TW201137975A - Silicon carbide semiconductor device and its manufacturing method - Google Patents

Silicon carbide semiconductor device and its manufacturing method Download PDF

Info

Publication number
TW201137975A
TW201137975A TW99112237A TW99112237A TW201137975A TW 201137975 A TW201137975 A TW 201137975A TW 99112237 A TW99112237 A TW 99112237A TW 99112237 A TW99112237 A TW 99112237A TW 201137975 A TW201137975 A TW 201137975A
Authority
TW
Taiwan
Prior art keywords
layer
metal
sic
semiconductor device
metal element
Prior art date
Application number
TW99112237A
Other languages
Chinese (zh)
Inventor
Hideto Tamaso
Original Assignee
Sumitomo Electric Industries
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 Sumitomo Electric Industries filed Critical Sumitomo Electric Industries
Priority to TW99112237A priority Critical patent/TW201137975A/en
Publication of TW201137975A publication Critical patent/TW201137975A/en

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/04Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer
    • H01L21/0445Manufacture 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 crystalline silicon carbide
    • H01L21/048Making electrodes
    • H01L21/0485Ohmic electrodes

Landscapes

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Electrodes Of Semiconductors (AREA)

Abstract

The present invention provides a silicon carbide semiconductor device having an ohmic electrode that can achieve improved adhesion to a wiring line by suppressing deposition of carbon without forming a Schottky contact, and also provides its manufacturing method. When an ohmic electrode is formed for a SiC semiconductor device, a first metal layer (12) that is formed of a first metal element is formed on one main surface of an SiC layer (11). Meanwhile, a Si layer (13) that is formed of Si is formed on a surface of the first metal layer, wherein the surface is on the reverse side of the surface facing the SiC layer (11). A laminated structure (10A) formed by the aforementioned process is subjected to a heat treatment. Consequently, there can be obtained a silicon carbide semiconductor device which includes an ohmic electrode that has good adhesion to a wiring line, while being suppressed in deposition of carbon atoms on the surface layer of the electrode and formation of a Schottky contact between Si and SiC.

Description

201137975 六、發明說明: 【發明所屬之技術領域】 本發明係關於一種碳化矽半導體裝置及其製造方法者, 更特定而言,其係關於一種可提高電極與佈線之密著性之 碳化矽半導體裝置及其製造方法。 【先前技術】 寬帶隙半導體之一的碳化矽(Sic)係作為用以實現高頻 功率元件或耐熱•耐放射線元件之材料而受到矚目。由於 碳化矽可藉由與矽(si)相同之手法而形成氧化膜(si〇2),故 對碳化♦半導體裝置、例如M〇SFET(Metal 〇xide_ Semiconductor Field_Effect Transist〇r,金屬氧化物半導體 場效電晶體)之研究逐漸盛行。另外,與Si相比,SiC之帶 隙(禁止帶寬)較寬’絕緣破壞電場強度較大。因此,例如 與使用有Si之半導體裝置相比,使用有Sic之半導體裝置 了成為切換特性優異、对電壓較大之半導體裝置。 —般而言,於SiC半導體裝置中具備用以容易地自形成 於基板上之電極取出電氣信號之佈線(塾)。並不限於sic半 導體裝置,對於Si半導體裝置’亦可經由該佈線而順利地 與·外部進行電氣信號之交換。 圖35係表示一般的SiC半導體裝置之電極與佈線之狀態 之概略剖面圖。如圖35所示,於SiC半導體基板99之一個 主表面上配置有電極98的SiC半導體裝置99A中,存在如下 情形:於電極98中之與SiC半導體基板99不對向之一個主 表面上析出碳97。此處,電極98係與SiC半導體基板99進 147784.doc 201137975 行馱姆接觸之歐姆電極。再者,此處所謂主表面,係指形 成表面之平面中之面積最大的面。在析出該碳97(c)後, 於形成配置在電極98中之與Sic半導體基板99不對向之一 個主表面上的佈線96時,在介入有碳97之區域,佈線96無 法與電極98直接接觸。因此,碳97會使電極⑽與佈線%之 密著性惡化。因此,其成為導致佈線96自電極98剝離之不 良情形之原因,從而有可能對Sic半導體裝置99八之耐久性 或電氣特性造成影響。 為了解決上述問題,例如日本專利特開平7_99169號公 報(以下,稱為「專利文獻1」)所揭示,考慮使用於SiC基 板95上形成Νι與Si之合金層之結構。圖36係表示於sic半 導體層上形成有Ni與Si之合金層之結構的概略剖面圖。如 圖36所示之電子元件95A係於Sic基板%上形成Ni si合金 層94後實施熱處理,藉此,Ni_Si合金層%具有作為歐姆電 極之功旎,此被揭示於專利文獻丨中。又,於專利文獻工 中,亦揭不有如下情形:對在SiC基板95上形成有Si層、 且在Si層上形成有州層之積層結構進行熱處理藉以形成歐 姆電極^ 已判明圖3 5所示之碳97係由以下方法而形成:因形成電 極98時之熱處理而使構成電極98之金屬與sic半導體基板 99之SiC發生反應,藉此自sic產生c(碳)作為殘渣並使 其於電極98之表面析出。因此’為了構成歐姆接觸,專利 文獻1中揭示:於Sic半導體基板99之车表面上,形成作為 金屬(Ni)與Si之合金之犯_8丨合金層94,藉此形成如圖刊所 147784.doc 201137975 示之電子元件95A。專利文獻1中亦有如下揭示:或者於BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a niobium carbide semiconductor device and a method of manufacturing the same, and more particularly to a niobium carbide semiconductor capable of improving adhesion between an electrode and a wiring. Device and method of manufacturing the same. [Prior Art] A tantalum carbide (Sic) which is one of wide-bandgap semiconductors has been attracting attention as a material for realizing a high-frequency power element or a heat-resistant/radiation-resistant element. Since the tantalum carbide can form an oxide film (si〇2) by the same method as the bismuth (si), the carbonization semiconductor device, for example, M〇SFET (Metal 〇xide_ Semiconductor Field_Effect Transist〇r, metal oxide semiconductor field Research on electro-optical crystals has become increasingly popular. In addition, compared with Si, the SiC band gap (prohibited bandwidth) is wider, and the dielectric breakdown electric field strength is large. Therefore, for example, a semiconductor device using Sic is used as a semiconductor device having excellent switching characteristics and a large voltage compared with a semiconductor device using Si. In general, a SiC semiconductor device is provided with a wiring for easily taking out an electrical signal from an electrode formed on a substrate. It is not limited to the sic semiconductor device, and the Si semiconductor device can be smoothly exchanged with the external electric signal via the wiring. Fig. 35 is a schematic cross-sectional view showing the state of electrodes and wirings of a general SiC semiconductor device. As shown in FIG. 35, in the SiC semiconductor device 99A in which the electrode 98 is disposed on one main surface of the SiC semiconductor substrate 99, there is a case where carbon is deposited on one main surface of the electrode 98 which is not opposed to the SiC semiconductor substrate 99. 97. Here, the electrode 98 is an ohmic electrode which is in contact with the SiC semiconductor substrate 99 in 147784.doc 201137975. Further, the term "main surface" as used herein refers to a surface having the largest area among the planes forming the surface. After the carbon 97(c) is precipitated, when the wiring 96 disposed on one main surface of the electrode 98 which is not opposed to the Sic semiconductor substrate 99 is formed, the wiring 96 cannot be directly connected to the electrode 98 in the region where the carbon 97 is interposed. contact. Therefore, the carbon 97 deteriorates the adhesion of the electrode (10) to the wiring %. Therefore, it is a cause of an unfavorable situation in which the wiring 96 is peeled off from the electrode 98, and it is possible to affect the durability or electrical characteristics of the Sic semiconductor device 99. In order to solve the above problems, for example, a structure in which an alloy layer of Νι and Si is formed on the SiC substrate 95 is considered in the publication of Japanese Laid-Open Patent Publication No. Hei 7-99169 (hereinafter referred to as "Patent Document 1"). Fig. 36 is a schematic cross-sectional view showing a structure in which an alloy layer of Ni and Si is formed on a sic semiconductor layer. The electronic component 95A shown in Fig. 36 is subjected to heat treatment after forming the Ni si alloy layer 94 on the Sic substrate %, whereby the Ni_Si alloy layer % has a function as an ohmic electrode, which is disclosed in the patent document. Further, in the work of the patent literature, there is also a case where a layered structure in which a Si layer is formed on a SiC substrate 95 and a state layer is formed on the Si layer is subjected to heat treatment to form an ohmic electrode. The carbon 97 shown is formed by reacting a metal constituting the electrode 98 with SiC of the sic semiconductor substrate 99 by heat treatment at the time of forming the electrode 98, whereby c (carbon) is generated as a residue from sic Precipitated on the surface of the electrode 98. Therefore, in order to constitute an ohmic contact, Patent Document 1 discloses that on the surface of the vehicle of the Sic semiconductor substrate 99, an alloy layer 94 which is an alloy of metal (Ni) and Si is formed, thereby forming a picture 147784. .doc 201137975 shows the electronic component 95A. Patent Document 1 also discloses the following:

SiC半導體基板99之主表面上形成Si層,且於si層上積層沁 層,之後實施熱處理,藉此形成如圖36所示之電子元件 95A 〇 通常,和SiC與金屬之矽化物化(合金化)反應溫度相 比,Si與金屬之矽化物化反應溫度較低。在SiC與金屬進 行矽化物化反應時,必需切斷Sic之“與(:之鍵結,相對於 此,在Si與金屬進行矽化物化反應時,無需切斷上述之鍵 結’故認為Si與金屬之矽化物化反應所必要之能量較小, 相應地反應溫度變低。 因此’在為了實施熱處理而升高溫度之中途,於結構之 上側(圖36中SiC基板95之上表面側)si與Ni進行反應。在Si 與Nl進行反應而完成矽化物化後’ SiC之Si與Ni發生反應 之石夕化物化受到阻礙’故可抑制由SiC之以與⑹反應而產 生C °因此’ C基本不會到達結構之最上面(圖36中Ni_Si合 金層之最上面)。所以’認為C基本不會於歐姆電極之表面 析出。 先前技術文獻 專利文獻 專利文獻1:日本專利特開平7_99169號公報 【發明内容】 發明所欲解決之問題 然而’如專利文獻i所揭示,於siC基板95上形成Ni-Si 合金層94(含有Si之合金層)或以層時,siC半導體層與含有 147784.doc 201137975An Si layer is formed on the main surface of the SiC semiconductor substrate 99, and a tantalum layer is laminated on the Si layer, followed by heat treatment, whereby an electronic component 95A as shown in FIG. 36 is formed, and SiC and metal are eutectic (alloyed). The reaction temperature of Si and metal is lower than that of the metal. When the SiC and the metal are subjected to a cerium formation reaction, it is necessary to cut off the "S:" bond, and in contrast, when the Si and the metal undergo a bismuthation reaction, it is not necessary to cut the above-mentioned bond. The energy necessary for the hydration reaction is small, and accordingly the reaction temperature becomes low. Therefore, 'in the middle of raising the temperature for the purpose of heat treatment, on the upper side of the structure (the upper surface side of the SiC substrate 95 in Fig. 36) si and Ni The reaction is carried out. After Si and Nl are reacted to complete the hydration, the Si chemistry of the reaction between Si and Ni is inhibited, so that the reaction between SiC and (6) can be suppressed to produce C °. Therefore, C is not substantially It is the uppermost part of the structure (the uppermost layer of the Ni_Si alloy layer in Fig. 36). Therefore, it is considered that C does not substantially precipitate on the surface of the ohmic electrode. PRIOR ART DOCUMENT Patent Document Patent Document 1: Japanese Patent Laid-Open No. Hei 7-99169 The problem to be solved by the invention However, as disclosed in Patent Document i, when a Ni-Si alloy layer 94 (an alloy layer containing Si) or a layer is formed on the siC substrate 95, the siC semiconductor layer and There 147784.doc 201137975

Si之層會直接接觸。於該情形時,本發明者發現存在如下 問題。以下進行說明。 一般而言,Si相對於SiC係作為蕭特基電極而工作。因 此在使Si層或含有Si之合金層與Sic層直接接觸後,存在如 下可能性:Si並不維持原樣地合金化,而是以與SiC接觸 之狀態殘存。於該情形時,由於該部分係作為蕭特基電極 而作用,故認為會對碳化矽半導體裝置之電氣特性造成影 響。 ’、/ 例如若Si與形成合金層之Ni或存在於以層上之Ni層之Ni 發生完全反應而形成合金(矽化物),則有可能以與以〇並不 形成蕭特基電極,而是如專利文獻i所揭示具有作為良好 歐姆電極之功能。然而,例如在8丨之量相對於可發生2應 之Ni之量而過剩時,或者在因製程條件之不均而導致局部 性地存在有si濃度較高之區域等情形時,未反應之si會析 出。於專利文獻1所揭示之結構中,該已析出之si有可能 與Sic直接接觸而如上所述形成蕭特基接觸。 本發明係鑒於以上問題而完成者,其目的在於提供一種 具有不會產生蕭特基接觸之藉由抑制碳之析出而提高佈線 密著性之歐姆電極的碳化矽半導體裝置及其製造方法。 解決問題之技術手段 本發明之碳化矽半導體裝置之製造方法係具有歐姆電極 之碳化矽半導體裝置之製造方法。該製造方法包括以下步 驟:形成包含碳化矽之SiC層;於上述Sic層之一個主表面 上,形成包含一種第丨金屬元素且不含有碳原子之第丨金屬 147784.doc -6 · 201137975 層;於上述第1金屬層之與上述sic層對向之表面為相反側 之表面上,形成包含矽(Si)且不含有碳原子之si層;及為 了形成歐姆電極,對上述Sic層、上述約金屬層、及上述 Si層進行熱處理。 根據上述方法,在sic層與Si層之間形成有第丨金屬層 後,可抑制藉由未反應之Si與Sic層直接接觸而形成蕭特 基接觸。纟中,構成P金屬元素之金屬宜為一種。其原 因在於,如以下詳細描述,例如若以2種金屬構成第丨金屬 層,則於熱處理之步驟中,難以形成符合目標之反應狀 態,即,最初視條件,Si與2種中之一種金屬發生二元反 應,其後視條件,Si與2種金屬發生三元反應等。 進而,在SiC層與Si層之間形成第1金屬層後,進行熱處 理之步驟,藉此於進行該熱處理之步驟之升溫時,首先第 1金屬層會與Si層優先發生反應而進行合金化(矽化物化卜 其係起因於如上所述Si與金屬層之反應溫度低於sic與金 屬層之反應溫度。在該反應中形成第丨金屬層之第丨金屬元 素完全消耗後,SiC與第1金屬層之反應受到阻礙。又,即 便在第1金屬層之金屬元素與⑴層反應後仍殘存之情形 時,於第1金屬層之上部(與SiC層相反側之表面層)亦會形 成由Si層之Si與金屬層之金屬原子合金化後(矽化物化後) 之層。因此,可抑制由SiC層與第1金屬層反應所產生之作 為殘渣之C析出於電極(由第丨金屬層與以層發生反應所形 成之歐姆電極)之表面(由Si層之Si與金屬層之金屬原子合 金化後(矽化物化後)之層之表面)的現象。因此,可抑制由 147784.doc 201137975 於連接有佈線之歐姆電極之表面層上析出c而導致的佈線 密著性之惡化。 再者,此處所謂「不含有碳原子」之第丨金屬層(或si 層)係私貫質上不含有碳原子、或碳原子之濃度以原子 數計已成為1%以下之第】金屬層或81層。 再者,本發明之碳化矽半導體裝置之製造方法更好的是 進而包括以下步驟:於進行上述熱處理之步驟之前,在上 述Si層之與上述第i金屬層對向之表面為相反側之表面 上,形成包含一種第2金屬元素且不含有碳原子之第2金屬 層。 藉此,根據進行熱處理之步驟之條件,於已形成之歐姆 電極之表面(與SiC層對向之面為相反側之表面),可形成使 包含構成第2金屬層之金屬之層殘存的層,或者可形成高 濃度地含有構成第2金屬層之金屬的層。因此,在相對於 該種歐姆電極之表面而連接佈線後,與相對於完全石夕化物 化之歐姆電極之表面層而連接佈線之情形相比,可使佈線 與歐姆電極更良好地密著。,可進__步提高佈線之密著 陡再者,此處所謂表面層,例如係指自歐姆電極之表面 (與SiC層對向之面為相反側之表面)起算的lOnm以内之區 域,且該區域中不含有碳原子為宜。 又,若存在第2金屬層,則可降低伴隨Sic與第i金屬層 反應而產生之碳(C)析出於上述第2金屬層之表面層的可能 性。 又,本發明之碳化矽半導體裝置之製造方法係具有歐姆 147784.doc 201137975 電極之碳化矽半導體裝置之製造方法。該製造方法包括以 下步驟:形成包含碳化矽之SiC層;於上述SiC層之一個主 表面上,形成包含一種第1金屬元素且不含有碳原子之第1 金屬層;於上述第1金屬層之與上述SiC層對向之表面為相 反侧之表面上,形成包含矽(Si)及上述一種第i金屬元素且 不含有碳原子之Si金屬層;及為了形成歐姆電極,對上述 SiC層、述第1金屬層、及上述8丨金屬層進行熱處理。 根據上述方法,在SiC層與Si金屬層之間形成有第1金屬 層後’可抑制因Si金屬層中含有之未反應之si與SiC層直接 接觸而形成蕭特基接觸。 進而,在SiC層與Si金屬層之間形成第1金屬層後,執行 熱處理之步驟,藉此於該熱處理之步驟中之升溫時,第1 金屬層會先與Si金屬層中含有之Si優先發生反應而進行合 金化(矽化物化)。在該反應中形成第1金屬層之第1金屬元 素元全消耗後,S i C與第1金屬層之反應受到阻礙。又,即 便在第1金屬層之金屬元素與Si金屬層之Si反應後仍殘存之 情形時’於第1金屬層之上部(Si金屬層側之表面層)亦會形 成由Si金屬層之Si與金屬層之金屬原子合金化後(矽化物化 後)之層。因此’可抑制由SiC層與第1金屬層反應所產生 之作為殘渣之C析出於電極(第1金屬層與Si金屬層發生反 應所形成之歐姆電極)之表面的現象。因此,可抑制由於 連接有佈線之歐姆電極之表面層上析出碳(C)而導致的佈 線密著性之惡化。 本發明之碳化矽半導體裝置之製造方法更好的是進而包 147784.doc 201137975 括以下步驟:在如上所述形成Si金屬層之情形時,亦於進 行熱處理之步驟之前,在上述^金屬層之與上述第1金屬 層對向之表面為相反侧之表面上’形成包含一種第2金屬 元素且不含有碳原子之第2金屬層。 藉此’根據進行熱處理之步驟之條件,於已形成之歐姆 電極之表面(與SiC層對向之面為相反側之表面),可形成使 包含構成第2金屬層之金屬之層殘存的層,或者可形成高 濃度地含有構成第2金屬層之金屬的層。因此,在相對於 該種歐姆電極之表面而連接佈線後,與相對於完全矽化物 化之歐姆電極之表面層而連接佈線之情形相比,可使佈線 與歐姆電極更良好地密著。即,可進一步提高佈線之密著 性。 本發明之碳化矽半導體裝置之製造方法中.,於進行熱處 理之步驟中,在上述SiC層之一個主表面上,亦可形成包 含一種箄1金屬元素與矽(Si)之合金且含有碳原子之含碳矽 化物層。此處’關於SiC層之SiC,亦與第1金屬層接觸, 故透過提高熱處理之溫度而與第1金屬層發生反應從而引 起矽化物化》如此一來’在SiC與第1金屬層反應所成之矽 化物層上,成為含有來自SiC之C之狀態。其結果形成含有 碳原子之含碳矽化物層。 然而’因上述矽化物化而隨附產生之C若不呈現於積層 結構之表面層(此處為所形成之歐姆電極之最表面),則在 將佈線連接於歐姆電極之表面層上之方面不會有障礙。因 此,於SiC層之一個主表面上,亦可形成包含一種第〗金屬 147784.doc •10· 201137975 元素與Si之合金且含有碳原子之含碳矽化物層。 為了形成良好的歐姆接觸,第1金屬元素較好的是選自 由鎳(Ni)、鈦(Ti)、鋁(A1)、鉑(Pt)、鎢(W)、及鈀(Pd)所組 成之群中之一種元素。又,第2金屬元素較好的是選自由 由鈦(Ti)、鋁(A1)、及鉻(Cr)所組成之群中之一種元素。藉 由使第2金屬元素為如上所述之元素而可確實提高歐姆電 極與佈線之密著性。 本發明之碳化矽半導體裝置係可使用上述本發明之碳化 矽半導體裝置之製造方法進行製造者,其包括:包含碳化 矽之SiC層;及矽化物層,其配置於上述Sic層之一個主表 面上’包含一種第丨金屬元素與矽(si)之合金,且在與上述 SiC層對向之表面為相反側之表面層不含有碳原子。而 且,上述SiC層與上述石夕化物層係進行歐姆接觸。 藉此,於成為歐姆電極之矽化物層之表面層不含有碳原 子,故在將佈線連接於該矽化物層之表面上時,可防止由 於該碳原子之存在而導致矽化物層(歐姆電極)與佈線之密 著性劣化。The layer of Si will be in direct contact. In this case, the inventors found that the following problems exist. The following is explained. In general, Si operates as a Schottky electrode with respect to the SiC system. Therefore, after the Si layer or the alloy layer containing Si is directly contacted with the Sic layer, there is a possibility that Si does not remain alloyed as it is, but remains in contact with SiC. In this case, since this portion functions as a Schottky electrode, it is considered to have an influence on the electrical characteristics of the tantalum carbide semiconductor device. ', / For example, if Si forms an alloy (deuteride) with Ni which forms an alloy layer or Ni which exists in a Ni layer on a layer, it is possible that the Schottky electrode is not formed by 〇 and It has a function as a good ohmic electrode as disclosed in Patent Document i. However, for example, when the amount of 8 相对 is excessive with respect to the amount of Ni which may occur 2, or when there is a localized region having a high concentration of si due to uneven process conditions, unreacted Si will precipitate. In the structure disclosed in Patent Document 1, it is possible that the precipitated si is in direct contact with Sic to form a Schottky contact as described above. The present invention has been made in view of the above problems, and an object thereof is to provide a tantalum carbide semiconductor device having an ohmic electrode which can improve wiring adhesion by suppressing precipitation of carbon without causing Schottky contact, and a method of manufacturing the same. Means for Solving the Problem A method of manufacturing a niobium carbide semiconductor device of the present invention is a method of manufacturing a niobium carbide semiconductor device having an ohmic electrode. The manufacturing method includes the steps of: forming a SiC layer comprising tantalum carbide; forming a layer of a second metal 147784.doc -6 · 201137975 comprising a second metal element and containing no carbon atoms on one major surface of the Sic layer; Forming a Si layer containing germanium (Si) and not containing carbon atoms on a surface of the first metal layer opposite to the surface opposite to the sic layer; and forming an ohmic electrode to the Sic layer, the above-mentioned The metal layer and the Si layer are heat treated. According to the above method, after the second metal layer is formed between the sic layer and the Si layer, the Schottky contact can be suppressed by direct contact of the unreacted Si with the Sic layer. In the crucible, the metal constituting the P metal element is preferably one. The reason is that, as described in detail below, for example, if the second metal layer is composed of two kinds of metals, it is difficult to form a reaction state in accordance with the target in the heat treatment step, that is, the first condition, Si and one of the two metals. A binary reaction occurs, and after that, Si reacts with two metals in a ternary reaction. Further, after the first metal layer is formed between the SiC layer and the Si layer, a heat treatment step is performed, whereby when the temperature is raised in the step of performing the heat treatment, first, the first metal layer is preferentially reacted with the Si layer to be alloyed. The bismuth compound is caused by the reaction temperature of Si and the metal layer being lower than the reaction temperature of the sic and the metal layer as described above. After the cerium metal element forming the second metal layer in the reaction is completely consumed, SiC and the first The reaction of the metal layer is hindered. Further, even when the metal element of the first metal layer remains after the reaction with the layer (1), the upper portion of the first metal layer (the surface layer opposite to the SiC layer) is formed. a layer in which Si of the Si layer is alloyed with a metal atom of the metal layer (after mashing). Therefore, it is possible to suppress precipitation of C as a residue generated by the reaction between the SiC layer and the first metal layer (by the second metal layer) The surface of the ohmic electrode formed by the reaction with the layer (the surface of the layer after the Si of the Si layer is alloyed with the metal atom of the metal layer (after mashing)). Therefore, it can be suppressed by 147784.doc 2011379 75. The wiring adhesion is deteriorated by the precipitation of c on the surface layer of the ohmic electrode to which the wiring is connected. Further, the second metal layer (or the Si layer) which is "having no carbon atoms" is a private quality. The metal layer or the 81 layer which does not contain a carbon atom or has a carbon atom concentration of 1% or less in terms of the number of atoms. Further, the method for producing a niobium carbide semiconductor device of the present invention further preferably includes the following steps. Before the step of performing the heat treatment, a second metal layer containing a second metal element and containing no carbon atoms is formed on the surface of the Si layer opposite to the surface opposite to the surface of the i-th metal layer. Therefore, depending on the conditions of the step of performing the heat treatment, a layer remaining on the layer including the metal constituting the second metal layer may be formed on the surface of the formed ohmic electrode (the surface opposite to the surface facing the SiC layer). Alternatively, a layer containing a metal constituting the second metal layer may be formed at a high concentration. Therefore, after the wiring is connected to the surface of the ohmic electrode, the ohmic electrode is formed with respect to the complete cerium. Compared with the case where the surface layer is connected to the wiring, the wiring can be more closely adhered to the ohmic electrode. The adhesion of the wiring can be improved in a stepwise manner. Here, the surface layer is, for example, a self-ohmic electrode. It is preferable that the surface (the surface opposite to the surface facing the SiC layer) is within 1 nm, and the region does not contain carbon atoms. Further, if the second metal layer is present, the Sic and the i-th can be reduced. The carbon (C) generated by the reaction of the metal layer may be deposited on the surface layer of the second metal layer. Further, the method for manufacturing the niobium carbide semiconductor device of the present invention has a niobium 147784.doc 201137975 electrode of a niobium carbide semiconductor device. a manufacturing method comprising the steps of: forming a SiC layer comprising tantalum carbide; forming a first metal layer containing a first metal element and not containing a carbon atom on one main surface of the SiC layer; Forming a Si metal layer containing germanium (Si) and the above-mentioned one i-th metal element and containing no carbon atoms on a surface of the metal layer opposite to the surface opposite to the SiC layer; and forming an ohmic layer , The above SiC layer, said first metal layer, and said metal layer is heat-treated 8 Shu. According to the above method, after the first metal layer is formed between the SiC layer and the Si metal layer, the unreacted Si contained in the Si metal layer can be prevented from directly contacting the SiC layer to form a Schottky contact. Further, after the first metal layer is formed between the SiC layer and the Si metal layer, a heat treatment step is performed, whereby the first metal layer is first and the Si contained in the Si metal layer is preferentially heated during the heat treatment step. The reaction is carried out to carry out alloying (deuteration). After the first metal element forming the first metal layer in the reaction is completely consumed, the reaction between S i C and the first metal layer is inhibited. Further, even when the metal element of the first metal layer remains after reacting with the Si of the Si metal layer, the Si portion of the Si metal layer is formed on the upper portion of the first metal layer (the surface layer on the Si metal layer side). A layer that is alloyed with a metal atom of the metal layer (after mashing). Therefore, it is possible to suppress the phenomenon that C, which is a residue generated by the reaction between the SiC layer and the first metal layer, is deposited on the surface of the electrode (the ohmic electrode formed by the reaction between the first metal layer and the Si metal layer). Therefore, it is possible to suppress the deterioration of the wiring adhesion due to the precipitation of carbon (C) on the surface layer of the ohmic electrode to which the wiring is connected. Preferably, the method for producing a tantalum carbide semiconductor device of the present invention further comprises the following steps: in the case of forming a Si metal layer as described above, and also before the step of performing the heat treatment, in the above metal layer On the surface opposite to the surface facing the first metal layer, a second metal layer containing a second metal element and containing no carbon atoms is formed. Thereby, a layer remaining on the layer including the metal constituting the second metal layer can be formed on the surface of the formed ohmic electrode (the surface opposite to the surface opposite to the SiC layer) according to the conditions of the step of performing the heat treatment Alternatively, a layer containing a metal constituting the second metal layer in a high concentration may be formed. Therefore, after the wiring is connected to the surface of the ohmic electrode, the wiring can be more closely adhered to the ohmic electrode than when the wiring is connected to the surface layer of the completely erbium-doped ohmic electrode. That is, the adhesion of the wiring can be further improved. In the method for producing a tantalum carbide semiconductor device according to the present invention, in the step of performing heat treatment, an alloy containing a metal element of lanthanum and cerium (Si) may be formed on one main surface of the SiC layer and contains carbon atoms. A carbon-containing telluride layer. Here, the SiC of the SiC layer is also in contact with the first metal layer, so that it reacts with the first metal layer by increasing the temperature of the heat treatment to cause deuteration. Thus, the reaction between SiC and the first metal layer is achieved. On the telluride layer, it is in a state containing C derived from SiC. As a result, a carbon-containing telluride layer containing carbon atoms is formed. However, if the C produced by the above hydration is not present on the surface layer of the laminate structure (here, the outermost surface of the formed ohmic electrode), the wiring is not connected to the surface layer of the ohmic electrode. There will be obstacles. Therefore, on one main surface of the SiC layer, a carbon-containing telluride layer containing a metal of 147784.doc •10·201137975 element and Si and containing carbon atoms may be formed. In order to form a good ohmic contact, the first metal element is preferably selected from the group consisting of nickel (Ni), titanium (Ti), aluminum (A1), platinum (Pt), tungsten (W), and palladium (Pd). An element of a group. Further, the second metal element is preferably one selected from the group consisting of titanium (Ti), aluminum (A1), and chromium (Cr). By making the second metal element an element as described above, it is possible to surely improve the adhesion between the ohmic electrode and the wiring. The tantalum carbide semiconductor device of the present invention can be manufactured by using the above-described method for producing a tantalum carbide semiconductor device of the present invention, comprising: a SiC layer containing tantalum carbide; and a telluride layer disposed on one main surface of the Sic layer The upper layer contains an alloy of a cerium metal element and cerium (si), and the surface layer on the opposite side to the surface opposite to the SiC layer does not contain carbon atoms. Further, the SiC layer is in ohmic contact with the above-mentioned layer. Thereby, the surface layer of the telluride layer which becomes the ohmic electrode does not contain carbon atoms, so that when the wiring is connected to the surface of the germanide layer, the vaporized layer (ohmic electrode) due to the presence of the carbon atom can be prevented. ) The adhesion to the wiring deteriorates.

層之表面起算的10 nm以内之區域。The area within 10 nm from the surface of the layer.

碳矽化物層係配置於SiC層之— 及不含有碳原子之矽化物層。含 層之一個主表面上,其包含一種 147784.doc • 11 · 201137975 第1金屬元素與矽(Si)之合金,且含有碳原子。不含有碳原 子之石夕化物層係配置於含碳;s夕化物層之與Sic層對向之表 面為相反側之主表面上,其包含一種第丨金屬元素與“之 合金,且在與含碳矽化物層對向之表面為相反側之表面層 不含有碳原子。SiC層與含碳矽化物層係進行歐姆接觸。 藉此,在成為歐姆電極之與含碳矽化物層連接之不含有 碳原子之矽化物層的表面層不含有碳原子,故在將佈線連 接於该矽化物層之表面上時,可防止由於碳原子之存在而 導致矽化物層與佈線之密著性劣化。 上述碳化矽半導體裝置可進而包含上部矽化物層,其係 形成於矽化物層之表面層上,包含一種第2金屬元素與Si 之合金,且在與上述矽化物層對向之表面為相反侧之表面 層不含有碳原子。 於該情形時,可與構成上述矽化物層之第丨金屬元素獨 立地選擇第2金屬元素,故在形成將佈線連接於上部矽化 物層之表面之構成時,可增大將能提高與佈線之密著性之 金屬元素選作第2金屬元素時之選擇的自由度。 於本發明之碳化矽半導體裝置中,第1金屬元素較好的 是選自由鎳、鈦、鋁、鉑、鎢、及鈀所組成之群中之一種 元素。於該情形時,可實現SiC層與矽化物層之良好的歐 姆接觸。又,第2金屬元素較好的是選自由鈦、鋁、及鉻 所組成之群中之一種元素。藉由使第2金屬元素為如上所 述之元素而可確實提高上部矽化物層(歐姆電極)與佈線之 密著性。 147784.doc •12- 201137975 發明之效果 種不會產生蕭特基接觸地藉由抑 密著性的碳化矽半導體裝置及其 根據本發明,可提供— 制碳之析出而提高佈線之 製造方法。 【實施方式】 ’—邊說明本發明之實施形態。再 對於發揮相同功能之部位附以相同 以下,一邊參照圖式 者’於各實施形態中 之參照符號’在無特別需要時不重複進行其說明。 (實施形態1) 圖1係表不用以形成本發明實施形態丨之碳化矽半導體裝 置之歐姆電極之積層結構的概略剖面圖。又,圖2係表示 本發明實施形態1之碳化矽半導體裝置之形成順序的流程 圖。此處,圖1所示之積層結構10A係表示用以形成本發明 實施形態1之碳化矽半導體裝置之於進行熱處理前的積層 結構。 如圖1之積層結構10A所示,為了形成本發明實施形態1 之碳化矽半導體裝置之歐姆電極,首先於Sic基板1〇之一 個主表面上形成包含碳化矽之SiC層11。然後,於siC層11 之一個主表面上,形成包含一種第1金屬元素之第1金屬層 12 ^又,在第1金屬層之與SiC層11對向之表面為相反側之 表面上(圖1中之上側),形成包含Si之Si層13。 根據該種構成,在對積層結構10 A進行熱處理時,構成 第1金屬層12之第1金屬元素會與反應溫度相對低於SiC層 11之Si的Si層13之Si優先發生反應而得以合金化(矽化物 147784.doc • 13- 201137975 化)。因此,透過熱處理,“層13與第丨金屬層12發生反應 而矽化物化。進而,隨著熱處理溫度之上升,在加熱溫度 達到第1金屬層12與SiC層11之Si之反應溫度後,由第1金 屬元素與SiC層11之Si發生反應而開始矽化物化。此時, 由於SiC之Si與第1金屬元素發生反應,故所殘存之碳(c)會 作為殘渣而滞留。若其於例如積層結構1〇A之表面層(圖1 中之最上面)析出,則在將佈線連接於積層結構丨〇A之表面 層上時’構成佈線之材質與積層結構1〇八之表面層之密著 性會惡化,從而引起佈線剝離等現象。然而,在所產生之 c於積層結構10A之表面層析出之前,由於先前剛開始反 應的Si層13之Si與第1金屬元素之反應而使第丨金屬元素全 部消耗後’用以與SiC層11之SiC發生反應之第1金屬元素 便不存在。因此,可抑制SiC與第1金屬元素發生反應,從 而可抑制影響佈線密著性之C之產生。 又,對於積層結構10A中之第1金屬層丨2之上侧區域(與 Si層13接近之區域),為了至少先與以層13之以進行矽化物 化,首先形成矽化物層。因此,由第i金屬層丨2之第i金屬 元素與SiC層11之SiC發生反應所形成之矽化物層到達積層 結構10A中之第1金屬層12之上端部的可能性較低。因此, 由一部分SiC與第1金屬層12反應之結果所產生之c即便滞 留於積層結構10 A之内部,該c到達積層結構1 〇 a之表面層 之可能性亦較低。由此亦可謂:藉由使第1金屬層12介於 SiC層11與Si層13之間而可抑制積層結構10A之表面層之c 的析出。 147784.doc 201137975 又,如上所述’ Si相對於SiC係作為蕭特基電極而工 作。因此使Si層或含有Si之合金層直接接觸於siC之層為不 佳。然而若如積層結構10A般在SiC層11與Si層13之間配置 第1金屬層12 ’則可抑制SiC層11之SiC與Si層13之Si直接接 觸而形成蕭特基接觸。 即’如上所述,對於直接接觸之SiC層11與Si層13,第1 金屬層12之第1金屬元素會與反應溫度較低之8丨層丨3之si優 先發生反應而得以合金化(矽化物化)。在Si層13之Si超越 進行熱處理前的積層結構10A中之第1金屬層12之區域而到 達SiC層11之前,若Si層13與第1金屬層12之反應完成後全 部得以石夕化物化’則無助於石夕化物化之Si與SiC層11之SiC 直接接觸而形成蕭特基接觸之可能性變得極低。根據以上 所述,藉由介在有第1金屬層12而可降低形成蕭特基接觸 之可能性。 其中,例如在使構成第1金屬層12之第1金屬元素進行二 元素配置時’由對其進行熱處理而引起之與Si之反應成為 三元素之反應。例如,若在含有金屬A、金屬B及Si該三元 素之狀態下進行加熱,則在金屬A與金屬B之反應溫度接 近於金屬B與Si之反應溫度時,成為金屬A、金屬B、及Si 之三元反應。又,於初始狀態下例如Si僅與金屬A直接接 觸’從而導致金屬B與Si並未直接接觸。因此,初始狀態 下為金屬A與Si之二元反應,但隨著反應之進展,認為反 應之形態變化為金屬A、金屬B及Si之三元反應。對於該種 反應’例如使用標記等予測反應,難以形成符合目標之反 147784.doc -15· 201137975 應狀態。 根據以上事由’構成第1金屬層12之第1金屬元素之金屬 較好的是一種。具體而言’較好的是選自由鎳、鈦、鋁、 翻、鎢、及鈀所組成之群中之一種元素。藉此,根據SiC 及Si之功函數之相關關係’在對該積層結構1〇A進行熱處 理時可形成良好之歐姆電極。 此處’使用圖2來說明本發明實施形態1之碳化石夕半導體 裝置之製造方法。首先,實施準備基板之步驟(sl〇)。具 體而言,準備形成有碳化矽半導體裝置之基板、即、圖1 所示之SiC基板10。例如n型之SiC晶圓可用作sic基板1〇, 亦可用作p型之SiC晶圓。 其次’實施形成SiC層之步驟(S2〇)。具體而言’如圖i 之積層結構10A中之SiC層11般,於siC基板10之一個主表 面上,形成包含碳化矽之SiC層11。 形成SiC層之步驟(S20)例如係為了藉由磊晶成長而使得 用以形成積層結構10A之主表面之結晶面之方向一致,以 確保所形成之半導體裝置之良好的電氣特性,且為了填補The carbon telluride layer is disposed on the SiC layer - and the vaporized layer containing no carbon atoms. On one of the major surfaces of the layer, it contains a 147784.doc • 11 · 201137975 alloy of the first metal element and bismuth (Si) and contains carbon atoms. The layer of the lithium-free layer containing no carbon atoms is disposed on the main surface of the carbon-containing s-slung layer opposite to the surface opposite to the Sic layer, and comprises a bismuth metal element and an alloy thereof. The surface layer on the opposite side of the surface opposite to the carbon-containing telluride layer does not contain carbon atoms. The SiC layer is in ohmic contact with the carbon-containing telluride layer. Thereby, it is connected to the carbon-containing telluride layer as the ohmic electrode. Since the surface layer of the telluride layer containing a carbon atom does not contain a carbon atom, when the wiring is connected to the surface of the vaporized layer, the adhesion of the vaporized layer to the wiring due to the presence of carbon atoms can be prevented from deteriorating. The tantalum carbide semiconductor device may further include an upper germanide layer formed on the surface layer of the germanide layer, including an alloy of the second metal element and Si, and opposite to the surface opposite to the germanide layer The surface layer does not contain carbon atoms. In this case, the second metal element can be selected independently of the second metal element constituting the telluride layer, so that the wiring is connected to the upper germanide layer. In the case of the surface structure, the degree of freedom in selecting a metal element capable of improving the adhesion to the wiring as the second metal element can be increased. In the silicon carbide semiconductor device of the present invention, the first metal element is preferable. It is one element selected from the group consisting of nickel, titanium, aluminum, platinum, tungsten, and palladium. In this case, good ohmic contact between the SiC layer and the telluride layer can be achieved. Preferably, it is one element selected from the group consisting of titanium, aluminum, and chromium. By making the second metal element an element as described above, it is possible to surely improve the adhesion of the upper vapor layer (ohmic electrode) to the wiring. 147784.doc • 12-201137975 The effect of the invention is that the Schottky-based semiconductor device does not generate Schottky contact, and according to the present invention, the deposition of carbon can be provided to improve the manufacture of wiring. [Embodiment] [Embodiment] The embodiments of the present invention will be described with reference to the embodiments of the present invention, and the same reference numerals will be used to refer to the figures in the embodiments. (Embodiment 1) FIG. 1 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode in which a tantalum carbide semiconductor device according to an embodiment of the present invention is not formed, and FIG. 2 shows an embodiment of the present invention. A flow chart of the formation sequence of the tantalum carbide semiconductor device of the first embodiment is shown in Fig. 1. The laminated structure 10A shown in Fig. 1 is a laminated structure for forming the tantalum carbide semiconductor device according to the first embodiment of the present invention before heat treatment. As shown in the laminated structure 10A of the first embodiment, in order to form the ohmic electrode of the tantalum carbide semiconductor device according to the first embodiment of the present invention, first, a SiC layer 11 containing tantalum carbide is formed on one main surface of the Sic substrate 1A. Then, in the siC layer 11 On one main surface, a first metal layer 12 including a first metal element is formed, and on the surface opposite to the surface of the first metal layer opposite to the SiC layer 11 (upper side in FIG. 1), A Si layer 13 containing Si is formed. According to this configuration, when the laminated structure 10 A is heat-treated, the first metal element constituting the first metal layer 12 is preferentially reacted with Si of the Si layer 13 having a reaction temperature lower than that of Si of the SiC layer 11 to be alloyed. (矽 147784.doc • 13-201137975). Therefore, by the heat treatment, "the layer 13 reacts with the second metal layer 12 to be deuterated. Further, as the heat treatment temperature rises, after the heating temperature reaches the reaction temperature of the Si of the first metal layer 12 and the SiC layer 11, The first metal element reacts with Si of the SiC layer 11 to start bismuthation. At this time, since Si of SiC reacts with the first metal element, the remaining carbon (c) is retained as a residue. When the surface layer of the laminated structure 1A (the uppermost one in FIG. 1) is precipitated, the material constituting the wiring and the surface layer of the laminated structure are densely connected when the wiring is connected to the surface layer of the laminated structure 丨〇A. The property is deteriorated, causing a phenomenon such as peeling of the wiring, etc. However, before the generated c is chromatographed on the surface of the laminated structure 10A, the reaction of Si of the Si layer 13 which has just started the reaction with the first metal element is caused. After the entire third metal element is consumed, the first metal element that reacts with the SiC of the SiC layer 11 does not exist. Therefore, it is possible to suppress the reaction between SiC and the first metal element, thereby suppressing the influence of the wiring adhesion. Production Further, in the upper layer region of the first metal layer 丨2 in the laminated structure 10A (the region close to the Si layer 13), the bismuth layer is first formed in order to at least first mash the layer with the layer 13. The telluride layer formed by the reaction of the ith metal element of the i-th metal layer 丨2 with the SiC of the SiC layer 11 reaches the upper end portion of the first metal layer 12 of the build-up structure 10A. Therefore, As a result of the reaction of a part of SiC with the first metal layer 12, even if it stays inside the laminated structure 10A, the possibility that the c reaches the surface layer of the laminated structure 1a is also low. The first metal layer 12 is interposed between the SiC layer 11 and the Si layer 13 to suppress precipitation of the surface layer c of the buildup structure 10A. 147784.doc 201137975 Further, as described above, 'Si is relative to the SiC system as Schottky. Therefore, it is not preferable to make the Si layer or the alloy layer containing Si directly contact the layer of siC. However, if the first metal layer 12' is disposed between the SiC layer 11 and the Si layer 13 as in the laminated structure 10A, The SiC of the SiC layer 11 is inhibited from being in direct contact with the Si of the Si layer 13 to form a Schottky That is, as described above, for the SiC layer 11 and the Si layer 13 which are in direct contact, the first metal element of the first metal layer 12 preferentially reacts with the Si of the 8 丨 layer 3 having a lower reaction temperature to be alloyed. After the Si layer 13 reaches the SiC layer 11 beyond the region of the first metal layer 12 in the laminated structure 10A before the heat treatment, if the reaction between the Si layer 13 and the first metal layer 12 is completed, The fact that all of them have been etched into the 'thirsty' does not contribute to the direct contact between the Si of the Sihua compound and the SiC of the SiC layer 11 to form a Schottky contact. According to the above, the possibility of forming the Schottky contact can be reduced by interposing the first metal layer 12. In the case where the first metal element constituting the first metal layer 12 is arranged in a two-element, for example, a reaction in which the reaction with Si is caused by the heat treatment is a three-element reaction. For example, when heating is performed in a state in which the three elements of the metal A, the metal B, and the Si are contained, when the reaction temperature of the metal A and the metal B is close to the reaction temperature of the metal B and Si, the metal A, the metal B, and The ternary reaction of Si. Further, in the initial state, for example, Si is only in direct contact with metal A, resulting in that metal B and Si are not in direct contact. Therefore, in the initial state, the binary reaction of metal A and Si is carried out, but as the reaction progresses, the morphological change of the reaction is considered to be a ternary reaction of metal A, metal B, and Si. For such a reaction, for example, using a pre-measurement reaction such as a label, it is difficult to form a target that meets the target 147784.doc -15·201137975. It is preferable that the metal constituting the first metal element of the first metal layer 12 is one of the above. Specifically, 'it is preferably one element selected from the group consisting of nickel, titanium, aluminum, turned, tungsten, and palladium. Thereby, a good ohmic electrode can be formed by heat-treating the laminated structure 1A according to the correlation of the work functions of SiC and Si. Here, a method of manufacturing the carbonized carbide semiconductor device according to the first embodiment of the present invention will be described with reference to Fig. 2 . First, a step (sl〇) of preparing a substrate is carried out. Specifically, a substrate on which a tantalum carbide semiconductor device is formed, that is, a SiC substrate 10 shown in Fig. 1 is prepared. For example, an n-type SiC wafer can be used as the sic substrate 1 〇, and can also be used as a p-type SiC wafer. Next, the step of forming a SiC layer (S2〇) is carried out. Specifically, the SiC layer 11 containing tantalum carbide is formed on one main surface of the siC substrate 10 as in the SiC layer 11 in the laminated structure 10A of Fig. i. The step of forming the SiC layer (S20) is, for example, to make the directions of the crystal faces for forming the main surface of the laminated structure 10A uniform by epitaxial growth to ensure good electrical characteristics of the formed semiconductor device, and to fill

SiC基板10之厚度而進行。根據所使用之基板、或所形成 之半導體裝置之用途,可形成n型磊晶層,亦可形成p型磊 晶層。 為了形成磊晶層,較好的是使用氣相磊晶成長法,即, 例如向含有構成SiC之5丨及(:之材料氣體即矽烷(SiHj或丙 烧(C^8)中’混入用以具有n型或p型之半導體特性之雜質 源即紹(Α1)或磷(Ρ)等以進行氣相成長。再者,作為用以形 147784.doc -16· 201137975 成P型磊晶層之p型雜質 甲基鋁(TMA),作為用 採用例如氮(N2)氣體。 源’可採用例如 以形成η型磊晶層 —烧(Β2Η6)或三 之η型雜質源,可 然後’實施形成第1金屬層之步驟叫具體而言,盆 係於圖1所示之训層11之-個主表面上,形成包含一種第 1金屬元素且不含有碳原子之^金屬層12的步驟。該第! 金屬層12係於之後進行熱處理時用以形成歐姆電極者。 此處,若第i金屬層12中含有碳原+,則於之後進行熱 處理之步驟中’有可能會於圖⑽示之積層結構ΜΑ之表面 層析出碳原子。gu匕’較好的是P金屬層12中不含有碳 原子。此處所謂不含有碳原子,係指例如碳原子之量以原 子數計為1%。再者,第i金屬層12較好的是例如藉由麟 或真空蒸鏟、離子束蒸鍍或讀法而形成。如上所述,構 成第1金屬層12之金屬元素較好的是選自由錄、鈦、铭、 鉑、鎢、及鈀所組成之群中之一種元素。 繼而,實施形成Si層之步驟(S40)e具體而言,如圖以斤 示,其係在第1金屬層12之與“(:層“對向之表面為相反側 之表面上,形成包含si且不含有碳原子之“層13的步驟。 該8:層13係在之後進行熱處理時用以形成歐姆電極者。“ 層13之形成較好的是例如採用濺鍍法等方法進行。 隨後,實施進行熱處理之步驟(S5〇)。具體而言,其係 為了形成上述歐姆電極而對已形成之圖丨所示之包含第1金 屬層12及Si層13之積層結構10A全體進行加熱處理,將構 成積層結構10A之第1金屬層12與以層13合金化的步驟。 147784.doc -17· 201137975 例如在對圖1所不之積層結構i 〇A進行熱處理而形成歐姆 電極時’作為進行熱處理之環境,較好的是使用例如氬 (Ar)條件之環境,此外亦可使用例如氮等之惰性氣體 之環境。又,進行熱處理之溫度較佳者為8〇〇<t以上且為 iioo°c以下,更佳者為900〇c以上且為l〇5〇ec以下,且宜 進行30秒以上5分以下之時間的加熱。在進行該熱處理 後Sl層13之以與構成第1金屬層12之第1金屬元素得以矽 化物化(合金化)。伴隨該合金化,該部分與sic層Η形成歐 姆接觸’從而形成歐姆電極。 圖3係表示進行熱處理後之作為本發明實施形態丨之歐姆 電極之積層結構的概略剖面圖。又,圖4係表示進行熱處 理後之作為本發明實施形態丨之歐姆電極之另一形態之積 層結構的概略剖面圖。 圖3所示之歐姆電極丨丨a、圖4所示之歐姆電極1 2 a皆係 顯示在對圖1所示之積層結構10A進行熱處理之步驟(s5〇) 後的形態者。例如圖3所示之歐姆電極11A係包含:含碳矽 化物層41,其配置於包含碳化矽之31(:層u之一個主表面 上,包含一種第1金屬元素與Si之合金,且含有碳原子; 及矽化物層42,其配置於上述含碳矽化物層41之與sic層 11對向之表面為相反側之主表面上,包含一種第丨金屬元 素與si之合金,且在與含碳矽化物層41對向之表面為相反 側之表面層不含有碳原子。而且,“(^層u與含碳矽化物 層41係進行歐姆接觸,再者,此處之表面層如上所述,係 指例如自圖3所示之歐姆電極丨丨a之最上層、即矽化物層42 147784.doc -18- 201137975 之與含碳矽化物層41對向之表面為相反側(圖3中之上側)之 表面起算的1 〇 nm以内之區域。 又,例如圖4所示之歐姆電極12A係包含矽化物層42,其 配置於形成於SiC基板10之一個主表面上之包含碳化石夕之 SiC層11的一個主表面上,包含一種第丨金屬元素與si之合 金’且在與SiC層11對向之表面為相反側之表面層不含有 碳原子。而且,SiC層11與矽化物層42係進行歐姆接觸。 在對圖1所示之積層結構10A進行熱處理後,構成第1金 屬層12之一種第1金屬元素首先會與8丨層13之“反應而得以 矽化物化。其原因在於,較SiC而言,第丨金屬元素更會與 Si之間以較低之溫度引起矽化物化之反應。此處,由於構 成為於第1金屬層12、Si層13中皆不含有碳原子,故由此 反應而形成之合金、即矽化物層42不含有碳原子。然而, 隨著熱處理之加熱溫度之上升’加熱溫度會達到使第1金 屬元素與SiC之Si—同矽化物化之溫度。因此,會引起第i 金屬層12與Si層13之Si及SiC層11之Si該兩方之矽化物化。 藉由與Si層13之Si之矽化物化而形成上述不含有碳原子之 矽化物層42,藉由與SiC層11之Si之矽化物化而形成含有 於其過程中剩餘之碳原子的含碳矽化物層41。該矽化物化 一直持續到第1金屬元素全部得以矽化物化為止。然後, 在第1金屬元素已全部矽化物化從而反應完成後,如圖3所 示之歐姆電極11A般’於Sic層11之一個主表面上,形成由 SiC層11之Si與第1金屬元素進行矽化物化所形成之含碳矽 化物層41,而且於含碳矽化物層4丨之與Sic層丨丨對向之表 147784.doc •19- 201137975 面為相反侧之表面上(圖3之上側),形成由8丨層13之8丨與第 1金屬元素進行矽化物化所形成之矽化物層42。 其中,例如熱處理之加熱溫度在達到使第〖金屬元素與 SiC之Si得以矽化物化之溫度前,當第丨金屬元素已全部與 Si層13之Si進行矽化物化之情形時,如圖4所示之歐姆電極 12A般,在SiC層11之與SiC基板1〇對向之表面為相反側之 表面上(圖4之上側),形成由81層1;3之81與第1金屬元素進 行石夕化物化所形成之不含有碳原子的矽化物層42。 即便形成有圖3、圖4之任一構成之歐姆電極,亦會由於 石夕化物層42之存在,故含碳矽化物層4〗及sic層丨丨之碳原 子不會到達歐姆電極Π a、12A之表面層即矽化物層42之 表面層。因此’在使用本發明之歐姆電極之製造方法時, 於所形成之歐姆電極11A、12A之表面層即石夕化物層42之 表面層不會析出碳原子。由此可使連接於矽化物層42之表 面層之佈線的密著度良好。 於I成姆電極後,最後實施形成佈線部之步驟 (S60)。具體而言,其係於歐姆電極之表面層上,即,於 圖3、圖4之歐姆電極11A、12A之矽化物層42之表面層 上’形成用作用以取出圖3及圖4中未圖示之電氣信號之佈 線之金屬層(墊)的步驟。佈線部例如可藉由真空蒸鍍、離 子束蒸鍵、賤鍵等而形成。如上所述,於圖3、圖4中之歐 姆電極11A、12A之矽化物層42之表面層,不存在碳原子 或已析出之碳97(參照圖35)。因此,可使連接於矽化物層 42之表面層之佈線部的密著度良好。 I47784.doc •20- 201137975 (實施形態2) 圖5係表示用以形成本發明實施形態2之碳化石夕半導體裝 置之歐姆電極之積層結構的概略剖面圖。又,圖6係表示 本發明實施形態2之碳化矽半導體裝置之形成順序的流程 圖。此處’圖5所示之積層結構10B係表示用以形成本發明 實施形態2之碳化矽半導體裝置之進行熱處理前的積層結 構。 如圖5之積層結構1 〇B所示,為了形成本發明實施形態2 之石反化妙半導體裝置之歐姆電極而準備的積層結構係具備 與本發明實施形態1之積層結構丨〇 A(參照圖丨)相同的形 態。然而,於積層結構10B中,在Si層13之與第1金屬層12 對向之表面為相反側之表面上(圖5中之上側),形成有包含 一種第2金屬元素碳且不含有碳原子之第2金屬層14。僅於 該所述之點,積層結構10B與積層結構i〇A不同。 進行熱處理步驟之後形成佈線部係於歐姆電極之表面層 進行。例如本發明實施形態1之歐姆電極11A、12八中,形 成有佈線部之表面層係矽化物層42之表面層。然而,於本 發明實施形態2中,於積層結構1〇B之最上層形成有第2金 屬層Μ。因此’根據進行熱處理後之狀況,所形成之歐姆 電極之表面層亦會成為第2金屬層14。因此,若相對於第2 金屬層14而形成佈線部,則與相對於矽化物層而形成佈線 4之If形相比,形成佈線部之金屬元素與歐姆電極表面層 之親和性會更佳,從而可提高佈線部與歐姆電極之密著 147784.doc 201137975 其人說明本發明實施形態2之碳化矽半導體裝置之製 =方法如圖6之流程圖所示,本發明實施形態2之碳化矽 半導體裝置之製造方法係與本發明實施形態1之碳化石夕半 導體裝置之製造方法基本相同。然而,如@6所示,本發 明實施形態2之碳切半導體裝置之製造方法進而包括步 驟(S45) ’其係於形成以層之步驟(S4〇)中形成圖5之§丨層u 後在進行熱處s之步驟(S5〇)之前在㈣Η之與第1金 屬層12對向之表面為相反側之表面上,形成包含一種第2 金屬元素且不含有碳原子之第2金屬層。 此處,若第2金屬層14中含有碳原子,則於之後進行熱 處理之步驟中,碳原子會擴散,由此導致圆5所示之積層 、·口構10B之表面層含有碳原子。因此,較好的是第2金屬層 14中不含有碳原子。此處所謂不含有碳原+,係指例如碳 原子之量以原子數計為1%以下。再者,第2金屬層14亦與 第1金屬層12同樣較好的是例如藉由濺鍍或真空蒸鍍、離 子束蒸鍍或電鍍法而形成。如上所述,構成第丨金屬層之 兀素較好的是選自由鎳、鈦、铭、始、鶴、及把所組成之 群中之一種元素。對於第2金屬層14,亦與第丨金屬層12同 樣較好的是由一種金屬元素所構成。又,構成第2金屬層 14之第2金屬元素較好是選自由鈦、鋁、及鉻所組成之群 中之一種元素》藉此,根據SiC及si之功函數之相關關 係,在對該積層結構10B進行熱處理時可形成良好之歐姆 電極》 圖6之流程圖僅於以上所述之點與圖2之流程圖不同。 147784.doc •22· 201137975 即,圖6之步驟(S10)係與圖2之步驟(S 10)為相同之步驟。 以下,圖 6之步驟(S20)、(S30)、(S40)、(S50)、(S60)各自 亦係與圖2之各步驟為相同之步驟。 其中,進行熱處理之步驟(S50)後之積層結構1〇b之取得 態樣與本發明之實施形態丨不同。圖7係表示進行熱處理後 之作為本發明實施形態2之歐姆電極之積層結構的概略剖 面圖。又’圖8係表示進行熱處理後之作為本發明實施形 態2之歐姆電極之另一形態之積層結構的概略剖面圖。圖$ 係表示進行熱處理後之作為本發明實施形態2之歐姆電極 之又一形態之積層結構的概略剖面圖。 圖7所示之歐姆電極iiB、圖8所示之歐姆電極12B、及 圖9所示之歐姆電極13B皆係表示對圖1所示之積層結構 10A進行熱處理之步驟(S50)後的形態。例如在圖5所示之 Si層13之Si首先與第1金屬層12之第丨金屬元素及第2金屬層 14之第2金屬元素進行石夕化物化之情形時,會形成混合有 Si、第1金屬元素、及第2金屬元素該三元素之矽化物層 43。然而,例如在第1金屬元素全部與以層13發生反應而 得以石夕化物化之前,若第1金屬元素與8丨(:層^之以一同開 始矽化物化’則如圖7所示會形成歐姆電極丨丨b,其係在 SiC層11之與SiC基板1〇對向之表面為相反側之表面上(圖7 之上側),配置有包含第1金屬元素與Si之合金且含有碳原 子之含碳矽化物層41,而且在含碳矽化物層41之與Sic層 11對向之表面為相反侧之主表面上,配置有包含第1金屬 元素、第2金屬元素與Si之合金且在與含碳石夕化物層41對 147784.doc -23- 201137975 向之表面為相反側之表面層不含有碳原子的矽化物層4 3。 其中,例如熱處理之加熱溫度在達到使第〗金屬元素與 S!C之Si得以矽化物化之溫度前,當第丨金屬元素已全部與The thickness of the SiC substrate 10 is performed. An n-type epitaxial layer may be formed depending on the substrate to be used or the use of the formed semiconductor device, and a p-type epitaxial layer may be formed. In order to form the epitaxial layer, it is preferred to use a vapor phase epitaxial growth method, that is, for example, a mixture containing 5 Å which constitutes SiC and (a material gas which is cerane (SiHj or propylene (C^8)) The gas phase growth is carried out by using an impurity source having an n-type or p-type semiconductor property, that is, Α1 or phosphorus, etc. Further, as a shape 147784.doc -16·201137975 into a P-type epitaxial layer a p-type impurity, methyl aluminum (TMA), for example, using a nitrogen (N 2 ) gas. The source ' can be used, for example, to form an n-type epitaxial layer-fired (Β2Η6) or three n-type impurity source, which can then be implemented The step of forming the first metal layer is specifically a step of forming a metal layer 12 containing a first metal element and containing no carbon atoms on a main surface of the training layer 11 shown in FIG. The first metal layer 12 is used to form an ohmic electrode after heat treatment. Here, if the ith metal layer 12 contains carbonogen +, then in the subsequent heat treatment step, it may be shown in FIG. The surface of the laminated structure is chromatographed to emit carbon atoms. gu匕' is preferably not contained in the P metal layer 12. The term "a carbon atom alone" means, for example, that the amount of carbon atoms is 1% in terms of the number of atoms. Further, the i-th metal layer 12 is preferably, for example, by a lining or vacuum shovel, ion beam evaporation. Or a read method is formed. As described above, the metal element constituting the first metal layer 12 is preferably one element selected from the group consisting of titanium, quartz, tungsten, and palladium. The step (S40) of the Si layer is specifically shown in the figure, and is formed on the surface of the first metal layer 12 opposite to the surface of the ("layer" opposite to the surface, and contains Si and does not contain The step of "layer 13 of carbon atoms. The 8: layer 13 is used to form an ohmic electrode when heat treatment is subsequently performed." The formation of the layer 13 is preferably carried out, for example, by a sputtering method or the like. Step (S5). Specifically, in order to form the ohmic electrode, the entire laminated structure 10A including the first metal layer 12 and the Si layer 13 shown in the figure is heat-treated to form a laminate. The step of alloying the first metal layer 12 of the structure 10A with the layer 13. 147 784.doc -17·201137975 For example, when heat treatment is performed on the laminated structure i 〇A of FIG. 1 to form an ohmic electrode, as an environment for performing heat treatment, it is preferable to use an environment such as argon (Ar), and An environment of an inert gas such as nitrogen may be used. Further, the temperature for heat treatment is preferably 8 〇〇 < t or more and iioo ° c or less, more preferably 900 〇 c or more and l 〇 5 〇 ec In the following, it is preferable to perform heating for 30 seconds or more and 5 minutes or less. After the heat treatment, the first layer of the first metal element constituting the first metal layer 12 is mashed (alloyed). Along with this alloying, this portion forms an ohmic contact with the sic layer ’ to form an ohmic electrode. Fig. 3 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode which is an embodiment of the present invention after heat treatment. Further, Fig. 4 is a schematic cross-sectional view showing a laminated structure of another embodiment of the ohmic electrode which is an embodiment of the present invention after the heat treatment. The ohmic electrode 丨丨a shown in Fig. 3 and the ohmic electrode 1 2 a shown in Fig. 4 are all shown in the step (s5〇) of heat-treating the laminated structure 10A shown in Fig. 1. For example, the ohmic electrode 11A shown in FIG. 3 includes a carbon-containing telluride layer 41 disposed on a main surface including a tantalum carbide 31 (on one surface of the layer u), containing an alloy of a first metal element and Si, and containing a carbon atom; and a telluride layer 42 disposed on a main surface of the carbon-containing vapor-deposited layer 41 opposite to the surface opposite to the sic layer 11, comprising an alloy of a third metal element and si, and The surface layer on the opposite side of the surface of the carbon-containing telluride layer 41 does not contain carbon atoms. Further, "(the layer u is in ohmic contact with the carbon-containing telluride layer 41, and further, the surface layer here is as above For example, the uppermost layer of the ohmic electrode 丨丨a shown in FIG. 3, that is, the vaporized layer 42 147784.doc -18-201137975 is opposite to the surface opposite to the carbon-containing telluride layer 41 (FIG. 3). Further, for example, the ohmic electrode 12A shown in FIG. 4 includes a vaporized layer 42 which is disposed on one main surface of the SiC substrate 10 and contains carbon fossils. On one major surface of the SiC layer 11, a third metal element is contained The surface layer opposite to the surface of the alloy SiC layer and the SiC layer 11 does not contain carbon atoms. Moreover, the SiC layer 11 is in ohmic contact with the vaporized layer 42. The laminated structure shown in Fig. 1 After the heat treatment of 10A, the first metal element constituting the first metal layer 12 firstly undergoes a "reaction with the 8" layer 13 to be deuterated. The reason is that, compared with SiC, the second metal element is more compatible with Si. The reaction in which the bismuth is formed at a lower temperature. Since the first metal layer 12 and the Si layer 13 do not contain carbon atoms, the alloy formed by the reaction, that is, the vaporized layer 42 is not formed. The carbon atom is contained. However, as the heating temperature of the heat treatment rises, the heating temperature reaches a temperature at which the first metal element and the Si of the SiC are homogenized. Therefore, the Si of the i-th metal layer 12 and the Si layer 13 are caused. The SiC layer 11 and the Si of the SiC layer 11 are decidized. The bismuth compound layer 42 containing no carbon atoms is formed by crystallization of Si with the Si layer 13, and is formed by crystallization of Si with the SiC layer 11. Carbon containing carbon atoms remaining in the process The telluride layer 41. The hydrazine formation continues until all of the first metal elements are deuterated. Then, after the first metal element has been completely deuterated and the reaction is completed, the ohmic electrode 11A as shown in FIG. On one main surface of the layer 11, a carbon-containing vaporized layer 41 formed by mash-forming Si of the SiC layer 11 and the first metal element is formed, and the carbon-containing telluride layer 4 is opposed to the Sic layer. Table 147784.doc • 19-201137975 The surface is on the opposite side (the upper side of Fig. 3), and a vaporized layer 42 formed by mashing the 8 丨 layer 13 and the first metal element is formed. Wherein, for example, when the heating temperature of the heat treatment reaches a temperature at which the metal element and the Si of the SiC are deuterated, when the second metal element has been mashed with the Si of the Si layer 13, as shown in FIG. In the same manner as the ohmic electrode 12A, on the surface of the SiC layer 11 opposite to the surface on which the SiC substrate 1 is opposed (the upper side in FIG. 4), the formation of 81 layers 1; 3 of 81 and the first metal element is performed. The vaporized layer 42 formed of the chemical conversion does not contain carbon atoms. Even if the ohmic electrode having any of the configurations of FIG. 3 and FIG. 4 is formed, the carbon-containing telluride layer 4 and the carbon atoms of the sic layer do not reach the ohmic electrode. The surface layer of 12A is the surface layer of the telluride layer 42. Therefore, when the ohmic electrode manufacturing method of the present invention is used, carbon atoms are not deposited on the surface layer of the surface layer of the formed ohmic electrodes 11A and 12A, i.e., the lithium layer 42. Thereby, the adhesion of the wiring connected to the surface layer of the telluride layer 42 can be made good. After the I-made electrode, the step of forming the wiring portion is finally performed (S60). Specifically, it is formed on the surface layer of the ohmic electrode, that is, on the surface layer of the bismuth layer 42 of the ohmic electrodes 11A, 12A of FIGS. 3 and 4, which is formed to be used for taking out FIG. 3 and FIG. The step of the metal layer (pad) of the wiring of the electrical signal shown. The wiring portion can be formed, for example, by vacuum deposition, ion beam evaporation, 贱 key or the like. As described above, in the surface layer of the telluride layer 42 of the ohmic electrodes 11A and 12A in Figs. 3 and 4, carbon atoms or precipitated carbon 97 are not present (see Fig. 35). Therefore, the adhesion of the wiring portion connected to the surface layer of the telluride layer 42 can be made good. [Embodiment 2] Fig. 5 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode for forming a carbonized carbide semiconductor device according to Embodiment 2 of the present invention. Fig. 6 is a flow chart showing the procedure for forming a tantalum carbide semiconductor device according to Embodiment 2 of the present invention. Here, the laminated structure 10B shown in Fig. 5 shows a laminated structure before heat treatment for forming the tantalum carbide semiconductor device according to the second embodiment of the present invention. As shown in the laminated structure 1 〇B of FIG. 5, the laminated structure prepared to form the ohmic electrode of the stone reversal semiconductor device according to the second embodiment of the present invention includes the laminated structure 丨〇A according to the first embodiment of the present invention (see Figure 丨) the same form. However, in the laminated structure 10B, on the surface opposite to the surface opposite to the first metal layer 12 of the Si layer 13 (the upper side in FIG. 5), a carbon containing a second metal element is formed and contains no carbon. The second metal layer 14 of the atom. Only at the point described, the laminated structure 10B is different from the laminated structure i〇A. The formation of the wiring portion after the heat treatment step is performed on the surface layer of the ohmic electrode. For example, in the ohmic electrodes 11A and 12 of the first embodiment of the present invention, the surface layer of the surface layer-based telluride layer 42 of the wiring portion is formed. However, in the second embodiment of the present invention, the second metal layer 形成 is formed on the uppermost layer of the laminated structure 1B. Therefore, the surface layer of the formed ohmic electrode also becomes the second metal layer 14 depending on the state after the heat treatment. Therefore, when the wiring portion is formed with respect to the second metal layer 14, the affinity between the metal element forming the wiring portion and the surface layer of the ohmic electrode is better than the If shape in which the wiring 4 is formed with respect to the vaporized layer. The adhesion between the wiring portion and the ohmic electrode can be improved. 147784.doc 201137975 The method for manufacturing the silicon carbide semiconductor device according to the second embodiment of the present invention is as shown in the flowchart of FIG. 6. The silicon carbide semiconductor device according to the second embodiment of the present invention is shown. The manufacturing method is basically the same as the method of manufacturing the carbonized carbide semiconductor device according to the first embodiment of the present invention. However, as shown in @6, the method of manufacturing the carbon-cut semiconductor device according to the second embodiment of the present invention further includes the step (S45) of "forming the layer uu of FIG. 5 in the step of forming the layer (S4〇). A second metal layer containing a second metal element and containing no carbon atoms is formed on the surface opposite to the surface on the opposite side of the first metal layer 12 before the step (S5〇) of the hot spot s. When the second metal layer 14 contains a carbon atom, the carbon atom is diffused in the subsequent step of heat treatment, whereby the layer shown by the circle 5 and the surface layer of the mouth structure 10B contain carbon atoms. Therefore, it is preferred that the second metal layer 14 does not contain carbon atoms. The term "carbonogen-free" as used herein means that the amount of carbon atoms is, for example, 1% or less in terms of the number of atoms. Further, the second metal layer 14 is also preferably formed, for example, by sputtering, vacuum deposition, ion beam evaporation or electroplating, similarly to the first metal layer 12. As described above, the halogen constituting the second metal layer is preferably one selected from the group consisting of nickel, titanium, melamine, crane, and handle. The second metal layer 14 is also preferably composed of a metal element as well as the second metal layer 12. Further, the second metal element constituting the second metal layer 14 is preferably one element selected from the group consisting of titanium, aluminum, and chromium. Thus, according to the correlation between the work functions of SiC and si, The laminated structure 10B can form a good ohmic electrode when heat-treated. The flowchart of FIG. 6 differs from the flowchart of FIG. 2 only in the above points. 147784.doc •22· 201137975 That is, the step (S10) of FIG. 6 is the same step as the step (S 10) of FIG. 2 . Hereinafter, the steps (S20), (S30), (S40), (S50), and (S60) of Fig. 6 are also the same steps as the steps of Fig. 2 . Here, the obtained structure of the laminated structure 1b after the step of heat treatment (S50) is different from the embodiment of the present invention. Fig. 7 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode according to a second embodiment of the present invention after heat treatment. Further, Fig. 8 is a schematic cross-sectional view showing a laminated structure of another embodiment of the ohmic electrode which is the second embodiment of the present invention after the heat treatment. Fig. $ is a schematic cross-sectional view showing a laminated structure of still another embodiment of the ohmic electrode according to the second embodiment of the present invention after the heat treatment. The ohmic electrode iiB shown in Fig. 7, the ohmic electrode 12B shown in Fig. 8, and the ohmic electrode 13B shown in Fig. 9 each show a state in which the step (S50) of heat-treating the laminated structure 10A shown in Fig. 1 is performed. For example, when Si of the Si layer 13 shown in FIG. 5 is firstly cleaved with the second metal element of the first metal layer 12 and the second metal element of the second metal layer 14, Si is mixed. The first metal element and the second metal element are the three element germanide layer 43. However, for example, before all of the first metal elements are reacted with the layer 13 to be etched, if the first metal element and the 8 丨 (the layer bismuth is formed together), it is formed as shown in FIG. The ohmic electrode 丨丨b is disposed on a surface of the SiC layer 11 opposite to the surface opposite to the surface of the SiC substrate 1 (on the upper side in FIG. 7), and is provided with an alloy containing the first metal element and Si and contains carbon atoms. The carbon-containing telluride layer 41 is provided on the main surface opposite to the surface of the carbon-containing telluride layer 41 opposed to the Sic layer 11 and is provided with an alloy containing a first metal element, a second metal element, and Si. The surface layer on the opposite side to the surface of the carbonaceous-ceramic layer 41 to 147784.doc -23- 201137975 is a vaporized layer 43 containing no carbon atoms. Among them, for example, the heating temperature of the heat treatment is attained to make the metal Before the temperature of the element and the S!C Si is mashed, when the third metal element has been

Si層13之Si進行矽化物化之情形時,如圖8所示之歐姆電極 12B般,在SiC層11之與SiC基板10對向之表面為相反側之 表面上(圖8之上侧),形成由Si層13之Si、第丨金屬元素、 及第2金屬元素進行矽化物化所形成之不含有碳原子的矽 化物層43。 又,例如當Si層Π之厚度為某程度以上、例如根據金屬 之種類而為圖5所示之第1金屬層12與第2金屬層14之共計 厚度的2倍以上時,亦會引起“層13之“與第丨金屬元素及 第2金屬元素各自獨立地發生矽化物化之反應。再者,此 處所謂厚度,係指對向之主表面間之距離。例如,對於si 層13之Si,於積層結構ι〇Β之下方(圖5之下側),以與第!金 屬元素及第2金屬元素該三元素混合而得以石夕化物化,於 積層結構10B之上方(圖5之上側),由於81層13較厚,故亦 有如下情形:第1金屬元素並不到達,而僅第2金屬元素與 Si層13之Si得以矽化物化。該結果如圖9所示,亦會有形成 進而包含上部矽化物層44之歐姆電極13B之情形,該上部 矽化物層44係包含一種第2金屬元素與以之之合金,且在 與矽化物層43對向之表面為相反側之表面層不含有碳原 子。又,雖未圖示’但亦可形成包含以下構成之歐姆電 極:例如由包含一種第1金屬元素與Si之合金之矽化物 層、及包含一種第2金屬元素與si之合金之石夕化物層積層 147784.doc -24- 201137975 而成的構成。 具有該種構成之歐姆電極係由一種金屬元素與si該二元 素而進行矽化物化,故與由三元素混合所形成矽化物層43 之情形相比,藉由使用標記等可較容易地予測反應。又, 於本發明之實施形態2中,於積層結構1〇B中包含第2金屬 層丨4,故自SiC層丨丨至積層結構最上層之表面層為止的距 離(厚度)大於本發明實施形態丨之情形。因此,可使siC層 11之C到達表面層之可能性更小。 進而,例如當第2金屬元素與Si之反應溫度較第丨金屬元 素與si之反應溫度而大幅升高時,或者當第2金屬層“之 厚度非常厚等之情形時,亦考慮第2金屬元素並未全部與When the Si of the Si layer 13 is decidualized, as in the ohmic electrode 12B shown in FIG. 8, on the surface of the SiC layer 11 opposite to the surface opposite to the SiC substrate 10 (the upper side of FIG. 8), A vaporized layer 43 containing no carbon atoms formed by mash formation of Si, a second metal element, and a second metal element of the Si layer 13 is formed. Further, for example, when the thickness of the Si layer is more than a certain degree, for example, twice the total thickness of the first metal layer 12 and the second metal layer 14 shown in FIG. 5 depending on the type of the metal, it may cause " The layer 13 "reacts with the second metal element and the second metal element independently of the cerium formation. Furthermore, the term "thickness" as used herein refers to the distance between the opposing major surfaces. For example, for the Si layer of the Si layer 13, below the layered structure ι〇Β (the lower side of Figure 5), with the first! The metal element and the second metal element are mixed by the three elements to form a Sihua compound. Above the layered structure 10B (on the upper side of FIG. 5), since the 81 layer 13 is thick, there are cases where the first metal element is not Arrived, and only the second metal element and the Si of the Si layer 13 are deuterated. As a result, as shown in FIG. 9, there is also a case where the ohmic electrode 13B including the upper germanide layer 44 is formed, and the upper germanide layer 44 contains a second metal element and an alloy thereof, and is in the form of a telluride. The surface layer on the opposite side of the layer 43 opposite thereto does not contain carbon atoms. Further, although not shown, an ohmic electrode having a structure in which a telluride layer containing an alloy of a first metal element and Si and an alloy containing a second metal element and an alloy of Si may be formed. The composition of the laminated layer 147784.doc -24- 201137975. Since the ohmic electrode having such a composition is deuterated by a metal element and the two elements of Si, it is easier to predict the reaction by using a label or the like than in the case of forming the vaporized layer 43 by mixing three elements. . Further, in the second embodiment of the present invention, since the second metal layer 丨4 is included in the laminated structure 1B, the distance (thickness) from the SiC layer 丨丨 to the surface layer of the uppermost layer of the laminated structure is larger than the implementation of the present invention. The situation is ambiguous. Therefore, it is less likely that the C of the siC layer 11 reaches the surface layer. Further, for example, when the reaction temperature of the second metal element and Si is greatly increased as compared with the reaction temperature of the second metal element and si, or when the thickness of the second metal layer is extremely thick, the second metal is also considered. The elements are not all

Si發生反應而得以矽化物化之情形。此時例如在圖7至 圖9所示之各歐姆電極之表面層(圖7、圖8中矽化物層之 最上層,圖9中上部石夕化物層料之最上層)上殘存有第2 金属層14(未圖不)。於該情形時關於該歐姆電極,亦以 與第2金屬層14之表面層接觸之方式而形成佈線部。因 此”相董十於矽 匕物層之表面層而形成佈線部之情形相 比,可保持良好之密著性。 本發明之實施形態2僅於以上所述之各點與本發明之實 施形態1不同。即,本發明之實施形態2中,以上未描述之 構成或條件、順序或效果等全部以本發明之實施形態 準。 •一 (實施形態3) 圖10係表Μ以形成本發明實施㈣3之碳切半導 147784.doc 201137975 裝置之歐姆電極之積層結構的概略剖面圖。此處,圖丨0所 示之積層結構1 〇c係表示用以形成本發明實施形態3之碳化 矽半導體裝置之進行熱處理前的積層結構。 如圖10之積層結構10C所示,為了形成本發明實施形態3 之碳化石夕半導體裝置之歐姆電極而準備的積層結構係具備 與本發明實施形態1之積層結構10 Α(參照圖i)相同的形 態。然而,於積層結構10C中,取代積層結構丨〇八之8丨層i 3 而形成包含si及一種第1金屬元素且不含有碳原子的Si金屬 層15。僅於該所述之點,積層結構i 〇c與積層結構丨〇A不 同。 本發明實施形態3之碳化矽半導體裝置之形成順序係以 圖2所示之本發明實施形態1之碳化石夕半導體裝置之形成順 序為準。但是’如上所述,於積層結構1〇c中,取代積層 結構10A之Si層13而形成S.i金屬層15。因此,圖2中之形成 Si層之步驟(S40)成為形成Si金屬層之步驟(S4〇>。 如此,作為Si金屬層15 ’若使含有Si之層中在進行熱處 理刖之初始狀態下含有第1金屬元素,則由於S丨元素存在 於較第1金屬元素更近之位置’故在進行熱處理時,可使 Si與第1金屬元素更迅速地進行石夕化物化。因此,可抑制 第1金屬元素與SiC層11之Si之石夕化物化,且可抑制sic層 11之C析出之現象。再者,作為構成Si金屬層15之第丨金屬 元素,可使用由鎳、鈦、銘、銘、鎢、及叙所組成之群中 之一種元素,但亦可與上述實施形態2中所示之第2金屬元 素同樣地使用選自由鈦、鋁、及鉻所組成之群中之一種元 147784.doc -26- 201137975 素。 本發明之實㈣態3僅於以上所述之各點與本發明之實 施形態1不同。即’本發明之實施形態3中,以上未描述之 構成或條件、順序或效果等全部以本發明之實施形態】為 準〇 (實施形態4) 圖11係表不用以形成本發明實施形態4之碳化矽半導體 裝置之歐姆電極之積層結構的概略剖面圖。此處,圖丨ι所 不之積層結構10 D係表示用以形成本發明實施形態4之碳化 矽半導體裝置之進行熱處理前的積層結構。 如圖11之積層結構10D所示,為了形成本發明實施形態4 之碳化矽半導體裝置之歐姆電極而準備之積層結構係具備 與本發明實施形態2之積層結構丨0B(參照圖丨)相同的形 態。但是,於積層結構10D中,取代積層結構丨0B之Si層 13 ’與本發明實施形態3之積層結構i〇c同樣地形成包含以 及一種第1金屬元素且不含有碳原子的si金屬層15。即, 進而包括以下步驟:於積層結構l〇D中,在si金屬層15之 與第1金屬層12對向之表面為相反側之表面上,形成包含 一種第2金屬元素且不含有碳原子之第2金屬層14。如此, 亦可於Si金屬層15之主表面上形成第2金屬層14。僅於以 上所述之點,積層結構10D與積層結構10B不同。 本發明實施形態4之碳化矽半導體裝置之形成順序係以 圖6所示之本發明實施形態2之碳化矽半導體裝置之形成順 序為準。但是,如上所述,於積層結構1〇D中,取代積層 147784.doc •27· 201137975 結構10B之Si層I3而形成Si金屬層15。因此,圖6之形成Si 層之步驟(S40)成為形成Si金屬層之步驟(s4〇)。 如此,即便於Si金屬層I5之主表面上形成有第2金屬層 14之情形時,只要如Si金屬層15般使含有“之層於進行熱 處理前之初始狀態下含有第丨金屬元素,則由於Si元素存 在於較第1金屬元素更近之位置,故在進行熱處理時,可 使Si與第1金屬元素更迅速地進行矽化物化。 本發明之實施形態4僅於以上所述之各點與本發明之實 施形態2不同。即,本發明之實施形態4中,以上未描述之 構成或條件、順序或效果等全部以本發明之實施形態2為 準。 實施例1 圖12〜圖19係表示利用本發明之實施形態2形成pn二極體 時執行各步驟後之狀態的概略剖面圖。更具體而言,圖a 係表示為了形成pn二極體而執行圖6之步驟(sl〇)之狀態的 概略剖面圖。圖13係表示為了形成卯二極體而執行圖6之 步驟(S20)之狀態的概略剖面圖。圖⑷系表示為了形成 二極體而進行離子注人之狀態的概略剖面圖。0i5係表示 為了形成pn二極體而形成場氧化膜之狀態的概略剖面圖。 圖16係表不為了形成pn二極體而執行圖6之步驟(sc)之狀 態的概略剖面圖。圖17係表示為了形成Pn二極體而執行圖 6之步驟(S50)之狀態的概略剖面圖。圖_表示為了形成 Pn二極體而執行圖6之步驟(S6〇)之狀態的概略剖面圖。圖 B係已完成之pn二極體之概略剖面圖。參照_〜圖19, 147784.doc •28- 201137975 說明應用有本發明之pn二極體之製造方法。 首先,如圖12所示,作為圖6之準備基板之步驟(sl〇), 準備例如η型之SiC基板20。其次,作為圖6之形成Sic層之 步驟(S20) ’於SiC基板20之一個主表面上形成-型蟲晶層 叫參照圖⑺。又,於n•型^層21之與训基板灣j 表面為相反側之表面上,形成〆型磊晶層22(參照圖13)。 藉此,形成如圖13所示之η·型磊晶層21與〆型磊晶層22之 積層結構。η·型磊晶層21之雜質濃度為lel6 em·3,膜厚為 10 μηι,p型磊晶層22之雜質濃度為2ei7 cm_3,膜厚為〇 8 μιη。 繼而,如圖14所示,向ρ+型磊晶層22之内部注入A】離 子,藉此形成Α1離子注入區域23。形成該A1離子注入區域 23之步驟可使所形成之歐姆電極與基板之電性接觸良好, 故其係形成雜質濃度較p+型磊晶層22之雜質濃度高出2位 〜3位左右之區域的步驟。此處離子注入之A1離子之劑量為 lel5 cm2。又,如圖14所示,進行八丨離子注入之深度較好 的是淺於P+型磊晶層22之厚度。 為了形成如圖14所示之Ai離子注入區域23 ,首先在/型 磊晶層22之與η·型磊晶層21不對向之主表面上’藉由例如 熱氧化而形成具有一定厚度之氧化矽膜(Si〇2膜)^然後, 於該Si〇2膜上,塗佈一定厚度之抗蝕劑。於該狀態下,例 如以光微影技術使上述抗蝕劑圖案化。接下來,將形成有 圖案之上述抗蝕劑用作遮罩,藉由例如RIE蝕刻將si〇2膜 以蝕刻而部分地除去(圖案化)。其結果可獲得具有使八丨離 147784.doc -29- 201137975 子注入區域23露出之開口部的8丨〇2膜。接著在去除抗蝕劑 後,自Ρ+型磊晶層22之與η·型磊晶層21不對向之主表面側 對Si〇2膜之開口部進行八丨離子注入◦其後,去除“ο〗膜。 藉此’形成顯示圖14所示之態樣之A1離子注入區域23。該 A1離子注入區域23中,雜質濃度高於P+型磊晶層22,且電 氣電阻較小。因此,可使其後形成之歐姆電極與基板之電 性接觸良好。 此處’為了使A1離子注入區域23之雜質活化,以17〇(Γ(: 實施30分鐘之活化退火(熱處理)。其後,如圖15所示,於 P+型磊晶層22及A1離子注入區域23之主表面上(圖15之上 側)’藉由例如於濕式環境中之熱氧化而形成包含Si〇2之 場氧化膜24(厚度50 nm)。該場氧化膜24係為了保護p+型磊 晶層22及A1離子注入區域23之主表面而形成者。 其-人’例如使用光微影技術而形成於場氧化膜2 4上具有 開口圖案之遮罩。藉由進行使用有該遮罩之蝕刻等而去除 A1離子注入區域23之與p+型磊晶層22不對向之主表面上所 形成的場氧化膜24。由此,使A1離子注入區域23之與p+型 蟲晶層22不對向之主表面露出。然後於該狀態下,作為圖 6所示之形成第1金屬層之步驟(S30),如圖16所示於八丨離 子注入區域23上形成例如厚度1〇 nm之Ti薄膜25。再者, 亦可取代Ti(鈦)而使用例如A1(鋁)或Ni(鎳)、Pt(鉑)、 W(鎢)、Pd(鈀)等。 繼而’作為圖6所示之形成Si層之步驟(S40),如圖16所 示於Ti薄膜25之主表面上形成例如厚度50 nm之Si層27。 147784.doc -30· 201137975 其次,作為圖6所示之形成第2金屬層之步驟(S45),如圖 16所示於Si層27之主表面上形成例如厚度5〇 nmiTi薄膜 25再者,亦可取代Ti(鈦)而使用例如A1(紹)或Cr(鉻)等。 於。亥狀態下,作為圖6所示之進行熱處理之步驟(S5〇), 對圖16所示之系統全體以1〇〇〇。〇進行2分鐘之熱處理。由 此,作為第1金屬層之Ti薄膜25之Ti、以層27之以、及作為 第2金屬層之Ti薄膜25之丁丨進行矽化物化,從而如圖丨了所 不形成作為矽化物層之電極5丨。該電極5丨係歐姆電極,其 可形成為由作為第1金屬層之卩薄膜25與“層27之以進行矽 化物化後之區域、及作為第2金屬層之丁丨薄膜^與以層” 之Si進行矽化物化後之區域獨立形成者積層而成的形態, 亦可為由作為第i金屬層之耵薄膜25、以層27之§1、及作為 第2金屬層之Ti薄膜25該三元素混合並矽化物化而成之^層 矽化物層。或者,電極51之與A1離子注入區域23不對向之 表面層中亦可殘存有未得以石夕化物化之Ti薄膜2 5。 然後,藉由形成佈線部之步驟(S6〇)而於作為歐姆電極 之電極51之表面層上,如圖18所示形成例如厚度5〇 之 Τι薄膜25及厚度3 nm2A1薄膜26作為佈線(墊)。 根據以上順序,完成Pn二極體之1個歐姆電極,但為了 心揮作為實際之pn二極體之功能,必需有2個歐姆電極(2 極)。因此,如圖19所示,例如於sic基板2〇之與n-型磊晶 層21不對向之主表面(背面)上亦形成歐姆電極(電極η), 藉此凡成如圖19所示之ρη二極體1〇〇。該ρη二極體1〇〇包含 八有在著!·生良好之佈線之歐姆電極,其能抑制碳原子於電 147784.doc •31 · 201137975 極51之表面層析出或抑制由Si與Sic形成蕭特基電極。再 者,Sic基板20之背面上之歐姆電極(電極51)、Ti薄膜25及 A1薄膜26之製造方法基本上與p+型磊晶層22上之電極5 i、 Ti薄膜25及A1薄膜26之製造方法為相同。 再者’上述的SiC基板20之與n-型磊晶層21不對向之主 表面上所形成之歐姆電極如圖19所示係與本發明之實施形 態2同樣地形成,但亦可採用其他手段形成。此時,如圖 19所不,在以使SiC基板20與所形成之歐姆電極進行良好 的電性接觸之目的下,無需進行用以使雜質高濃度地摻雜 之離子注入。其原因在於:Sic基板2〇一般會高濃度地含 有雜質,故與P+型磊晶層22相比接觸電阻較小,於該狀態 下可取得良好的電性接觸。 又,已例示本發明實施形態2之形成方法作為以上所述 之pn二極體1〇〇之歐姆電極的形成方法,但並不限於此, 作為本發明之其他實施形態,例如亦可使用實施形態^、 3、4之形成方法而形成歐姆電極。使用任一實施形態時, 圖17〜19之電極51皆可藉由將Si與一種或2種金屬元素進行 矽化物化所成之1層或2層之矽化物層而形成。 實施例2 圖20〜圖26係表示利用本發明之實施形態2 JFET(Reduce Surface Field- junction field effect transistor ’低表面電場_接面場效電晶體)時執行各步驟後 之狀態之概略剖面圖。更具體而言,圖2〇係表示為了形成 RESURF·JFET而執行圖6之步驟(s丨…之狀態的概略剖面 147784.doc •32- 201137975 圖。圖21係表示為了形成RESURF-JFET而執行圖6之步驟 (S20)之狀態的概略剖面圖。圖22係表示為了形成 RESURF-JFET而進行離子注入之狀態的概略剖面圖》圖23 係表示為了形成RESURF-JFET而形成場氧化膜之狀態的概 略剖面圖。圖24係表示為了形成RESURF-JFET而執行圖6 之步驟(S45)之狀態的概略剖面圖。圖25係表示為了形成 RESURF-JFET而執行圖6之步驟(S50)之狀態的概略剖面 圖。圖26係表示為了形成RESURF-JFET而執行圖6之步驟 (S60)後已完成之RESURF-JFET之狀態的概略剖面圖。參 照圖20〜圖26,說明利用有本發明實施形態2之RESURF-JFET之製造方法。 首先,如圖20所示,作為圖6之準備基板之步驟(S10), 準備例如η型之SiC基板20。其次,作為圖6之形成SiC層之 步驟(S20),於SiC基板20之一個主表面上形成p+型磊晶層 22(參照圖21) »又,在p+型磊晶層22之與SiC基板20對向之 表面為相反側之表面上,形成n+型磊晶層32(參照圖21)。 在n+型磊晶層32之與p+型磊晶層22對向之表面為相反側之 表面上,進而形成P+型磊晶層22(參照圖21)。根據以上所 述,形成有如圖21所示之p+型磊晶層22、n+型磊晶層32、 及p+型磊晶層22之積層結構。與SiC基板2〇對向之p+型磊 晶層22之雜質濃度為2el7 cm_3,膜厚為10 μη^ n+型蟲晶 層32之雜質濃度為2el7 cm·3,膜厚為0.4 μπι。最上層之p+ 型磊晶層22之雜質濃度為2el7 cnT3,膜厚為0.2 μπι。 繼而,如圖22所示,向最上層之ρ +型磊晶層22及η+型磊 147784.doc -33- 201137975 晶層32之内部注入p離子及A1離子,藉此形成源極區域 33、閘極區域34、及汲極區域35。形成源極區域33及汲極 區域35之步驟係為了使所形成之歐姆電極與基板之電性接 觸良好而形成雜質濃度較n+型磊晶層32之雜質濃度高出2 位~3位左右之區域的步驟。又,形成閘極區域34之步驟係 為了提高控制所形成之電晶體之通道之閘極電極的電性特 性’而形成雜質濃度較P+型磊晶層22及n+型磊晶層32之雜 質濃度高出1位〜3位左右之區域的步驟。此處,為了形成 源極區域33及汲極區域35,以離子注入將p(磷)離子以6el4 cm之劑里進行主入,且為了形成閘極區域34,將Ai離子 以8el4 cm·2之劑量進行注入。再者,為了使作為resurf_ JFET之源極區域33與汲極區域35之間之區域的電場強度分 佈均勻化以具有抑制電場集中之功能,如圖22所示,較好 的是源極區域33、閘極區域34、及汲極區域35之深度較最 上層之p+型磊晶層22之厚度更深,且較p+型磊晶層^與^ 型磊晶層32之共計厚度更淺。 為了形成圖22所示之源極區域33、閘極區域34、及j:及極 區域35,較好的是與上述的如圖14所示之形成μ離子注入 區域23之情形同樣地例如併用光微影技術與離子注入法。 此處’為了使源極區域33、閘極區域34、及汲極區域35 之雜質活化,以1700°C實施30分鐘之活化退火(熱處理)。 其後’如圖23所示’於p+型磊晶層22及源極區域33、閑極 區域34、汲極區域35之主表面上(圖23之上側),藉由例如 於濕式環境中之熱氧化而形成100 nm之包含Si〇2的場氧化 147784.doc -34· 201137975 膜24。該場氧化膜係為了保護p+型磊晶層22、源極區域 33、閘極區域34及汲極區域35之主表面而形成者。 例如使用光微影技術而形成於場氧化膜24上具有開口圖 案之遮罩。#由進行使用有該遮罩之钮料而去除源極區 域33、閘極區域34、及汲極區域35之與p+型磊晶層。不對 向之主表面上所形成的場氧化膜24。由此’使源極區域 33、閘極區域34、及汲極區域35之與〆型磊晶層。不對向 之主表面露出。然後於該狀態下,作為圖6所示之形成第丄 金屬層之步驟(S30),如圖24所示於源極區域33、閘極區 域34、及汲極區域35上形成例如厚度5〇 nm2Ni薄膜%。 再者,亦可取代Ni(鎳)而使用例如Ai(鋁)或们(鈦)、 Pt(鉑)、w(鎢)、Pd(鈀)等。接著,作為圖6所示之形成si 層之步驟(S40) ’如圖24所示於Ni薄膜36之主表面上形成 例如厚度1〇〇 nm之Si層27。其次作為圖6所示之形成第2金 屬層之步驟(S45) ’如圖24所示於Si層27之主表面上形成例 如厚度20 nm之Ni薄膜36。再者,亦可取代见(鎳)而使用 例如Ti(鈦)或A1(銘)、Cr(鉻)等。 於該狀態下,作為圖6所示之進行熱處理之步驟(S5〇), 對圖24所示之系統全體以1〇〇(rc進行2分鐘之熱處理。由 此,作為第1金屬層之Ni薄膜36之Ni、Si層27之Si、及作為 第2金屬層之Ni薄膜36之Ni進行矽化物化’從而如圖25所 示形成作為矽化物層之電極52。該電極52係歐姆電極。電 極52可形成為由作為第i金屬層之犯薄膜36與以層27之以進 行矽化物化後之區域、及作為第2金屬層之Ni薄膜刊與以 147784.doc •35· 201137975 層27之Si進行矽化物化後之區域獨立形成者積層而成的形 態。又,電極52亦可為由作為第1金屬層之n丨薄膜%、si 層27之Si、及作為第2金屬層之Ni薄膜36該三元素混合並 矽化物化而成之1層矽化物層。或者,電極52之例如與源 極區域33不對向之表面層中’亦可殘存有未得以矽化物化 之Ni薄膜36。 然後’藉由形成佈線部之步驟(S60)而於作為歐姆電極 之電極52之表面層上,如圖26所示形成例如厚度5〇 nm2 Ti薄膜25及厚度3 nm之A1薄膜26作為佈線(墊)。 根據以上順序形成之圖26所示之RESURF-JFET 200包含 具有密著性良好之佈線之歐姆電極(電極52),其能抑制碳 原子於電極52之表面層析出或抑制由8丨與Sic形成蕭特基 電極。 再者,已例示本發明實施形態2之形成方法作為以上所 述之RESURF-JFET 200之歐姆電極之形成方法,但並不限 於此,作為本發明之其他實施形態,例如亦可使用實施形 態1、3、4之形成方法而形成歐姆電極。使用任一實施形 態時,圖25〜26之電極52皆可藉由將Si與一種或2種金屬元 素進行矽化物化所成之丨層或2層之矽化物層而形成。 實施例3 圖27〜圖34係表示利用本發明之實施形態3形成橫型 MOSFET時執行各步驟後之狀態的概略剖面圖。更具體而 言’圖27係表示為了形成橫型m〇sfet而執行圖6之步驟 (sio)之狀態的概略剖面圖。圖28係表示為了形成橫型 147784.doc •36· 201137975 MOSFET而執行圖6之步驟(S20)之狀態的概略剖面圖。圖 29係表示為了形成橫型MOSFET而進行離子注入之狀態的 概略剖面圖。圖30係表示為了形成橫型MOSFET而形成場 氧化膜之狀態的概略剖面圖。圖3 1係表示為了形成橫型 MOSFET而執行圖6之步驟(S45)之狀態的概略剖面圖。圖 32係表示為了形成橫型MOSFET而執行圖6之步驟(S50)之 狀態的概略剖面圖。圖33係表示為了形成橫型m〇SFET而 形成閘極電極之狀態的概略剖面圖。圖34係表示為了形成 橫型MOSFET而執行圖6之步驟(S60)後已完成之橫型 MOSFET之狀態的概略剖面圖。參照圖27〜圖34,說明利 用有本發明實施形態3之橫型MOSFET之製造方法。 首先,如圖27所示,作為圖6之準備基板之步驟(sl〇), 準備例如η型之SiC基板20。其次,作為圖形成siC層之 步驟(S20),於SiC基板20之一個主表面上形成p-型磊晶層 31。由此,可形成如圖28所示之p-型磊晶層31。再者,該 P型蟲晶層31之雜質濃度為lel6 cm·3,膜厚為1〇 μιη。 繼而’如圖29所示’向ρ-型磊晶層31之内部注入ρ離 子’藉此形成導電型為η型之源極區域33、汲極區域35。 形成源極區域33及汲極區域35之步驟係為了使所形成之歐 姆電極與基板之電性接觸良好,且為了提高控制所形成之 電晶體之通道之閘極電極的電性特性,而形成雜質濃度較 Ρ型蟲晶層31之雜質濃度高出2位〜3位左右之區域的步 驟。此處’為了形成源極區域33及汲極區域35,以離子注 入將例如Ρ(磷)離子以5e 14 cm·2之劑量進行注入。再者, 147784.doc •37- 201137975 如圖29所示’較好的是源極區域33及汲極區域35之深度較 P‘型磊晶層31之厚度更淺。 為了形成圖29所示之源極區域33、汲極區域35,較好的 是與上述的如圖14所示之A1離子注入區域23之情形同樣例 如併用光微影技術與離子注入。 此處’為了使源極區域33、汲極區域35之雜質活化,例 如以1750°C實施30分鐘之活化退火(熱處理)後,如圖3〇所 示’於P型蟲晶層31及源極區域33、汲極區域35之主表面 上(圖30之上側)’藉由例如於濕式環境中之熱氧化而形成 50 nm之包含Si〇2的場氧化膜24。該場氧化膜24係使通道 區域上之部分作為閘極絕緣膜而發揮作用,且係為了保護 P·型蟲晶層31、源極區域33及汲極區域35之主表面而形成 者。 例如使用光微影技術而形成於場氧化膜24上具有開口圖 案之遮罩。藉由進行使用有該遮罩之蝕刻等而去除源極區 域33及汲極區域35之與p-型磊晶層3丨不對向之主表面上所 形成的場氧化膜24之一部分。由此,使源極區域33及汲極 區域35之與p-型磊晶層31不對向之主表面之一部分露出。 然後於該狀態下,作為圖6所示之形成第丨金屬層之步驟 (S30) ’如圖31所示於源極區域33及汲極區域35上之已去 除場氧化膜24之區域,形成例如厚度5〇 nmiNi薄膜%。 再者,亦可取代Ni(錄)而使用例如A!(銘)或丁丨(鈦)、 Pt(始)、W(鎢)、Pd(纪)等。接著,作為圖6所示之形成以 層之步驟(S40),如圖31所示於Ni薄膜刊之主表面上形成 147784.doc -38 - 201137975 例如厚度H)〇腿之Si層27。其次,作為圖6所示之形成第2 金屬層之步驟師),如圖31所示於叫27之主表面上形成 例如厚度20 nm之W薄膜37。再者,亦可取❹⑷而使用 例如Ti(鈦)或A1(鋁)、Cr(鉻)等。 於該狀態下,作為圖6所示之進行熱處理之步驟(S50), 對圖31所示之系統全體例如以1〇〇〇c>c進行2分鐘之熱處 理。由此,作為第1金屬層之川薄膜36之沁、以層”之&、 及作為第2金屬層之W薄膜37之霤進行矽化物化,從而如圖 32所示形成作為矽化物層之電極53。該電極^係歐姆電 極,其可形成為由作為第1金屬層之沁薄膜36與以層27之以 進行矽化物化後之區域、及作為第2金屬層之w薄膜37與以 層27之Si進行石夕化物化後之區域獨立形成者積層而成的形 態。又,該電極53亦可為由作為第i金屬層之犯薄膜36、Si is catalyzed by the reaction. At this time, for example, the surface layer of each of the ohmic electrodes shown in FIGS. 7 to 9 (the uppermost layer of the telluride layer in FIGS. 7 and 8 and the uppermost layer of the upper lithi layer in FIG. 9) remains on the second layer. Metal layer 14 (not shown). In this case, the wiring portion is formed in contact with the surface layer of the second metal layer 14 with respect to the ohmic electrode. Therefore, it is possible to maintain good adhesion compared to the case where the wiring layer is formed on the surface layer of the coating layer. Embodiment 2 of the present invention is only the above-described points and embodiments of the present invention. In the second embodiment of the present invention, the configuration, the condition, the sequence, the effect, and the like, which are not described above, are all based on the embodiment of the present invention. 1. One (Embodiment 3) FIG. 10 is a representation to form the present invention. (C) A carbon-cut tantalum of 147784.doc 201137975 A schematic cross-sectional view of a laminated structure of an ohmic electrode of the device. Here, the laminated structure 1 〇c shown in Fig. 0 represents a tantalum carbide used to form the third embodiment of the present invention. The laminated structure of the semiconductor device before the heat treatment is performed. The laminated structure prepared to form the ohmic electrode of the carbonized carbide semiconductor device according to the third embodiment of the present invention is provided in the first embodiment of the present invention. The laminated structure 10 Α (refer to FIG. i) has the same form. However, in the laminated structure 10C, instead of the 8 丨 layer i 3 of the laminated structure, the formation of si and a first metal element is not performed. The Si metal layer 15 having carbon atoms. The layered structure i 〇c is different from the layer structure 丨〇A only in the above-described point. The order of formation of the tantalum carbide semiconductor device according to the third embodiment of the present invention is as shown in FIG. The order of formation of the carbonized carbide semiconductor device according to the first embodiment of the present invention is as follows. However, as described above, the Si metal layer 15 is formed in place of the Si layer 13 of the buildup structure 10A in the buildup structure 1〇c. Therefore, Fig. 2 The step of forming the Si layer in the step (S40) is a step of forming the Si metal layer (S4〇>. Thus, the Si metal layer 15' contains the first metal in the initial state in which the Si-containing layer is subjected to heat treatment. In the element, since the S element is present in a position closer to the first metal element, when the heat treatment is performed, Si and the first metal element can be more rapidly formed. Therefore, the first metal element can be suppressed. It is possible to suppress the precipitation of C of the sic layer 11 by the Si of the SiC layer 11. Further, as the second metal element constituting the Si metal layer 15, nickel, titanium, Ming, Ming, and One of the group consisting of tungsten and Syria However, in the same manner as the second metal element shown in the second embodiment, one element selected from the group consisting of titanium, aluminum, and chromium may be used as the element 147784.doc -26-201137975. (4) The state 3 differs from the first embodiment of the present invention only in the points described above. That is, in the third embodiment of the present invention, the configuration, the condition, the procedure, the effect, and the like which are not described above are all the embodiments of the present invention. (Embodiment 4) FIG. 11 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode in which a tantalum carbide semiconductor device according to Embodiment 4 of the present invention is not formed. Here, the laminated structure 10 D is not shown. The laminated structure before heat treatment for forming the tantalum carbide semiconductor device according to the fourth embodiment of the present invention. As shown in the laminated structure 10D of Fig. 11, the laminated structure prepared to form the ohmic electrode of the tantalum carbide semiconductor device according to the fourth embodiment of the present invention is the same as the laminated structure 丨0B (see Fig. 本) of the second embodiment of the present invention. form. However, in the laminated structure 10D, the Si layer 13' of the laminated structure 丨0B is formed in the same manner as the laminated structure i〇c of the third embodiment of the present invention, and the Si metal layer 15 containing a first metal element and containing no carbon atoms is formed. . In other words, the method further includes the step of forming a second metal element and not containing a carbon atom on the surface of the Si metal layer 15 opposite to the surface opposite to the first metal layer 12 in the laminated structure 10D. The second metal layer 14 is formed. Thus, the second metal layer 14 can also be formed on the main surface of the Si metal layer 15. The laminated structure 10D is different from the laminated structure 10B only in the points described above. The order of formation of the tantalum carbide semiconductor device according to the fourth embodiment of the present invention is based on the order of formation of the tantalum carbide semiconductor device according to the second embodiment of the present invention shown in Fig. 6. However, as described above, in the laminated structure 1A, the Si metal layer 15 is formed instead of the Si layer I3 of the structure 147784.doc • 27·201137975 structure 10B. Therefore, the step (S40) of forming the Si layer of Fig. 6 becomes the step (s4) of forming the Si metal layer. As described above, even when the second metal layer 14 is formed on the main surface of the Si metal layer I5, the layer containing "the third layer is contained in the initial state before the heat treatment as in the case of the Si metal layer 15," Since the Si element is present in a position closer to the first metal element, Si can be cyanated more rapidly with the first metal element during the heat treatment. Embodiment 4 of the present invention is only at the above points. In the fourth embodiment of the present invention, the configuration, the condition, the procedure, the effect, and the like which are not described above are all based on the second embodiment of the present invention. Embodiment 1 FIG. 12 to FIG. A schematic cross-sectional view showing a state in which each step is performed when a pn diode is formed in the second embodiment of the present invention. More specifically, Fig. a shows that the step of Fig. 6 is performed in order to form a pn diode (sl〇 Fig. 13 is a schematic cross-sectional view showing a state in which the step (S20) of Fig. 6 is performed in order to form a ruthenium diode. Fig. 4(4) shows a state in which ion implantation is performed in order to form a diode. Rough Fig. 16 is a schematic cross-sectional view showing a state in which a field oxide film is formed in order to form a pn diode. Fig. 16 is a schematic cross-sectional view showing a state in which the step (sc) of Fig. 6 is performed in order to form a pn diode. Fig. 17 is a schematic cross-sectional view showing a state in which the step (S50) of Fig. 6 is performed in order to form a Pn diode. Fig. _ shows a schematic diagram of a state in which the step (S6〇) of Fig. 6 is performed in order to form a Pn diode. Fig. B is a schematic cross-sectional view of a completed pn diode. Referring to _~19, 147784.doc • 28-201137975, a manufacturing method of the pn diode of the present invention is applied. As shown in the step (s10) of preparing the substrate of Fig. 6, for example, an n-type SiC substrate 20 is prepared. Secondly, a step (S20) of forming a Sic layer as shown in Fig. 6 is formed on one main surface of the SiC substrate 20. - The type of insect crystal layer is referred to as Fig. 7 (7). Further, on the surface opposite to the surface of the n-type layer 21 and the surface of the substrate j, a 磊-type epitaxial layer 22 is formed (refer to Fig. 13). The laminated structure of the η· type epitaxial layer 21 and the 〆 type epitaxial layer 22 as shown in FIG. 13 is mixed with the η· type epitaxial layer 21 The concentration is lel6 em·3, the film thickness is 10 μηι, the impurity concentration of the p-type epitaxial layer 22 is 2 ei7 cm_3, and the film thickness is 〇8 μιη. Then, as shown in FIG. 14, the p+ type epitaxial layer 22 is Internally implanting A] ions, thereby forming a Α1 ion implantation region 23. The step of forming the A1 ion implantation region 23 can make the formed ohmic electrode have good electrical contact with the substrate, so that the impurity concentration is higher than that of the p+ type epitaxial layer. The impurity concentration of the layer 22 is higher than the region of about 2 to 3 positions. Here, the dose of the ion implanted A1 ion is le5 cm 2 . Further, as shown in FIG. 14 , the depth of the helium ion implantation is preferably It is shallower than the thickness of the P+ type epitaxial layer 22. In order to form the Ai ion implantation region 23 as shown in FIG. 14, first, an oxidation having a certain thickness is formed on the main surface of the /-type epitaxial layer 22 opposite to the n-type epitaxial layer 21 by, for example, thermal oxidation. A ruthenium film (Si 〇 2 film) ^ Then, a resist of a certain thickness is applied onto the SiO 2 film. In this state, the resist is patterned by, for example, photolithography. Next, the above-described resist formed with a pattern is used as a mask, and the si〇2 film is partially removed (patterned) by etching, for example, by RIE etching. As a result, an 8 丨〇 2 film having an opening portion in which the erbium 147784.doc -29-201137975 sub-injection region 23 is exposed can be obtained. Then, after the resist is removed, the opening of the Si〇2 film is performed on the main surface side of the Ρ+-type epitaxial layer 22 opposite to the η·-type epitaxial layer 21, and then the erbium ion implantation is performed, and then the removal is performed. ο ing the film, thereby forming an A1 ion implantation region 23 showing the aspect shown in Fig. 14. In the A1 ion implantation region 23, the impurity concentration is higher than that of the P+ type epitaxial layer 22, and the electrical resistance is small. The ohmic electrode formed later can be electrically contacted with the substrate. Here, 'in order to activate the impurity of the A1 ion implantation region 23, 17 〇 (: 30 minutes of activation annealing (heat treatment) is performed. Thereafter, As shown in FIG. 15, on the main surface of the P+ type epitaxial layer 22 and the A1 ion implantation region 23 (on the upper side of FIG. 15), field oxidation including Si〇2 is formed by thermal oxidation in, for example, a wet environment. The film 24 (thickness 50 nm) is formed by protecting the p + -type epitaxial layer 22 and the main surface of the A1 ion implantation region 23. The human-film is formed by field oxidization using, for example, photolithography. a mask having an opening pattern on the film 24, which is removed by etching using the mask or the like The field oxide film 24 formed on the main surface of the A1 ion implantation region 23 and the p+ type epitaxial layer 22 is not opposed. Thereby, the main surface of the A1 ion implantation region 23 which is opposite to the p+ type crystal layer 22 is exposed. Then, in this state, as a step (S30) of forming the first metal layer shown in Fig. 6, a Ti film 25 having a thickness of, for example, 1 nm is formed on the gossip ion implantation region 23 as shown in Fig. 16. Instead of Ti (titanium), for example, A1 (aluminum) or Ni (nickel), Pt (platinum), W (tungsten), Pd (palladium), or the like may be used. Then, as a step of forming a Si layer as shown in FIG. (S40), for example, a Si layer 27 having a thickness of 50 nm is formed on the main surface of the Ti film 25 as shown in Fig. 16. 147784.doc -30·201137975 Next, as a step of forming a second metal layer as shown in Fig. 6 ( S45), for example, a film 5 〇nmiTi film 25 is formed on the main surface of the Si layer 27 as shown in Fig. 16. Alternatively, Ti (titanium) may be used instead of Ti (titanium), for example, A1 (Cr) or Cr (Chromium). In the sea state, as a heat treatment step (S5〇) shown in Fig. 6, the entire system shown in Fig. 16 was heat-treated for 2 minutes. As a result, Ti which is the Ti film 25 of the first metal layer, and the ruthenium of the Ti film 25 which is the second metal layer and the Ti thin film 25 which is the second metal layer are formed into a bismuth telluride layer. The electrode 5 is an ohmic electrode, which can be formed by a tantalum film 25 as a first metal layer, a region of the layer 27 for mash-forming, and a butadiene film as a second metal layer. The form formed by laminating the regions formed by the bismuthization of Si in the layer may be the ruthenium film 25 as the i-th metal layer, the § 1 of the layer 27, and the second metal layer. The Ti film 25 is a layered telluride layer in which the three elements are mixed and deuterated. Alternatively, a Ti film 25 which is not etched into the surface layer of the electrode 51 and the A1 ion implantation region 23 may be left. Then, on the surface layer of the electrode 51 as the ohmic electrode by the step of forming the wiring portion (S6〇), as shown in FIG. 18, for example, a film 5 having a thickness of 5 Å and a film 26 having a thickness of 3 nm 2 A1 are formed as wirings (pads). ). According to the above procedure, one ohmic electrode of the Pn diode is completed, but in order to function as an actual pn diode, two ohmic electrodes (two poles) are necessary. Therefore, as shown in FIG. 19, for example, an ohmic electrode (electrode η) is formed also on the main surface (back surface) of the sic substrate 2 and the n-type epitaxial layer 21 which are not opposed to each other, whereby The ρη diode is 1〇〇. The ρη diode 1〇〇 contains eight in existence! An ohmic electrode with a good wiring that inhibits the formation of a Schottky electrode from Si and Sic by suppressing the carbon atom from the surface of the electrode 147784.doc •31 · 201137975. Furthermore, the ohmic electrode (electrode 51), the Ti film 25, and the A1 film 26 on the back surface of the Sic substrate 20 are substantially the same as the electrode 5 i, the Ti film 25, and the A1 film 26 on the p + type epitaxial layer 22. The manufacturing method is the same. Further, the ohmic electrode formed on the main surface of the SiC substrate 20 and the n-type epitaxial layer 21 which are not opposed to each other is formed in the same manner as in the second embodiment of the present invention, as shown in FIG. Means formed. At this time, as shown in Fig. 19, in order to make the SiC substrate 20 be in good electrical contact with the formed ohmic electrode, it is not necessary to perform ion implantation for doping impurities with a high concentration. The reason for this is that the Sic substrate 2 〇 generally contains impurities at a high concentration, so that the contact resistance is smaller than that of the P + -type epitaxial layer 22, and good electrical contact can be obtained in this state. Further, the method for forming the second embodiment of the present invention is exemplified as the method of forming the ohmic electrode of the above-described pn diode 1 ,. However, the present invention is not limited thereto, and other embodiments of the present invention may be implemented, for example. An ohmic electrode is formed by the formation of the forms ^, 3, and 4. When any of the embodiments is used, the electrodes 51 of Figs. 17 to 19 can be formed by one or two layers of a telluride layer formed by mashing Si with one or two metal elements. (Embodiment 2) FIG. 20 to FIG. 26 are schematic cross-sectional views showing a state in which each step is performed when a JFET (Red Surface Field-Connected Field Effect Transistor) according to Embodiment 2 of the present invention is used. . More specifically, FIG. 2B shows a schematic cross-section of 147784.doc • 32-201137975 in which the steps of FIG. 6 are performed in order to form a RESURF·JFET. FIG. 21 is a diagram showing execution for forming a RESURF-JFET. Fig. 22 is a schematic cross-sectional view showing a state in which ion implantation is performed to form a RESURF-JFET. Fig. 23 is a view showing a state in which a field oxide film is formed in order to form a RESURF-JFET. Fig. 24 is a schematic cross-sectional view showing a state in which the step (S45) of Fig. 6 is performed in order to form a RESURF-JFET, and Fig. 25 is a view showing a state in which the step (S50) of Fig. 6 is performed in order to form a RESURF-JFET. Fig. 26 is a schematic cross-sectional view showing a state in which the RESURF-JFET which has been completed after the step (S60) of Fig. 6 is formed in order to form a RESURF-JFET. With reference to Fig. 20 to Fig. 26, the use of the present invention will be described. First, as shown in Fig. 20, as shown in Fig. 20, a step (S10) of preparing a substrate in Fig. 6 is prepared, for example, an η-type SiC substrate 20 is prepared. Next, a step of forming a SiC layer as Fig. 6 is provided. (S20), in SiC A p + -type epitaxial layer 22 is formed on one main surface of the board 20 (refer to FIG. 21). Further, on the surface of the p + type epitaxial layer 22 opposite to the surface opposite to the SiC substrate 20, n + type epitaxial is formed. Layer 32 (see Fig. 21). On the surface of the n + -type epitaxial layer 32 opposite to the surface opposite to the p + -type epitaxial layer 22, a P + -type epitaxial layer 22 is further formed (see Fig. 21). The laminated structure of the p+ type epitaxial layer 22, the n+ type epitaxial layer 32, and the p+ type epitaxial layer 22 as shown in FIG. 21 is formed, and the p+ type epitaxial layer 22 opposite to the SiC substrate 2 is formed. The impurity concentration is 2el7 cm_3, and the film thickness is 10 μηη n + type worm layer 32 has an impurity concentration of 2el7 cm·3 and a film thickness of 0.4 μm. The uppermost p+ type epitaxial layer 22 has an impurity concentration of 2el7 cnT3, film. The thickness is 0.2 μm. Then, as shown in FIG. 22, p ions and A1 ions are implanted into the innermost layer of the p + type epitaxial layer 22 and the n + type bar 147784.doc -33 - 201137975 crystal layer 32. The source region 33, the gate region 34, and the drain region 35 are formed. The steps of forming the source region 33 and the drain region 35 are for electrically connecting the formed ohmic electrode to the substrate. Preferably, the impurity concentration is higher than the impurity concentration of the n+ type epitaxial layer 32 by a region of about 2 to about 3. Further, the step of forming the gate region 34 is to increase the gate of the transistor formed by the control. The electrical property of the electrode of the electrode is a step of forming a region in which the impurity concentration is higher than the impurity concentration of the P + -type epitaxial layer 22 and the n + -type epitaxial layer 32 by about 1 to 3 bits. Here, in order to form the source region 33 and the drain region 35, p (phosphorus) ions are mainly implanted in a dose of 6 el 4 cm by ion implantation, and in order to form the gate region 34, Ai ions are made at 8 el 4 cm · 2 The dose is injected. Further, in order to homogenize the electric field intensity distribution as the region between the source region 33 and the drain region 35 of the resurf_JFET to have a function of suppressing electric field concentration, as shown in Fig. 22, the source region 33 is preferable. The depth of the gate region 34 and the drain region 35 is deeper than the thickness of the p+ type epitaxial layer 22 of the uppermost layer, and is shallower than the total thickness of the p+ epitaxial layer and the epitaxial layer 32. In order to form the source region 33, the gate region 34, and the j: and the polar region 35 shown in Fig. 22, it is preferable to use, for example, the same as the case of forming the μ ion implantation region 23 as shown in Fig. 14 described above. Photolithography and ion implantation. Here, in order to activate the impurities in the source region 33, the gate region 34, and the drain region 35, activation annealing (heat treatment) was performed at 1700 ° C for 30 minutes. Thereafter, 'as shown in FIG. 23' on the main surface of the p+ type epitaxial layer 22 and the source region 33, the idle region 34, and the drain region 35 (on the upper side of FIG. 23), for example, in a wet environment Thermal oxidation to form a 100 nm field oxide containing Si 〇 2 147784.doc -34·201137975 Membrane 24. The field oxide film is formed to protect the principal surfaces of the p + -type epitaxial layer 22, the source region 33, the gate region 34, and the drain region 35. For example, a mask having an opening pattern formed on the field oxide film 24 is formed using photolithography. # The p+ type epitaxial layer is removed from the source region 33, the gate region 34, and the drain region 35 by using the mask material. The field oxide film 24 formed on the surface of the main surface is not opposed. Thus, the source region 33, the gate region 34, and the drain region 35 are combined with the 磊-type epitaxial layer. The surface of the main surface is not facing. Then, in this state, as a step (S30) of forming the second metal layer shown in Fig. 6, as shown in Fig. 24, for example, a thickness of 5 Å is formed on the source region 33, the gate region 34, and the drain region 35. % of nm2Ni film. Further, in place of Ni (nickel), for example, Ai (aluminum) or (titanium), Pt (platinum), w (tungsten), Pd (palladium) or the like may be used. Next, as the step (S40) of forming the Si layer shown in Fig. 6, as shown in Fig. 24, a Si layer 27 of, for example, a thickness of 1 〇〇 nm is formed on the main surface of the Ni thin film 36. Next, as a step (S45) of forming the second metal layer shown in Fig. 6, as shown in Fig. 24, a Ni film 36 having a thickness of, for example, 20 nm is formed on the main surface of the Si layer 27. Further, instead of seeing (nickel), for example, Ti (titanium) or A1 (inger), Cr (chromium) or the like may be used. In this state, as a heat treatment step (S5〇) shown in FIG. 6, the entire system shown in FIG. 24 is heat-treated at 1 Torr for 2 minutes. Thus, Ni as the first metal layer. The Ni of the thin film 36, the Si of the Si layer 27, and the Ni of the Ni thin film 36 as the second metal layer are germanium-formed to form an electrode 52 as a telluride layer as shown in Fig. 25. The electrode 52 is an ohmic electrode. 52 can be formed by the film 36 as the i-th metal layer and the region after the layer 27 is deuterated, and the Ni film as the second metal layer is published as 147784.doc •35·201137975 layer 27 Si The electrode 52 may be formed by laminating the region formed by the bismuthization, and the electrode 52 may be an n 丨 film % as the first metal layer, Si of the Si layer 27, and a Ni film 36 as the second metal layer. The three elements are mixed and eutecticized into a single layer of germanide layer. Alternatively, the surface of the electrode 52, which is not opposite the source region 33, may also have a Ni film 36 that has not been deuterated. The electric power as the ohmic electrode by the step of forming the wiring portion (S60) On the surface layer of 52, for example, a thin film 25 of a thickness of 5 〇 nm 2 and a thin film 26 of a thickness of 3 nm are formed as wirings (pads) as shown in Fig. 26. The RESURF-JFET 200 shown in Fig. 26 formed in accordance with the above sequence includes An ohmic electrode (electrode 52) of a wiring having good adhesion, which can suppress carbon atoms from being crystallized on the surface of the electrode 52 or suppress formation of a Schottky electrode from 8 Å and Sic. Further, Embodiment 2 of the present invention has been exemplified. The method of forming the ohmic electrode of the RESURF-JFET 200 described above is not limited thereto, and as another embodiment of the present invention, for example, the formation methods of the first, third, and fourth embodiments may be used to form ohmic. When any of the embodiments is used, the electrodes 52 of FIGS. 25 to 26 can be formed by a bismuth layer or a two-layer bismuth layer formed by mashing Si with one or two metal elements. 27 to 34 are schematic cross-sectional views showing a state in which each step is performed when a lateral MOSFET is formed in the third embodiment of the present invention. More specifically, Fig. 27 shows that Fig. 6 is performed in order to form a horizontal type m〇sfet. Step (sio) Fig. 28 is a schematic cross-sectional view showing a state in which the step (S20) of Fig. 6 is performed to form a lateral type 147784.doc • 36·201137975 MOSFET. Fig. 29 shows an ion for forming a horizontal MOSFET. Fig. 30 is a schematic cross-sectional view showing a state in which a field oxide film is formed in order to form a lateral MOSFET. Fig. 31 shows a state in which the step (S45) of Fig. 6 is performed in order to form a lateral MOSFET. A schematic cross-sectional view. Fig. 32 is a schematic cross-sectional view showing a state in which the step (S50) of Fig. 6 is performed in order to form a lateral MOSFET. Fig. 33 is a schematic cross-sectional view showing a state in which a gate electrode is formed in order to form a lateral m〇SFET. Fig. 34 is a schematic cross-sectional view showing the state of the horizontal MOSFET which has been completed after the step (S60) of Fig. 6 is performed to form a horizontal MOSFET. A method of manufacturing a horizontal MOSFET according to a third embodiment of the present invention will be described with reference to Figs. 27 to 34. First, as shown in FIG. 27, as the step (s10) of preparing the substrate of FIG. 6, for example, an n-type SiC substrate 20 is prepared. Next, as a step of forming a siC layer (S20), a p-type epitaxial layer 31 is formed on one main surface of the SiC substrate 20. Thereby, the p-type epitaxial layer 31 as shown in FIG. 28 can be formed. Further, the P type worm layer 31 has an impurity concentration of le6 cm·3 and a film thickness of 1 〇 μιη. Then, as shown in Fig. 29, a p-ion is implanted into the p-type epitaxial layer 31 to form a source region 33 and a drain region 35 of a conductivity type. The steps of forming the source region 33 and the drain region 35 are performed in order to improve the electrical contact between the formed ohmic electrode and the substrate, and to improve the electrical characteristics of the gate electrode of the channel formed by the controlled transistor. The impurity concentration is higher than the impurity concentration of the ruthenium-type worm layer 31 by a region of about 2 to 3 positions. Here, in order to form the source region 33 and the drain region 35, for example, cerium (phosphorus) ions are implanted at a dose of 5e 14 cm·2 by ion implantation. Further, 147784.doc • 37- 201137975 is shown in Fig. 29. It is preferable that the depths of the source region 33 and the drain region 35 are shallower than the thickness of the P' type epitaxial layer 31. In order to form the source region 33 and the drain region 35 shown in Fig. 29, it is preferable to use photolithography and ion implantation in the same manner as the above-described A1 ion implantation region 23 shown in Fig. 14. Here, in order to activate the impurities of the source region 33 and the drain region 35, for example, after activation annealing (heat treatment) at 1750 ° C for 30 minutes, as shown in FIG. 3A, the P-type crystal layer 31 and the source are shown. On the main surface of the polar region 33 and the drain region 35 (upper side of FIG. 30), a 50 nm field oxide film 24 containing Si〇2 is formed by thermal oxidation in, for example, a wet environment. The field oxide film 24 functions as a gate insulating film and is formed to protect the main surfaces of the P-type crystal layer 31, the source region 33, and the drain region 35. For example, a mask having an opening pattern formed on the field oxide film 24 is formed using photolithography. A portion of the field oxide film 24 formed on the main surface of the source region 33 and the drain region 35 which is not opposed to the p-type epitaxial layer 3 is removed by etching using the mask or the like. Thereby, the source region 33 and the drain region 35 are partially exposed to one of the main surfaces on which the p-type epitaxial layer 31 is not opposed. Then, in this state, as the step (S30) of forming the second metal layer shown in FIG. 6, the region of the source region 33 and the drain region 35 on which the field oxide film 24 has been removed is formed as shown in FIG. For example, a thickness of 5 〇nmiNi film%. Further, in place of Ni, it is also possible to use, for example, A! (Ming) or Ding (Titanium), Pt (Start), W (Tungsten), Pd (Ji), and the like. Next, as a step of forming a layer as shown in Fig. 6 (S40), as shown in Fig. 31, a Si layer 27 of 147784.doc - 38 - 201137975, for example, a thickness H) of the leg is formed on the main surface of the Ni film. Next, as a step of forming the second metal layer shown in Fig. 6, as shown in Fig. 31, a W film 37 having a thickness of, for example, 20 nm is formed on the main surface of Fig. 27. Further, for example, Ti (titanium) or Al (aluminum), Cr (chromium) or the like may be used. In this state, as a step (S50) of performing heat treatment as shown in Fig. 6, the entire system shown in Fig. 31 is subjected to heat treatment for 2 minutes, for example, at 1 〇〇〇 c > c. As a result, the first metal layer is formed into a tantalum film 36, and the layer "and" and the W film 37 as the second metal layer are mashed to form a telluride layer as shown in FIG. An electrode 53. The electrode is an ohmic electrode which can be formed by a germanium film 36 as a first metal layer and a germanium layer formed by the layer 27, and a w film 37 and a layer as a second metal layer. The Si of 27 is formed by laminating the regions formed by the Sihua compoundization. Further, the electrode 53 may be a film 36 as the ith metal layer.

Si層27之Si '及作為第2金屬層之评薄膜37該三元素混合並 矽化物化而成之1層矽化物層。或者,電極53之例如與源 極區域33不對向之表面層中,亦可殘存有未得以矽化物化 之W薄膜37。 其次,例如使用光微影法形成抗蝕劑遮罩後,使用真空 蒸鑛、離子束瘵鍍或減鍍而形成作為導電體膜之A丨薄膜, 並將應成為該導電體膜之閘極電極之部分以外的部分與抗 蝕劑遮罩一併去除(剝離),藉此如圖33所示於場氧化膜^ 上形成作為閘極電極之A1薄膜26,此處,A1薄膜26之厚度 為200 nm,A卜薄膜26係以跨及源極區域33與汲極區域35之 方式(於通道區域上)形成。 147784.doc -39- 201137975 然後,藉由形成佈線部之步驟(S60),於作為歐姆電極 之電極53之表面層上’如圖34所示形成例如厚度5〇⑽之 W薄膜37及厚度3 nm之Ai薄膜26作為佈線(墊)。再者,於 作為閘極電極之…薄膜26上亦同樣地形成…薄膜”及…薄 膜26。 根據以上順序所形成之圖所示之橫型mosfet 300包 έ 〃、有迕著性良好之佈線之歐姆電極,其能抑制碳原子於 電極53之表面層析出或抑制由81與81<::形成蕭特基電極。 再者,已例示本發明實施形態2之形成方法作為以上所 述之橫型MOSFET 300之歐姆電極之形成方法,但並不限 於此作為本發明之其他實施形態,例如亦可使用實施形 態1、3、4之形成方法而形成歐姆電極。使用任一實施形 態時,圖32〜34之電極53皆可藉由將Si與一種或2種金屬元 素矽化物化所成之1層或2層之矽化物層而形成。 此外不限於檢型MOSFET,例如對於縱型MOSFET、 MESFET、IGBT等具有歐姆電極之碳化石夕半導體裝置,亦 可使用本發明之實施形態。 當認為’本次揭示之各實施形態及各實施例之所有方面 皆為例示而非限制性者。本發明之範圍係由申請專利範圍 而非上述說明所揭示,其設法包含與請求項均等之意義及 範圍内之所有變更。 產業上之可利用性 本發明作為可提供一種具有不會形成蕭特基接觸地藉由 抑制喊之析出而提高與佈線之密著性之歐姆電極的碳化石夕 147784.doc 201137975 半導體裝置之技術為特別優異。 【圖式簡單說明】 圖1係表示用以形成本發明實施形態丨之碳化石夕半導體裝 置之歐姆電極之積層結構的概略剖面圖。 圖2係表示本發明實施形態1之碳化石夕半導體裝置之形成 順序的流程圖。 圖3係表示進行熱處理後之作為本發明實施形態」之歐姆 電極之積層結構的概略剖面圖。 圖4係表示進行熱處理後之作為本發明實施形態」之歐姆 電極之另一形態之積層結構的概略剖面圖。 圖5係表示用以形成本發明實施形態2之碳化石夕半導體裝 置之歐姆電極之積層結構的概略剖面圖。 圖6係表示本發明實施形態2之碳化石夕半導體裝置之形成 順序的流程圖。 圖7係表示進行熱處理後之作為本發明實施形&之歐姆 電極之積層結構的概略剖面圖。 圖8係表示進行熱處理後之作為本發明實施形態之之歐姆 電極之另一形態之積層結構的概略剖面圖。 圖9係表示進行熱處理後之作為本發明實施形態2之歐姆 電極之又一形態之積層結構的概略剖面圖。 圖10係表示用以形成本發明實施形態3之碳化矽半導體 裝置之歐姆電極之積層結構的概略剖面圖。 圖11係表示用以形成本發明實施形態4之碳化矽半導體 農置之歐姆電極之積層結構的概略剖面圖。 147784.doc 41 201137975 圖12係表子& '、馮了形成pn二極體而執行圖6之步驟(sl〇)之 狀態的概略剖面圖。 圖13係表矛& 、為了形成pn二極體而執行圖6之步驟(S2〇)之 狀態的概略剖面圖。 圖14 4系表^ -达 不為了形成ρη二極體而進行離子注入之狀態的 概略剖面圖。 圖係表7為了形成ρη二極體而形成場氧化膜之狀態的 概略剖面圖。 系表示為了形成ρη二極體而執行圖6之步驟(S45)之 狀態的概略剖面圖。 廣I 17 李霸 '、 示為了形成ρη二極體而執行圖6之步驟(S50)之 狀態的概略剖面圖。 圖1 8传矣—、 ’、 示為了形成Pn二極體而執行圖6之步驟(S60)之 狀態的概略剖面圖。 圖19係已完成之pn二極體之概略剖面圖。 圖2〇係表示為了形成RESURF-JFET而執行圖6之步驟 (S 10)之狀態的概略剖面圖。 圖21在皂-& 承衣不為了形成RESURF-JFET而執行圖6之步驟 (S20)之狀態的概略剖面圖。 圖22係表示為了形成rESURF_JFET而進行離子注入之狀 態的概略剖面圖。 圖23係表示為了形$resurf_jfet而形成場氧化膜之狀 態的概略剖面圖。 圖24係表示為了形成rESURF_jFEt而執行圖6之步驟 147784.doc • 42- 201137975 (S45)之狀態的概略剖面圖。 圖25係表示為了形成RESURF-JFET而執行圖6之步驟 (S50)之狀態的概略剖面圖。 圖26係表示為了形成RESURF-JFET而執行圖6之步驟 (S60)後已完成之RESURF-JFET之狀態的概略剖面圖。 圖27係表示為了形成橫型MOSFET而執行圖6之步驟 (S 10)之狀態的概略剖面圖。 圖28係表示為了形成橫型MOSFET而執行圖6之步驟 (S20)之狀態的概略剖·面圖。 圖29係表示為了形成橫型MOSFET而進行離子注入之狀 態的概略剖面圖。 圖3 0係表示為了形成橫型MOSFET而形成場氧化膜之狀 態的概略剖面圖。 圖3 1係表示為了形成橫型MOSFET而執行圖6之步驟 (S45)之狀態的概略剖面圖。 圖32係表示為了形成橫型MOSFET而執行圖6之步驟 (S50)之狀態的概略剖面圖。 圖33係表示為了形成橫型MOSFET而形成閘極電極之狀 態的概略剖面圖。 圖34係表示為了形成横型MOSFET而執行圖6之步驟 (S60)後已完成之橫型MOSFET之狀態的概略剖面圖。 圖35係表示一般的SiC半導體裝置之電極與佈線之狀態 的概略剖面圖。 圖36係表示於SiC半導體層上形成有Ni與Si之合金層之 147784.doc -43 -The Si' of the Si layer 27 and the thin film 37 which is the second metal layer are mixed and eutecticized to form a single telluride layer. Alternatively, in the surface layer of the electrode 53 which is not opposed to the source region 33, for example, the W film 37 which is not cleaved may remain. Next, after forming a resist mask by, for example, photolithography, an A丨 film as a conductor film is formed by vacuum evaporation, ion beam plating or deplating, and should be the gate of the conductor film. A portion other than the portion of the electrode is removed (peeled) together with the resist mask, whereby an A1 film 26 as a gate electrode is formed on the field oxide film as shown in Fig. 33, where the thickness of the A1 film 26 is At 200 nm, the A film 26 is formed across the source region 33 and the drain region 35 (on the channel region). 147784.doc -39- 201137975 Then, by the step of forming the wiring portion (S60), a W film 37 having a thickness of 5 〇 (10) and a thickness of 3 are formed as shown in FIG. 34 on the surface layer of the electrode 53 as the ohmic electrode. The Ai film 26 of nm is used as a wiring (pad). Further, the film 26 and the film 26 are formed similarly on the film 26 as the gate electrode. The horizontal mosfet 300 packaged according to the above-described sequence is formed, and the wiring having good adhesion is good. An ohmic electrode capable of suppressing carbon atoms from being crystallized on the surface of the electrode 53 or suppressing the formation of a Schottky electrode from 81 and 81. Further, the formation method of the second embodiment of the present invention has been exemplified as described above. Although the method of forming the ohmic electrode of the horizontal MOSFET 300 is not limited to this, as another embodiment of the present invention, for example, the ohmic electrode may be formed by the formation methods of the first, third, and fourth embodiments. The electrodes 53 of FIGS. 32 to 34 can be formed by mashing one or two layers of Si and one or two metal elements. Further, it is not limited to a type MOSFET, for example, a vertical MOSFET. Embodiments of the present invention may be used in a carbon carbide-based semiconductor device having an ohmic electrode such as a MESFET or an IGBT. It is to be understood that the embodiments of the present disclosure and all embodiments are illustrative and not limiting. The scope of the present invention is disclosed by the scope of the claims and not the foregoing description, which is intended to include all modifications within the meaning and scope of the claims. Industrial Applicability The present invention A carbon carbide which forms an ohmic electrode which is improved in adhesion to wiring by suppressing the release of shouting. 147784.doc 201137975 The technology of the semiconductor device is particularly excellent. [Simplified description of the drawing] FIG. A schematic cross-sectional view showing a laminated structure of an ohmic electrode of a carbonized carbide semiconductor device according to an embodiment of the present invention. Fig. 2 is a flow chart showing a procedure for forming a carbonized carbide semiconductor device according to Embodiment 1 of the present invention. A schematic cross-sectional view of a laminated structure of an ohmic electrode as an embodiment of the present invention after heat treatment. Fig. 4 is a schematic cross-sectional view showing a laminated structure of another embodiment of an ohmic electrode which is an embodiment of the present invention after heat treatment. Fig. 5 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode for forming a carbonized carbide semiconductor device according to a second embodiment of the present invention. Fig. 6 is a flow chart showing the procedure for forming a carbonized carbide semiconductor device according to Embodiment 2 of the present invention. Fig. 7 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode of the embodiment of the present invention after heat treatment. Fig. 8 is a schematic cross-sectional view showing a laminated structure of another embodiment of the ohmic electrode according to the embodiment of the present invention after heat treatment. Fig. 9 is a schematic cross-sectional view showing a laminated structure of still another embodiment of the ohmic electrode according to the second embodiment of the present invention after the heat treatment. Fig. 10 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode for forming a tantalum carbide semiconductor device according to a third embodiment of the present invention. Fig. 11 is a schematic cross-sectional view showing a laminated structure of an ohmic electrode for forming a tantalum carbide semiconductor according to a fourth embodiment of the present invention. 147784.doc 41 201137975 Fig. 12 is a schematic cross-sectional view showing a state in which the pn diode is formed and the step (s1) of Fig. 6 is performed. Fig. 13 is a schematic cross-sectional view showing a state in which the step (S2) of Fig. 6 is performed in order to form a pn diode. Fig. 14 is a schematic cross-sectional view showing a state in which ion implantation is not performed to form a ρη diode. Fig. 7 is a schematic cross-sectional view showing a state in which a field oxide film is formed in order to form a ρη diode. A schematic cross-sectional view showing a state in which the step (S45) of Fig. 6 is performed in order to form a ρη diode.广 I 17 李霸', a schematic cross-sectional view showing a state in which the step (S50) of Fig. 6 is performed in order to form a ρη diode. Fig. 1 is a schematic cross-sectional view showing a state in which the step (S60) of Fig. 6 is performed in order to form a Pn diode. Figure 19 is a schematic cross-sectional view of a completed pn diode. Fig. 2 is a schematic cross-sectional view showing a state in which the step (S 10) of Fig. 6 is performed in order to form a RESURF-JFET. Fig. 21 is a schematic cross-sectional view showing a state in which the soap-& garment is not subjected to the step (S20) of Fig. 6 in order to form a RESURF-JFET. Fig. 22 is a schematic cross-sectional view showing a state in which ion implantation is performed in order to form an rESURF_JFET. Fig. 23 is a schematic cross-sectional view showing a state in which a field oxide film is formed in order to form $resurf_jfet. Fig. 24 is a schematic cross-sectional view showing a state in which steps 147784.doc • 42-201137975 (S45) of Fig. 6 are executed in order to form rESURF_jFEt. Fig. 25 is a schematic cross-sectional view showing a state in which the step (S50) of Fig. 6 is performed in order to form a RESURF-JFET. Fig. 26 is a schematic cross-sectional view showing the state of the RESURF-JFET which has been completed after the step (S60) of Fig. 6 is performed to form the RESURF-JFET. Fig. 27 is a schematic cross-sectional view showing a state in which the step (S 10) of Fig. 6 is performed in order to form a lateral MOSFET. Fig. 28 is a schematic cross-sectional view showing a state in which the step (S20) of Fig. 6 is performed to form a lateral MOSFET. Fig. 29 is a schematic cross-sectional view showing a state in which ion implantation is performed to form a lateral MOSFET. Fig. 30 is a schematic cross-sectional view showing a state in which a field oxide film is formed in order to form a lateral MOSFET. Fig. 3 is a schematic cross-sectional view showing a state in which the step (S45) of Fig. 6 is performed in order to form a lateral MOSFET. Fig. 32 is a schematic cross-sectional view showing a state in which the step (S50) of Fig. 6 is performed in order to form a lateral MOSFET. Fig. 33 is a schematic cross-sectional view showing a state in which a gate electrode is formed to form a lateral MOSFET. Fig. 34 is a schematic cross-sectional view showing the state of the horizontal MOSFET which has been completed after the step (S60) of Fig. 6 is performed to form a lateral MOSFET. Fig. 35 is a schematic cross-sectional view showing the state of electrodes and wirings of a general SiC semiconductor device. Figure 36 shows the formation of an alloy layer of Ni and Si on the SiC semiconductor layer 147784.doc -43 -

201137975 結構的概略剖面圖。 【主要元件符號說明】 10 、 20 ' 95 10A、10B、10C、10D 11 11A、11B、12A、 12B 、 13B 12 13、27 14 15 21 22 23 24 25 26 31 32 33 34 35 36 37A schematic cross-sectional view of the structure of 201137975. [Description of main component symbols] 10, 20 '95 10A, 10B, 10C, 10D 11 11A, 11B, 12A, 12B, 13B 12 13, 27 14 15 21 22 23 24 25 26 31 32 33 34 35 36 37

SiC基板 積層結構 SiC層 歐姆電極 第1金屬層 Si層 第2金屬層 Si金屬層 η-型蟲晶層 Ρ +型蠢晶層 Α1離子注入區域 場氧化膜 Ti薄膜 A1薄膜 p-型蟲晶層 n+型遙晶層 源極區域 閘極區域 >及極區域 Ni薄膜 W薄膜 147784.doc -44- 201137975SiC substrate laminated structure SiC layer ohmic electrode first metal layer Si layer second metal layer Si metal layer η-type worm layer Ρ + type stray layer Α 1 ion implantation region field oxide film Ti film A1 film p-type worm layer n+ type remote crystal layer source region gate region> and polar region Ni film W film 147784.doc -44- 201137975

41 含碳矽化物層 42 ' 43 石夕化物層 44 上部石夕化物層 51 、 52 、 53 、 98 電極 94 Ni-Si合金層 95A 電子元件 96 佈線 97 碳 99 SiC半導體基板 99A Sic半導體裝置 100 pn二極體 200 RESURF-JFET 300 橫型MOSFET 147784.doc -45-41 carbon-containing telluride layer 42 ' 43 lithium layer 44 upper lithium layer 51, 52, 53 , 98 electrode 94 Ni-Si alloy layer 95A electronic component 96 wiring 97 carbon 99 SiC semiconductor substrate 99A Sic semiconductor device 100 pn Diode 200 RESURF-JFET 300 Horizontal MOSFET 147784.doc -45-

Claims (1)

201137975 七、申請專利範圍: 1. 一種碳化矽半導體裝置之製造方法,該碳化矽半導體裝 置具有歐姆電極,該製造方法係包括: 步驟(S20) ’形成包含碳化矽之Sic層(11); 步驟(S30),於上述siC層(11)之一個主表面上,形成 包含一種第1金屬元素且不含有碳原子之第1金屬層 (12); 步驟(S4〇) ’於上述第1金屬層(12)之與上述以(2層(11) 對向之表面為相反側之表面上,形成包含矽且不含有碳 原子之Si層(13,15,27);及 步驟(S5 0),為了形成歐姆電極,對上述sic層(u)、 上述第1金屬層(I2)、及上述Si層(13)進行熱處理。 2. 如請求項1之碳化矽半導體裝置之製造方法,其係進而 包括: 步驟(S45) ’於進行上述熱處理之步驟(S5〇)之前,在 上述Si層(13)之與上述第丨金屬層(12)對向之表面為相反 側之表面上,形成包含一種第2金屬元素且不含有碳原 子之第2金屬層(14)。 3. 如請求項2之碳化矽半導體裝置之製造方法,其中 上述第2金屬元素(M)係選自由鈦、鋁、及鉻所組成之 群中之一種元素。 4·如請求項1之碳化矽半導體裝置之製造方法,其中 於進行上述熱處理之步驟(S5〇)中,在上述Sic層(u) 之一個主表面上,形成包含上述一種第丨金屬元素與矽 147784.doc 201137975 之合金且含有碳原子之含碳矽化物層(41)。 5. 如請求項1之碳化矽半導體裝置之製造方法,其中上述 第1金屬元素係選自由鎳、鈦、鋁、鉑、鎢、及鈀所組 成之群中之一種元素。 6. —種碳化矽半導體裝置之製造方法,該碳化矽半導體裝 置具有歐姆電極’該製造方法係包括以下步驟: 步驟(S20),形成包含碳化石夕之ye層(11); 於上述SiC層(11)之一個主表面上,形成包含一種第i 金屬元素且不含有碳原子之第1金屬層(12); 於上述第1金屬層(12)之與上述SiC層(u)對向之表面 為相反側之表面上’形成包含矽及上述一種第1金屬元 素且不含有碳原子之Si金屬層(15);及 為了形成歐姆電極,對上述Sic層(11)、上述第丨金屬 層(12)、及上述Si金屬層(15)進行熱處理。 7·如請求項6之碳化矽半導體裝置之製造方法,其係進而 包括以下步驟: 於進行上述熱處理之步驟之前,在上述以金屬層(15) 之與上述第1金屬層(12)對向之表面為相反側之表面上, 形成包含一種第2金屬元素且不含有碳原子之第2金 (14)。 曰 8·如請求項7之碳化矽半導體裝置之製造方法,其中上述 第2金屬元素係選自由鈦、鋁、及鉻所組成之群中之一 種元素。 9.如請求項6之碳化矽半導體裝置之製造方法,其中於上 147784.doc 201137975 述熱處理之步驟中,在上述SiC層(11)之一個主表面上, 形成包含上述一種第1金屬元素與石夕之合金且含有碳原 子之含碳石夕化物層。 10. 如請求項6之碳化矽半導體裝置之製造方法,其中上述 第1.金屬元素係選自由錄、鈦、紹、始、鎢、及把所組 成之群中之一種元素。 11. 一種碳化矽半導體裝置,其包括: 包含碳化矽之SiC層(11);及 矽化物層(41,43) ’其配置於上述SiC層(11)之一個主 表面上,包含一種第1金屬元素與矽之合金,且在與上 述SiC層(11)對向之表面為相反側之表面層不含有碳原 子;且 上述SiC層(11)與上述矽化物層(41,43)係進行歐姆接 觸。 12. 如請求項11之碳化矽半導體裝置,其係進而包含上部石夕 化物層(44),其形成於上述矽化物層(41,43)之表面層 上,包含一種第2金屬元素與矽之合金,且在與上述矽 化物層(41,43)對向之表面為相反側之表面層不含有碳 原子。 13. 如請求項12之碳化矽半導體裝置,其中上述第2金屬元 素係選自由鈦、鋁、及鉻所組成之群中之一種元素。 14. 如請求項11之碳化矽半導體裝置,其中上述第丨金屬元 素係選自由鎳、鈦、鋁、鉑、鎢、及鈀所組成之群十之 一種元素。 147784.doc 201137975 15. —種碳化矽半導體裝置,其包括: 包含碳化矽之SiC層(11); 含碳矽化物層(41),其配置於上述SiC層(11)之一個主 表面上’包含一種第1金屬元素與矽之合金,且含有碳 原子;及 矽化物層(43),其配置於上述含碳矽化物層(41)之與 上述SiC層對向之表面為相反側之主表面上,包含上述 一種第1金屬元素與矽之合金,且在與上述含碳矽化物 層(41)對向之表面為相反側之表面層不含有碳原子;且 上述SiC層(11)與上述含碳矽化物層(41)係進行歐姆接 觸。 16. 如請求項15之碳化矽半導體裝置,其係進而包含上部石夕 化物層(44) ’其形成於上述矽化物層(43)之表面層上, 包含一種第2金屬元素與矽之合金,且在與上述碎化物 層(43)對向之表面為相反側之表面層不含有碳原子。 17. 如請求項16之碳化矽半導體裝置,其中上述第2金屬元 素係選自由鈦、鋁、及鉻所組成之群中之一種元素。 18. 如請求項15之碳化石夕半導體裝置,其中上述第1金屬元 素係選自由鎳、鈦、鋁、鉑、鎢、及鈀所組成之群中之 一種元素。 147784.doc201137975 VII. Patent application scope: 1. A method for manufacturing a tantalum carbide semiconductor device, the tantalum carbide semiconductor device having an ohmic electrode, the manufacturing method comprising: step (S20) 'forming a Sic layer (11) comprising tantalum carbide; (S30) forming a first metal layer (12) containing a first metal element and containing no carbon atoms on one main surface of the siC layer (11); and step (S4〇) 'to the first metal layer (12) forming a Si layer (13, 15, 27) containing germanium and containing no carbon atoms on the surface opposite to the surface opposite to the (2) layer (11); and the step (S5 0), The sic layer (u), the first metal layer (I2), and the Si layer (13) are heat-treated to form an ohmic electrode. 2. The method for manufacturing a silicon carbide semiconductor device according to claim 1, further The method includes the following steps: Step (S45): forming a surface on the surface opposite to the surface of the Si layer (13) opposite to the second metal layer (12) before the step of performing the heat treatment (S5〇) The second metal element does not contain the carbon atom The metal layer (14). The method for producing a silicon carbide semiconductor device according to claim 2, wherein the second metal element (M) is one element selected from the group consisting of titanium, aluminum, and chromium. The method of manufacturing a tantalum carbide semiconductor device according to claim 1, wherein in the step (S5) of performing the heat treatment, a first metal element and a tantalum are formed on one main surface of the Sic layer (u). 147784.doc 201137975 An alloy comprising a carbon atom-containing carbide layer (41). The method of manufacturing a silicon carbide semiconductor device according to claim 1, wherein the first metal element is selected from the group consisting of nickel, titanium, aluminum, An element of a group consisting of platinum, tungsten, and palladium. 6. A method of manufacturing a tantalum carbide semiconductor device having an ohmic electrode. The manufacturing method includes the following steps: Step (S20), forming a layer of carbonaceous stone ye (11); on a main surface of the SiC layer (11), forming a first metal layer (12) containing an ith metal element and containing no carbon atoms; Forming a Si metal layer (15) comprising germanium and one of the first metal elements and not containing carbon atoms on the surface of the layer (12) opposite to the surface opposite to the SiC layer (u); and forming an ohmic layer The electrode is heat-treated to the Sic layer (11), the second metal layer (12), and the Si metal layer (15). The method for manufacturing a silicon carbide semiconductor device according to claim 6, further comprising the following Step: before the step of performing the heat treatment, forming a second metal element and not containing carbon on the surface of the metal layer (15) opposite to the surface opposite to the first metal layer (12) The second gold of the atom (14). The method of producing a silicon carbide semiconductor device according to claim 7, wherein the second metal element is one selected from the group consisting of titanium, aluminum, and chromium. 9. The method of manufacturing a silicon carbide semiconductor device according to claim 6, wherein in the step of heat treatment of the above 147784.doc 201137975, forming a first metal element and a first metal element on one main surface of the SiC layer (11) A stone-like alloy layer containing carbon atoms. 10. The method of manufacturing a niobium carbide semiconductor device according to claim 6, wherein the first metal element is one selected from the group consisting of ruthenium, titanium, samarium, tungsten, and the like. A silicon carbide semiconductor device comprising: a SiC layer (11) comprising tantalum carbide; and a vaporized layer (41, 43) disposed on one major surface of the SiC layer (11), comprising a first An alloy of a metal element and ruthenium, and a surface layer on the opposite side of the surface opposite to the SiC layer (11) does not contain carbon atoms; and the SiC layer (11) and the vaporization layer (41, 43) are Ohmic contact. 12. The tantalum carbide semiconductor device of claim 11, further comprising an upper layer (44) formed on a surface layer of the germanide layer (41, 43), comprising a second metal element and germanium The alloy layer does not contain carbon atoms on the surface layer opposite to the surface opposite to the above-described telluride layer (41, 43). 13. The silicon carbide semiconductor device according to claim 12, wherein the second metal element is one selected from the group consisting of titanium, aluminum, and chromium. 14. The tantalum carbide semiconductor device of claim 11, wherein the second metal element is selected from the group consisting of nickel, titanium, aluminum, platinum, tungsten, and palladium. 147784.doc 201137975 15. A silicon carbide semiconductor device comprising: a SiC layer (11) comprising tantalum carbide; a carbon-containing germanide layer (41) disposed on a major surface of the SiC layer (11) An alloy comprising a first metal element and ruthenium and comprising a carbon atom; and a ruthenide layer (43) disposed on the opposite side of the surface of the carbon-containing ruthenide layer (41) opposite to the SiC layer a surface layer comprising the first metal element and the bismuth alloy, and the surface layer opposite to the surface opposite to the carbon-containing hydrazine layer (41) does not contain carbon atoms; and the SiC layer (11) and The carbon-containing telluride layer (41) is subjected to ohmic contact. 16. The silicon carbide semiconductor device of claim 15, further comprising an upper layer (44) formed on the surface layer of the vapor layer (43), comprising a second metal element and an alloy of tantalum The surface layer on the opposite side to the surface opposite to the above-mentioned fragment (43) does not contain carbon atoms. 17. The silicon carbide semiconductor device according to claim 16, wherein the second metal element is one selected from the group consisting of titanium, aluminum, and chromium. 18. The carbonized carbide semiconductor device according to claim 15, wherein the first metal element is one element selected from the group consisting of nickel, titanium, aluminum, platinum, tungsten, and palladium. 147784.doc
TW99112237A 2010-04-19 2010-04-19 Silicon carbide semiconductor device and its manufacturing method TW201137975A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW99112237A TW201137975A (en) 2010-04-19 2010-04-19 Silicon carbide semiconductor device and its manufacturing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW99112237A TW201137975A (en) 2010-04-19 2010-04-19 Silicon carbide semiconductor device and its manufacturing method

Publications (1)

Publication Number Publication Date
TW201137975A true TW201137975A (en) 2011-11-01

Family

ID=46759720

Family Applications (1)

Application Number Title Priority Date Filing Date
TW99112237A TW201137975A (en) 2010-04-19 2010-04-19 Silicon carbide semiconductor device and its manufacturing method

Country Status (1)

Country Link
TW (1) TW201137975A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI777752B (en) * 2020-08-27 2022-09-11 日商新電元工業股份有限公司 wide bandgap semiconductor device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI777752B (en) * 2020-08-27 2022-09-11 日商新電元工業股份有限公司 wide bandgap semiconductor device

Similar Documents

Publication Publication Date Title
US9437682B2 (en) Semiconductor device and semiconductor device manufacturing method
US9129804B2 (en) Silicon carbide semiconductor device and method for manufacturing same
JP5391643B2 (en) Silicon carbide semiconductor device and manufacturing method thereof
US8487318B2 (en) Semiconductor device and manufacturing method thereof
JP6099298B2 (en) SiC semiconductor device and manufacturing method thereof
JP5728339B2 (en) Semiconductor device and manufacturing method of semiconductor device
JP5286677B2 (en) Method for forming ohmic electrode on P-type 4H-SiC substrate
JP5449786B2 (en) Silicon carbide semiconductor device and method for manufacturing silicon carbide semiconductor device
US9793354B2 (en) Semiconductor device and method of manufacturing the same
TW201133573A (en) Method for manufacturing semiconductor device
TW201222820A (en) Semiconductor device and manufacturing method therefor
JP2015076592A (en) Silicon carbide semiconductor device and manufacturing method of the same
JP2020129583A (en) Semiconductor device and semiconductor device manufacturing method
JP2009049130A (en) Silicon carbide semiconductor device, method of manufacturing the same, and silicon carbide device
JP6160541B2 (en) Method for manufacturing silicon carbide semiconductor device
JP2014216529A (en) Silicon carbide semiconductor device manufacturing method
JP2010062348A (en) Silicon carbide semiconductor substrate and method of manufacturing the same
TW201137975A (en) Silicon carbide semiconductor device and its manufacturing method
JP2017028219A (en) Silicon carbide semiconductor device and manufacturing method of the same
US8633101B2 (en) Semiconductor device and manufacturing method of semiconductor device
JP6724444B2 (en) Silicon carbide semiconductor device and method for manufacturing silicon carbide semiconductor device
JP2017168685A (en) Method for manufacturing silicon carbide semiconductor device