation was performed, but the ion doping concentration was about 1.2 × 1016 ions / cm 2 this time, and the energy was about 150keV. This is shown in Figure 26b. It should be understood that all components on this plane, except for the area of high impedance resistance Rhi, are all (谙 read the precautions on the back before filling this page). 丨 salt ____- 43 · This paper is applicable to niT Mound standard (CNS) M specification (ϋ97 瘳 A 2 3 ◦ 7 9 a? _____B7 V. Description of the invention (W) The above two implantations were performed. Next, the lithography etching layout of the plane was performed, and then The area where the low-resistance resistance RL0 is defined 'in the last-mentioned case' will be left on all parts of the planar surface except for the low-resistance resistance R1, which photoresist layer is applied during this layout step and is not shown. Beyond the opening of 〇 "using this photoresist layer as a mask, implanting the field with an energy of about 25keV and a doping concentration of about 4 X 1 015 ions / cm 2 'These operating methods are not shown in the figure, but Comparison 26b. After completing the fabrication steps described in this article, the implantation step as described above can achieve a high impedance resistance Rhi with a surface impedance of about 500 ohms / square and a surface impedance of about 100 ohms / square. Low impedance resistance RL0. To compensate other Cheng can change the doping concentration and / or energy to the implant do a little adjustment Central Bureau of Standards, consumer cooperatives HIGHLAND Ministry of Economic Affairs print

(請先閱讀背面之注意事項再填寫本頁J —旅 冗成聚乙醯胺層46的不同摻雜步驟之後,以一已 知的微影蝕刻術法對此層進行佈圖,如先前已說明 者。接著界定用於欲被製作的ΝΡΝ電晶體的射極 49’和集極50’的接觸面積用於欲被製作的平面電容 之底部電極的接觸面積51,(見圖26a所示)、與欲被 製作的橫向PNP電晶體之基極連接5丨,,(見圖26b所示 )以及低阻抗5 2 ’和高阻抗5 3 ’電阻(見圖2 6 b所示)。 在聚乙醯胺層46與射極開口 49”内的單晶带表面直 接接觸的位置上’該高度摻雜的聚乙醯層將在後來 的製程級中作為驅入本質之基極區域中之射極時的 摻雜源。使用其光阻層佈圖作為一罩幕的方式(圖 -44- 本纸張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 經濟部中央標準局貝工消f合作.杜印製 |ΤΛ 9 3 Ο 7 9__β7 五、發明説明(42 ) 式中未顯示’但其結果可部分見於圖26a和26c中) ’該被摻雜成N +的聚乙臨胺層4 6被姓刻掉,直到 其正下方的場氧化物1 8之表面露出為止。此蚀刻步 驟最好由在Ch、HBr和02组成的電漿中以RIE進行 之。蝕刻完此聚乙醯胺層4 6之後,即以一已知的方 式將光阻層去除。 接著將在先前產生之P +型聚乙醯胺層3 8上方的氧 化層39蝕刻掉’此蝕刻步騾最好以乾蝕刻進行,它 可以全面性地對整個表面為之,或是較好的方式在 以微影蝕刻方式界定相關部份後為之,此將於下文 說明。 、 因此在較佳實施例中,平面首先以微影蝕刻術佈 圖,因而在聚乙醯胺層38上以及後來施加的聚乙驢 胺層46之區域上的其他位置之間的其他區域上產生 開口 ,見圖27a和27b所示。接下來以例如Αι_、 CHF3和CF4組成的電漿中以rie之乾蝕刻方式將光 阻層52内的開口中之氧化層39去除掉。當聚乙酸胺 38由該開口露出時即停止此蝕刻過程,本不包含全 面性飯刻而是在蝕刻之前做微影蝕刻式佈圖的步驟 之優點在於場氧化物區1 8會受到光阻層52的保護( 否則在乾蝕刻步驟中它已經被腐蝕了),因而可維 持其完整度。另一項優點在於氧化層39會被留在不 希望有碎化物(下文將說明)(例如用於欲被製作的横 向PNP電晶體的射極之區域52,上)的部份上,見^ ---------<------II------^ (請先聞讀背面之注意事項再填寫本頁) -45-(Please read the notes on the back before filling in this page J-After different doping steps to form the polyacetamide layer 46, pattern this layer using a known lithography method, as previously The explainer then defines the contact area of the emitter 49 'and the collector 50' for the NPN transistor to be fabricated, and the contact area 51 of the bottom electrode of the planar capacitor to be fabricated (see Fig. 26a). Connect to the base of the lateral PNP transistor to be fabricated, as shown in Figure 26b, and low resistance 5 2 'and high resistance 5 3' resistance (see Figure 2 6 b). At the position where the acetamide layer 46 is in direct contact with the surface of the single crystal strip in the emitter opening 49 ", the highly doped polyacetamide layer will be driven into the base region of the essence in a later process level. Extremely doped source. Use its photoresist layer layout as a cover (Figure-44- This paper size applies to China National Standard (CNS) A4 specification (210X297 mm)) Eliminate cooperation. Du printed | ΤΛ 9 3 Ο 7 9__β7 V. Description of the invention (42) Not shown in the formula ' But the results can be seen in part in Figures 26a and 26c) 'The polyethylene doped layer 46, which is doped with N +, is engraved by the last name until the surface of the field oxide 18 directly below it is exposed. This etching The step is preferably performed by RIE in a plasma consisting of Ch, HBr, and O2. After the polyethylenimine layer 46 is etched, the photoresist layer is removed in a known manner. It will then be generated previously The P + -type polyethylenamine layer 38 over the oxide layer 39 is etched away. This etching step is best performed by dry etching. It can be used for the entire surface in a comprehensive manner, or in a better way. After the relevant parts are defined by the shadow etching method, this will be described below. Therefore, in the preferred embodiment, the plane is first laid out by lithography, so the polyimide layer 38 and the polymer layer applied later Openings are created in other areas between other locations on the area of the Ethylamine layer 46, as shown in Figures 27a and 27b. Next, the light is etched in a dry manner in a plasma composed of, for example, Ai_, CHF3 and CF4 The oxide layer 39 in the opening in the resist layer 52 is removed. This etching process is stopped when exposed. The advantage of this step is that it does not include a full-scale engraving but a lithographic etching pattern layout before etching. The advantage is that the field oxide region 18 will be protected by the photoresist layer 52. It has been etched during the etching step), so it can maintain its integrity. Another advantage is that the oxide layer 39 will be left undesired with debris (explained below) (for example for lateral PNP to be fabricated) On the part of the emitter region 52, above) of the crystal, see ^ --------- < -------- II ------ ^ (Please read the note on the back first (Fill in this page again) -45-

A7 B7 Λ23019 五、發明説明(43 27b所示。如此給了製程一較佳的再生性。根據前 述的製程,蝕刻氧化物後的結果如圖27a和27b所示 〇 在完成触刻製程以後,以一已知的方式將光阻 52去除,然後將一約3〇nm厚的薄氧化層%沉積於 平面的表面上,見圖28a和28b所示者。在本較佳 施例中,此氧化物之沉積是以TE数 。但是’該氧化層也可以其他的方法為之 或PECVD。接下來在現在施加的氧化層%之上方以 ,好LPCVD的方法沉積約1〇〇nm厚的氮矽層58 ’此 氮化層5 8係均勻地施於平面的表面上。 在做完該氮化層之沉積以後,平面在高溫下進 退火,以驅入和致能先前植入的離子。.在本較佳實 施例中,此驅入步驟是以兩個步觸進行。首先, =面在85〇°C下在氧氣和氮氧的混合氣體中於烤箱 =火處理约30分鐘,以使挣雜離子可更均勾“ :在所植入的層中。接下來對此平面再做退火 氮氣大氣下,約1 075。(:達約16秒, 經濟部中央標準局員工消費合作社印製 設備”(“快速熱退火”)。在一較佳實施例中 ^設備中使用一所稱的,,熱線器”(h〇t]iner)控制 驅。步驟期間的溫度。在該RTA設備中處理A7 B7 Λ23019 V. Description of the invention (shown in 43 27b. This gives the process a better reproducibility. According to the aforementioned process, the results after etching the oxide are shown in Figures 27a and 27b. After the contact etching process is completed, The photoresist 52 is removed in a known manner, and then a thin oxide layer of about 30 nm thick is deposited on a flat surface, as shown in Figures 28a and 28b. In the preferred embodiment, this The oxide is deposited with TE number. But 'the oxide layer can also be other methods or PECVD. Next, over 100% of the currently applied oxide layer, a good LPCVD method is used to deposit about 100 nm of nitrogen. The silicon layer 58 is uniformly applied to the surface of the plane. After the deposition of the nitride layer is completed, the plane is annealed at a high temperature to drive and enable the previously implanted ions. In the preferred embodiment, this driving step is performed in two steps. First, the surface is treated in a mixed gas of oxygen and nitrogen and oxygen at 85 ° C in an oven for about 30 minutes in order to Make the impurity ions more uniform ": in the implanted layer. Next on this plane Annealing in nitrogen atmosphere, about 1 075. (: up to about 16 seconds, employees of the Central Standards Bureau of the Ministry of Economic Affairs consumer equipment printing equipment "(" rapid thermal annealing "). In a preferred embodiment, a so-called ", Hotline" (Hot) iner control drive. Temperature during the step. Processed in the RTA equipment

St:間的聯合有時可以依欲被製作的電晶體 料而改變。應了解在此退火操作期間,氮 母〇氧矽層會維持為該平面上的保護層,以防止 本紙狀CNS ) -46- 經濟部中央橾準局員工消費合作社印製 4 2 3 0 7 9 ΑΊ _____Β7 五、發明説明(44 ) 所植入的離子被向外擴散到其周圍中。 在此退火操作期間’被植入上方的Ν+型聚乙醯胺 層46中之坤將經由擴散而穿透本質的基極,然後形 成射-基接面6 Γ。在本文所描述的整個製程中,射 極深度約為60nm,且在此射極下方之本質基極的 剩下厚度約為1 OOnm。在該磊晶的單晶矽層9之表 面和N+型聚乙醯胺49,之間的接觸區域内的射極開 口中之砷濃度基本上約為4X 1〇20原子/cm3。在本質 基極中之射基極接面内的硼之對應濃度基本上約為 8X1017 原子/ cm3。 同樣地’因擴散而已被植入p +型的聚乙醢胺層38 中之硼將會穿透和連接至該本質的基極上。對本文 所述的整個製程而言,該本質的基極深度約為 200nm’在P+型的聚乙醯胺層38和磊晶的單晶碎層 9之間的邊界表面内之對應的硼濃度基本上約為2 X 1〇19原子/cm3。因此產生的P+型之高度摻雜區即稱 為本質的基極。基質接點6 0 ’係類似於以观|自該p + 型的聚乙醯胺層3 8向外擴散的方式形成。同樣地可 形成用於欲製作的橫向PNP電晶體的射極電極62,,, 和集極電極62”,如圖28b所示。 圖2 9顯示以SIM S測得之在N +型聚乙酿胺射極下 方的摻雜子外形。聚乙醯胺的厚度是由圖29之左邊 的陰影部份指示。為分析起見,某些自射極導出的 绅信號被加大,因此,所看到的神邊緣將比實際上 -47- 本紙張尺度適用中國國家標準{ CNS ) A4規格(2S0X297公釐) -----I - I — ϋ - ----丁 _ i n _ ] 1 泉 -1二5 务 {請先閲讀背面之注意事項再填寫本頁} 經濟部中央橾準局員工消費合作社印裝 f 4 2 3 0 ^9^ A7 _ 87 五、發明説明(Μ ) 更深入所攝取的硼信號中(指示基極的延伸)。 驅入後,平面被微影蝕刻式佈圖,使得經過佈圖 後將有一光阻保護層60保留在電阻Rm和RL0的電阻 本體上,如圖30所示。該電阻的終端部份將曝露出 來。經過佈圖該光阻層60後,以各向異性的乾蝕刻 步驟將未為光阻層60所覆蓋的表面部份内之氮矽層 58和氧化矽層56蝕刻掉,使得所稱之間隔物或侧條 54可沿著N+型的聚乙醯胺層46之邊緣形成,見圖 28 a所示。若為避免矽形成,使氧化層39保留在(例 如)欲製作的橫向PNP電晶體p射極上,此氮化層之 蝕刻步驟會在去除了層56和這些部份曝露之前停止 ’如圖28b所示。本文所描述之用以在一薄氧化矽 層上方製作所稱的氮矽間隔物之流程的一本質部份 類似於H.Norstrom等人的美國專利第4,470,484號。 在本較佳實施例中’使用一各向異性的(亦即方向 相依的)的電漿蝕刻過程去除該氮矽層。當在平面 的水平場氧化物層1 8上的所有氮矽物已被去除以後 ’邊蚀刻步驟(使用氣體SF6、HBr和〇2)即停止。既 然氮矽層5 8已以一良好均勻的方式沉積,亦即同質 性的覆蓋在所有表面,在此蝕刻步驟完成後,仍有 氮矽串、間隔物保留在沿著由已佈圖之N+型聚乙 酿胺層46產生的平面表面上的尖銳階級或遮蔽的位 置上。接下來以RIE银刻該薄氧化;?夕層5 6,然後該 侧條或間隔物5 4得到其最後的形狀。當該n +型的 -48- 本紙張尺度適用中國圉家標^準(CNS ) A4規格(210X297公釐) ------ ί i .^1 n I— T ------象 (請先聞讀背面之注意事項再填寫本頁:> A7 423079 B7 五、發明説明(46 ) 聚乙驢胺層46和P +型的聚乙醯胺層38的表面皆曝露 出來時,該最好是使用氣體Ar' CHF3和的蝕刻 過程即停止。 接下來以已知的方法去除該光阻層6〇。其結果如 圖28a、28b和31所示。圖31是顯示在完成驅又射極 和基極的掺雜子和蝕刻以產生側條以後,要製作 NPN電晶體的區域之放大圖像。由此圖中可看出該 由氮碎層34和氧矽層24中之開口以微影蝕刻術界定 的本質之基極區域由場氧化物區丨8的最相鄰邊緣予 以分開。如稍早在圖丨6中所描述者,在要製作的 NPN電晶體中的集極和基極之間的電容值將會降低 〇 若需要的話,在去除了光阻層6〇以後,該N+型的 聚乙臨胺層46和P +型的聚乙醯胺層38可具有一薄的 硬化層’以降低與欲製作的元件之不同電極區之導 體的阻抗’這些導體將由此矽化層予以分接。此矽 化層可以由例如PtSi、CoSi2或TiSi2構成。在較佳 實施例中,使用二矽化鈦TiSi2,它是以所稱的”自 行對準方法”在曝露的矽表面上形成的。既然電阻 主體未露出’而是受到氮矽層5 8的剩下部份之保護 ,因而其上不會有石夕。 在這類的自行對準矽化過程(“SACICIDE”)中(見 Brington等人的美國專利第4,789,995號和311比&13的 美國專利第4,622,735號),最好以濺鍍法在平面的 土衣 訂 ·線--ί .— (請先Μ讀背面之注意事項再填寫本頁) 經濟部中央樣率局員Η消費合作社印製The combination of St: can sometimes be changed depending on the transistor material to be made. It should be understood that during this annealing operation, the oxynitride layer will be maintained as a protective layer on this plane to prevent paper-like CNS.) ΑΊ _____ Β7 5. Description of the Invention (44) The implanted ions are diffused out into the surroundings. During this annealing operation, the substrate implanted in the N + -type polyethyleneamine layer 46 above will penetrate the essential base via diffusion, and then form an emitter-base junction 6 Γ. In the entire process described in this article, the emitter depth is about 60nm, and the remaining thickness of the essential base below this emitter is about 100nm. The arsenic concentration in the emitter opening in the contact area between the surface of the epitaxial single-crystal silicon layer 9 and the N + -type polyacetamidine 49, is substantially about 4 × 1020 atoms / cm3. The corresponding concentration of boron in the emitter base junction in the essential base is approximately 8 × 1017 atoms / cm3. Similarly, boron, which has been implanted into the p + -type polyamidamine layer 38 due to diffusion, will penetrate and connect to the base of this essence. For the entire process described herein, the base depth of this essence is about 200 nm. The corresponding boron concentration in the boundary surface between the P + polyacetamide layer 38 and the epitaxial single crystal chip layer 9 It is basically about 2 × 1019 atoms / cm3. The resulting highly doped P + -type region is called the essential base. The matrix contact 6 0 ′ is formed in a manner similar to the manner in which outward diffusion from the p + -type polyethyleneamine layer 38 is performed. Similarly, emitter electrodes 62 ,, and collector electrodes 62 "for a lateral PNP transistor to be fabricated can be formed, as shown in Fig. 28b. Fig. 29 shows the N + type polyethylene measured by SIM S. The shape of the dopants below the amine emitter. The thickness of polyethylene is indicated by the shaded part on the left side of Figure 29. For analysis purposes, some of the gentle signals derived from the emitter are amplified, so Seen at the edge of the god than actually -47- This paper size applies the Chinese National Standard {CNS) A4 specification (2S0X297 mm) ----- I-I — ϋ----- 丁 _ in _] 1 Izumi-12.5 Service {Please read the notes on the back before filling out this page} Printed by the Consumer Cooperatives of the Central Procurement Bureau of the Ministry of Economic Affairs f 4 2 3 0 ^ 9 ^ A7 _ 87 V. Description of Invention (Μ) Further In the captured boron signal (indicating the extension of the base). After driving in, the plane is patterned by lithographic etching, so that after the layout, a photoresist protective layer 60 will remain on the resistor body of the resistors Rm and RL0, As shown in Figure 30. The terminal portion of the resistor will be exposed. After the photoresist layer 60 is patterned, an anisotropic dry etching step is performed. The nitrogen silicon layer 58 and the silicon oxide layer 56 in the surface portion not covered by the photoresist layer 60 are etched away, so that the so-called spacers or side strips 54 can follow the edges of the N + type polyethyleneamine layer 46 The formation is shown in Figure 28a. If the oxide layer 39 is to be left on, for example, the lateral PNP transistor p-emitter to be formed in order to avoid silicon formation, the etching step of this nitride layer will remove the layer 56 and these Partially stop before exposure 'as shown in Figure 28b. An essential part of the process described herein for making a so-called nitrogen-silicon spacer over a thin silicon oxide layer is similar to H. Norstrom et al. No. 4,470,484. In the preferred embodiment, 'an anisotropic (ie, direction-dependent) plasma etching process is used to remove the nitrogen silicon layer. When all the nitrogen on the planar horizontal field oxide layer 18 After the silicon has been removed, the 'edge-etching' step (using the gases SF6, HBr, and 〇2) is stopped. Now that the silicon-nitrogen layer 58 has been deposited in a good and uniform manner, that is, it covers all surfaces homogeneously, here After the etching step is completed, nitrogen silicon strings and spacers remain. On the sharp class or shaded position along the plane surface produced by the laid-out N + -type polyethylenamine layer 46. Next, the thin oxide is etched with RIE silver; the layer 5 6 and then the side strip Or spacer 5 4 to get its final shape. When the n + -48- this paper size applies the Chinese standard ^ standard (CNS) A4 (210X297 mm) ------ ί i. ^ 1 n I— T ------ Elephant (please read the notes on the back before filling in this page: > A7 423079 B7 V. Description of the invention (46) Polyethyleneamine layer 46 and P + type polymer When the surface of the acetamide layer 38 is exposed, it is preferable to stop the etching process using the gas Ar'CHF3 and. Next, the photoresist layer 60 is removed by a known method. The results are shown in Figures 28a, 28b, and 31. FIG. 31 is an enlarged image showing a region where an NPN transistor is to be produced after the doping and etching of the emitter and base electrodes are completed to generate side stripes. It can be seen from this figure that the essential base region defined by the lithographic etching of the openings in the nitrogen fragmentation layer 34 and the silicon oxide layer 24 is separated by the most adjacent edge of the field oxide region 8. As described earlier in Figure 6-6, the capacitance between the collector and base in the NPN transistor to be fabricated will decrease. If necessary, after removing the photoresist layer 60, the N + -type polyethyleneimine layer 46 and P + -type polyamidamine layer 38 may have a thin hardened layer 'to reduce the impedance of the conductors in different electrode regions from those of the component to be fabricated' These conductors will be silicided from this layer To be tapped. This silicide layer may be composed of, for example, PtSi, CoSi2, or TiSi2. In the preferred embodiment, titanium disilicide TiSi2 is used, which is formed on the exposed silicon surface by a so-called "self-alignment method". Since the main body of the resistor is not exposed ', it is protected by the remaining portion of the silicon-silicon layer 58, so there will be no stone evening on it. In this type of self-aligned silicidation process ("SACICIDE") (see U.S. Pat. Nos. 4,789,995 to Briton et al. And U.S. Pat. No. 4,622,735 to 311 & 13), it is preferred to sputter the Binding · Thread--ί .— (Please read the notes on the back before filling this page) Printed by the Central Sample Rate Bureau of the Ministry of Economic Affairs and printed by the Consumer Cooperative

本紙伕尺度適用中國國家標準(CNS ) Α4規格(210Χ297公着) 經濟部中央橾牟局員工消费合作社印裝 ΓΛ23〇^9 a? ______Β7 五、發明説明(47 ) 表面上沉積一薄的金屬層70(本例中為厚度約為 50nm的鈦層),如圖32a和32b所示此金屬層可在 一 RT Α設備下,在氮氣大氣下約7丨5的預估溫度 下,與曝露的矽反應一短暫時間。在某些情況中, 也可以使用乳乳與尿素之混合物。接著以濕姓刻方 法將尚未與矽反應的鈦蝕刻掉,亦即在金屬尚未曝 露矽以前的部份上的鈥蝕刻掉。此可選擇性地去除 尚未反應的鈥之姓刻步驟對碎化鈇的影響程度相當 地小。在此化學蝕刻過程後,在875°c下對該平面 做退火約3 0秒’以致於形成二矽化鈦的低阻抗形式 。然後如此產生的矽化層(其表面阻抗約為2_5歐姆/ 平方)只會出現在先前露出的平面之矽表面上,亦 即與這些表面對齊。 經過矽化後,沉積一氧化矽的被動層8 〇,如圖3 3 所示。此氧化層80最好是以—TEOS為基礎的氧化 物組成’此氧化物是以熱分解或是以PECVD沉積而 成°此氧化層80將於後來以光阻回蝕刻(reB)予以 平坦化之沉積厚度約為1私m,接著施加亦為1 # m 厚度(在大平坦表面上測量)的光阻層(圖未顯示)於 平面表面上。此光阻層接著在约1 901下退火一段 時間。因為光阻的表面平坦之特性,其上表面將相 當平坦’雖然其下方表面的地形相當不均勻或粗键 。然後此平面被電漿蝕刻,以利用相同的速度去除 此光阻層和該二氧化矽被動層8〇的突出或延伸部份 -50- 本紙張尺度適用中國國家標準(CNS }ϋ見格(21〇ί 297公# ) ---------:衣------iT,----,~ 線 (請先閱讀背面之注意事項再填寫本頁) 經濟部中央標準局員工消費合作社印製 Γ 4 23〇 7 9 五、發明説明(48) 。在完全去除光阻層以後,最後的結果是該被動層 8 0的表面可得到均勻的地形,亦即,其表面變的非 常平坦和水平。此平坦化方法(REB)如A.C. Adams 、C.D· Capio 在 Journal of the Electrochem,Soc., Vol. 128 , 1981 年第 423 頁之”PUnarization phosphorous doped silicon-dioxide,,中所描述者。 上述已平坦化的氧化層80再覆以一厚約4〇〇nm的 摻雜氧化層82。此由TEOS-基礎的氧化物形成的氧 化層82最好被摻雜約4%的磷,以得到所稱之吸氣 劑(gettering) ’因而可容易地固定擴散的離子。 其他摻離劑之聯合亦是可行的,例如3%的硼和6〇/〇 的鱗。接著在此摻雜的氧化層82上沉積(最好以 PECVD為之)一約250nm厚的未摻雜氧化層82。此 氧化層在後來的作用如同一所稱的硬罩幕。然後將 此氧化層放在700。(:之氮氣下退火約4〇分鐘,使之 密集化。可也改以一 RTA製程在875它下進行約 秒 。此 RTA製 程亦可 取代較 早前為 產生低 阻抗的 二矽化鈦時所進行的退火步驟。 在較佳實施例中,接著對平面做微影蝕刻式佈圖 ,藉此界定出深基質接點。這些接點的製成方式是 先利用各向異性電漿蝕刻法將一施加的光阻層8 1之 佈圖轉換成其下方的氧化層84(82、84),然^以一 ^知=方法將此光阻層81去除,並以乾蝕^的方式 向基質(9、1)下做約7 a m的深孔85。此過程類似於 --—_ -51- 本紙張尺度 ---------A------IT------^ (請先閲讀背面之注意事項再填寫本頁) 罗42307 9 五、發明説明(49 ) 經濟部中央標準局員工消費合作社印製 在蝕刻用以絕緣的溝槽22時所述之過程。當蝕刻孔 85以與基質接觸,最上方的氧化層84(即所稱的硬 罩幕)將會%全地或部份地消耗掉。經過蝕刻基質 連接後的結果如圖3 3所示。 姓刻了該基質連接孔8 5之後,以如圖3 3之箭頭所 示的方向,以約3X1015離予/平方公分之挣雜濃度 和約30keV之能量在將硼植入平面内。該植入能$ 是可調整的,以致於硼原子可受到被動氧化摻雜層 82之阻隔’且它只會藉著首先經過孔8S之開口而穿 透進入矽基質中β在完成此植入步驟以後,在基本 上875 t:的氧氣中對此平面做退火達約3 〇秒。在用 於射頻之1C電路製程中產生和實施如上所逑之深基 负接點的優點已描述於國際專利申請案 PCT/SE97/00487案中。 完成上述之植入和退火後,該平面再度被覆以一 光阻層,並在此時佈圖用於主動和被動元件的接觸 孔,見圖34a和34b所示。然後使用各向異性電漿蝕 刻法在疊置的氧化層82、80中製作接觸孔86、87。 因為接觸孔的深度不同、並因為下方之拓樸圖形, 在做此姓刻時要使用固定的時間。因而某些連接層 依照其拓樸地形的不同而會比其他的層受到更強烈 的過度蝕刻。經過蝕刻此接觸孔以後,即以一已知 的方式將該光阻層去除。因此在此情況下,將可界 定出對基質之接觸孔86以及用以連接被動和主動元 (請先閔讀背面之注意事項再填寫本頁j 丁 、-0 .丨線 -52- 本紙張尺度遥用中國國家標準(CNS ) A4規格(:!!0 X 297公釐} ---- 1423079 A7 B7 五 '發明説明(5〇) 件的接觸孔8 7。其結果如圖3 4 a和3 4 b所示β . 平面上可以濺鍍的方式覆以一雙層結構或是一三 明治的結構,所覆蓋之結構之组成是底部為約 100nm之Τί層’其上部則為約5〇nm厚的TiN層。在 本較佳實施例中,該Ti層是利用在所稱之”離子金 屬電漿設備(IMP-設備)中(例如,’vectra Source” ’ 此為Applied Materials公司之商標名)濺鍍沉積而成 ’以致於它可以覆蓋該深基質接觸孔85的底部。該 TiN層是以反應性濺鍍沉積而成,例如在心和心的 混合氣體中完成者,它最好使用俗稱的瞄準方法( 亦稱為一貫性濺鍍)。此TiN層的沉積方式也可以類 似於Ti層的方式,由使用IMP Vectra Source之反應 性賤鐘法達成。 在沉積完Ti接點金屬層和氮化金屬TiN後(該TiN 層可構成其下方金屬層的屏障層),該平面在一預 估的溫度下進行退火,因而使在其上具有自由矽表 面的位置處之Ti層與下方之矽反應,或是與矽化鈦 反應(亦即在用於元件的接觸孔内之矽化鈦反應)s 在本較佳實施例中,此退火作用是在烤箱中,以約 為600t:的N2和H2混合氣體中進行約半小時。也可 以較高的溫度在一 RTA設備中進行退火,但持續較 短的時間,例如在N2或尿素大氣下為之。它也可以 用增強深基質接點内的屏障_,使得所植入的硼原子 可以擴散至基質材料中。 _— -53- 本紙張尺度朝+~^|家& ( CNS ) Α4· ( 21〖χ297公 ----------------ir------線 ί請先閱讀背面之注意事項再填寫本ί ) 經濟部中央揼準局負工消費合作社印策 rd23〇79 * 五、發明説明(51) M濟部中央標準局員工消費合作社印繁 矣下來以C VD沉積約1以m的鎮層。此沉積步驟可 艮好的均勻度,它是在平面的整個表面上進行的 因此,所有的接觸孔皆可以完全地的均勻地充填 以鎢。跟在此鎢沉積過程後者為—再蝕刻過程, 以去除平面的平坦(亦即水平)部份上的所有鎢。當 該ΤιΝ層被曝露出來時,則此蝕刻過程停止。因而 可使鎢保留在該接觸孔内,並且形成所謂的接觸插栓。 接下來沉積一由約5〇11111的TiN組成的第一導電層 ,其沉積方式和上述之方式,同,並且被覆蓋以約 600nm厚的鋁層。此鋁層是以濺鍍方式形成’最好 包含0.5-2_0°/。的銅,以壓抑電子遷移。這些金屬層 所使用的厚度可以依施加的不同而改變。在此鋁層 上方以反應性賤鍍法沉積基本厚度約為5〇ηιη的TiN 層,以助於後續的佈圖作用和壓抑所謂的,,土丘,,( 向上隆起),接著此以Ti/TiN/ANCu/TiN組成的金屬 層結構被微影蝕刻方式佈圖,然後以乾蝕刻方式界 定元件之連接。 本製程中可加入更多的金屬層,其方式是沉積一 被動層於第一連接層的上方,然後使用微影蝕刻術 和乾钱刻法界定通孔連接。接著根據前述的濺鍍法 沉積一Ti/TiN之雙層結構,然後根據前述之流程, 使用鎢插入該通孔開口中。接著以濺鍍法沉積由 TiN/Al-Cu/TiN疊層組成的第二金屬層結構。然後 -54- 本紙張尺度適用中國园家標準{ CNS ) A4規格(210X297公f ) (請先聞讀背面之注意事項再填寫本頁) 訂 線 「9五、發明説明(52 A7 B7 經濟部中央標準局員工消費合作社印裝 術和乾姓刻方式在其上界定連接層,若 而更,連接層時,此流程重覆進行。所使用的 A1 Cu層之厚度可以由幾百㈣變化到幾個"爪,此 依金屬系統和電路施加的複雜度而定。若希望在電 路中整合-平坦的線圈,則具有相當厚的導電層( 例如Al-Cu)之多層式金屬系統是有利的。在製造用 於RF-IC施加的平坦式線圈時,一使用幾個平行連 接的金屬層且設置在溝槽所分割的基質上方之製程 可見於國際專利申請案第PCT/SE9/0〇954號中。此 已知的製程可以在前述製程宁使用。 本製心中加入幾層金屬層後之結構如圖35a所示 。圖35a之左上角顯示製作的平面電容(稱為 ’’CapDn”)之截面圖,其電極是由其下方之被摻雜成 N+的單晶矽層和位於氮化物介電層上方的p+型聚 乙酿胺層構成,一橫向PNP電晶體直接設在電容的 右邊’其使用P +型的聚乙醯胺形成射極和集極。其 基極是由來自與N+型的底部擴散層串聯的表面之 N+型插栓擴散的方式形成的。其右上方顯示一製 得的NPN電晶體之截面圖,以及由型的聚乙醯胺 產生的電阻"應了解設置在矽基質内的所有元件皆 因深溝槽而彼此絕緣。在孔8 5内之充填有鎢的深基 質接觸孔和/或由P+聚乙醯胺驅入的基質連接被適 當地放置在圍繞每個元件區域的絕緣溝槽22之間, 以得最佳的可能電解耦合。 -55- 本纸張尺度通用中國國家標準(CNS > Α4規格(210Χ297公釐) (請先閲讀背面之注意事項再填寫本頁)The size of this paper is in accordance with the Chinese National Standard (CNS) A4 specification (210 × 297). Printed by the Consumer Cooperatives of the Central Ministry of Economic Affairs of the Ministry of Economic Affairs ΓΛ23〇 ^ 9 a? ______ Β7 5. A thin metal layer is deposited on the surface 70 (in this example, a titanium layer with a thickness of about 50 nm), as shown in Figs. 32a and 32b, this metal layer can be exposed to the exposed temperature at an estimated temperature of about 7 丨 5 under a nitrogen atmosphere under an RT A device. The silicon reacts for a short time. In some cases, a mixture of milk and urea can also be used. Then, the titanium that has not reacted with silicon is etched away by a wet method, that is, the etched away portion of the metal before the silicon has been exposed. This selectively removes the unreacted step, which has a relatively small effect on the fragmentation. After this chemical etching process, the plane is annealed at 875 ° C for about 30 seconds' so as to form a low-resistance form of titanium disilicide. Then the silicidation layer thus produced (its surface impedance is about 2-5 ohms / square) will only appear on the silicon surface of the previously exposed plane, that is, aligned with these surfaces. After silicidation, a passive layer 80 of silicon monoxide is deposited, as shown in FIG. 3 3. The oxide layer 80 is preferably composed of a TEOS-based oxide. The oxide is formed by thermal decomposition or deposited by PECVD. The oxide layer 80 will be planarized by photoresistive etching (reB) later. The deposited thickness is about 1 μm, and then a photoresist layer (not shown) with a thickness of 1 # m (measured on a large flat surface) is applied on a flat surface. This photoresist layer is then annealed at about 1 901 for a period of time. Because of the flat surface of the photoresist, its upper surface will be relatively flat, although the topography of its lower surface is quite uneven or rough. This plane is then etched by plasma to remove the protruding or extended portion of the photoresist layer and the silicon dioxide passive layer 80 at the same speed. -50- This paper is in accordance with the Chinese National Standard (CNS) 21〇ί 297 公 #) ---------: clothing ------ iT, ----, ~ line (please read the precautions on the back before filling this page) Central Standard of the Ministry of Economic Affairs Printed by the Consumer Cooperatives of the Bureau Γ 4 23〇7 9 V. Invention Description (48). After the photoresist layer is completely removed, the final result is that the surface of the passive layer 80 can obtain a uniform topography, that is, its surface It becomes very flat and horizontal. This flattening method (REB) such as AC Adams, CD · Capio in Journal of the Electrochem, Soc., Vol. 128, p. 423, 1981, "PUnarization phosphorous doped silicon-dioxide," The above-described planarized oxide layer 80 is covered with a doped oxide layer 82 having a thickness of about 400 nm. The oxide layer 82 formed of a TEOS-based oxide is preferably doped with about 4%. Phosphorus to get the so-called gettering 'so it can easily fix the diffused ions. A combination of dopants is also feasible, such as 3% boron and 60/0 scale. Next, a doped oxide layer 82 (preferably PECVD) is deposited on the doped oxide layer 82 to a thickness of about 250 nm. Oxidation layer 82. This oxide layer functions as the same as a hard mask in the later. Then this oxide layer is placed at 700. (: Annealed under nitrogen for about 40 minutes to make it denser. It can also be changed to one The RTA process is performed at about 875 seconds. This RTA process can also replace the annealing step previously performed to generate low-resistance titanium disilicide. In a preferred embodiment, the plane is then lithographically etched. Figure, thereby defining the deep matrix contacts. These contacts are made by first using an anisotropic plasma etching method to convert an applied photoresist layer 81 layout into the underlying oxide layer 84 (82 , 84), then ^ remove this photoresist layer 81 by a known method, and make a deep hole 85 about 7 am under the substrate (9, 1) by dry etching. This process is similar to- —_ -51- Dimensions of this paper --------- A ------ IT ------ ^ (Please read the precautions on the back before filling this page) Luo 42307 9 V. hair Note (49) The process described by the Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs when printing the trench 22 for insulation. When the etching hole 85 is in contact with the substrate, the uppermost oxide layer 84 (the so-called The hard mask) will be completely or partially consumed. The results after the etching substrate connection are shown in Figure 33. After engraving the matrix connection hole 85, the boron is implanted into the plane at a concentration of about 3 × 1015 ion / cm 2 and an energy of about 30 keV in the direction shown by the arrow in FIG. 33. The implantation energy is adjustable so that the boron atom can be blocked by the passive oxidation doped layer 82 'and it will only penetrate into the silicon matrix by first passing through the opening of the hole 8S. After the step, the plane is annealed in oxygen at substantially 875 t: for about 30 seconds. The advantages of generating and implementing a deep-based negative contact as described above in a 1C circuit process for radio frequency have been described in the international patent application PCT / SE97 / 00487. After the above implantation and annealing are completed, the plane is covered with a photoresist layer again, and the contact holes for active and passive components are patterned at this time, as shown in Figs. 34a and 34b. Anisotropic plasma etching is then used to make contact holes 86, 87 in the stacked oxide layers 82, 80. Because the depth of the contact hole is different, and because of the topological figure below, it takes a fixed time to make this last name. Therefore, depending on the topology of some connecting layers, they will be over-etched more strongly than others. After the contact hole is etched, the photoresist layer is removed in a known manner. Therefore, in this case, the contact hole 86 to the substrate and the connection between the passive and active elements (please read the precautions on the back before filling in this page) j Ding, -0. Line -52- This paper The scale uses the Chinese National Standard (CNS) A4 specification (: !! 0 X 297 mm) ---- 1423079 A7 B7 Five 'invention description (50) The contact hole 8 7 of the piece. The result is shown in Figure 3 4 a And β shown in 3 4 b. The surface can be sputter-coated with a double-layer structure or a sandwich structure. The structure of the covered structure is a layer with a bottom of about 100 nm, and its upper part is about 50. nm-thick TiN layer. In the preferred embodiment, the Ti layer is used in a so-called "ion metal plasma equipment (IMP-equipment)" (eg, 'vectra Source' 'This is a trademark of Applied Materials (Named) deposited by sputtering so that it can cover the bottom of the deep matrix contact hole 85. The TiN layer is deposited by reactive sputtering, such as completed in a mixed gas of heart and heart, it is best to use Commonly known as aiming method (also known as consistent sputtering). This TiN layer can also be deposited by In a similar manner to the Ti layer, it is achieved by the reactive base clock method using IMP Vectra Source. After the Ti contact metal layer and the nitrided metal TiN are deposited (the TiN layer can form a barrier layer for the metal layer below it), The plane is annealed at an estimated temperature, so that the Ti layer at a position with a free silicon surface thereon reacts with the underlying silicon or with titanium silicide (that is, in the contact hole for the component) (Titanium silicide reaction) In this preferred embodiment, the annealing is performed in an oven in a mixture of N2 and H2 at a temperature of about 600t: for about half an hour. It can also be performed at a higher temperature in an RTA equipment Annealing, but for a short period of time, such as under N2 or urea atmosphere. It can also be used to strengthen the barrier in the deep matrix junction _, so that the implanted boron atoms can diffuse into the matrix material. _- -53- The size of this paper is toward + ~ ^ | 家 & (CNS) Α4 · (21 〖χ297 公 ---------------- ir ------ line ί Please Read the precautions on the back before filling out this one.) Instruction for Consumer Work Cooperatives, Central Government Standards Bureau, Ministry of Economic Affairs, rd23〇79 * V. Description of the invention (51) M Fan Ji, the Central Standards Bureau employee consumer cooperative, India Fanfan, deposited a town layer of about 1 m in C VD. This deposition step can be a good uniformity, it is performed on the entire surface of the plane Therefore, all contact holes can be completely and uniformly filled with tungsten. The latter is followed by a tungsten-re-etching process to remove all tungsten on the flat (ie horizontal) portion of the plane. When this TiN When the layer is exposed, the etching process stops. This allows tungsten to remain in the contact hole and form a so-called contact plug. Next, a first conductive layer composed of TiN of about 5011111 is deposited in the same manner as described above, and is covered with an aluminum layer having a thickness of about 600 nm. This aluminum layer is formed by sputtering 'and preferably contains 0.5-2_0 ° /. Copper to suppress electron migration. The thickness of these metal layers can vary depending on the application. A TiN layer with a basic thickness of about 50 nm is deposited on the aluminum layer by a reactive base plating method to help the subsequent layout and suppress the so-called, mound, (upward bulge), and then Ti The metal layer structure composed of / TiN / ANCu / TiN is patterned by lithographic etching, and then the component connections are defined by dry etching. More metal layers can be added in this process. The method is to deposit a passive layer over the first connection layer, and then use lithography and dry money engraving to define the via connection. Next, a Ti / TiN two-layer structure is deposited according to the aforementioned sputtering method, and then tungsten is inserted into the opening of the through hole according to the aforementioned procedure. Then, a second metal layer structure composed of a TiN / Al-Cu / TiN stack is deposited by sputtering. Then -54- This paper size applies the Chinese Gardener Standard {CNS) A4 specification (210X297 male f) (please read the precautions on the back before filling this page). "9 V. Description of Invention (52 A7 B7 Ministry of Economic Affairs) The Central Standards Bureau's consumer co-operative printing method and dry name engraving method define the connection layer on it. If more, the process is repeated. The thickness of the A1 Cu layer used can be changed from several hundred ㈣ to A few "claws, depending on the complexity imposed by the metal system and the circuit. If you want to integrate a flat coil in the circuit, a multilayer metal system with a relatively thick conductive layer (such as Al-Cu) is advantageous When manufacturing a flat coil for RF-IC application, a process using several metal layers connected in parallel and disposed above the substrate divided by the trench can be found in International Patent Application No. PCT / SE9 / 0. No. 954. This known process can be used in the aforementioned process. The structure of the core after adding several metal layers is shown in Figure 35a. The upper-left corner of Figure 35a shows the manufactured planar capacitor (called `` CapDn '') ) Cross-section It consists of a single-crystal silicon layer doped with N + underneath it and a p + -type polyethyleneamine layer above the nitride dielectric layer. A lateral PNP transistor is directly located on the right side of the capacitor. Polyethyleneamine forms an emitter and a collector. Its base is formed by diffusion of N + plugs from the surface in series with the bottom diffusion layer of N + type. An NPN transistor is shown on the upper right Cross-sectional view, and the resistance generated by the type of polyacetamide " It should be understood that all the elements arranged in the silicon matrix are insulated from each other by deep trenches. The deep matrix contact holes filled with tungsten in holes 85 and The matrix connection driven by P + polyamidamine is appropriately placed between the insulation grooves 22 surrounding each element area to obtain the best possible electrolytic coupling. -55- This paper standard is common in China Standard (CNS > Α4 specification (210 × 297 mm) (Please read the precautions on the back before filling this page)

*1T A7 B7 423〇^ 9 * 五、發明説明(53 ) 由圖35a中亦可看出除了前述tCapDn以外, =電容CapMm整合於最上方的金屬層之間 '在= 二此金屬-金屬電容CapMlN的製程係使用PECVD· 乳化物作為-介電層,此製程已被調整以使用於鎢 η孔中。後者之電容的製造者將其整合於整 個1社中的優點已描述於瑞典專 97〇1618_2號。 〒叫莱弟 圖3 5b顯π和圖353相同的結構’但其中無 絕緣不同元件的溝槽22,電路平面之表面二的不= 凡件之間的絕緣是在根據圖35之結構中,其只由 的埋入式區域8和Ρ-區域或p—穴14組成。在此情況 下’當實施硼以產生該ρ_區域14時,使用上述範圍 的較尚限制之摻雜濃度,亦即,約丨χ丨〇"離子/平方 公分,以獲得Ρ-區域中之夠高的電導性,因而可在 不同元件區域之間的電路平面表面上得到一良好的 絕緣性。當產生如圖35b所示的結構時,上述之步 驟被省略,此步驟只有在產生溝槽22時才需要,亦 即在施加光阻層2 1和對它做適當的開口、乾蝕刻以 在硬罩幕内產生向下通過聚乙醯胺屏障層19和場氧 化物層18之開口(亦即在氧化層2〇内)、去除該光阻 層21、蝕刻溝槽22、在蝕刻溝槽後做可能的離予植 入步驟,以預估電場臨界值、蝕刻氧化層2〇的其餘 部份、餘刻掉聚乙醯胺層19 >沉積為單晶矽或^聚 乙醯胺的層25,以充填該溝槽22 '將矽層27蝕刻掉 -56- 决乐尺度適用中國國家標準(CNS)A4说格(2ΐ〇χ297公« ) I - rut 1^1 —^^1 .^ϋ -衣 1 I f請先聞讀背面之注意事項再填疼本頁} 訂 ---線1 經濟部中央標準局員工消费合作社印製 PA23079 五、發明説明(54 ) A7 B7 、熱氧化珍平面以氧化在溝槽22内的矽27,以在漠 槽内產生二氧化矽的絕緣層或帽-氧化層28。但是 ’當產生層23時,如上所述之用以使溝槽22之壁氧 化的平面表面之熱氧化(見圖1 8)則必須執行,因為 在此氧化步驟中,K00卜氧化層24是在主動表面上 形成的。 圖36<圖像係顯示最後產生之電路的顯微鏡圖傳 ’其中可看到充填有鎢的深基質接點、聚乙醯胺電 阻和一由溝槽絕緣的ΝΡΝ電晶體。 由圖37中可看出不同元件夯水平地沿平面表面 亩ΓΝ電晶體在其深度方向有延伸,其截面方 向垂直於紙面,而橫向ΡΝΡ電晶體有一相反 形形狀,其中射極放置於中央。 麵濟部中央標準局員工消費合作杜印製 57 度 尺 張 紙 一率 榇 家 國 0* 1T A7 B7 423〇 ^ 9 * V. Description of the invention (53) It can also be seen in Figure 35a that in addition to the aforementioned tCapDn, the = capacity CapMm is integrated between the uppermost metal layers, 'in = this metal-metal capacitor The process of CapMlN uses PECVD · emulsion as the dielectric layer. This process has been adjusted for use in tungsten n-holes. The advantages of the latter capacitor manufacturer who integrated it into the entire 1st company have been described in Swedish Patent No. 97〇1618_2. Howling is shown in Figure 3 and 5b, which shows the same structure as that in Figure 353, but in which there are no grooves 22 of different components for insulation, and the surface of the circuit plane is not equal to the insulation between all parts in the structure according to Figure 35. It consists only of an embedded region 8 and a P-region or p-hole 14. In this case, 'when boron is implemented to produce the ρ_region 14, the more limited doping concentration of the above range is used, that is, about 丨 χ 丨 " ions / cm2 to obtain the P-region The electrical conductivity is sufficiently high, so that a good insulation can be obtained on the plane surface of the circuit between different element regions. When the structure shown in FIG. 35b is generated, the above steps are omitted. This step is only required when the trench 22 is generated, that is, after applying the photoresist layer 21 and making appropriate openings and dry etching to the In the hard mask, openings are created that pass downward through the polyethylene barrier layer 19 and the field oxide layer 18 (that is, within the oxide layer 20). The photoresist layer 21 is removed, the trench 22 is etched, and after the trench is etched, Do a possible implantation step to estimate the critical value of the electric field, etch the rest of the oxide layer 20, and remove the polyacetamide layer 19 > a layer deposited as single crystal silicon or polyacetamide 25, to fill the trench 22 'to etch away the silicon layer 27 -56- The decision standard is applicable to Chinese National Standard (CNS) A4 (2ΐ〇χ297 公 «) I-rut 1 ^ 1 — ^^ 1. ^ ϋ-clothing 1 I f Please read the notes on the back before filling this page} Order --- line 1 Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs PA23079 V. Description of the invention (54) A7 B7, thermal oxidation The plane is oxidized with silicon 27 in the trench 22 to generate a silicon dioxide insulating layer or cap-oxide layer 28 in the trench. However, when the layer 23 is generated, the thermal oxidation of the planar surface (see FIG. 18) used to oxidize the wall of the trench 22 as described above must be performed, because in this oxidation step, the K00 oxide layer 24 is Formed on the active surface. Figure 36 < Image is a microscope image showing the resulting circuit ' In it, a deep matrix contact filled with tungsten, a polyethylenimide resistor, and an pnn transistor insulated by a trench can be seen. It can be seen from Fig. 37 that the different elements ram horizontally extend along the plane surface of the transistor, and the cross-section direction is perpendicular to the paper surface, and the transverse PN transistor has an opposite shape, in which the emitter is placed in the center. The consumer cooperation of the Central Bureau of Standards of the Ministry of Noodles Du printed 57-degree ruled paper One rate 榇 Home country 0

Ns 婦 一釐 公 7 9 2 (讀先閲讀背面之注意事項再填寫本頁〕Ns Women 1 cm 7 9 2 (Read the notes on the back before filling in this page)

Claims (1)

99 C8 D8 2. 3. 4. 經濟部十央標车局員工消費合作社印裝 —種半導體元件,特別是一雙極性NpN型電晶體,在此 疋件表面上之一主動區圍繞有厚的場氧化物區,如由元 件表面視之者,其特徵在於主動區是部份由一電絕緣表 面層覆蓋,以及在此主動區内的一基極區由該電絕緣表 面層内之以微影蝕刻術界定的開口界定之。 如申請專利範圍第丨項之半導體元件,特別是一雙極 性NPN型電晶體,沿著該元件的表面視之,可看見該 元件表面上的射極和集極區域由厚的場氧化物區包圍 其特徵在於一射極區域及/或一集極區域是由一延伸 過和通過該圍繞的場氧化物區之電絕緣表面層内之以 微影蝕刻術所界定的開口予以界定,使得一條狀的電 緣表面層可存在於個別的射極或集極區和最接近此 區的場氧化物區之間。 如申請專利範圍第1或2項的半導體元件,其特徵在於 該電絕緣層包括一氮化矽和氧化矽之疊層。 一種用以製造一雙極性電晶體特別是一 NpN電晶體之 製程,其中在基質上之摻雜有第一摻雜型式的一下方 挣雜區的表面上, 厚場氧化物’其係產生作為環繞一開口的邊緣; 在該場氧化物和開口上^第一摻雜型式特別是N 摻雜層的已摻雜、電子式電矽層,和在其上部產 生第一電絕緣層; : ^該開口中產生一貫穿該髮1和第一絕緣層的孔,使 知场孔會設置在與開口之邊緣和因此該場氧化物相距 {請先閱讀背面之注意事項再填寫本頁) 訂 -58 r4 230 7 9 ABCD 申請專利範圍 經濟部令央榡牟局員工消費合作社印繁 一距離的位置處; 在該開口中摻雜和第—摻雜型式相反的第二摻雜型式( 特別是指口-型)和摻雜第一摻雜型式(特別是指N-型), 以產生一射極和一基極; 其特徵在於 在產生該導電的矽層和第一絕緣層之前,將一第二電 絕緣層施加於該場氧化物和開口上; 由此第二電絕緣層’將該開口内的部份去除,使得它 只以一條狀形式保留在該開口的邊緣上; 菘穿透該矽層和第一絕緣層0孔之製作方式使得此孔 會设置在與該條狀之内緣相距一般距離的位置上,藉 此可在第一次提到的條狀内形成由該第二絕緣層產生 的另一條狀。 如申請專利範圍第4項之製程,其特徵在於第二電絕 緣層的施加是首先施加一底部次層,然後是一上部次 層’以獲得一疊置的第二層。 6·如申請專利範圍第5項之製程,其特徵在於底部次層 的施加是由在表面上產生一薄氧化層的方式達成的。 如申請專利範圍第5或6項之製程,其特徵在於該上部 次層的施加是由在表面上施加一氮化層的方式達成的 〇 如申請專利範圍第4項之製程,其特徵在於摻雜第二 摻雜型式和第一摻雜型式之後,特別是指在K和队摻 雜以後,進行一退火過程,以使摻雜子自導電矽層向 5. 7. 8. 59- ί請先聞讀背面之a意事項再填寫本頁} 訂 本紙張从朝t關轉丰(CNS ) ( 21GX297公釐 P4230 7 9 經濟部中央揉準局舅工消費合作社印«. A8 B8 C8 D8 六、申請專利範圍 一 外擴散,以形成本質的基極,和用以自當摻雜第二摻 雜型和第一摻雜型時(特別是指在?_和沁摻雜中得到者 )所件到的捧雜面積中形成一本質的基極和本質的射極 0 9. 如申請專利範圍第4項之製程’其特徵在於在產生第 =絕緣層時,此製程亦被施加在被作為一電容内的介 電層的區域上,此區域亦設置在基質的第一摻雜型的 挣雜區域上。 10. =申請專利範圍第9項之製程,其特徵在於在產生導 電矽層時,是將該導電矽層f蓋於用以形成介電層的 區域内之第二絕緣層上,以致於該導電矽層可構成電 容的上部電極。 11. 如申請專利範圍第9或10項之製程,其特徵在於第二 電絕緣層的施加是首先施加一下部的薄氧化層,和在 其上施加一上部氮化層的方式進行的’該下部的氧化 層在將上部氮化層施加於一用以形成電容的—埋入式 電極的接點插栓之區域内之前被去除。 !2· —種用以製造一雙極性橫向電晶體的製程,其中在— 基質的下方區域被摻雜以第一摻雜型式的表面上, 厚場氧化物,其以作為環繞一射極和一集極的邊緣之 方式產生; 在该用於射極和集極的開口内摻雜以和第一刑 反的第二摻雜型; ’土相 做在該用於射極和集極的開口内之具有第二摻雜型式 -60- (請先聞讀背面之注意事項再填寫本頁}C8 D8 2. 3. 4. Printed by the Consumer Cooperative of Shiyang Standard Vehicle Bureau of the Ministry of Economics—a kind of semiconductor component, especially a bipolar NpN transistor. A thick field surrounds one of the active areas on the surface of this component. The oxide region, as viewed from the surface of the element, is characterized in that the active region is partially covered by an electrically insulating surface layer, and a base region in the active region is lithographed by the electrically insulating surface layer. Etching defines the opening. For example, for a semiconductor device under the scope of patent application, especially a bipolar NPN transistor, viewed along the surface of the device, it can be seen that the emitter and collector regions on the surface of the device consist of thick field oxide regions. It is surrounded by a feature in which an emitter region and / or a collector region is defined by an opening defined by lithography in an electrically insulating surface layer extending through and through the surrounding field oxide region, so that a An electrical edge surface layer may exist between an individual emitter or collector region and the field oxide region closest to this region. For example, the semiconductor device according to item 1 or 2 of the patent application scope is characterized in that the electrical insulating layer includes a stack of silicon nitride and silicon oxide. A process for manufacturing a bipolar transistor, particularly an NpN transistor, in which a thick field oxide is formed on a surface of a substrate doped with a first doped pattern below a doped region. Around the edge of an opening; on the field oxide and the opening ^ a first doped pattern, especially a doped, electronic silicon layer of an N-doped layer, and a first electrically insulating layer on top of it; ^ A hole is formed in the opening to penetrate the hair 1 and the first insulating layer, so that the field hole will be set at a distance from the edge of the opening and therefore the field oxide (please read the precautions on the back before filling this page). 58 r4 230 7 9 The scope of patent application for ABCD is that the Ministry of Economic Affairs and the Central Government Office of the Ministry of Economic Affairs employee consumer cooperatives print a distance away from each other; in this opening, a second doping pattern opposite to the first doping pattern (especially refers to Mouth-type) and doped first doped pattern (especially N-type) to generate an emitter and a base; characterized in that before the conductive silicon layer and the first insulating layer are generated, a A second electrically insulating layer is applied to the field oxide and On the opening; the second electrical insulation layer 'removes the part inside the opening, so that it only remains on the edge of the opening in a stripe form; 菘 through the silicon layer and the first insulating layer 0 hole making The method is such that the hole is arranged at a general distance from the inner edge of the strip, so that another strip generated by the second insulating layer can be formed in the first-mentioned strip. For example, the process of applying for the fourth item of the patent scope is characterized in that the application of the second electrically insulating layer is firstly applying a bottom sublayer, and then an upper sublayer 'to obtain a stacked second layer. 6. The process of claim 5 in the scope of patent application, characterized in that the application of the bottom sublayer is achieved by generating a thin oxide layer on the surface. For example, the process of applying for item 5 or 6 of the patent scope is characterized in that the application of the upper sublayer is achieved by applying a nitrided layer on the surface. The process of applying for item 4 of the patent scope is characterized by doping After doping the second doping pattern and the first doping pattern, especially after K and team doping, an annealing process is performed to make the dopants from the conductive silicon layer to 5. 7. 8. 59- First read the meanings on the back and then fill out this page} The bound paper is printed from Chaosguan to CNS (21GX297 mm P4230 7 9) Printed by the Central Consumers ’Bureau of the Ministry of Economic Affairs, Masonry Consumer Cooperative«. A8 B8 C8 D8 6 1. The scope of the application for a patent is diffused to form an essential base, and is used when doping the second doping type and the first doping type (especially those obtained in? _ And Qin doping). An essential base and an essential emitter are formed in the area where the component is received. 9. The process of the fourth scope of the patent application is characterized in that when the third insulation layer is generated, this process is also applied to be used as The area of the dielectric layer in a capacitor is also located in the Doped type doped region. 10. = The process of item 9 of the scope of patent application, characterized in that when a conductive silicon layer is generated, the conductive silicon layer f is covered in a region for forming a dielectric layer. On the second insulating layer, so that the conductive silicon layer can constitute the upper electrode of the capacitor. 11. The process of item 9 or 10 of the scope of patent application is characterized in that the application of the second electrical insulating layer is firstly applying the thin part of the lower part. An oxide layer, and an upper nitride layer applied thereon, the lower oxide layer is in the region where the upper nitride layer is applied to a contact plug of a buried electrode used to form a capacitor It was previously removed.! 2 · A process for manufacturing a bipolar lateral transistor, in which the lower region of the substrate is doped with a first doped pattern on the surface, a thick field oxide, which acts as a surround An emitter and a collector are generated by the edge; Doped in the openings for the emitter and collector to a second doping type that is opposite to the first penalty; And a second doping pattern in the opening of the collector -60- (Please read the notes on the back before filling this page} 經濟部t央標牟局員工消費合作社印製 ,23〇79 A8 —_ 群 -~------^ 穴、申請專利範圍 的面積和具有第—摻雜型式的下方摻耗之導電連接 1 其特徵在於 f產生菘厚的場氧化物層之後和摻雜第一摻雜型式之 刖,在該場氧化物和開口上產生一電絕緣層; 由該電絕緣層使開口内的部份去除,使得在開口區域 内的電絕緣層只會以條狀形式保留在開口的邊緣上, 接下來對該開口做第二摻雜型式的摻雜。 13.請專利範園第12項之製程,其特徵在於開口内的 第二摻雜型式之摻雜是以將摻雜物自導電連接線向外 擴散而得。 如申請專利範圍第12項之製程,其特徵在於該電絕緣 層的施加首先是施加一下部次層、然後施加一上部次 層的方式達成,以致於可得到一疊置的第二層。 15. 如申請專利範圍第14項之製程,其特徵在於%下部次 層的施加是由在表面上產生一薄氧化層的方式達成。 16. 如申請專利範圍第14或15項之製程,其特徵在於該上 邵次層的施加是由在表面上施加一氮化層的方式達成 ό令。, 17. —種在一基質的表面上之電容,其特徵在於: 一介電層被配置於一基質表面之一部份上; 一導電層被配置於介電層上,以形成電容之一電極; 以及 一導電連接,係由基質之未由介電層覆蓋的部份之表 -61 - 本紙張尺度適用中囷囷家標準< CNS )八4说格(2丨0X297公釐) ---------衣------tr------^ (婧先閱讀背面之注意事項再填寫本頁) 9 " 六、申請專利範圍 面配置到該介電層下方的區域上。 18. 如申請專利範圍第17項之電容,其特徵在於該電容係 放置在基貝的一摻雜區域上,該區域亦包括該導電連 接線由其配置的基質部份之表面。 19. ,申請專利範圍第17項之電容,其特徵在於該介電層 是一氮矽層》 20. =申請專利範圍第17項之電容,其特徵在於該介電層 疋配置在一埋入式、高摻雜的區域上,一具有高摻雜 的接觸插栓是由基質之未為介電層所覆蓋的該部份内 之表面配置而成。 21. 如申請專利範圍第20項之電容,其特徵在於一具有高 接#之用以形成電容的底部電極之電極插栓,該電極 插拴係自介電層的下側向該埋入式、高摻雜的區域延 伸。 22. —種用以製造在一基質之表面上的電容之製程,其特 徵在於: 介電層,被加在該基質的表面之一部餘上; 經濟部中夫標準局員工消費合作社印^ (請先聞讀背面之注意事項再填寫本頁) —導電層,被加在該介電層上,以形成電容的一電極 :和 與介電層下方的區域之導電連接,它是由基質上未 被該介電層覆蓋的部份之表面配置而成。 23. 如_請專利範圍第22項之製程,其特徵在於在施加該 介電層前,將一具有一摻雜的層設置在基質的表面上 ,接著,此介電層被配置於該已摻雜層的上部,且訪 -62- A8 B8 C8 D8 廖4 2 3 ◦ 7 _________ 申請專利範圍 導⑨連接係由此已摻雜層的表面配置而成。 从如_請專利範圍第22項之製程,其特徵在於在施加該 介電層時’它是以一氮石夕層的方式施加的。 仏如申請專利範圍第h項之製程,其特徵在於在施加該 介電層之前,將-埋入式、高摻雜的區域產生在基質 的表面上,此區域與基質的表面相距一距離,且位於 要施加介電層的位表面下方’接下來自其上將不會施 加該介電層的基質部份之表面產生一具高度接雜的接 觸插栓。 26. 如申請專利範圍第25項之製其特徵在於在施加該 介電層之前,先產生一具有高度摻雜的電極插栓,以 形成電容内的底部電極,該電極插栓是由一基質的一 部份之表面(在要形成該介電層的部份上部)延伸至該 埋入式、高度摻雜的區域上。 27. —種用以在一基質的表面上產生一自由區域之製程, 該區域是由一電絕緣層(特別是指—氮化層)的邊緣 限制,其特徵在於 ’ 將一材料層施加於一用以形成該自由區域的第—區域 上方的表面上: °° 將一第一氧化層均勻地施加於該材料層上; 做一開口貫穿該第一氧化層和材料層,向下到達唁基 質的表面,該開口比第一區域稍大,且具有實質^ 垂直的邊緣’以致於該開口的邊緣與第一區域的相鄰 邊緣相距一實質上為常數的距離; 63 本紙乐尺度適用中国國家梯準(CNS ) A4現格(210 X 297公釐) .衣1T------線 ____(請先聞讀背面之注意事項再填寫本瓦) 經濟部t央橾隼局員工消費合作杜印製 Γ 4 2 3 ο 7 9 AS Β8 C8 D8 '中請專利範圍 經濟部中央標準局員工消費合作衽印製 將不同於孩氧化層的另一種型式之電絕緣層均勾地施 加於整個表面上; 將其蝕刻特性不同於該第一氧化層的第二氧化層均勻 地施加於電絕緣層的整個表面上; θ 進行一第一各向異性蝕刻,以去除只在本質上全部是 平坦的、水平的表面上之第二氧化層,藉此使該電絕 緣層曝露於這些表面上,且第二氧化層的側條被保留 在該垂直表面上; 進行第一蝕刻,以去除只有在非為該第二氧化層覆 蓋的表面上之電絕緣層,藉$使只有在平坦的、水平 的表面上之電絕緣層被去除厂 利用1^第二氧化層的蝕刻特性不同於第一氧化層的蝕 刻特性的事實,進行第三次姓刻,以只去除該第二層 =未與第一氧化層的自由表面部份踫撞的剩下部份, 藉此使得忒電絕緣層的邵份實質上只會保留在該材料 層以及圍繞該第一面積的第一氡化層的垂直邊緣上。 28. 如申請專利範圍第27項之製程,其特徵在於該材料層 包含一導電材料層,其至少與該第一面積接近的部份 會與基質的表面電接觸。 29. 如申請專利範圍第28項之製程,其特徵在於該導電材 料層形成其垂直邊緣上之材料層的所有厚度。 3〇·二種用以在一基質的表面上產生一自由區域的製程, 該基質在該自由區域的邊緣上具有垂直部份,其特徵 在於首先將第一電絕緣層均勻地施加在基質的表面上 請 先 聞 讀 背 * i I 再 填 寫 本 頁 訂 •银 64-Printed by the Consumers' Cooperative of the Central Government Bureau of the Ministry of Economic Affairs, 23〇79 A8 —_ group-~ ------ ^ holes, area covered by patents, and conductive connection with the first doped type below the doped consumption 1 It is characterized in that after a thick field oxide layer is generated by f and a first doping pattern is doped, an electrical insulating layer is generated on the field oxide and the opening; and the part inside the opening is made by the electrical insulating layer It is removed so that the electrical insulation layer in the opening region will only remain on the edge of the opening in a stripe form, and then the opening is doped with a second doping type. 13. The process of item 12 of the patented patent park is characterized in that the doping of the second doping pattern in the opening is obtained by diffusing the dopant from the conductive connection line outward. For example, the process of claim 12 is characterized in that the application of the electrical insulating layer is firstly achieved by applying a lower sublayer and then an upper sublayer, so that a stacked second layer can be obtained. 15. The process of claim 14 in the scope of patent application is characterized in that the application of the lower sublayer is achieved by a thin oxide layer on the surface. 16. The process of claim 14 or 15 is characterized in that the application of the upper sub-layer is achieved by applying a nitrided layer on the surface. 17. A capacitor on the surface of a substrate, characterized in that: a dielectric layer is disposed on a portion of a substrate surface; a conductive layer is disposed on the dielectric layer to form one of the capacitors Electrodes; and a conductive connection, which is the part of the substrate that is not covered by the dielectric layer. -61-This paper applies the standard of China Standard < CNS), 8 and 4 (2 丨 0X297 mm)- -------- 衣 ------ tr ------ ^ (Jing first read the precautions on the back and then fill out this page) 9 " VI. Patent Application Scope Configuration to the dielectric On the area below the layer. 18. The capacitor according to item 17 of the scope of patent application, characterized in that the capacitor is placed on a doped region of the base, and the region also includes the surface of the substrate portion where the conductive connection is arranged. 19. The capacitor in the scope of patent application No. 17 is characterized in that the dielectric layer is a silicon nitride layer. 20. = The capacitor in the scope of patent application No. 17 is characterized in that the dielectric layer is disposed in a buried state. A highly-doped contact plug is formed from a surface in the part of the matrix that is not covered by the dielectric layer on the highly doped region. 21. The capacitor according to item 20 of the patent application is characterized by an electrode plug having a high connection to form a bottom electrode of the capacitor, and the electrode plug is connected from the lower side of the dielectric layer to the buried type. Highly doped regions extend. 22. —A manufacturing process for manufacturing a capacitor on the surface of a substrate, characterized in that: a dielectric layer is added to one of the surfaces of the substrate; (Please read the precautions on the back before filling out this page) — A conductive layer is added to the dielectric layer to form an electrode of the capacitor: it is conductively connected to the area below the dielectric layer. It is formed by the substrate The surface of the portion not covered by the dielectric layer is arranged. 23. For example, the process of claim 22 is characterized in that a layer with a doping is disposed on the surface of the substrate before the dielectric layer is applied, and then the dielectric layer is disposed on the substrate. The upper part of the doped layer, and visit -62- A8 B8 C8 D8 Liao 4 2 3 ◦ 7 _________ The scope of patent application for the conductive connection is configured from the surface of the doped layer. The process according to item 22 of the patent scope is characterized in that when the dielectric layer is applied, it is applied in the form of a nitrogen stone layer. For example, the process of item h in the scope of patent application is characterized in that, before applying the dielectric layer, a -buried, highly doped region is generated on the surface of the substrate, and this region is a distance from the surface of the substrate. It is located below the bit surface to which the dielectric layer is to be applied. 'Next, a highly doped contact plug is created from the surface of the substrate portion on which the dielectric layer will not be applied. 26. The system of claim 25 is characterized in that before applying the dielectric layer, a highly doped electrode plug is generated to form the bottom electrode in the capacitor. The electrode plug is made of a substrate. A portion of the surface (on top of the portion where the dielectric layer is to be formed) extends onto the buried, highly doped region. 27. A process for creating a free area on the surface of a substrate, the area being constrained by the edges of an electrically insulating layer (especially a nitride layer), characterized in that 'a layer of material is applied to A surface above the first area for forming the free area: °° A first oxide layer is uniformly applied to the material layer; an opening is made through the first oxide layer and the material layer, and reaches down to 唁On the surface of the substrate, the opening is slightly larger than the first region, and has a substantially vertical edge, so that the edge of the opening and the adjacent edge of the first region are at a substantially constant distance; 63 This paper scale is applicable to China National Ladder Standard (CNS) A4 is now available (210 X 297 mm). Clothing 1T -------- line ____ (please read the notes on the back before filling in this tile) Staff of the Central Government Bureau of the Ministry of Economic Affairs Consumption cooperation Du printed Γ 4 2 3 ο 7 9 AS Β8 C8 D8 'Please request the patent scope Employees of the Central Standards Bureau of the Ministry of Economic Affairs Consumer Cooperation 衽 Printing will apply another type of electrical insulation layer different from the oxide layer On the entire surface; A second oxide layer having characteristics different from the first oxide layer is uniformly applied to the entire surface of the electrically insulating layer; θ performs a first anisotropic etching to remove only the substantially flat, horizontal surface A second oxide layer, thereby exposing the electrically insulating layer to the surfaces, and the side strips of the second oxide layer are retained on the vertical surface; a first etching is performed to remove only the second oxide The electrically insulating layer on the surface covered by the layer, so that only the electrically insulating layer on the flat, horizontal surface is removed. The fact that the etching characteristic of the second oxide layer is different from that of the first oxide layer is used. Carry out the third surname engraving to remove only the remaining part of the second layer = that does not collide with the free surface portion of the first oxide layer, so that only a small portion of the galvanic insulating layer will remain. On the material layer and on the vertical edges of the first halide layer surrounding the first area. 28. The process of claim 27, wherein the material layer includes a conductive material layer, and at least a portion close to the first area is in electrical contact with the surface of the substrate. 29. The process of claim 28, characterized in that the conductive material layer forms all thicknesses of the material layer on its vertical edges. 30. Two processes for generating a free area on the surface of a substrate, the substrate having a vertical portion on the edge of the free area, which is characterized in that a first electrical insulating layer is first uniformly applied to the substrate. On the surface, please read it first * i I, then fill in this page to order • Silver 64- A3 B8 C8 D8 '423079 -*---- 申請專利範圍 將第二電絕緣層均勾地施加在該第—層的表运 ’這些層的材料之選擇係使其具有不同的蝕刻特担 沾f下來進行選擇性地蝕刻’以首先去除實質上垂直 些部::::::所有第-和第二層,然後亦去除這 31·=請專利範圍第3Q項H其特徵在於該第ϋ 1是以兩步驟芫成,第—次步驟中,其中只有在水平 ,,部份上的第二階梯被移除,在第二次步驟中,其 只有在水平表面部份上的第一階梯被移除。 32. :種位於一摻雜基質之表面的主動區上之側條結構, 其特徵在於導切基質之配置與主動區之-邊界區接 觸和電絕緣層本質上只均勻地配置在該導電矽部份 文垂直表面部份上,藉此使各處之該電絕緣層本質上 具有相同的厚度。 33. 如,請專利範圍第32項之侧條結構,其特徵在於—絕 緣氧化層實質上配置在該導電矽層的整個表面上,且 在該電絕緣層的下方。 34. 如申請專利範圍第32項之側條結構,其特徵在於該電 絕緣層是一氮矽層。 35. 種用以在一基質的表面上製造一電晶體的製程,其 中。導電珍層被加在該表面上方,以與該表面上的捧 雜區接觸,然後將該電絕緣層直接施加於該導電秒層 的上方,再將另一導電層加在絕緣層的上方,接下來 將未為此額外導電層所覆蓋的區域内之電絕緣層去除 -65 - 从適用TSIBI家標率(CNS )从祕(2敝加公着) 农-------- (請先閣讀背面之注意事項再填莴本頁) 張 紙 I__一本 經濟部宁央標率局眉工消費合作社印製 14 2 3 Ο 7 9 ABCD 中請專利範圍 經濟部中央標隼局員工消費合作社印製 /以保留與該後來施加的導電金屬層的電接觸,其特 徵在於只有在該選定區域内的電絕緣層被去除。 36.如申請專利範園第35項之製程,其特徵在於當做去除 步驟時’施加一其上以微影蝕刻術製作有開口的罩幕 ’使得該罩幕之其餘部份可覆蓋先前施加的絕緣層之 選定區域和在後來施加電絕緣層時避免被矽化的區域 ,以致於在完成該去除步驟後,這些層仍會由該絕緣 層覆蓋。 3入—種積體電路,包含在一基質之表面上的元件,其特 徵在於每一元件實質上是以亨構在該基質之表面層的 Ν-摻雜區域上的方式產生’且一埋入式、高度摻雜的 區域位於其下方和基質内且具有和其上方區域相同的 摻雜$式,以底部接觸該队摻雜區域,和在元件之間 在基質的表面層内存在有ρ_摻雜的區域,其下方具有 和,上方區域相同的摻雜型式之埋入式區域,以作為 此區域和基質之間的接觸,藉此可得到元件的橫向電 絕緣。 38.=申請專利範圍第37項之積體電路,其特徵在於在基 質之表面層内之Ρ·摻雜區域上方配置有一良好導電層 ,該層係用以與電氣接地面連接。 39_ 了種用以製造一積體電路之製程,該積體電路在基質 之表面上包含有元件,其特徵在於 在欲製作元件的第一區域上,製作一具有和基質之表 面層相同的第一摻雜型式之摻雜區,一具有相同型式 請 先 聞 之 注 2 旁 訂 -66 - Α4说格(210x297公釐) A8 B8 C8 D8 423079 6、申請專利範圍 區域位於其下方,它的摻雜型式 型式的摻雜面積接觸:和 .万…、有弟-摻雜 =====製作摻雜區域,此區 接其下万、其摻雜型式與在其上方的區域之 .同:區’以使它和基質電接觸,藉此 .Η于到兀件又横向電絕緣。 圍第39項之製程,其特徵在於完成摻雜 使弟二摻雜進入基質,表面層内後,在其表面 上她加一艮好導電的層,以與電接地連接。 41_ =申請專利範圍第40項之製程,其特徵在於該良好導 是—接雜有該第二摻雜型式的高度接雜層,Ϊ =層,受-退火操作,以使摻雜子由其中擴散到一基 ^又表面層内之具有第二摻雜的摻雜區域内,以減少 该艮好導電層和具有第二摻雜的接雜區域之間的阻抗 42. :種用以製造用於配置在_基質表面之元件的基質連 經 濟 部 中 央 準 局 Ά 工 消 合 作 社 印 製 接之氣心,產生一深孔洞並充填以導電材料(最好是一 金屬)’其特徵在於在充填該導電材料之前,先在該深 孔洞底部進行一摻雜,以降低在該導電材料和基質之 間的阻抗。 43,如申請專利範圍第42項之製程,其特徵在於該捧雜是 經由植入和接下來做退火操作,以擴散植入原子之方 -67- 本紙浪尺度適用中國國家梯準(CNS > A4規格(2!OX297公;§ ) A8 B8 C8 D8 4230 7 9 '申請專利範圍 式達成。 44. 如申請專利範圍第43項之製程,其特徵在於在該植入 步驟中係植入硼。 45. 如申請專利範圍第42項之製程’其特徵在於在完成摻 雜步驟後,將一含鈦的薄層加在該深孔洞的壁和底部 上P 46. 如申請專利範圍第衫項之製程,其特徵在於在施加至 47 少一薄層後,進行一退火操作,以在該至少一薄層和 基質之間的邊界表面上形成矽化物。 如申請專利範圍第45項之製程,其特徵在於該至少— 薄層疋作為只有鈇的第一層之方式形成,且在其上方 為一薄的氮化鈇層。 48 —種用於配置在一基質表面上之半導體元件的基質連 接,該基質上具有一深孔洞,其内充填以導電材料(特 別是一金屬),其特徵在於在基質内配置一摻雜於該深 孔洞之底部’以降低導電材料和基質之間的阻抗。 49.如申請專利第48項之基質連接,其特徵在於該摻雜包 含植入之硼原子。 經濟部中央榡率局員工消費合作社印製 5〇·如申請專利第48項之基質連接,其特徵在於至少一含 鈦的薄層被設置在基質與導電材料之間的深孔洞之壁 和底部上。 51_ ^申請專利第50項之基質連接,其特徵在於該至少一 薄層包含形成在該至少一薄層和基質之間的邊界表面 上的矽化物。 -68- 本紙張尺度逍用中國國家標準(CNS ) A4说格(210X297公釐 423079¾ 六、申請專利範圍 種形成在一半導體平面的表面上之雙極性裝置, 含一電絕緣層,此層係至少部份地包圍位於裝置表面 亡的王動區’(且被充填以電絕緣或半電絕緣材料,^ 特徵在於該電絕緣材料或半電絕緣材料直接在溝槽;; 土包含-氧化層和其上有一氮化層,該溝槽的主^ 刀被充填以一均勻的或是同質性的電絕緣或半電絕 材料。 ,, 53,如申請專利範圍第52項之雙極性裝置,其特徵在於該 均勻的或同質性的電絕緣材料或半電絕緣材料本質上 包含未摻雜的碎,最好是微晶碎。 54_如申請專利範圍第52或53項‘雙極性裝置,其特徵在 $該均勻的或同質性的電絕緣材料或半電絕緣材料是 電子半絕緣的材料,且在該半電絕緣材料和氮化層之 間具有另一個氧化層。 55. 如申請專利範圍第52項之雙極性裝置’其特徵在於該 均勻或同質性的電絕緣材料或半電絕緣材料包含氧化 砂。 56. —種用以於一半導體平面之表面上製造一雙極性元件 的製程’包含下列步驟: 經濟部中央標隼局員工消費合作社印製 產生一電絕緣溝槽,其部份圍繞位於該元件表面上之 主動區; 以電絕緣材料或半電絕緣材料充填該溝槽: 其特徵在於該充填溝槽的步驟是由下列步驟達成: 在該溝槽的壁上施加一氧化層.; 在該氧化層上施加一氮化層; -69- 本紙浪尺度賴t S國家標率(€叫八4胁(2丨0父297公董) 膠42307 Q A8 η C8 D8 '申請專利範圍 =一均句的或同質性的電絕緣或半電絕緣材料充填該 溝槽的其餘部份。 57.=申請專利範圍第56項之製程,其特徵在於當以該均 二的或同質性的電絕緣材料或半電絕緣材料充填時, 使用一本質上未摻雜的矽,最好是單晶矽。 8·如申凊專利範圍第56或57項之製程,其特徵在於於當 以孩均勻的或同質性的電絕緣材料或半電絕緣材料充 填時,使用一半電絕緣材料,和在於以該半電絕緣材 料充填以如時,先將—氧化層施加在氮化層的上方。 如申請專利範圍第56項之製程,其特徵在於當充填該 溝槽的其他主要部份時,使用氧化矽作為該均勻的或 同質性的電絕緣材料。 (靖先閎讀背面之注意事項再填寫本K) t 經濟部_央榡率局員工消費合作社印製 70-A3 B8 C8 D8 '423079-* ---- The scope of the patent application applies the second electrical insulation layer to the surface of the first layer. The materials of these layers are selected so that they have different etching characteristics. f. Selectively etch down to remove the substantially vertical portions first :::::: all the first and second layers, and then also remove the 31 · = patent scope item 3Q H, which is characterized by the 1 is formed in two steps. In the first step, only the second step on the horizontal part is removed. In the second step, it is only the first step on the horizontal surface part. Was removed. 32. A side strip structure on the active region on the surface of a doped substrate, which is characterized in that the arrangement of the lead-through substrate is in contact with the boundary region of the active region and the electrical insulating layer is essentially uniformly arranged only on the conductive silicon The partial vertical surface portion is thereby used to make the electrical insulating layer substantially the same thickness everywhere. 33. For example, the sidebar structure of item 32 of the patent scope is characterized in that-an insulating oxide layer is substantially disposed on the entire surface of the conductive silicon layer and under the electrically insulating layer. 34. The sidebar structure of item 32 in the scope of patent application, characterized in that the electrical insulation layer is a silicon silicon layer. 35. A process for making a transistor on the surface of a substrate, wherein. A conductive layer is added over the surface to make contact with the doped region on the surface, and then the electrical insulating layer is directly applied over the conductive second layer, and another conductive layer is added over the insulating layer. Next, remove the electrically insulating layer in the area not covered by this additional conductive layer -65-From the applicable TSIBI family standard rate (CNS) from the secret (2 敝 plus public) Nong -------- ( Please read the notes on the back first and then fill out the lettuce page) Sheet of paper I__ A copy printed by the Ningyang Standards Bureau of the Ministry of Economic Affairs and the Consumer Cooperatives 14 2 3 Ο 7 9 The employee consumer cooperative prints / retains electrical contact with the subsequently applied conductive metal layer and is characterized in that only the electrically insulating layer in the selected area is removed. 36. The process of claim 35 in the patent application park, characterized in that when the removal step is performed, 'applying a mask with openings made by lithographic etching thereon' so that the rest of the mask can cover the previously applied mask Selected areas of the insulating layer and areas that are prevented from being silicified when the electrical insulating layer is subsequently applied, so that after the removal step is completed, these layers will still be covered by the insulating layer. 3-input—a kind of integrated circuit, comprising elements on the surface of a substrate, characterized in that each element is essentially generated in a manner that is structured on the N-doped region of the surface layer of the substrate, and is buried An in-fill, highly doped region is located below and in the matrix and has the same doping pattern as the region above it, contacts the doped region with the bottom, and there is ρ in the surface layer of the matrix between the elements The doped region has a buried region with the same doping pattern as the upper region as a contact between the region and the substrate, thereby obtaining lateral electrical insulation of the element. 38. = The integrated circuit of item 37 of the scope of patent application is characterized in that a good conductive layer is arranged above the P · doped region in the surface layer of the substrate, and this layer is used to connect to the electrical ground plane. 39_ A process for manufacturing an integrated circuit that includes components on the surface of a substrate is characterized in that a first region having the same surface layer as that of the substrate is made on the first region of the component to be fabricated. A doped area of the doped type, please note first with the same type. 2 Side-66-Α4 grid (210x297 mm) A8 B8 C8 D8 423079 6. The area of the patent application area is located below it. Heterotype pattern of doped area contact: and ... million, and have-doped ===== make a doped region, this region is connected to the next, its doped pattern is the same as the region above it. Zone 'so that it is in electrical contact with the substrate, thereby allowing the element to be laterally electrically insulated. The process of item 39 is characterized in that the doping is performed so that the second dopant enters the substrate. After the surface layer is added, a conductive layer is added to the surface to connect it to the electrical ground. 41_ = The process of applying for the scope of patent application No. 40, which is characterized in that the good conductivity is-a highly doped layer doped with the second doping type, Ϊ = layer, subjected to -annealing operation, so that the dopants pass through it Diffusion into a doped region with a second doping in a base layer and a surface layer to reduce the impedance between the conductive layer and the doped region with the second doping The components placed on the surface of the substrate are connected to the Central Quasi Bureau of the Ministry of Economic Affairs and printed by the Industrial Cooperative, creating a deep hole and filling it with a conductive material (preferably a metal). Prior to the conductive material, a doping is performed at the bottom of the deep hole to reduce the impedance between the conductive material and the substrate. 43, if the process of applying for the scope of the patent No. 42 is characterized in that the impurity is implanted and then annealed to diffuse the implanted atoms -67- The standard of this paper is applicable to China ’s national standard (CNS & gt A4 specification (2! OX297 public; §) A8 B8 C8 D8 4230 7 9 'The scope of patent application was reached. 44. If the process of applying for the scope of patent item 43 is characterized in that the implantation step is implanted with boron 45. If the process of applying for the scope of patent No. 42 is characterized in that after the doping step is completed, a thin layer containing titanium is added to the wall and bottom of the deep hole. The process is characterized in that after applying to at least one thin layer, an annealing operation is performed to form a silicide on the boundary surface between the at least one thin layer and the substrate. It is characterized in that the at least-thin layer of gadolinium is formed as the first layer of gadolinium, and a thin gadolinium nitride layer is formed above it. 48-a matrix connection for a semiconductor element arranged on a substrate surface , Which has A deep hole is filled with a conductive material (especially a metal), and is characterized in that a bottom doped with the deep hole is disposed in the substrate to reduce the impedance between the conductive material and the substrate. 49. For example, applying for a patent The matrix connection of item 48 is characterized in that the doping contains implanted boron atoms. Printed by the Consumer Cooperatives of the Central Government Bureau of the Ministry of Economic Affairs 50. If the matrix connection of item 48 of the patent application is characterized by at least one A thin layer of titanium is provided on the wall and bottom of the deep hole between the substrate and the conductive material. 51_ ^ The patent application No. 50 substrate connection, characterized in that the at least one thin layer includes the at least one thin layer and Silicide on the surface of the boundary between the substrates. -68- This paper size is in accordance with Chinese National Standard (CNS) A4 (210X297 mm 423079¾) 6. Scope of patent application Bipolar formed on the surface of a semiconductor plane The device includes an electrically insulating layer which at least partially surrounds the active area on the surface of the device '(and is filled with an electrically insulating or semi-electrically insulating material, which is characterized by The electrically insulating material or semi-electrically insulating material is directly in the trench; the soil contains an oxide layer and a nitride layer thereon, and the main blade of the trench is filled with a uniform or homogeneous electrically insulating or semi-insulating material. Electrical insulation materials, 53, such as the bipolar device of the 52th scope of the patent application, characterized in that the homogeneous or homogeneous electrical insulation material or semi-electrical insulation material essentially contains undoped fragments, preferably Microcrystalline fragments. 54_ If the patent application scope is 52 or 53 'Bipolar device, it is characterized in that the uniform or homogeneous electrically insulating material or semi-electrically insulating material is an electronic semi-insulating material, and There is another oxide layer between the electrically insulating material and the nitride layer. 55. The bipolar device 'according to item 52 of the scope of the patent application is characterized in that the uniform or homogeneous electrically insulating material or semi-electrically insulating material contains oxidized sand. 56. —A process for manufacturing a bipolar element on the surface of a semiconductor plane 'includes the following steps: Printed by the Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs to produce an electrically insulating trench, a portion of which surrounds the element Active area on the surface; filling the trench with an electrically insulating material or semi-electrically insulating material: characterized in that the step of filling the trench is achieved by the following steps: applying an oxide layer on the wall of the trench; A nitride layer is applied on the oxide layer; -69- The standard of this paper is t S national standard (called eight 4th (2 丨 0 parent 297 public director) glue 42307 Q A8 η C8 D8 'Scope of patent application = one uniform Sentence or homogeneous electrically insulating or semi-electrically insulating material fills the rest of the trench. 57. = The process of applying for item 56 of the scope of patent application is characterized by the homogeneous or homogeneous electrically insulating material When filling with semi-electrically insulating materials, a substantially undoped silicon, preferably single crystal silicon, is used. 8. The process of item 56 or 57 of the patent application scope is characterized by a uniform or Homogeneous electrically insulating material or When filling the electrically insulating material, half of the electrically insulating material is used, and the semi-electrically insulating material is filled with the semi-electrically insulating material as before, and an oxide layer is first applied on the nitrided layer. For example, the process of applying for the scope of patent application No. 56 features The reason is that when filling the other main parts of the trench, use silicon oxide as the uniform or homogeneous electrical insulation material. (Jingxian first read the precautions on the back and then fill out this K) t Ministry of Economy _ 阳 榡 率 局Printed by Employee Consumer Cooperatives 70-
TW087107482A 1997-07-11 1998-05-14 A process for manufacturing ic-components to be used at radio frequencies TW423079B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
SE9702693A SE9702693L (en) 1997-07-11 1997-07-11 Process for manufacturing IC components for radio frequency use

Publications (1)

Publication Number Publication Date
TW423079B true TW423079B (en) 2001-02-21

Family

ID=20407731

Family Applications (1)

Application Number Title Priority Date Filing Date
TW087107482A TW423079B (en) 1997-07-11 1998-05-14 A process for manufacturing ic-components to be used at radio frequencies

Country Status (2)

Country Link
SE (1) SE9702693L (en)
TW (1) TW423079B (en)

Also Published As

Publication number Publication date
SE9702693L (en) 1999-01-12
SE9702693D0 (en) 1997-07-11

Similar Documents

Publication Publication Date Title
US6610578B2 (en) Methods of manufacturing bipolar transistors for use at radio frequencies
TW569451B (en) Process for fabricating a vertical bipolar junction transistor, process for fabricating an integrated circuit structure, and integrated circuit structure
US6118171A (en) Semiconductor device having a pedestal structure and method of making
JP2009141375A (en) Semiconductor process and integrated circuit
KR100691036B1 (en) Lateral bipolar transistor and method of making same
JPH0376576B2 (en)
JP2002512736A (en) Bipolar device isolated by trench
JPH0693462B2 (en) Bipolar transistor and manufacturing method thereof
JPH07193155A (en) Varactor and manufacture thereof
JP2003524875A5 (en)
JP2002252230A (en) Heterojunction bipolar transistor
KR20040071949A (en) Method for manufacturing self-aligned BiCMOS
JPH06342802A (en) High-performance semiconductor device and its manufacture
US20030080394A1 (en) Control of dopant diffusion from polysilicon emitters in bipolar integrated circuits
JPH11330084A (en) Manufacture of bipolar transistor and its structure
US6180478B1 (en) Fabrication process for a single polysilicon layer, bipolar junction transistor featuring reduced junction capacitance
JPH0744160B2 (en) Method of forming self-aligned contact
JPH0646638B2 (en) Vertical bipolar transistor and method of manufacturing the same
TW423079B (en) A process for manufacturing ic-components to be used at radio frequencies
US6475848B1 (en) Polysilicon-edge, low-power, high-frequency bipolar transistor and method of forming the transistor
TW417275B (en) Method for manufacturing bipolar transistor capable of supressing deterioration of transistor characteristics
JPH07335664A (en) Preparation of bipolar junction transistor with polysilicon emitter
US20040082135A1 (en) Structure and method for forming self-aligned bipolar junction transistor with expitaxy base
JP3456864B2 (en) Semiconductor device and manufacturing method thereof
JP3537494B2 (en) Semiconductor device and method of manufacturing semiconductor device

Legal Events

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