TWI277139B - Improved process for deposition of semiconductor filme - Google Patents

Improved process for deposition of semiconductor filme Download PDF

Info

Publication number
TWI277139B
TWI277139B TW91101956A TW91101956A TWI277139B TW I277139 B TWI277139 B TW I277139B TW 91101956 A TW91101956 A TW 91101956A TW 91101956 A TW91101956 A TW 91101956A TW I277139 B TWI277139 B TW I277139B
Authority
TW
Taiwan
Prior art keywords
film
ruthenium
depositing
temperature
trioxane
Prior art date
Application number
TW91101956A
Other languages
Chinese (zh)
Inventor
Michael A Todd
Mark Hawkins
Original Assignee
Asm Inc
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 Asm Inc filed Critical Asm Inc
Application granted granted Critical
Publication of TWI277139B publication Critical patent/TWI277139B/en

Links

Landscapes

  • Chemical Vapour Deposition (AREA)
  • Electrodes Of Semiconductors (AREA)

Abstract

Chemical vapor deposition processes utilize chemical precursors that allow for the deposition of thin films to be conducted at or near the mass transport limited regime. The processes have high deposition rates yet produce more uniform films, both compositionally and in thickness, than films prepared using conventional chemical precursors. In preferred embodiments, trisilane is employed to deposit thin films containing silicon are useful in the semiconductor industry in various applications such as transistor gate electrodes.

Description

12771 ^826pifl doc/012 玖、發明說明: 發明領域 本發明是有關於沉積半導體薄膜譬如含矽、鍺或是碳 之薄膜的積體電路製造法,且特別是有關於一種半導體薄 fe之改良式沉積製程’以於化學氣相沉積系統(chemical vapor deposition system)中製造具有較大厚度與組成均勻性 (compositional uniformity)的材料。 發明背景 畠微 fe子兀件(microelectroic device)的尺寸(dimension) 愈變愈小時,用於其製造中之材料的物理與化學特性的重 要性也就愈變愈重要。特別合適的是那些使用已有的製造 工具之先進的可結合現行世代的元件之材料。舉例來說, 於雙載子(Bipolar)與雙載子互補式金氧半導體(BiCMOS)元 件製程中適於使用混合磊晶矽鍺(SinGeJ與矽鍺碳化合物 (Su+yGe/y)合金。這些先進的合金材料作爲異質接合雙載 子電晶體(heterojunction bipolar transistor,簡稱 HBT)之基 底(base)、作爲雙載子互補式金氧半導體(BiCMOS)元件的 電阻(resistor)以及作爲互補式金氧半導體(CMOS)元件的閘 電極(gate electrode)或用於其他各種積體電路元件均具有功 效。 習知用於單晶(single crystal)、非晶(amorphous)或複晶 砂(polycrystalline silicon)、砂鍺(SiGe)與砂鍺碳(SiGeC)合 金的製程典型上是用不是低壓(low pressure,簡稱LP)就是 超高真空(ultra-high vacuum,簡稱UHV)批(batch)熱製程或 12771¾ 46pifl.doc/012 疋:單晶圓製程(single wafer process)。單晶圓製程變得逐漸 重要’但仍存在一些問題。例如,晶圓中(within-wafer)與 晶圓間(wafer-to-wafer)的均勻性、沉積速率(deposition rate) 與製程再現性(repeataMlhy)仍然是傳統單一晶圓製程所關 注的事’特別是臨場(in Situ)摻雜的半導體膜。當晶圓在尺 寸上持續增加時(現在300mm的晶圓已被整合於製造製程 中),持續的均勻性以變得更具挑戰性。 曰本專利申請公開號S60-43485已揭露在300°C使用 三矽烷製造非晶矽薄膜以用於光電流應用(photovoltaic application)。而日本專利申請公開號H5-6291 1已揭露在500 °〇或更低的溫度下使用三矽烷與鍺烷製造薄膜。另外,曰 本專利申請公開號 H3-91239、H3-185817、H3-187215 與 H02-155225均揭露使用二矽烷,或是有些提及三矽烷。 技術方面皆已注意到於相當低的沉積溫度下使用二矽 烷與三矽烷製造非晶氫化矽。然而,最好能不犧牲優良的 均勻性於較高沉積速率下在表面上沉積半導體材料如沉積 的矽、低氫含量非晶矽與矽鍺之製程仍是必要的。 發明槪要 本發明已揭露一種半導體薄膜之改良式沉積製程,以 製造含矽與含鍺膜。此方法教示於化學氣相沉積製程中使 用如較高級砂院(higher-order silane)與/或較高級鍺院,已 提供含矽膜之改良製程,特別適用於半導體產業的矽、矽 鍺(SiGe)、矽鍺碳(SiGeC)合金薄膜。這些化學前驅物 (chemical precursor)已降低有關砂院、鍺院與習知碳源分子 6 12771说 pifl.doc/012 (carbon_source molecule)的熱穩定度(thermal stability)。 因此,本發明的目的在提供一種半導體薄膜之改良式 沉積製程,係使用特殊的前驅物,以使沉積製程在與習知 前驅物相同溫度下被引導至較接近或是於一質量傳輸限定 成長規律(mass transport limited growth regime)的範圍內。 於此規範內可避免從屬於譬如討厭的元素濃度梯度 (concentration gradient)與變化的膜沉積速率之不均句性 (non-uniformity)以及後續厚度的不一致性的溫度。較佳的 化學前驅物包括三矽烷與三矽烷結合二鍺烷。在低於習知 化學前驅物所採用的溫度下,可以較高的膜沉積速率獲致 均勻的沉積。 本發明的另一目的在提供一種半導體薄膜之改良式沉 積製程,係調整較佳前驅物的流速與溫度之關係’以獲得 較使用習知的前驅物(如矽烷)沉積更高沉積速率下相同或 更佳的均勻性。已知三矽烷優於矽烷,尤其是可應用於如 積集電晶體中的主動層(active layer)之含矽層的沉積。 本發明的又一目的在提供一種半導體薄膜之改良式沉 積製程,係教示一種方法用以逐步或是不斷變化地改變如 溫度、溫度分布、壓力、反應物流速與反應物分壓的製程 參數,藉以降低或消除討厭的元素濃度梯度、厚度的不一 致性與變化的膜沉積速率。這些方法可用以結合較高級矽 烷與/或鍺烷的使用。 圖式之簡單說明 爲讓本發明之上述和其他目的、特徵和優點能更明顯 12771撚 pifl.doc/012 易懂,下文特舉一較佳實施例,並配合所附圖式,作詳細 說明如下: 弟1圖所不係依照本發明—^較佳實施例之一種形成一 閘極疊層的製造流程步驟圖; 第2圖所示係依照本發明一較佳實施例之一閘極疊層 的示意圖; 第3圖所示係依照本發明一較佳實施例之一種於沉積 期間改變溫度設定點(set pomt)的製造流程步驟圖; 第4圖所示係一較佳矽鍺膜的膜厚與測量點 (measurement site)之關係圖; 第5圖所示係用矽烷與鍺烷沉積的一矽鍺膜的掃描式 電子顯微相片(scanning electron photomicrograph); 第6圖所示係依照第5圖之矽鍺膜剖面的掃描式電子 顯微相片; 第7圖所示係用三矽烷與鍺烷沉積的一矽鍺膜的掃描 式電子顯微相片; 第8圖所示係依照第7圖之矽鍺膜剖面的掃描式電子 顯微相片; 第9圖所示係一較佳氮化矽膜剖面的穿透式電子顯微 相片(transmission electron photomicrograph); 第10圖所示爲矽烷、二矽烷與三矽烷於下述條件下 得到的Arrhenius圖; 第11圖所示爲在600°C、40Torr下,於氧化矽基底上 月旲沉積速率與二砂院(SilcoreTMMjfE速的關係圖, 8 12771涊6 pifl.doc/012 第12圖所示爲在650°C、40ΤΟΙΓ下使用三矽烷(Silcore™) 於各個時間中膜厚與位置(posmon)的關係圖; 第13圖所示爲使用三矽烷(trisUane)沉積之沉積速率 與二硼烷流量(diboraneflow)的關係圖;以及 第14圖所示爲在600°C、40ΤΟΠ*下用三矽烷沉積的非 晶石夕膜之拉塞福回向散射光譜(Rutherford Backscattering Spectrometry,簡稱 RBS)彈性反衝偵測(Elastic Recoil Detection,簡稱 ERD)光譜; 第15圖所示爲一系列在60(TC、650°C、700°C與750 它下(分別從底部至頂部)用三矽烷沉積的膜之χ光繞射圖 (X-ray diffraction pattern); 第16圖所示係一複晶矽膜剖面的穿透式電子顯微相 片; 第17圖所示係一複晶矽膜剖面的選區繞射(selected area diffraction,簡稱 SAD)圖; 第18圖所示係一保角非晶矽膜剖面的掃描式電子顯 微相片; 第19圖所示係一氮化矽膜之拉塞福回向散射光譜; 以及 第20圖所示係一氮化矽膜之拉塞福回向散射光譜彈 性反衝偵測光譜。 標記之簡單說明: 100 :形成閘介電層 110 :淸潔閘介電層表面 12771l ifl.doc/012 120 :沉積含矽薄膜 130 :沉積金屬層 140 :圖案化閘電極 150 :繼續積體電路的製造 200 :閘極疊層 210 :閘介電層 220 :半導體基底 230 :含矽膜 240 :金屬層 300 :準備基底 310 :調整設定點至600°C 320 :用98%二矽烷與2°/〇二鍺烷沉積矽鍺 330 :用85%三矽烷與15%二鍺烷沉積矽鍺 340,360 :降低3°C設定點 350 :用75%三矽烷與25%二鍺烷沉積矽鍺 370 :用65%三矽烷與35%二鍺烷沉積矽鍺 380 :用85%三矽烷、12%二鍺烷、2°/〇二硼烷與1°/〇 二矽甲烷沉積摻雜硼與摻雜碳的矽鍺 390 :淸潔反應器 400 :用90%三矽烷與10%二鍺烷沉積矽鍺BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of fabricating a semiconductor circuit, such as a thin film containing germanium, germanium or carbon, and in particular to an improved semiconductor thin film. The deposition process 'produces a material having a large thickness and compositional uniformity in a chemical vapor deposition system. BACKGROUND OF THE INVENTION As the dimensions of microelectroic devices become smaller and smaller, the importance of the physical and chemical properties of the materials used in their manufacture becomes more and more important. Particularly suitable are those materials that use existing manufacturing tools to incorporate elements of the current generation. For example, mixed epitaxial germanium (SinGeJ and tantalum carbon compound (Su+yGe/y) alloys are suitable for use in bipolar and bipolar complementary metal oxide semiconductor (BiCMOS) devices. These advanced alloy materials serve as a base for a heterojunction bipolar transistor (HBT), a resistor for a bipolar complementary metal oxide semiconductor (BiCMOS) device, and as a complementary gold. A gate electrode of an oxygen semiconductor (CMOS) device or a device for use in various other integrated circuit components is known to be useful for single crystal, amorphous or polycrystalline silicon. The process of sand germanium (SiGe) and sand germanium carbon (SiGeC) alloys is typically a low pressure (LP) or ultra-high vacuum (UHV) batch thermal process or 127713⁄4 46pifl.doc/012 疋: Single wafer process. Single-wafer process is becoming more important' but there are still some problems. For example, in-wafer and wafer-to-wafer (wafer-to -wafer) Uniformity, deposition rate, and process reproducibility (repeataMlhy) remain the focus of traditional single-wafer processes, especially in Situ-doped semiconductor films. As wafers continue to increase in size Time (now 300mm wafers have been integrated into the manufacturing process), continuous uniformity to become more challenging. 专利 Patent Application Publication No. S60-43485 has disclosed the use of trioxane to make amorphous germanium at 300 ° C The film is used for a photovoltaic application, and Japanese Patent Application Laid-Open No. H5-6291 1 discloses that a film is produced using trioxane and decane at a temperature of 500 ° C or lower. No. H3-91239, H3-185817, H3-187215 and H02-155225 both disclose the use of dioxane, or some mention of trioxane. Technical aspects have been noted using dioxane and trioxane at relatively low deposition temperatures. Amorphous hydrogenated ruthenium. However, it is preferred to deposit semiconductor materials such as deposited tantalum, low hydrogen content amorphous tantalum and tantalum on the surface at a higher deposition rate without sacrificing good uniformity. It is still necessary. SUMMARY OF THE INVENTION The present invention has disclosed an improved deposition process for semiconductor thin films to produce germanium-containing and germanium-containing films. This method teaches the use of higher-order silanes and/or higher-grade brothels in chemical vapor deposition processes, and has provided improved processes for ruthenium-containing films, particularly for the semiconductor industry. SiGe), tantalum carbon (SiGeC) alloy film. These chemical precursors have reduced the thermal stability of pifl.doc/012 (carbon_source molecule) in sandstones, brothels, and conventional carbon source molecules. Accordingly, it is an object of the present invention to provide an improved deposition process for a semiconductor thin film using a special precursor such that the deposition process is directed closer to the same temperature as the conventional precursor or to a mass transfer limited growth. Within the scope of the mass transport limited growth regime. Temperatures that are subordinate to the non-uniformity of the concentration gradient and the varying film deposition rate, as well as subsequent thickness inconsistencies, can be avoided within this specification. Preferred chemical precursors include trioxane and trioxane in combination with dioxane. At a lower temperature than that used in conventional chemical precursors, uniform deposition can be achieved at higher film deposition rates. Another object of the present invention is to provide an improved deposition process for a semiconductor film that adjusts the flow rate versus temperature of a preferred precursor to achieve the same higher deposition rate than conventional precursors (e.g., decane) deposition. Or better uniformity. It is known that trioxane is superior to decane, and in particular, it can be applied to the deposition of a ruthenium-containing layer such as an active layer in an integrated crystal. It is yet another object of the present invention to provide an improved deposition process for a semiconductor thin film that teaches a method for progressively or continually changing process parameters such as temperature, temperature profile, pressure, reactant flow rate, and reactant partial pressure. In order to reduce or eliminate the objection concentration gradient, thickness inconsistency and varying film deposition rate. These methods can be used in conjunction with the use of higher decane and/or decane. BRIEF DESCRIPTION OF THE DRAWINGS The above and other objects, features, and advantages of the present invention will become more apparent from the description of the appended claims. The following is a schematic diagram of a manufacturing process for forming a gate stack according to the preferred embodiment of the present invention; and FIG. 2 is a schematic diagram of a gate stack according to a preferred embodiment of the present invention. FIG. 3 is a schematic diagram of a manufacturing process for changing a set pomt during deposition according to a preferred embodiment of the present invention; FIG. 4 is a preferred enamel film. The relationship between the film thickness and the measurement site; Figure 5 shows a scanning electron photomicrograph of a film deposited with decane and decane; Figure 6 shows Scanning electron micrograph of the ruthenium film section of Figure 5; Figure 7 is a scanning electron micrograph of a ruthenium film deposited with trioxane and decane; Figure 8 shows Scanning electron microscopic phase of the 矽锗 film profile of Figure 7 Figure 9 shows a transmission electron photomicrograph of a preferred tantalum nitride film profile; Figure 10 shows Arrhenius obtained from decane, dioxane and trioxane under the following conditions. Fig. 11 is a graph showing the deposition rate of the ruthenium on the ruthenium oxide substrate at 600 ° C and 40 Torr, and the relationship between the rate of the SilicaTM MjfE, 8 12771涊6 pifl.doc/012, Fig. 12 The relationship between film thickness and position (posmon) at various times using silane (SilcoreTM) at 650 ° C, 40 ;; Figure 13 shows the deposition rate and diborane using trisane deposition Diagram of flow (diboraneflow); and Figure 14 shows the Rutherford Backscattering Spectrometry (RBS) elasticity of amorphous austenite deposited with trioxane at 600 ° C, 40 °* Elastic Recoil Detection (ERD) spectrum; Figure 15 shows a series of depositions with trioxane at 60 (TC, 650 ° C, 700 ° C and 750 (from bottom to top, respectively) X-ray diffraction patter n); Figure 16 is a transmission electron micrograph of a polycrystalline tantalum film section; Figure 17 is a selected area diffraction (SAD) diagram of a polycrystalline tantalum film section; Figure 18 is a scanning electron micrograph of a conformal amorphous ruthenium film; Figure 19 is a Latifah backscattering spectrum of a tantalum nitride film; and Figure 20 is a The elastic backlash detection spectrum of the Latifah backscattered spectrum of the tantalum nitride film. Brief description of the mark: 100: forming the gate dielectric layer 110: the surface of the dielectric layer of the stencil gate 12771l ifl.doc/012 120: depositing the germanium-containing film 130: depositing the metal layer 140: patterning the gate electrode 150: continuing the integrated circuit Fabrication 200: gate stack 210: gate dielectric layer 220: semiconductor substrate 230: germanium containing film 240: metal layer 300: preparation substrate 310: adjustment set point to 600 ° C 320: with 98% dioxane and 2 ° /〇Dioxane deposit 矽锗330: Deposition with 85% trioxane and 15% dioxane 矽锗340,360: Lower 3 °C set point 350: Deposition with 75% trioxane and 25% dioxane 370: Deposition of 矽锗380 with 65% trioxane and 35% dioxane: boron doping with 85% trioxane, 12% dioxane, 2°/〇 diborane and 1°/〇 dimethane Carbon-doped 矽锗390: 淸洁反应400: deposited with 90% trioxane and 10% dioxane

410 :增加設定點至650°C 420 :用100%三矽烷沉積矽頂蓋層 430 :將晶圓移出反應器 10 I2771396pin,〇c/ol2 實施例 由於熱與溫度的控制系統的限制,動態的溫度變化 (dynamic temperature variation)在以化學氣相沉積法 (chemical vapor deposition,簡稱CVD)於基底表面上膜之沉 積的均勻性(uniformity)上扮演重要角色。通常沉積膜最好 能夠在厚度與元素組成上都儘可能的均勻,但是現有的製 程趨向製造不均勻的膜以改善程度(degree)。這種非均勻性 總是導因於橫越基底表面之溫度變動,因爲基底表面溫度 影響沉積速率與最終膜(the resultmg fUm)的組成。而且於 其他製程參數不完美的控制包括氣體流速與總壓也被相信 是促使膜物理特性上的不均勻。 追求均勻性往往是憑經驗調整沉積條件例如氣體流 速、基底轉速、加熱元件的能量分布等等,以達到一全面 均勻厚度的要求膜。這在不同基底上且在每一基底需在不 同的預選沉積條件(pre-selected set of deposition condition) 下藉由第一沉積大量的膜來達成。然後,量測每一個膜的 厚度變化並分析結果,以確定能消除厚度變化的條件。然 而,已瞭解憑經驗的製程在整個製程中並不需要達到均勻 的溫度分布,或者應該說是於一特殊反應溫度'設定點(set point/製程有效地平均時間(time-average)由溫度變化引起 的厚度變化。 因此,經驗法則不需要於整個沉積製程中產生均勻的 溫度橫越基底。換句話說,因爲在包括橫過膜面與穿過膜 厚的立體中要求組成均質性(homogeneity)或者至少控制, ►pifl.doc/012 而引起組成變化的爭議。這是因爲很多膜含有摻質 (dopant),而這些摻質的程度(level)會影-膜的電子特性 (electronic property)。同樣也會招致組成中其他的非均勻 性。 較佳實施例提供解決上述問題的製程,而這些製程可 單獨使用或是最好是能一起使用。一個影響化學前驅物 (chemical precursor)使用的製程相聿交方令習知前趨物在相同溫 度下提供膜沉積,以被實際於一質量傳輸限定成長規範 (mass transport limited growth regime)中處理。對一特定的 化學前驅物而言,質量傳輸限定規範是一種溫度範圍,且 於此溫度範圍中膜沉積速率是與溫度無關的。沉積速率實 質上在這個溫度範圍中不被橫過基底表面的小小溫度變化 所影響,而這些變化導致溫度維持在或是接近於質量傳輸 限定規範。這使得膜的製作比使用習知化學前驅物於相同 溫度下沉積膜更爲均勻,例如其擁有較高組成上的均勻性 與厚度均勻性。這是因爲習知的前驅物,以於質量傳輸限 定規範中沉積。 熟悉此技術者應可理解質量傳輸限定規範的溫度範圍 可藉特定的前驅物與反應條件來決定,並且以一 Arrhenius 圖作圖解。以化學前驅物三矽烷(trisilane)來說,從根據溫 度的'沉積速率(temperature-dependent deposition rate)至無關 溫度的沉積速率(temperature-independent deposition rate)之 轉變點(traimUon pomt)較矽烷或二矽烷的轉變點低得多, 如第10圖所示之Arrhenius圖。圖中較低區域至轉變點具 12 12771¾ 846pifl.doc/012 有一明顯向上的線性斜面(upward linear slope),這代表三 矽烷在較低的溫度範圍中的沉積與溫度有極大關聯,因此 其並非於質量傳輸限定規範中。舉例來說,如第10圖所 示三矽烷在每公分25標準立方公分(sccm)的流速與40Torr 壓力之條件,且於溫度低於525°C下沉積並不是質量傳輸 限定(例如在動力規範中)。相較之下,在圖中之轉變點以 上的區域幾乎是平的,這代表三矽烷在此溫度範圍中的沉 積與溫度無關,因此其是於質量傳輸限定規範中。例如, 第10圖顯示在溫度約620°C或以上時,三矽烷沉積明顯是 質量傳輸限定的。可知轉變將發生在Arrhenius圖中超過 一具有下降斜面(declining slope)之溫度範圍,即代表在這 個溫度範圍中三矽烷之沉積實際上是無關於溫度的,而是 接近質量傳輸限定規範。例如,第10圖係顯示在溫度約525 °C或大於此溫度下三矽烷的沉積質量上是質量傳輸限定 的。可知轉變點在較高流速下也許能稍微增加並且在較低 流速下稍微減少。例如,從根據溫度的沉積至實際上質量 傳輸限定沉積之轉變點於三矽烷流速增加時會轉移至較高 溫已被驗證過。因此,在當代製造中實際上爲其他理由(例 如爲維持晶體特性之熱預算的保持、控制摻質輪廓等等) 而在合意的溫度下能夠使三矽烷的使用爲質量傳輸限定沉 積。 種種含砂與含鍺化學前驅物可適用於在此揭示的膜沉 積製程’以預備包含矽與鍺之含矽膜、含鍺膜與合金膜, 例如砂鍺(SiGe ’於此並非意指等化學計量)膜。這些化學 13 ►pifl.doc/012 則驅物也可用以與碳源(carbon source)相結合以預備合金 膜’例如砂鍺碳(SiGeC,於此並非意指等化學計量)合金薄 膜°適用於本發明的較佳含矽化學前驅物包括矽的較高級 非鹵化氫化物(higher-order,non-halogenated hydride),特別 是分子式爲SinH2n+2的矽烷,其中n zz 2〜6。獨特的範例包 括二矽烷(H3SiSiH3)、三矽烷(H3SiSiH2SiH3)與四矽烷 (HeSiHJiHAHD。三矽烷(也寫作Sl3H8)是最佳用於達到 揮發性(volatility)與活性(reactivity)的平衡。於相對低溫下 進行實質上或是接近質量傳輸限定的沉積對矽鍺沉積而言 是較佳的(但非必要的)。適用於急迫的發明中隻較佳含鍺 化學前驅物包括分子式爲GenH2n+2的較高級鍺烷,其中n = 2〜3。其他配置(arrangement)中的鍺源可以包括 (H3Ge)(GeH2)x(GeH3),其中X = 0〜2。獨特的範例包括二鍺 烷(H3GeGeH3)、三鍺烷(H3GeGeH2GeH3)與四鍺烷 (H3GeGeH2GeH2GeH3)。 在一較佳實施例中,化學前驅物用以與一碳源相結 合。較佳的碳源包含矽基甲烷(silylmethane)[(H3Si)4_xCRx], 其中X = 0〜3、R = Η與/或D-構型。而較佳的矽基甲烷包 括二矽甲烷、三矽甲烷與四矽甲烷(X = 0〜2),而最佳者爲 四矽甲烷。附加的較佳碳源包括碳氫化合物如甲烷、乙烷、 丙烷、丁烷等等,以及一氧化碳、二氧化碳與氰化氫 (hydrogen cyanide,HCN)。這些化學前驅物與碳源可以從 商業性來源(commercial source)獲得或是由熟悉此技術者所 知的方法合成得到。含矽膜譬如MC、SiNC或是SiOCX上 14 12771¾ 46pifl.doc/012 述無一是意指獨特的等化學計量)在半導體製造業中具有 多種用途,例如作爲蝕刻中止層(etch stop layer)、硬罩幕 (hard mask)與鈍化層(passivation layer)。 月旲最好在貫際上是質量傳輸限定規範中的一溫度下用 獨特的化學前驅物進行沉積。對任何獨特的化學前驅物以 及反應條件而言’質量傳輸限定規範可在不同溫度下藉由 憑經驗源於沉積資料(empirically-derived from deposition data)的Airhenius圖來決定。而最佳矽前驅物-三矽烷於特 別條件下的Arrhenius圖被添加如上述第1〇圖。 除了使用於此描述的較佳化學前驅物(特別是三矽烷) 以及爲此前驅物選擇一於質量傳輸限定規範中的溫度之 外’用第一製程的沉積最好包含適當的其他沉積參數選 擇’特別是氣體流速。結合於質量傳輸限定規範中之較佳 的氣體流速選擇已知,可在與砂院相較高得多的沉積速率 下產生具有高度均勻性的膜。爲了在動力規範中的溫度下 用砂烷沉積’膜的均勻性主要依據溫度控制器上的設定點 與(一更小範圍)氣體流速控制設定點。相較之下,爲在質 量傳輸限定規範的溫度下進行包含較高級矽烷的沉積,已 知濫度控制器上的設定點與氣體流速控制設定點之敏感度 (sensiUWty)是相反的。舉例來說,在質量傳輸限定規範的 溫度下用三矽烷的沉積提昇溫度控制器上的設定點將會有 較提昇氣體流速控制器上的設定點小得多的影響。 當實施於此所述的沉積時,較佳的最終膜比對照膜 (comparable film)更均勻。除了用矽烷取代較高級矽烷以及 15 I2771^46pif,doc/〇, /或是用鍺烷取代較高級鍺烷之外,而用於此的“對照膜,,是 以與討論中發明的膜在全部有意義的觀點中實質相同的方 式製作,並且每一個膜的沉積製程是分別考慮上述於敏感 度對溫度與氣體流速控制設定點中之不同處進行調整。更 特別的是當比較過不同層的結果時,由下列規範(the following standard)測量所得之厚度均勻性··利用一橫過晶 圓的任選直徑並沿著直徑49個點測量沉積層厚度。而在 晶圓圓周3mm被排除區域(exclusion zone)中是不用測量 的。然後,這49個點中厚度測量所得的範圍(例如是± 6 埃)是受到由這49個點之間的最大厚度測量加上最小厚度 測量的總和所分配。而均勻性於此是以百分比表示。更好 的是已知此處所述使用前驅物的方法可導致異常高的沉積 速率,而且更驚人地是能獲得特別高的均勻性與平滑度 (smoothness) 〇 舉例來說,用三矽烷製作的較佳多晶矽膜在較高沉積 速率下有比由用矽烷於三矽烷所用之相同溫度下的製程所 製得的對照膜較大的均勻性。同樣地,發明人已分別用三 石夕院形成非晶砂(amorphous silicon’簡稱a-Si)層與晶晶石夕 (epitaxial silicon,簡稱epi-Si)層,以顯出與砂院沉積層相 較下較佳的均勻性。請見第15〜18圖以及下面對應的敘述。 同樣地,用較高級鍺烷製作較佳矽鍺膜以及此矽鍺膜具有 比由用鍺烷替代較高級鍺烷所製得的對照膜較大的均勻 性。而且,用所揭露的矽與鍺源也可在較低反應溫度下獲 得較高的沉積速率。 16 ►pifl.doc/012 第11圖所示爲在600°C沉積溫度、40 Torr壓力下, 沉積速率與三矽烷(於某些圖示中是以Silcore™表示之)流 速之線性關係圖。這個線性(lineanty)更表示再這些條件下 的三砂院沉積實質上是或接近質量傳輸限定,以及進一步 表示其較氧化物更低得多的成核時間(nucleation time)。第 12圖則是在相同條件下(650°C、40Torr)使用三矽烷沉積的 膜之膜厚與測量地點(she)的關係圖,只是沉積時間是從90 秒到15秒的範圍中作變化。第12圖顯示在一固定三砂院 流速下於一大範圍沉積時間可獲得卓越的膜均勻性,其結 果所代表的不僅僅是平均時間的(time-averaged)而是導因於 前驅物的性質(nature)與選擇的條件,以及因爲那些層不顧 厚度仍能維持是均勻的,所以發射率(或是其他根據厚度 的溫度控制)效應並不會改變均勻性。第13圖所示爲在600 °C沉積溫度、40 Torr壓力並在二硼烷流速的範圍(從0到18〇 seem)下使用三矽烷與二硼烷(作爲摻質前驅物)所獲得的沉 積速率圖。第13圖所示爲使用三矽烷的沉積速率相當無 動於衷於摻質前驅物的流速。 以更恰當於減小熱穩定性(thermal stability)的較低溫 而言,較佳溫度範圍有趨向依據獨特化學前驅物的傾向。 對較局級砂院與較高級鍺院而言,當鏈長(chain-length)增 加率時較佳的是比較低的溫度。所以,二矽烷沉積的較佳 溫度範圍將有趨於較三矽烷高的傾向,而依次趨於較四砂 烷相對高的傾向等等。而鍺烷也有類似的趨勢。沉積三砂 烷之較佳沉積溫度約高於35CTC,更佳則約高於45Qt, 127電—2 以使最終膜中氫含量縮至最少。爲達到接近或於質量傳輸 限定規範中,溫度更好的是要維持約高於525°C,甚至是 約高於550°C,而最號是約高於600°C。製程可於700°C溫 度以上施行,但是最好在約700°C或低於此溫度。因此較 佳溫度是在450°C到約700°C的範圍內,更佳溫度是在525 °C到約650°C的範圍內。對任何獨特的化學前驅物或其混 合物(mixture)而言,最佳溫度範圍可採用於此提供的指導 方針(guideline)而藉一般實驗得知。已知所列出之溫度最 好是用於熱化學氣相沉積法(thermal CVD)。而根據對應用 而言可接受的氫合倂(incorporation)程度,較低溫是適於電 獎促進沉積製程(plasma assisted deposition process)。 沉積溫度的選擇也可部分取決於沉積層所需之結晶度 (crystallinity)。例如佔優勢的結晶度矽可在約620°C〜80〇。(: 範圍中沉積,且此沉積顯然是於質量傳輸限定規範中。更 好的是在650°C〜750°C之間實施多晶矽沉積。較低的溫度 可用於非晶矽沉積,而溫度最好選擇以維持至少實質上爲 質量傳輸限定的(例如較佳條件在較高於525°C)。磊晶矽 大部分取決於發生沉積之表面的純度(purity)。那就是熟悉 此技術者可認知一極乾淨的單晶表面譬如先前沉積的磊晶 層或是單晶圓的上表面能夠在取決於流速、壓力等的大範 圍溫度下進行磊晶沉積。在一適當表面上磊晶沉積通常可 以在5〇(TC〜1160°C之間舉行。最好是在考慮熱預算的理由 下利用例如從約500°C到約750t的較低溫範圍。請見第 15〜18圖與以下相關文章。 18 lpifl.doc/012 相較於使用砂院以及/或是鍺院製作的對照膜,在能 有效地達到較高沉積速率以及/或是更均勻膜的溫度下實行 沉積最好使用例如較高級矽烷以及/或是較高級鍺烷的化學 前驅物。 根據熟悉此技術者所知的各種氣相沉積方法可適當地 實行這些化學前驅物的沉積,而當根據於此教示的改良式 化學氣相沉積製造.技術去施行沉積製程時卻能獲得最大利 益。這個被揭露的製程可藉由實行包括電漿增益化學氣相 沉積法(plasma-enhanced CVD)或熱化學氣相沉積法的化學 氣相沉積法來適當地實施,且使用由含矽或含鍺化學前驅 物之組成的進料氣體(feed gas)於化學氣相沉積腔體中的一 基底上沉積一含矽或含鍺膜。在一較佳實施例中,氣體是 由三砂院所組成以及沉積一含砂膜。於另一較佳實施例 中,氣體是由一較高級矽烷與一較高級鍺烷所組成並且沉 積一矽鍺膜(SiGe film)。 可以使用適當的多種方法以提供進料氣體到化學氣相 沉積腔體內。於此描述的實驗結果是於化學氣相沉積腔體 中用一水平氣流處理的,且較佳的腔體是一單一晶圓水平 氣流輻射加熱反應器。這種形式的適當反應器是可商業供 應的,且較佳型號包括由美國亞利桑那州鳳凰城的美國 ASM公司商業供應的Epsilon™系列單一晶圓磊晶反應器。 而於此描述之製程也可選擇使用如蓮蓬頭排列的反應器, 以利於增加的均勻性與已知在水平單一通過(single-pass)流 線氣流(laminar gas flow)分布的Epsilon™腔體中特別有效 19 I2771396pi,doc/012 地增加的沉積速率。 在用於沉積之溫度與壓力下提供至化學氣相沉積腔體 之化學前驅物最好是進料氣體的形式或進料氣體的化合 物。在化學氣相沉積腔體中的總壓較佳是在約〇.〇〇1 T〇rr 到約700 T〇rr的範圍內,更佳是在約01 T〇rr到約200 T〇rr 的範圍內,最佳是在約1 T〇rr到約60 Torr的範圍內。每 一含矽以及/或是含鍺化學前驅物的分壓較佳的是在約1χ 10-6%到約100%的總壓之範圍內,更佳的是在約ιχ 1〇_4 %到約100%的總壓之範圍內。如果有使用碳源的話,其 有效分壓較佳的是提供最終含矽以及/或是含鍺膜約2〇% 或更少的碳含量(對單晶材料則約10%或更少的碳含量), 甚至更佳的是約10%或更少的碳含量(對單晶材料則約5 %或更少的碳含量),此處的百分比係根據總膜重之重量 而言。 進料·體也可包含其他化學前驅物與碳源之氣體,例 如鈍載體虱體(inert,carrier gas)。示範的載體氣體包括氦 氣、氬氣、氪氣與氖氣。氫氣是於此描述最佳的製程載體 氣體’特別是對單晶材料而言。而氮氣也可用於多晶矽與 非晶砍膜的沉積。其他化合物可依所需存在於進料氣體 中。而較佳氣體更進一步是由一或多個化合物所組成,其 係選自於包括矽烷、二矽烷、四矽烷、鍺烷、二鍺烷、三 錯院、二氯化氮(nf3)、一矽甲烷(mon〇silyhnethane)、二矽 甲烷、二矽甲烷、四矽甲烷與一摻質前驅物的群組。 而慘質前驅物包括二硼烷、氘化二硼烷(deuterated 20 12771¾ 846pifl.doc/012 diborane)、磷化氫(phosphine)與砷化三氫(arsine)。矽基磷 化氫(silylphosphme)[(H3Si)3_xPRx]以及矽基砷化三氫(silyl arsine)[(H3Si)3_xAsRx]是較佳的磷與砷摻質源,其中X =〇〜2、 RX=H 與/或 D-構型。SbH3 與二甲基銦(trimethyiin(jium)分別 是鍊與銦的較佳來源。這些摻質與摻質來源對譬如以上述 方法製造的摻雜硼、磷、銻、銦與砷的矽、矽鍺與矽鍺碳 膜很有用。在摻雜時這些材料的摻質濃度較佳是在約lx 1014到約lx 1022原子/立方公分的範圍中。摻質可合倂使 用很低的摻質源濃度,例如在氫氣中用約lppm到約1% 總重的濃度範圍。然後根據欲摻雜的濃度與摻雜氣體濃 度,於設定點在約10到200 sccm範圍中經由一質流控制 器(mass flow controller)把這些稀釋的混合物傳送到反應 器。摻質源也可包括隨矽/鍺/碳源一起於載體氣體中被更 進一步稀釋而傳送至反應器中。因爲通常流速是在約每分 鐘20標準公升(SLM)到約180 SLM的範圍內,所以用於習 知製程的摻質濃度將會很小。 而在沉積含矽與含鍺膜的過程中,化學前驅物的相關 分壓(以及碳源,如果有的話)可被保持固定,或是可被改 變以製造出在膜厚中具有不同矽與/或鍺的數量如深度函數 般的分級的膜(graded film)。膜厚最好是在約10埃到5000 埃的範圍內。膜的元素組成可在逐步與/或連續的方法 (stepwise and/or continuous fashion)中變化。根據習知技術 預期的應用可藉著改變沉積時間與/或氣體流速改變膜厚。 無論是固定的或分級的,用於此所述之方法沉積的化合物 21 I2771396pifLd〇c/012 與摻雜膜具有在任何獨特設定的厚度下橫過一平面(Plane) 相對的固定組成。如果在一圖案化基底上沉積一膜,則“平 面”在此處的意義是起伏的。 於此描述之膜的沉積較佳的是在約每分鐘50埃(A/mm.) 的速率下或更高速率下實施,更佳的是在約的速 率或更高速率。而最終含矽膜最好是選自於包括矽鍺膜、 矽鍺碳膜、氮化矽膜(SiN,並非意指等化學計量)、氧化矽 膜(SiO,並非意指等化學計量)、氮氧化矽膜(SiON,並非 意指等化學計量)、摻雜硼膜、摻雜砷膜、摻雜磷膜以及 具有介電常數約2.2或更低之膜的群組。製造適合之低介 電常數膜的方法已揭露於西元2001年11月13日提出申請 的美國申請號第09/993,024號中。而含矽膜可以是非晶質 的、多晶的或是嘉晶的。已知三砂院對改良沉積速率與晶 晶之均勻性有特優的效果。 較佳實施例也提供另一製程,用以解決上述均勻性的 問題。這個製程之範例係於第3圖與範例39中提出,並 且於此作更詳盡的描述。在沉積層中穿過厚度之組成上之 不均勻性(non-uniformity)尤其被相信是導因於在基底表面 溫度中的動力學(相對於靜力學)改變。化學氣相沉積腔體 通常需要裝配一溫度控制器,用以提供一套溫度控制條件 以使整個獨特層的沉積過程保持固定。這個設定點濫度通 常在製程一開始就被選定並維持到層被完成。上述論述中 在過去已藉由憑經驗協調如氣體流速、基底轉速、加熱元 件的能量分布等沉積條件而發現膜厚問題,以有效地時間 22 而1涊6—2 平均溫度變化的厚度效應。 已知一溫度設定點或一組更常影響溫度控制的反應條 件會導致一第一 5埃至1000埃的膜之沉積在組成與厚度 上是相當均勻的膜,但是這個膜會隨沉積的不斷進行而變 得不均勻。而這種結果並不容易瞭解,而且本發明不限於 原理,而是發射率與基底的其它特性,這將與沉積時間一 同改變而影響溫度控制系統。相對來說,這產生導致組成 與厚度變化的溫度變化。 無論轉變至較少均勻沉積之原因是什麼,已知可用一 逐層方式去製造具較佳均勻性的膜。與本實施例一樣,在 逐層基礎上決定一組憑經驗決定的溫度設定點、T2、T3 等。在一積體電路中具有單一功能的特殊點之單一膜會在 憑經驗決定的過程中被破壞成數層以及每一層需決定最理 想的設定點。因此,導因於膜形成之厚度的溫度控制變化 可在沉積期間藉由分開有效進行的設定點的使用而被抵 銷。 可藉由使用各個溫度設定點於每一些不同的工件 (workpiece)上第一沉積一第一層實施憑經驗的決定,然後 測量每一工件上第一層之厚度與組成上的變化,以識別最 均勻層之設定點爲何。層的目標厚度可根據特殊應用所需 的均勻性程度而改變成想要的厚度,譬如從約50埃到約 1000埃,而較佳的是約100埃到約700埃。 接著,在識別過的設定點乃於數個工件上準備一第 一層,以作爲第二設定點T2憑經驗決定的基底。當Τι決 23 12771¾¾^ .doc/012 定後,用各個溫度設定點在每一工件的第一膜上沉積第二 層,然後測量其厚度與組成的變化,以判定導致最均勻的 第二層之第二設定點。上述第二層所需厚度可根據特殊應 用所需的均勻性程度而改變成想要的厚度,譬如從約5〇 埃到約1000埃,而較佳的是約100埃到約700埃。如果最 理想的第一層與第二層形成了具有所需厚度與均勻性的多 層膜(multi-layer film)則可終止製程。如果需要較厚的膜, 則可連續此製程,例如準備一組具有在最初的兩個識別設 定點、T2下沉積兩層的工件,然後用各個溫度設定點在 每一工件的第二層上沉積一第三層,接著測量其厚度與組 成的變化,以判定導致最均勻的第三層之第三設定點丁3, 依此類推。 在此使用的溫度設定點作爲一沉積製程期間可保持固 定之溫度控制變化的範例,但是藉由上述教示之憑經驗製 程於沉積期間可以是多變的。這個憑經驗製程也可供給其 他單一膜沉積製程期間保持固定的溫度控制變化,例如PID 控制器的溫度偏移量(offset)或PID係數。 如氣體流速、分壓與氣體組成之製程變數最好像上述 爲辨別溫度設定點或是在相同實驗期間一樣變化,以便釋 出每一層所需的沉積條件。最好採用實驗設計方法去決定 多種製程變化的效果與其均勻性或沉積速率上的組合。而 實驗設計方法已是眾所皆知的,請見Douglas C. Montgomery, "Design and Analysis of Experiments,w 2nd Ed., John Wiley and Sons,1984。對一獨特製程而言,藉由這些實驗設計方法 I2771l96plfl.doc/012 來決定層的均勻性以及/或是沉積速率上多種製程變化的效 果與組合之後,最好藉由電腦控制的自動製程去確認一組 組(batch-to-batch)或一晶圓一晶圓(wafer-to-wafer)的一致性 (consistency)。而最佳的製程改良處是導因於臨場⑻、 逐步或動力調節(dynamic adjustment)到上述製程變化。已 知用以分別改良層的特性之協調製程變化的憑經驗方法可 不顧上述原理,而改善整個單一構造的或功能膜的特性。 因此本實施例的運作並非根據原理的正確或不正確。 已決定設定點、T2、T3、T4等,較佳實施例也可用 需要裝配溫度控制器以提供單一方法(recipe)多溫度設定點 程式化的一化學氣相沉積腔體來蓮作。藉由輸入一溫度設 定點乃到溫度控制器以施行一較佳製程,並且導入一包括 X/ζό組成的第一氣體到化學氣相沉積腔體內,其中Xi的 範圍係約0到約100。然後在腔體中的一基底上沉積一第 一含矽層。此製程最好繼續輸入〜溫度設定點八到溫度控 制器,並導入一第二含矽化學前驅物的χ2%組成的第二氣 體到化學氣相沉積腔體內,然後在第一含矽層上沉積一第 二含矽層,藉以形成多層含矽膜。而第二含矽化學前驅物 可以是化性與第一含矽化學前驅物相同或不同的,且圖示 於第3圖與範例39。 這個製程更可繼續譬如輸入一溫度設定點T3到溫度 控制器,並導入一第三含矽化學前驅物Χ3%組成的第三氣 體到化學氣相沉積腔體內,然後在第二含矽層上沉積一第 三含矽層等等,以製造和所需一樣多層的含矽膜。 25 12 127711— 而較佳的化學前驅物包括於此描述的較高級矽烷,以 包括習知的化學前驅物如矽烷。第一與第二化學前驅物其 中之一最好至少是選自於包括矽烷、二矽烷、三矽烷的群 組。第一、第二與第三氣體其中之一最好至少是由選自於 包括鍺院、二鍺院、二鍺院、二氟化氮(nf3)、一砂甲院、 二矽甲烷、三矽甲烷、四矽甲烷與摻質前驅物的群組所構 成的混合物。在沉積製程的任何特別階段中於氣體中每一 含矽化學前驅物的含量χη如XA、χ2%、χ3%、χ4%等 是總體積約lx 到約100%的範圍內,較佳的是在約 lx 10,到約100%的範圍內。 而基底較佳的溫度是約350°C或更高,更佳的溫度是 在450°C到約700°C的範圍內。化學氣相沉積腔體最好是 一單一晶圓水平氣流反應器。而最終多層含矽膜最好是選 自於包括一微小縮點(rmcrodot)、矽鍺膜、矽鍺碳膜、氮化 石夕膜、砂-氧膜(silicon-oxygen film)、砂氧氮膜(silicon-oxygen-nitrogen film) 、 摻雜硼膜 、摻雑砷膜 、摻雜磷膜以 及具有介電常數約2.2或更低之膜的群組。使用合適的低 介電常數膜的方法已揭示於在此爲參考文獻的西元2001 年11月13日提出申請的美國申請號第09/993,024號中。 可藉由在逐步或持續的方法中沉積多層膜而實施較佳 實施例的製程。當暫停沉積以調整溫度設定點時,最好也 調整如流速、分壓與氣體組成之所需的製程參數,以製作 具有多種組成的膜。舉例來說,上述沉積膜可具有均質或 均勻的組成,或是在組成上逐步或持續的改變。而在暫停 26 ^771¾ 6pifl.doc/012 期間可修改含矽化學前驅物的特性(identity),並且/或者改 變其氣體含量X#、X2%、X3%、X4%等等。在一較佳實 施例中,製程會藉由不連續或逐步改變鍺濃度以影響分級 之鍺濃度層的成長’較佳的達成方式係藉由於其頂部互相 沉積選擇鍺濃度的層以提供具有間斷週期的超晶格 (supedamce)。範例39結合下列之範例43將闡明本實施例。 已知實施例的整個“膜”是在一積體電路中從其功能觀 點而言是組成一單一結構膜,並且通常貫穿其厚度都具有 類似的組成。所以’用以定義上述逐步沉積製程所形成的 單一膜之類似組成包含分等級的膜,其中相同組成 (constituent)於膜厚中的不同點具有不同的濃度。 而測定膜均勻性與沉積速率的方法已是眾所皆知的。 沉積速率可由測量膜的平均厚度與時間的關係來測知並且 以每分鐘幾埃(A/min.)作爲單位。較佳的沉積速率約爲20 A/min.或更高,而更佳的沉積速率約50 A/min.或更高,而 最佳的沉積速率約100 "min·或更高。測量膜厚的適當方 法包括多點式橢圓對稱法(multiple-point ellipsometric method)。而測量膜厚的儀器也是眾所皆知且爲商業尙可應 用的,其較佳者包括美國加州森尼維耳市(Sunnyvale)的 Nanometrics公司的NanoSpec®系列儀器。 用於此意指沉積膜均勻性的「均勻性」也用以代表厚 度均勻性與組成上之均勻性。膜厚均勻性之測定最好先做 多點式膜厚測量,再決定中間厚度(mean thickness),然後 決定不同於中間數的多次測量平均數。爲了能夠作比較, 27 12771說 pifl.doc/012 可用百分比之不均勻性(percent non-uniformity)表示其結 果。較佳的百分比之不均勻性約10%或更低,而更佳的約 約5%或更低,而最佳的約約2%或更低。組成上之均勻 性則用電性測量(如四點探針)、二次離子質譜儀(secondary ion mass spectrometry,簡稱SIMS)、拉賽福回向散射光譜 儀(Rutherford backscattering spectroscopy,簡稱 RBS)、光 譜橢圓對稱(spectroscopic ellipsometry)以及/或是高解析X 光繞射儀(high resolution X-ray diffractometry,簡稱 HR-XRD) 來測得。 第14圖所示爲在1306 A/min.沉積速率、40Torr壓力 與600°C沉積溫度下,用三矽烷沉積的非晶矽膜之拉塞福 回向散射光譜(RBS)[彈性反衝偵測(Elastic Rec〇1l Detecnon) 光譜’簡稱ERD]。其中實線代表膜的原始資料,而虛線 是一個資料模擬(data simulation)軟體RUMP™在假定殘留 氫濃度爲0.5 at.%下的模型(model)。原始資料指的是從被 吸收的碳氫化合物或水氣(m〇isture)而來的輕微表面污染, 但是能譜代表的是膜中在偵測限定下殘留的氫濃度,相當 於低於0.2 at·%的氯濃度。 第15圖所示爲在600°C、650°C、700°C與750°C下(分 別從第15圖的底部至頂部)用三矽烷沉積的一系列矽膜之 X光繞射圖。X光繞射圖顯示在60(TC沉積的膜是非晶質 的’在650°C沉積的膜是部分結晶的,而在7〇〇°c與75〇°C 沉積的膜則是漸增地結晶化。 第16圖所示係一在750°C沉積的膜(中間的層)剖面的 28 1277 1 ^46pifl.doc/012 穿透式電子顯微相片,其顯示用三砂院沉積的多晶膜中有 相當局程度之膜厚均句性與沉積薄度(thinness)。一個膜之 選區繞射(selected area diffraction,簡稱 SAD)圖(如第 17 圖 所不)顯示膜中並無優先方位(preferential orientation),即此 爲一多晶砂膜。 第18圖所示係在600°C、40ΤΟΙΓ下用三矽烷沉積的非 晶矽膜剖面之掃描式電子顯微相片。在曲面的基底上沉積 的膜顯不於深且窄縫中具有極佳的保角性(conformality)。 在另一較佳實施例中,也使用較高級矽烷於組成範圍 從幾乎純矽到Sl3N4之氮化矽(SiN)材料的低溫低壓化學氣 相沉積合成中。氮源最好包括例如三矽胺[trisilylamine, 也寫作(H3Si)3N]、氨、原子氮(atormc nitrogen)與三氟化氮 (NF3)之化學前驅物。而原子氮最好是用一遙控微波輻射產 生器(remote microwave radical generator)產生。導入化學氣 相沉積腔體的氮源與較高級矽烷之相關量最好是選擇用以 提供具有比使用矽烷替代較高級矽烷所製作之對照膜更高 等級的均勻性。於一較佳實施例中,連續導入原子氮,並 且不是連續導入就是脈衝地如一或多次脈衝導入三矽烷。 如下範例所作的說明,已知藉由脈衝導入較高級矽烷可獲 得較佳的膜均勻性,以及藉由間歇化學氣相沉積法可獲得 極薄且高均勻性的氮化矽膜。依照實施例所準備的氮化矽 膜具有約10埃到約300埃範圍的較佳厚度,更佳的厚度 在約15埃到150埃。 這些作爲化學前驅物並與三矽烷結合的氮源之使用特 29 1277139 08846pifl.doc/012 別在低溫下能使氮化矽材料的沉積於薄膜中具有極少量氮 -氫鍵,且其沉積速率比用傳統矽源如矽烷的製程要高得 多。而使用其他較高級矽烷也能獲得相同結果。沉積溫度 超過450°C時,氫含量較佳是少於4 at.% ,更較佳是約少 於2 at.% ’最佳是約少於1 at·%。而上述沉積最好於一質 量傳輸限定規範中實行。 於另一實施例中,也可使用較高級矽烷於氧化矽材料 與氮氧化矽材料的低溫低壓化學氣相沉積合成中。較高級 石夕院的低溫與局成長速率之優點特別在低壓化學氣相沉積 條件下提供超越使用矽烷之製程的製造優點。氧源可包括 臭氧(ozone)、氧、水、含氮的氧(nitric 0Xide)、過氧化氫與 同樣的氧源。用於引導氮到這些材料中的氮源包括如上述 的三砂胺、氨、原子氮與三氟化氮。這些氧源與氮源可以 連糸買地使用,或是以不連接的步驟抑或是以包含這些製程 的結合之方法來使用。而上述沉積最好於一質量傳輸限定 規範中或接近一質量傳輸限定規範來實行。用三矽胺與三 矽烷進行的沉積較佳是在約350°C到約750°C的範圍施行, 更佳是在約40(TC到約700°C,最佳是在約450°C到約650 °C。用三氟化氮與三矽烷進行的沉積較佳是在約300°C到 約750°C的範圍施行,更佳是在約350°C到約700°C,最佳 是在約400°C到約650°C。 雖然沒有提出氧化物和氮氧化物的個別範例,但是熟 悉此技術者將易於瞭解於此揭露的原理,以及體會上述氮 化矽與矽鍺化合層可等同應用於氧化矽的沉積。同樣地, 30 I2771^46pi,doc/012 二石夕院關於較低活化能與較低溫以實現質量傳輸限定ί几積 的優點對氣相沉積具有益處,尤其是多種矽化合材料的化 學氣相沉積。 一較佳實施例提供有用於微電子業中多種應用的膜。 而較佳含矽膜具有小於約2%厚度不均勻性以及小於約2 %之組成不均勻性。於此所述之膜可用於多種應用中,例 如電晶體閘電極。於此所述之層對於形成積體電路中的關 鍵元件層特別有用,譬如積集電晶體中的閘層。其他例子 包括在異質接合雙載子電晶體(簡稱HBT’s)中的半導體層。 而用以製作這樣的積體電路的製程已是孰悉此技術者眾所 皆知的。這些積體電路可由熟悉此技術者所知的方法加上 電腦系統,所以一更佳的實施例提供由一或多個積體電路 組成的電腦系統。 第1圖所示係於此描述之較佳製造流程步驟圖。步驟 100,形成閘介電層於一半導體基底上。如果需要的話則 如步驟110,淸潔閘介電層表面。然後如上所述於步驟120 中,沉積含矽薄膜,最好包括流動的三矽烷。接著,如果 想改善橫向訊號傳輸(lateral signal transmission),可於步驟 130中,沉積金屬層於含矽膜上。然後,於步驟140中, 圖案化閘電極。並於步驟150中,繼續積體電路的製造。 第2圖所示係依照第1圖之製程所形成的閘極疊層200 示意圖。在一半導體基底220上形成有一閘介電層210, 在閘介電層210形成有一電性摻雜含矽膜230,以及在含 矽膜230上放置有一非必要的金屬層240,藉以形成閘極 31 1277139 1 08^46pifl.doc/012 疊層200。然後疊層200被圖案化而形成閘電極(於第2 _ 中未繪示),並且繼續積體電路的製造。 閘介電層210能包括至少一高介電常數材料是較佳 的,且具有大於5的介電常數,更好的是具有大於1〇的 介電常數。範例材料包括氧化鋁、氧化鉛與氧化鍩,並且 較佳的是藉原子層沉積(atomic layer deposition,簡稱ALD) 形成局品質、無孔洞的層(pinhole free layer)。在質量傳_ 限定中或接近之的三矽烷的使用特別是結合較高級鍺院中 有利於補償對傳統矽沉積於這樣的高介電常數材料上的,辕 慢成核時間。 於另一範例中,於單晶基底上沉積磊晶含矽層並流靈力 三矽烷。矽層與異質磊晶矽鍺、碳化矽與矽鍺碳層可藉箸 於此所述之製程沉積。 另一較佳實施例提供用以於一表面上沉積一含矽材料 的儀器。這個儀器包括一化學氣相沉積腔體(CVD chamber)、一含三砂院的導管(vessel)、一進料管線(feed line),用以操作連接導管到化學氣相沉積腔體,使導管至fJ 化學氣相沉積腔體之三矽烷的通過,以及一溫度控制器, 其配置圍繞著導管並維持在約10°C到70°C溫度範圍中, 較佳是在15°C到約52t以控制三矽烷的蒸發率。適合的 溫度控制器範例包括熱電控制器(thermoelectric controller*;) 以及/或是富含液體的外罩(liquid-filled jacket)。而較佳的 化學氣相沉積腔體是一單一晶圓水平氣流反應器。而整個 儀器還可以包括一歧管(manifold),用以操作連接進料管 32 12771恐46 pifl.doc/012 線,以控制導管到化學氣相沉積腔體之三矽烷的通過。最 好配置一熱源圍繞進料管線並加熱氣體管線到約35°C到70 °C,更好的是約40°C到52°C,以避免在高氣體流速下凝 結。另外,三矽烷較佳係藉由一載體氣體來乘載三矽烷蒸 氣用氣泡的方式導入,最好是溫度控制的氣泡結合加熱的 氣體管線以傳送三矽烷。 範例 下列範例是用ASM EpsUon 2000™水平氣流磊晶反應 器系統實施的,並裝配Bernoulli棒(wand)晶圓傳輸系統、 只用於淸潔的負荷鎖(purge-only load lock)、不滑動凹面電 納器(non-slide concave susceptor)、預熱環、可調位置照射 器(adjustable spot lamp)以及可調整氣體入口注射器(tunable gas mlet m^jector)。在進料管線中將含矽與含鍺前驅物供應 到腔體內,且可含氫氣與二硼烷摻質。然後將氫氣中之120 seem的1%二硼烷(B2H6)稀釋於2 slm氫氣中,以及將120 seem的混合物導入混合有20 slm氫氣與前驅物之反應器 中,並且在下列顯示的範例中之流速條件下沉積在一轉動 基底上。用二次離子質譜儀測量與光學橢圓對稱測量器 (Nanometnc)從氧與硼的深度輪廓(depth profile)判斷沉積速 率。 mm 1-4 根據表一所示,使用三矽烷作爲化學前驅物沉積含矽 膜。沉積溫度爲適於三矽烷之質量傳輸限定規範的70(TC。 33 12771¾ 846pifl.doc/012 然而因爲流速(在這些特殊的沉積條件下)不適合提供一均 勻的膜,所以最終膜並不均勻反而在沉積輪廓有凹面(中 間較薄、邊緣較厚)。 表一 編號 溫度 re) 壓力 (Torr) 流速設定値 (seem) 前驅物 基底 沉積輪廓 1 700 40 50 Si,H« SiCh 凹面 2 700 40 45 SiA Si〇, 凹面 3 700 40 15 Si,Η, Si〇, 凹面 4 700 40 25 Si,Η, Si〇, 凹面 範例5〜15 根據表二所示之參數,使用三矽烷和矽烷作爲化學前 驅物以及用二硼烷作爲摻質去沉積含矽非晶膜。將氫氣中 120 seem的1%二硼烷(B2H6)稀釋於2 slm氫氣中,以及將 120 seem的混合物導入混合有20 slm氫氣、三矽烷和矽烷 之反應器中,其中三矽烷和矽烷之流速條件顯示於表二 中。這些結果顯示使用三矽烷在預設溫度下能獲致比使用 矽烷高的沉積速率,甚至在三矽烷流速比矽烷流速低的情 形下也一樣。 34 I27713?46pifl,oc/012 表二 編號 溫度 (°C) 壓力 (Torr) 流速値 (seem) 前驅物 基底 沉積速率 (A/min.) 5C 650 40 50 SiH4 _ Si09 46 6C 650 40 50 SiH4 Si<10〇〉 68 7 650 40 50 Si.H, Si<l〇〇> 462 8C 600 40 50 SiH4 Si02 19 9C 600 40 50 SiH4 Si<l〇〇> 9 10 600 40 20 Si,H8 [TsiO. 359 11 600 40 15 SiA Si<l〇〇> 181 12C 550 760 25 SiH4 Si〇2 <1 13C 550 40 50 SiH4 Si〇2 7 14 550 40 30 SiA —Si〇? 287 15C 550 40 50 SiH4 Si〇2 2 範例16〜19 根據表三所示之參數,使用三矽烷和矽烷作爲化學前 驅物沉積含矽膜。沉積時間可調整以使每一膜有約500埃 的平均厚度。然後使用Nanometnc之橢圓對稱測量器測量 平均厚度並由沉積時間分配這些數目,以判斷沉積速率。 膜的不均勻性可由膜厚的49點厚度圖來判別。結果顯示 在指定溫度下使用三矽烷替代矽烷,能在很高的沉積速率 下獲得更均勻的膜。在550°C即可,但在600°C更是引人 注目。 35 /012 /012 表三 編號 前驅物 溫度 (°C) %不均勻性 1 SiH4 600 5.93 2 Si,Η, 一 600 0.83 3 SiH4 550 8.5 ^ 4 Si,H8 一 550 7.31 ~~ 沉。積速率 _例20〜38410: increasing the set point to 650 ° C 420: depositing the dome layer 430 with 100% trioxane: removing the wafer from the reactor 10 I2771396pin, 〇c/ol2 embodiment due to thermal and temperature control system limitations, dynamic The dynamic temperature variation plays an important role in the uniformity of deposition of the film on the surface of the substrate by chemical vapor deposition (CVD). It is generally preferred that the deposited film be as uniform as possible in thickness and elemental composition, but existing processes tend to produce a non-uniform film to improve the degree. This non-uniformity is always due to temperature variations across the surface of the substrate because the surface temperature of the substrate affects the deposition rate and the composition of the final film (the result mg fUm). Moreover, imperfect control of other process parameters, including gas flow rate and total pressure, is also believed to contribute to the inhomogeneity of the physical properties of the film. The pursuit of uniformity is often based on empirical adjustment of deposition conditions such as gas flow rate, substrate rotation speed, energy distribution of the heating element, etc., to achieve a desired film of uniform thickness. This is achieved by depositing a large number of films on the different substrates and at each substrate under different pre-selected set of deposition conditions. Then, the thickness variation of each film was measured and the results were analyzed to determine the conditions under which the thickness variation could be eliminated. However, it has been learned that an empirical process does not require a uniform temperature distribution throughout the process, or should be said to be a special reaction temperature 'set point' (set point / process effective time-average time change from temperature The resulting thickness variation. Therefore, the rule of thumb does not require uniform temperature across the substrate throughout the deposition process. In other words, because of the requirement for composition homogeneity in a solid that includes crossing the film surface and passing through the film thickness. Or at least control, ►pifl. Doc/012 caused controversy over compositional changes. This is because many films contain dopants, and the level of these dopants affects the electronic properties of the film. It also incurs other non-uniformities in the composition. The preferred embodiment provides a process for solving the above problems, and these processes can be used alone or preferably together. A process that affects the use of chemical precursors allows conventional precursors to provide film deposition at the same temperature to be processed in a mass transport limited growth regime. For a particular chemical precursor, the mass transfer specification is a temperature range in which the film deposition rate is temperature independent. The deposition rate is substantially unaffected by this small temperature change across the surface of the substrate in this temperature range, and these changes result in temperatures that are maintained at or near the mass transfer specification. This allows the film to be made more uniform than the conventional chemical precursors deposited at the same temperature, e.g., having a higher composition uniformity and thickness uniformity. This is because conventional precursors are deposited in mass transfer specification. Those skilled in the art will appreciate that the temperature range of the mass transfer specification can be determined by the particular precursor and reaction conditions and is illustrated by an Arrhenius diagram. In the case of the chemical precursor trisilane, the transition point from the temperature-dependent deposition rate to the temperature-independent deposition rate (traimUon pomt) is higher than that of decane or two. The conversion point of decane is much lower, as shown in Figure 10 for the Arrhenius diagram. The lower area to the transition point in the figure has 12 127713⁄4 846pifl. Doc/012 has a clearly upward linear slope, which means that the deposition of trioxane in the lower temperature range is highly correlated with temperature, so it is not in the mass transfer specification. For example, as shown in Figure 10, trioxane is at a flow rate of 25 standard cubic centimeters per minute (sccm) and a pressure of 40 Torr, and deposition at temperatures below 525 °C is not a mass transfer limitation (eg, in dynamic specifications). in). In contrast, the area above the transition point in the figure is almost flat, which means that the deposition of trioxane in this temperature range is temperature independent, so it is in the specification of mass transfer. For example, Figure 10 shows that trioxane deposition is clearly defined by mass transport at temperatures of about 620 ° C or above. It can be seen that the transition will occur in the Arrhenius diagram over a temperature range with a declining slope, meaning that the deposition of trioxane in this temperature range is virtually temperature-independent, but rather close to the mass transfer specification. For example, Figure 10 shows that the deposition quality of trioxane at a temperature of about 525 ° C or greater is defined by mass transport. It can be seen that the transition point may increase slightly at higher flow rates and slightly decrease at lower flow rates. For example, a transition point from deposition based on temperature to actual mass transfer limiting deposition has been verified to shift to higher temperatures as the trioxane flow rate increases. Thus, in contemporary manufacturing, for practical reasons (e.g., to maintain thermal budget retention of crystal characteristics, control of dopant profile, etc.), the use of trioxane can limit the deposition of mass transfer at a desired temperature. Various sand and bismuth-containing chemical precursors can be applied to the film deposition process disclosed herein to prepare ruthenium-containing, ruthenium-containing and alloy films containing ruthenium and osmium, such as sand shovel (SiGe' does not mean, etc. Stoichiometric) membrane. These chemistry 13 ►pifl. Doc/012 The precursor may also be used in combination with a carbon source to prepare an alloy film, such as a sand germanium carbon (SiGeC, here not meant to be stoichiometric) alloy film. Suitable for use in the present invention. The ruthenium chemical precursor comprises a higher-order non-halogenated hydride of ruthenium, especially a decane of the formula SinH2n+2, wherein n zz 2-6. Unique examples include dioxane (H3SiSiH3), trioxane (H3SiSiH2SiH3) and tetraoxane (HeSiHJiHAHD. Trioxane (also written as Sl3H8) is best used to achieve a balance between volatility and activity. Deposition under substantial or near mass transfer definition is preferred (but not necessary) for deposition of tantalum. Suitable for use in urgent inventions, only preferred chemical precursors containing ruthenium include a molecular formula of GenH2n+2. Higher decane, where n = 2 to 3. The source of lanthanum in other arrangements may include (H3Ge)(GeH2)x(GeH3), where X = 0~2. A unique example includes dioxane (H3GeGeH3) , trioxane (H3GeGeH2GeH3) and tetraoxane (H3GeGeH2GeH2GeH3). In a preferred embodiment, the chemical precursor is used in combination with a carbon source. A preferred carbon source comprises silylmethane [( H3Si)4_xCRx], where X = 0~3, R = Η and/or D-configuration. Preferred mercapto methane includes dimethane, trimethane and tetramethane (X = 0~2), The best is tetrahydromethane. Additional preferred carbon sources include hydrocarbons such as Alkanes, ethane, propane, butane, etc., as well as carbon monoxide, carbon dioxide and hydrogen cyanide (HCN). These chemical precursors and carbon sources can be obtained from commercial sources or are familiar with this technology. Synthesized by a method known as a ruthenium film such as MC, SiNC or SiOCX 14 127713⁄4 46pifl. None of the doc/012 means unique iso-stoichiometry) has many uses in semiconductor manufacturing, for example as an etch stop layer, a hard mask and a passivation layer. It is best to deposit with a unique chemical precursor at a temperature consistent with the mass transfer specification. For any unique chemical precursor and reaction conditions, the mass transfer limit specification can be determined at different temperatures by empirically derived from the deposition data (Airhenius diagram). The Arrhenius plot of the best ruthenium precursor, trioxane, under special conditions was added as shown in Figure 1 above. In addition to using the preferred chemical precursors described herein (particularly trioxane) and selecting a temperature for the precursor in a mass transfer specification, the deposition with the first process preferably includes appropriate other deposition parameter options. 'Specially the gas flow rate. The selection of preferred gas flow rates in conjunction with mass transfer limiting specifications is known to produce highly uniform films at much higher deposition rates than sand yards. In order to be deposited with sand at temperatures in the power specification, the uniformity of the membrane is primarily controlled by the set point on the temperature controller and the (a smaller range) gas flow rate. In contrast, for depositions containing higher levels of decane at a temperature-limited specification, it is known that the set point on the abuse controller is opposite to the sensitivity of the gas flow rate control set point (sensiUWty). For example, raising the set point on the temperature controller with the deposition of trioxane at the temperature transfer specification will have a much smaller effect than the set point on the boost gas flow controller. The preferred final film is more uniform than the comparable film when performing the deposition described herein. In addition to replacing the higher decane with decane and 15 I2771^46pif, doc/〇, or replacing the higher decane with decane, the "control film" used for this is in the film of the invention in question. All meaningful views are made in substantially the same way, and each film deposition process is separately adjusted to account for the difference in sensitivity to temperature and gas flow rate control set points, respectively. More particularly when comparing different layers As a result, the thickness uniformity measured by the following standard was measured by using an optional diameter across the wafer and measuring the thickness of the deposited layer along the diameter of 49 points. (Exclusion zone) is not measured. Then, the thickness measurement of the 49 points (for example, ± 6 angstroms) is the sum of the maximum thickness measurement between the 49 points plus the minimum thickness measurement. The uniformity is here expressed as a percentage. It is better to know that the method of using the precursor described herein can result in an abnormally high deposition rate, and more surprisingly High uniformity and smoothness 〇 For example, a preferred polycrystalline germanium film made with trioxane has a higher deposition rate than the one prepared by the same temperature used for decane at trioxane. The film has a large uniformity. Similarly, the inventors have separately formed an amorphous silicon (abbreviated a-Si) layer and an epitaxial silicon (epi-Si) layer using Sanshi Xiyuan. The better uniformity is compared with the sediment layer of the sand courtyard. See Figures 15 to 18 and the corresponding description below. Similarly, a better ruthenium film is prepared with higher decane and the ruthenium film has a ratio The uniformity of the control film prepared by replacing the higher decane with decane is greater. Moreover, higher deposition rates can be obtained at lower reaction temperatures using the disclosed ruthenium and ruthenium sources. 16 ►pifl. Doc/012 Figure 11 shows a linear plot of the deposition rate versus the flow rate of trioxane (in some of the illustrations, SilcoreTM) at a deposition temperature of 600 °C and a pressure of 40 Torr. This linearity further indicates that the Sanshayuan deposition under these conditions is substantially or close to the mass transfer limit, and further indicates that it has a much lower nucleation time than the oxide. Figure 12 is a plot of the film thickness of the film deposited with trioxane under the same conditions (650 ° C, 40 Torr) and the measurement site (she), except that the deposition time is varied from 90 seconds to 15 seconds. . Figure 12 shows excellent film uniformity over a wide range of deposition times at a fixed three-sand flow rate. The results represent not only time-average but the nature of the precursor. The (nature) and selected conditions, and because those layers remain uniform regardless of thickness, the emissivity (or other temperature-dependent temperature control) effect does not change the uniformity. Figure 13 shows the use of trioxane and diborane (as a dopant precursor) at a deposition temperature of 600 °C, a pressure of 40 Torr, and a range of diborane flow rates (from 0 to 18 〇seem). Deposition rate map. Figure 13 shows that the deposition rate using trioxane is quite inconsistent with the flow rate of the dopant precursor. In the lower temperatures that are more appropriate to reduce thermal stability, the preferred temperature range tends to favor a unique chemical precursor. For lesser grade sand yards and higher level brothels, a lower chain-length increase rate is preferred for lower temperatures. Therefore, the preferred temperature range for dioxane deposition will tend to be higher than trioxane, and in turn tend to be relatively higher than tetrasane and the like. And decane has a similar trend. The preferred deposition temperature for the deposition of trisane is above about 35 CTC, more preferably above about 45 Qt, and 127 for 2 to minimize hydrogen content in the final film. In order to achieve close or mass transfer specifications, the temperature is preferably maintained above about 525 ° C, even above about 550 ° C, and most is about above 600 ° C. The process can be carried out at temperatures above 700 ° C, but is preferably at or below about 700 ° C. Therefore, the preferred temperature is in the range of from 450 ° C to about 700 ° C, and more preferably in the range of from 525 ° C to about 650 ° C. For any unique chemical precursor or mixture thereof, the optimum temperature range can be determined by general experimentation using the guidelines provided herein. It is known that the temperatures listed are preferably used for thermal CVD. The lower temperature is suitable for the plasma assisted deposition process, depending on the degree of incorporation of the acceptable acid for the application. The choice of deposition temperature may also depend in part on the crystallinity required for the deposited layer. For example, the predominant crystallinity 矽 can be from about 620 ° C to 80 〇. (: Deposition in the range, and this deposition is clearly in the mass transfer specification. It is better to carry out polycrystalline germanium deposition between 650 ° C and 750 ° C. Lower temperature can be used for amorphous germanium deposition, and the temperature is the most It is preferred to maintain at least substantially the mass transfer definition (e.g., preferably at a temperature above 525 ° C.) The epitaxial enthalpy is largely dependent on the purity of the surface on which the deposition takes place. That is, those skilled in the art can Cognizing a very clean single crystal surface, such as a previously deposited epitaxial layer or the upper surface of a single wafer, can be epitaxially deposited over a wide range of temperatures depending on flow rate, pressure, etc. Epitaxial deposition on a suitable surface is typically It can be held at 5 〇 (TC~1160 °C. It is best to use a lower temperature range of, for example, from about 500 ° C to about 750 t for consideration of the thermal budget. See Figures 15 to 18 and related articles below 18 lpifl. Doc/012 Compared to the use of sand and/or brothel-made control membranes, it is preferred to use, for example, higher decane and/or at temperatures effective to achieve higher deposition rates and/or more uniform membranes. Or a chemical precursor of higher decane. The deposition of these chemical precursors can be suitably carried out according to various vapor deposition methods known to those skilled in the art, and when modified chemical vapor deposition is performed according to the teachings herein. The technology can be used to maximize the benefits of the deposition process. The disclosed process can be suitably carried out by performing a chemical vapor deposition method including plasma-enhanced CVD or thermal chemical vapor deposition, and is used for containing or containing germanium. A feed gas of a composition of a chemical precursor deposits a ruthenium containing or ruthenium containing film on a substrate in the chemical vapor deposition chamber. In a preferred embodiment, the gas is comprised of a Sanshayuan and a sand-containing film is deposited. In another preferred embodiment, the gas is comprised of a higher decane and a higher decane and a SiGe film is deposited. A variety of suitable methods can be used to provide the feed gas to the chemical vapor deposition chamber. The experimental results described herein are treated with a horizontal gas stream in a chemical vapor deposition chamber, and the preferred chamber is a single wafer horizontal gas stream radiant heating reactor. Suitable reactors of this type are commercially available, and preferred models include the EpsilonTM series of single wafer epitaxial reactors commercially available from ASM, Inc. of Phoenix, Arizona, USA. The process described herein may also employ a reactor such as a showerhead arrangement to facilitate increased uniformity in an EpsilonTM chamber known to be distributed in a horizontal single-pass laminar gas flow. Particularly effective 19 I2771396pi, doc/012 increased deposition rate. The chemical precursor supplied to the chemical vapor deposition chamber at the temperature and pressure for deposition is preferably in the form of a feed gas or a compound of a feed gas. The total pressure in the chemical vapor deposition chamber is preferably about 〇. 〇〇1 T〇rr to a range of about 700 T rr, more preferably in the range of about 01 T rr to about 200 T rr, most preferably in the range of about 1 T rr to about 60 Torr. Inside. The partial pressure of each cerium-containing and/or cerium-containing chemical precursor is preferably in the range of from about 1 χ 10-6% to about 100% of the total pressure, more preferably in the range of about 1 〇 1 〇 _ 4 %. To the range of about 100% of the total pressure. If a carbon source is used, the effective partial pressure is preferably to provide a final cerium and/or a carbon content of about 2% or less of the cerium-containing film (about 10% or less for the single crystal material). Content), even more preferably a carbon content of about 10% or less (about 5% or less carbon content for a single crystal material), the percentage here being based on the weight of the total film weight. The feed body may also contain gases from other chemical precursors and carbon sources, such as an inert carrier gas. Exemplary carrier gases include helium, argon, helium and neon. Hydrogen is the preferred process carrier gas described herein, particularly for single crystal materials. Nitrogen can also be used for the deposition of polycrystalline germanium and amorphous dicing films. Other compounds may be present in the feed gas as desired. Preferably, the preferred gas is further composed of one or more compounds selected from the group consisting of decane, dioxane, tetraoxane, decane, dioxane, trisomid, nitrogen dichloride (nf3), A group of mon〇silyhnethane, dimethane, dimethane, tetramethane, and a dopant precursor. The tragic precursors include diborane and deuterated diborane (deuterated 20 127713⁄4 846pifl. Doc/012 diborane), phosphine and arsine. Silylphosphme [(H3Si)3_xPRx] and silyl arsine [(H3Si)3_xAsRx] are preferred sources of phosphorus and arsenic, where X = 〇 〜 2, RX =H and / or D-configuration. SbH3 and dimethyl indium (trimiumin (jium) are preferred sources of chain and indium, respectively. These dopants and dopant sources are, for example, doped with lanthanum, phosphorus, antimony, indium and arsenic, which are produced by the above method. Tantalum and tantalum carbon films are useful. The doping concentration of these materials during doping is preferably in the range of about 1 x 1014 to about 1 x 1022 atoms/cm 3 . The dopants can be combined with very low dopant sources. The concentration, for example, in hydrogen, is in a concentration range of from about 1 ppm to about 1% total weight. Then, depending on the concentration to be doped and the concentration of the dopant gas, the mass flow controller is passed through the mass flow controller at a set point in the range of about 10 to 200 sccm ( The mass flow controller) delivers the diluted mixture to the reactor. The source of dopants may also be passed to the reactor with further hydrazine/hydrazine/carbon source together in the carrier gas, as the flow rate is typically about The concentration of 20 standard liters (SLM) to about 180 SLM, so the concentration of dopants used in the conventional process will be small. In the process of depositing yttrium and yttrium containing membranes, the relevant partial pressure of chemical precursors (and carbon source, if any) can be protected Fixed, or can be modified to produce a graded film having a different number of turns and/or turns in the film thickness as a function of depth. The film thickness is preferably in the range of about 10 angstroms to 5000 angstroms. The elemental composition of the film can be varied in a stepwise and/or continuous fashion. The application contemplated by the prior art can vary the film thickness by varying the deposition time and/or gas flow rate. The fixed or fractionated compound 21 I2771396pifLd〇c/012 deposited for use in the method described herein has a fixed composition opposite the doped film across any plane (Plane) at any uniquely set thickness. The deposition of a film on a substrate, the meaning of "planar" here is undulating. The deposition of the film described herein is preferably about 50 angstroms per minute (A / mm. It is implemented at a rate or higher rate, and more preferably at a rate or higher. The final ruthenium-containing film is preferably selected from the group consisting of a ruthenium film, a ruthenium carbon film, a tantalum nitride film (SiN, not meant to be stoichiometric), a ruthenium oxide film (SiO, not meant to be stoichiometric), Niobium oxynitride film (SiON, does not mean equal stoichiometry), doped boron film, doped arsenic film, doped phosphorus film, and has a dielectric constant of about 2. Group of membranes of 2 or lower. A method of making a suitable low dielectric constant film is disclosed in U.S. Application Serial No. 09/993,024, filed on Nov. 13, 2001. The ruthenium-containing film may be amorphous, polycrystalline or Jiajing. It is known that Sanshayuan has an excellent effect on the improvement of deposition rate and the uniformity of crystallites. The preferred embodiment also provides another process for solving the above problem of uniformity. An example of this process is presented in Figure 3 and Example 39 and is described in more detail herein. The non-uniformity of the composition through the thickness in the deposited layer is especially believed to be due to changes in kinetics (relative to statics) in the surface temperature of the substrate. Chemical vapor deposition chambers typically require a temperature controller to provide a set of temperature control conditions to maintain the deposition process throughout the unique layer. This setpoint abuse is usually selected at the beginning of the process and maintained until the layer is completed. In the above discussion, in the past, the film thickness problem has been found by empirically coordinating deposition conditions such as gas flow rate, substrate rotation speed, and energy distribution of the heating element, to effectively effect the thickness effect of the average temperature change of time 22 and 1涊6-2. It is known that a temperature set point or a set of reaction conditions that more often affect temperature control results in a film of a first 5 angstrom to 1000 angstrom deposition that is fairly uniform in composition and thickness, but this film will continue to deposit with deposition. It becomes uneven and becomes uneven. While this result is not readily apparent, and the invention is not limited to the principle, but the emissivity and other characteristics of the substrate, which will vary with the deposition time to affect the temperature control system. Relatively speaking, this produces a temperature change that results in a change in composition and thickness. Regardless of the reason for the transition to less uniform deposition, it is known to use a layer-by-layer approach to fabricate films having better uniformity. As in the present embodiment, a set of empirically determined temperature set points, T2, T3, etc., are determined on a layer-by-layer basis. A single film with a single point of special function in an integrated circuit is broken into several layers in an empirically determined process and each layer determines the most desirable set point. Thus, temperature control changes due to the thickness of the film formation can be offset during deposition by the use of separately setpoints that are effectively performed. An empirical decision can be made by first depositing a first layer on each of the different workpieces using individual temperature set points, and then measuring the thickness and composition changes of the first layer on each workpiece to identify What is the set point of the most uniform layer? The target thickness of the layer can be varied to a desired thickness depending on the degree of uniformity desired for a particular application, such as from about 50 angstroms to about 1000 angstroms, and preferably from about 100 angstroms to about 700 angstroms. Next, a first layer is prepared on the identified set points on a plurality of workpieces as an empirically determined basis for the second set point T2. When Τι determines 23 127713⁄43⁄4^ . After doc/012 is set, a second layer is deposited on the first film of each workpiece with each temperature set point, and then the change in thickness and composition is measured to determine the second set point that results in the most uniform second layer. The thickness of the second layer described above may be varied to a desired thickness depending on the degree of uniformity desired for a particular application, such as from about 5 angstroms to about 1000 angstroms, and more preferably from about 100 angstroms to about 700 angstroms. The process can be terminated if the most desirable first and second layers form a multi-layer film of the desired thickness and uniformity. If a thicker film is required, the process can be continued, for example, preparing a set of workpieces having two layers deposited at the first two identification set points, T2, and then using the various temperature set points on the second layer of each workpiece. A third layer is deposited, followed by a change in thickness and composition to determine the third set point 3 that results in the most uniform third layer, and so on. The temperature set point as used herein serves as an example of a fixed temperature control change during a deposition process, but the empirical process by the above teachings can be varied during deposition. This empirical process can also be used to maintain fixed temperature control changes during other single film deposition processes, such as the temperature offset (offset) or PID coefficient of the PID controller. Process variables such as gas flow rate, partial pressure and gas composition are preferably varied as described above for discriminating temperature set points or during the same experiment to release the desired deposition conditions for each layer. It is best to use experimental design methods to determine the combination of multiple process variations and their uniformity or deposition rate. The experimental design method is well known, see Douglas C.  Montgomery, "Design and Analysis of Experiments,w 2nd Ed. , John Wiley and Sons, 1984. For a unique process, by these experimental design methods I2771l96plfl. Doc/012 to determine the uniformity of the layer and / or the effect and combination of various process changes in the deposition rate, it is best to confirm a set of batch-to-batch or wafer by computer-controlled automatic process Wafer-to-wafer consistency. The best process improvement is due to on-the-spot (8), stepwise or dynamic adjustment to the above process variations. It is known that empirical methods for coordinating process variations that individually improve the characteristics of the layers can improve the properties of the entire unitary or functional film regardless of the above principles. Therefore, the operation of this embodiment is not correct or incorrect according to the principle. The set point, T2, T3, T4, etc. have been determined, and the preferred embodiment can also be used to assemble a temperature controller to provide a single method of chemical vapor deposition chamber programmed with multiple temperature set points. A preferred process is performed by inputting a temperature set point to the temperature controller and introducing a first gas comprising a X/ζό composition into the chemical vapor deposition chamber, wherein Xi ranges from about 0 to about 100. A first layer of germanium is then deposited on a substrate in the cavity. Preferably, the process continues to input a temperature set point of eight to the temperature controller, and introduces a second gas containing χ2% of the second cerium-containing chemical precursor into the chemical vapor deposition chamber, and then on the first ruthenium layer. A second ruthenium containing layer is deposited to form a multilayer ruthenium containing film. The second ruthenium containing chemical precursor may be the same or different from the first ruthenium containing chemical precursor and is shown in Figure 3 and Example 39. The process can continue, for example, by inputting a temperature set point T3 to the temperature controller, and introducing a third cerium-containing chemical precursor Χ3% of the third gas into the chemical vapor deposition chamber, and then on the second ruthenium layer. A third ruthenium-containing layer or the like is deposited to produce a ruthenium-containing film as many layers as desired. 25 12 127711 - Preferred chemical precursors include the higher decane described herein to include conventional chemical precursors such as decane. Preferably, one of the first and second chemical precursors is at least selected from the group consisting of decane, dioxane, and trioxane. Preferably, at least one of the first, second and third gases is selected from the group consisting of a brothel, a second brothel, a second brothel, a nitrogen difluoride (nf3), a sandstone, a methane, a third a mixture of methane, tetramethane, and a mixture of dopant precursors. The content of each of the cerium-containing chemical precursors in the gas at any particular stage of the deposition process, such as XA, χ2%, χ3%, χ4%, etc., is in the range of from about 1x to about 100% by volume, preferably It is in the range of about lx 10 to about 100%. The preferred temperature of the substrate is about 350 ° C or higher, and more preferably the temperature is in the range of 450 ° C to about 700 ° C. The chemical vapor deposition chamber is preferably a single wafer horizontal gas flow reactor. Finally, the multilayer ruthenium-containing film is preferably selected from the group consisting of a rmcrodot, a ruthenium film, a ruthenium carbon film, a nitriding film, a silicon-oxygen film, a sand oxynitride film. (silicon-oxygen-nitrogen film), doped boron film, antimony-doped arsenic film, doped phosphor film, and have a dielectric constant of about 2. Group of membranes of 2 or lower. A method of using a suitable low dielectric constant film is disclosed in U.S. Application Serial No. 09/993,024, filed on Nov. 13, 2001. The process of the preferred embodiment can be carried out by depositing a multilayer film in a stepwise or continuous process. When the deposition is paused to adjust the temperature set point, it is preferred to also adjust the process parameters required for flow rate, partial pressure, and gas composition to produce a film having a variety of compositions. For example, the deposited film described above may have a homogeneous or uniform composition or a gradual or continuous change in composition. While on a pause of 26^7713⁄4 6pifl. The identity of the ruthenium containing chemical precursor may be modified during doc/012 and/or its gas content X#, X2%, X3%, X4%, etc. may be altered. In a preferred embodiment, the process may affect the growth of the graded germanium concentration layer by discontinuous or stepwise changes in the germanium concentration. A preferred way to achieve this is by providing a layer of germanium concentration on top of each other to provide discontinuity. Periodic superlattice (supedamce). Example 39 will be clarified in conjunction with Example 43 below. The entire "film" of the known embodiment is composed of a single structural film from its functional point of view in an integrated circuit, and typically has a similar composition throughout its thickness. Therefore, a similar composition for defining a single film formed by the above-described stepwise deposition process comprises a graded film in which the same composition has different concentrations at different points in the film thickness. Methods for determining film uniformity and deposition rate are well known. The deposition rate can be measured by measuring the average thickness of the film as a function of time and is several angstroms per minute (A/min. ) as a unit. The preferred deposition rate is about 20 A/min. Or higher, and a better deposition rate of about 50 A/min. Or higher, and the optimum deposition rate is about 100 "min· or higher. Suitable methods for measuring film thickness include the multiple-point ellipsometric method. Instruments for measuring film thickness are also well known and commercially available, and preferably include NanoSpec® series instruments from Nanometrics, Inc., Sunnyvale, California. The term "uniformity" as used herein to mean the uniformity of the deposited film is also used to represent thickness uniformity and composition uniformity. The uniformity of the film thickness is preferably measured by multi-point film thickness measurement, and then the intermediate thickness is determined, and then the multiple measurement averages different from the middle number are determined. In order to be able to compare, 27 12771 said pifl. Doc/012 The percentage non-uniformity is used to indicate the result. The preferred percentage non-uniformity is about 10% or less, and more preferably about 5% or less, and most preferably about 2% or less. The uniformity of composition is measured by electrical measurements (such as four-point probe), secondary ion mass spectrometry (SIMS), Rutherford backscattering spectroscopy (RBS), and spectroscopy. Spectroscopic ellipsometry and/or high resolution X-ray diffractometry (HR-XRD). Figure 14 shows the 1306 A/min. Laceford backscattering spectroscopy (RBS) of amorphous ruthenium film deposited with trioxane at deposition rate, 40 Torr pressure and 600 ° C deposition temperature [Elastic Rec〇1 Detecnon spectrum] ]. The solid line represents the raw material of the film, and the dotted line is a data simulation software RUMPTM with assuming a residual hydrogen concentration of 0. 5 at. Model under %. The raw data refers to the slight surface contamination from the absorbed hydrocarbons or water vapor, but the energy spectrum represents the concentration of hydrogen remaining in the membrane under the detection limit, which is equivalent to less than 0. . 2 at·% chlorine concentration. Figure 15 shows an X-ray diffraction pattern of a series of ruthenium films deposited with trioxane at 600 ° C, 650 ° C, 700 ° C and 750 ° C (from the bottom to the top of Figure 15 respectively). The X-ray diffraction pattern shows that the film deposited at 650 ° C is partially crystalline at 60 (the TC deposited film is amorphous), while the film deposited at 7 ° C and 75 ° C is incrementally Crystallization. Figure 16 shows a section of the film (intermediate layer) deposited at 750 ° C. 28 1277 1 ^46 pifl. Doc/012 Penetrating electron micrograph showing a considerable degree of film thickness and thinness in a polycrystalline film deposited with Sanshayuan. A selected area diffraction (SAD) pattern of a film (as shown in Figure 17) shows that there is no preferential orientation in the film, i.e., a polycrystalline film. Figure 18 is a scanning electron micrograph of a non-crystalline tantalum film profile deposited with trioxane at 600 ° C and 40 Torr. The film deposited on the curved substrate is not deep and has excellent conformality in the slit. In another preferred embodiment, higher order decane is also used in the low temperature low pressure chemical vapor deposition synthesis of a composition of tantalum nitride (SiN) material ranging from almost pure tantalum to Sl3N4. The nitrogen source preferably includes, for example, a chemical precursor of trisilylamine (also written as (H3Si)3N], ammonia, atomic nitrogen (nitrogen) and nitrogen trifluoride (NF3). Preferably, the atomic nitrogen is produced by a remote microwave radical generator. The amount of nitrogen source introduced into the chemical vapor deposition chamber associated with the higher decane is preferably selected to provide a higher level of uniformity than the control film made using decane instead of higher decane. In a preferred embodiment, the atomic nitrogen is introduced continuously and is not continuously introduced or pulsed into the trioxane as one or more pulses. As explained in the following examples, it is known that a higher film uniformity can be obtained by pulse introduction of higher decane, and a very thin and highly uniform tantalum nitride film can be obtained by batch chemical vapor deposition. The tantalum nitride film prepared in accordance with the embodiment has a preferred thickness in the range of from about 10 angstroms to about 300 angstroms, more preferably from about 15 angstroms to 150 angstroms. The use of these nitrogen sources as a chemical precursor and in combination with trioxane is particularly applicable to 29 1277139 08846pifl. Doc/012 does not have a very small amount of nitrogen-hydrogen bonds in the film deposited at low temperatures, and its deposition rate is much higher than that of conventional germanium sources such as decane. The same result can be obtained with other higher decane. When the deposition temperature exceeds 450 ° C, the hydrogen content is preferably less than 4 at. %, more preferably less than 2 at. % ‘best is less than 1 at·%. The above deposition is preferably carried out in a mass transfer limiting specification. In another embodiment, higher order decane may also be used in the low temperature low pressure chemical vapor deposition synthesis of a cerium oxide material and a cerium oxynitride material. Higher grades The advantages of the low temperature and local growth rate of Shixiyuan provide manufacturing advantages over the use of decane, especially under low pressure chemical vapor deposition. Oxygen sources may include ozone, oxygen, water, nitric 0Xide, hydrogen peroxide, and the same source of oxygen. The nitrogen source used to direct nitrogen into these materials includes trisalamine, ammonia, atomic nitrogen and nitrogen trifluoride as described above. These oxygen sources and nitrogen sources can be used in conjunction with each other, either in a non-joining step or in a combination comprising these processes. The above deposition is preferably carried out in a mass transfer limiting specification or close to a mass transfer limiting specification. The deposition with tridecylamine and trioxane is preferably carried out in the range of from about 350 ° C to about 750 ° C, more preferably from about 40 (TC to about 700 ° C, most preferably at about 450 ° C to Approximately 650 ° C. The deposition with nitrogen trifluoride and trioxane is preferably carried out in the range of from about 300 ° C to about 750 ° C, more preferably from about 350 ° C to about 700 ° C, most preferably From about 400 ° C to about 650 ° C. Although individual examples of oxides and oxynitrides have not been proposed, those skilled in the art will readily appreciate the principles disclosed herein, as well as the above-described layers of tantalum nitride and tantalum. Equivalently applied to the deposition of yttrium oxide. Similarly, the advantages of 30 I2771^46pi, doc/012, and the lower activation energy and lower temperature to achieve mass transfer are beneficial to vapor deposition, especially Chemical vapor deposition of various bismuth compounds. A preferred embodiment provides films for a variety of applications in the microelectronics industry. Preferably, the ruthenium containing film has a thickness non-uniformity of less than about 2% and a composition of less than about 2%. Uniformity. The films described herein can be used in a variety of applications, such as transistor gate electrodes. The layers are particularly useful for forming key component layers in integrated circuits, such as the accumulation of gate layers in transistors. Other examples include semiconductor layers in heterojunction bipolar transistors (abbreviated as HBT's). Processes for integrated circuits are well known to those skilled in the art. These integrated circuits can be coupled to computer systems by methods known to those skilled in the art, so a preferred embodiment is provided by one or more A computer system consisting of integrated circuits. Figure 1 is a diagram of a preferred manufacturing process step described herein. Step 100, forming a gate dielectric layer on a semiconductor substrate. If necessary, step 110 The surface of the dielectric layer. The ruthenium containing film, preferably including the flowing trioxane, is then deposited as described above in step 120. Next, if lateral signal transmission is desired, a metal layer may be deposited in step 130. On the germanium-containing film, then, in step 140, the gate electrode is patterned, and in step 150, the fabrication of the integrated circuit is continued. Figure 2 is in accordance with the process of Figure 1. A schematic diagram of a gate stack 200. A gate dielectric layer 210 is formed on a semiconductor substrate 220, an electrically doped germanium-containing film 230 is formed on the gate dielectric layer 210, and a non-essential layer is placed on the germanium-containing film 230. Metal layer 240, thereby forming a gate 31 1277139 1 08^46pifl. Doc/012 Stack 200. The laminate 200 is then patterned to form a gate electrode (not shown in the second _), and the fabrication of the integrated circuit continues. It is preferred that the gate dielectric layer 210 can comprise at least one high dielectric constant material and have a dielectric constant greater than 5, more preferably a dielectric constant greater than 1 。. Exemplary materials include aluminum oxide, lead oxide, and antimony oxide, and are preferably formed by atomic layer deposition (ALD) to form a pinhole free layer. The use of trioxane in or near the mass transfer limit, especially in combination with higher broths, facilitates compensating for the slow nucleation time of conventional tantalum deposited on such high dielectric constant materials. In another example, an epitaxial germanium-containing layer is deposited on a single crystal substrate and the fluid is trioxane. The tantalum layer and the heterogeneous epitaxial germanium, tantalum carbide and tantalum carbon layers can be deposited by the processes described herein. Another preferred embodiment provides an apparatus for depositing a ruthenium containing material on a surface. The apparatus includes a chemical vapor deposition chamber (CVD chamber), a vessel containing a triple sand chamber, and a feed line for operating the connecting conduit to the chemical vapor deposition chamber to cause the conduit to The passage of trioxane in the fJ chemical vapor deposition chamber, and a temperature controller disposed around the conduit and maintained at a temperature in the range of about 10 ° C to 70 ° C, preferably at 15 ° C to about 52 t Control the evaporation rate of trioxane. Examples of suitable temperature controllers include thermoelectric controllers (*) and/or liquid-filled jackets. The preferred chemical vapor deposition chamber is a single wafer horizontal gas flow reactor. The entire instrument may also include a manifold for operating the connection feed tube 32 12771 fear 46 pifl. Doc/012 line to control the passage of the conduit to the trioxane of the chemical vapor deposition chamber. Preferably, a heat source is disposed around the feed line and the gas line is heated to between about 35 ° C and 70 ° C, more preferably between about 40 ° C and 52 ° C, to avoid condensation at high gas flow rates. Further, trioxane is preferably introduced by bubbling a trioxane vapor by a carrier gas, preferably a temperature-controlled bubble in combination with a heated gas line for transporting trioxane. Examples The following examples were performed with the ASM EpsUon 2000TM horizontal airflow epitaxy reactor system and equipped with a Bernoulli wand wafer transfer system, pure-only load lock, non-sliding concave surface A non-slide concave susceptor, a preheating ring, an adjustable spot lamp, and a tunable gas mlet injector. The ruthenium containing and ruthenium containing precursors are supplied to the chamber in a feed line and may contain hydrogen and diborane dopants. Then 120 seeming 1% diborane (B2H6) in hydrogen was diluted in 2 slm of hydrogen, and 120 seem of the mixture was introduced into a reactor mixed with 20 slm of hydrogen and precursor, and in the example shown below The flow rate is deposited on a rotating substrate. The deposition rate was judged by a secondary ion mass spectrometer and an optical ellipsometer (Nanometnc) from the depth profile of oxygen and boron. Mm 1-4 According to Table 1, the ruthenium-containing film was deposited using trioxane as a chemical precursor. The deposition temperature is 70 (TC. 33 127713⁄4 846pifl.) for the mass transfer specification of trioxane. Doc/012 However, because the flow rate (under these special deposition conditions) is not suitable for providing a uniform film, the final film is not uniform but has a concave profile on the deposition profile (thinner, thicker edges). Table 1 No. Temperature re) Pressure (Torr) Flow Rate Setting 値 (seem) Precursor Base Deposition Profile 1 700 40 50 Si, H« SiCh Concave 2 700 40 45 SiA Si〇, Concave 3 700 40 15 Si, Η, Si〇 , concave 4 700 40 25 Si, Η, Si〇, concave example 5~15 According to the parameters shown in Table 2, using trioxane and decane as chemical precursors and diborane as dopant to deposit yttrium-containing amorphous film . Diluted 120 seem of 1% diborane (B2H6) in hydrogen to 2 slm of hydrogen, and introduced 120 seem of the mixture into a reactor mixed with 20 slm of hydrogen, trioxane and decane, wherein the flow rates of trioxane and decane The conditions are shown in Table 2. These results show that the use of trioxane at a preset temperature results in a higher deposition rate than the use of decane, even in the case where the trioxane flow rate is lower than the decane flow rate. 34 I27713?46pifl,oc/012 Table 2 No. Temperature (°C) Pressure (Torr) Flow rate see (seem) Precursor Base Deposition rate (A/min. 5C 650 40 50 SiH4 _ Si09 46 6C 650 40 50 SiH4 Si <10〇〉 68 7 650 40 50 Si.H, Si <l〇〇> 462 8C 600 40 50 SiH4 Si02 19 9C 600 40 50 SiH4 Si <l〇〇> 9 10 600 40 20 Si, H8 [TsiO. 359 11 600 40 15 SiA Si <l〇〇> 181 12C 550 760 25 SiH4 Si〇2 <1 13C 550 40 50 SiH4 Si〇2 7 14 550 40 30 SiA —Si〇? 287 15C 550 40 50 SiH4 Si〇2 2 Examples 16 to 19 According to the parameters shown in Table 3, tritane and decane were used as chemical precursors to deposit the ruthenium-containing film. The deposition time can be adjusted to give each film an average thickness of about 500 angstroms. The average thickness was then measured using an ellipsometric measure of Nanometnc and these numbers were assigned by the deposition time to determine the deposition rate. The film non-uniformity can be judged by the 49-point thickness map of the film thickness. The results show that the use of trioxane instead of decane at the specified temperature results in a more uniform film at very high deposition rates. It can be at 550 ° C, but it is more attractive at 600 ° C. 35 /012 /012 Table 3 No. Precursor Temperature (°C) % Inhomogeneity 1 SiH4 600 5.93 2 Si, Η, a 600 0.83 3 SiH4 550 8.5 ^ 4 Si, H8 a 550 7.31 ~~ Sink. Product rate _example 20~38

用80%三矽烷與20%二鍺烷的混合物代替只用三矽 烷以及用80%矽烷與20%二鍺烷的混合物代替只用矽烷, 以重覆範例1〜19而獲得矽鍺膜。這比只使用三矽烷與矽 烷能得到較高的沉積速率。 範例39A ruthenium film was obtained by substituting a mixture of 80% trioxane and 20% dioxane instead of trioxane alone and a mixture of 80% decane and 20% dioxane instead of decane alone to repeat Examples 1-19. This gives a higher deposition rate than using only trioxane and decane. Example 39

參考第3圖所示的流程並藉由下列不連續地週期超晶 格成長提供一矽鍺膜。於步驟300中,準備一矽<100>基 底,係於高流量超純氫氣下將基底置入一反應器中,再藉 由施行一 d-幻·^7氟化氣(HF)最近淸潔來去除原生氧化物 (native oxide)。當晶圓加熱至9〇〇°C時在一高流量氫氣下於 60 rpm旋轉(以去除基底表面上之任何污染物)。然後,冷 卻並提供晶圓穩定在700°C下,並且在質量傳輸限定條件 下用三砂院和三砂胂(trisilylarsine)成長厚度約300埃的摻 雜砷砂緩衝層。 於步驟310中,調整設定點至600°c ’係藉由在氫氣 流下冷卻,以調整晶圓溫度。於步驟320中,用98%二矽 烷與2%二鍺烷沉積第一階段矽鍺超晶格。於步驟330中, 36 127711·舰 用85%三矽烷與15%二鍺烷沉積第二階段矽鍺超晶格。 在氫氣流下,於步驟340中,降低3°C設定點溫度, 並且使晶圓穩定30秒。再於步驟350中,用75%三矽烷 與25%二鍺烷沉積第三階段矽鍺超晶格。 在氫氣流下,於步驟360中,降低3°C設定點溫度, 並且使晶圓穩定30秒。再於步驟370中,用65%三矽烷 與35%二鍺烷沉積第四階段矽鍺超晶格。於步驟380中, 用85°/。三矽烷、12%二鍺烷、2%二硼烷與1%二矽甲烷 沉積第五階段摻雜硼與摻雜碳的矽鍺超晶格。在氫氣流 下,於步驟390中,淸潔反應器30秒。再於步驟400中, 用90%三矽烷與二鍺烷沉積第六階段矽鍺超晶格。 在氫氣流下,於步驟410中,增加設定點至650°C, 並且稍微調節發射器組(bank)的相對能源’以最大化成長 的石夕頂蓋層之晶圓中心均勻性(within_wafer uniformity)。然 後提供晶圓穩定化30秒。於步驟420中’用100%三矽烷 沉積矽頂蓋層。最後’於步驟430中,將晶圓移出反應器, 再進行下一晶圓的製程。 範例40 用三矽烷與鍺烷作爲化學前驅物在沉積溫度650°C、 壓力40 Torr下沉積平均厚度1038埃的含矽膜。而在一系 列先前運轉中氣流注射器設定點已被憑經驗調整於平常方 法中。具有百分比不均勻性0.37% (8埃的範圍)的最終矽 鍺膜藉由用排除6釐米邊緣的49點線性尺寸掃描來測得。 第4圖所示係這個膜膜厚與測量site)之關 fl.doc/012 係圖。 範例41(對照版) 用矽烷與鍺烷作爲前驅物在60(TC溫度下於一二氧化 矽(Si02)基底上(沒有成核層)沉積一含矽膜。而最終矽鍺膜 的表面粗糙度用原子力顯微鏡(atomic force microscopy,簡 稱AFM)於10微米x 10微米掃描面積測量所得是226埃。 而砂鍺膜經由掃描式電子顯微鏡(scanning electron microscopy,簡稱SEM)揭示出表示爲島型沉積(island-type deposition)的角錐狀、有小平面晶粒(Pyramidal, faceted grain),如第5圖與第6圖所顯示的掃描式電子顯微相片。 範例42 如範例41所述在600°C沉積一含矽膜,但是用三矽烷 與鍺烷取代矽烷與鍺烷作爲前驅物。而最終矽鍺膜的表面 粗糙度用原子力顯微鏡(AFM)於10微米X 10微米掃描面積 測量所得是18.4埃。而矽鍺膜經由掃描式電子顯微鏡(SEM) 揭示出一均勻得多的表面,如第7圖與第8圖所顯示的掃 描式電子顯微相片(分別與第5圖、第6圖有相同放大倍率 與傾斜角)。 範例43〜63 用三矽烷與鍺烷在40 Torr壓力下於二氧化矽(Si02)基 底上(沒有成核層)沉積一系列含矽膜。而三矽烷的流速如 表四之範例是固定於77 seem(氫氣載體、氣泡)。於表四所 示尙有鍺烷流量(10%鍺烷、90%氫氣)與沉積溫度是可變 的。最終矽鍺膜的鍺濃度(atomic %)與厚度係藉由拉塞福 38 c/012 1277 ll.d。 回向散射光譜(RBS)測定的,而表面粗糙度是由原子力顯 微鏡(AFM)測定的。表四之結果顯示在溫度範圍上與流速 條件下且特別於一鍺濃度範圍內可準備高均勻性的膜。 表四 編號 溫度 CC) 鍺烷流量 (seem) %鍺 厚度(A) 沉^速率 (A/min.) 粗糙度 0 (A) 43 450 25 5.0 34* 8.5 3.2 44 450 50 7.5 34* 11 4.1 45 450 100 * 11 59* 15 3.7 46 450 100 11 53* 13 nd 47 500 25 6.0 190 63 7.8 48 500 50 10 230 77 9.1 49 500 100 13.5 290 97 8.3 50 500 100 13.5 380* 127 7.2 51 550 25 6.0 630 315 5.2 52 550 50 9.5 670 335 13.6 53 550 100 14 300 450 12.1 54 550 100 14 1016 508 9.4 55 600 25 7.0 1160 580 8.1 56 600 50 13 1230 615 25.7 57 600 100 19 1685 843 31.8 58 650 25 11 630 630 23.3 59 650 50 17 800 800 31.5 60 650 100 27 1050 1050 50.2 61 700 25 11 680 680 18.1 62 700 50 18 835 835 37.8 63 700 100 31 960 960 、 44.9 >代表其厚度是以光學技術測量的 nd :未測定的 39 12771¾¾ 46pifl.doc/012 範例64〜78 於表五中用三矽烷與氨(範例64〜77)或矽烷與氨(範例 78)在矽<100>基底的原生氧化層上沉積一系列含矽膜。而 載體氣體的流量是30 slm以及氨氣流速是7 slm。表五不 但顯示最終氮化矽(SiN)膜所測得的沉積速率與折射率 (refractive index,即“RI”),也有選出的膜之砂對氮的原子 比(atomic ratio of silicon to nitrogen,即“Si/N”)與氫含量(即 “%H”)。 表五 編 號 壓力 (Torr) 溫度 (V) 載體 矽源/流速 (seem) 沉積速率 (A/min.) Si/N %H RI 64 20 625 n2 三矽烷/20 124 0.85 4 2.074 65 20 725 n2 三矽烷/20 149 0.85 4 2.034 66 20 725 n2 三矽烷/80 585 0.95 4 2.182 67 20 725 h2 三矽烷/80 611 1.0 2.2 2.266 68 20 775 n2 三矽烷/20 158 0.88 4 2.010 69 20 775 h2 三矽烷/20 117 0.88 3 1.999 70 20 775 n2 三矽烷/40 308 0.85 4 2.053 71 20 775 n2 三矽烷/80 582 0.88 4 2.101 72 20 775 h2 三矽烷/80 600 0.88 3.5 2.146 73 20 775 n2 三矽烷/160 1050 0.88 4 2.141 74 20 775 h2 三矽烷/160 1283 0.92 3.5 2.281 75 20 775 n2 三矽烷/80 346 nd nd 2.006 76 100 775 n2 三矽烷/160 589 nd nd 2.028 77 100 775 h2 三矽烷/160 244 nd nd 2.012 78 100 775 n2 矽烷/40 208 nd nd 2.007 nd :未測定的 矽對氮的原子比(“Si/N”)與氫含量係藉由拉塞福 40 12771¾ ;46pifl.doc/012 回向散射光譜(RBS)測定的。第19圖所示係用三矽烷在775 °C、20T〇rr下沉積的氮化矽樣本的拉塞福回向散射光譜圖(2 MeV He++)。第20圖所示爲用反衝偵測(ERD)所獲得的彈性 反衝偵測光譜。這些圖均顯示原始資料與依照可作矽、氮、 氫濃度量化的RUMP製造模型程式(modeling program)而得 的模擬(smuilat1〇n)。而模擬圖代表那個膜有約爲Si45N51H4 的化學計量比。第20圖之拉塞福回向散射光譜彈性反衝 偵測光譜(RBS ERD spectrum)也揭露氫是均勻分布於整個 膜。 範例79〜82 用三矽烷與原子氮在矽<1〇〇>基底的原生氧化層上沉 積一系列含矽材料。用一商業上可供應的800 watt微波輻 射產生益(microwave radical generator,簡稱 MRG)去產生原 子氮’並且將其供應至化學氣相沉積腔體。利用一氮氣載 體氣體於5 slm流速(範例82中是1〇 sim)下將三矽烷與原 子氮一起經由一發泡器(bubbler)供應到化學氣相沉積腔 體’而其沉積溫度如表六所示。通入三矽烷至腔體不是如 範例79爲連續地就是如範例8〇〜82是脈衝的(in pulse)。脈 衝式導入是藉由連續導入原子氮以及在約1分30秒間隔 下脈衝導入三矽烷。在上述流動條件下,每三矽烷脈衝持 續約6秒。每一最終氮化矽膜具有範圍近乎Sl43N54 56H3 i 的化學計量比。 表六顯示了最終氮化矽膜中的厚度、折射率與氫程度 (hydrogen level)。範例79的氮化矽膜因爲中心明顯厚於邊 41 oc/012 I277l^46pifl.d 緣所以是不均句的’而且橫跨膜表面所其測得之折射率也 有變化(在中心高於邊緣)。而使用脈衝式製程的範例80〜82 則改善的均勻性。採取增加原子氮流速以及/或是減少三矽 院流速的連續製程也能獲致均勻的膜。 表六 編號 製程 沉積溫度 (°C) 折射率 %H 中心 邊緣 79 連續 650 869 J 510 1.97-2.2 2 80 脈衝 650 324 268 1.98 2 81 脈衝 600 635 655 1.96 3 82 脈衝 650 1115 1174 2.02 0.7 節例83 大體上如範例80〜82所述,在650°C、3 Torr壓力下 使用遙控產生的原子氮與單一六秒脈衝的三矽烷沉積一厚 度約18埃的連續氮化矽膜。這個膜塗上環氧物並剖面, 以使用穿透式電子顯微鏡(transmission electron microscopy,簡稱TEM)取像,如第9圖所示的穿透式電子 顯微相片。可發現膜與基底的界面沒有原生氧化物。 範例84〜87 以表七所示之沉積溫度與沉積速率在沉積壓力40Torr 下,用三矽烷於淸潔過的矽<100>基底上沉積一系列磊晶 矽膜。如表七所示,由拉塞福回向散射頻道光譜(channeling spectrum)得到的Z -mm値可知能製造出高品質磊晶矽膜。 42 12771¾ 46pifl.doc/012 表七 編號 沉積溫度 fc) 沉積速率 (A/min.) χ -min(%) 84 650 47 2.7 85 650 50 3.1 86 600 145 2.9 87 650 460 3.2 雖然本發明已以一較佳實施例揭露如上,然其並非用 以限定本發明,任何熟習此技藝者,在不脫離本發明之精 神和範圍內,當可作些許之更動與潤飾,因此本發明之保 護範圍當視後附之申請專利範圍所界定者爲準。 43Refer to the procedure shown in Figure 3 and provide a ruthenium film by the following discontinuous periodic superlattice growth. In step 300, a 矽 <100> substrate is prepared, and the substrate is placed in a reactor under high flow of ultrapure hydrogen, and then a d-magic fluorinated gas (HF) is applied. Clean to remove native oxide. Rotate at 60 rpm under a high flow of hydrogen (to remove any contaminants on the substrate surface) when the wafer is heated to 9 °C. Then, the wafer was cooled and stabilized at 700 ° C, and a arsenic-doped buffer layer having a thickness of about 300 angstroms was grown with tri-sand and trisilylarsine under mass transfer conditions. In step 310, the set point is adjusted to 600 ° c' to cool the wafer temperature by cooling under a stream of hydrogen. In step 320, a first stage ruthenium superlattice is deposited with 98% dioxane and 2% dioxane. In step 330, 36 127711. Marine 85% trioxane and 15% dioxane are deposited in a second stage ruthenium superlattice. Under a stream of hydrogen, in step 340, the set point temperature is lowered by 3 ° C and the wafer is allowed to stabilize for 30 seconds. Further in step 350, a third stage ruthenium superlattice is deposited with 75% trioxane and 25% dioxane. Under a stream of hydrogen, in step 360, the 3 °C set point temperature is lowered and the wafer is allowed to stabilize for 30 seconds. Further in step 370, a fourth stage ruthenium superlattice is deposited with 65% trioxane and 35% dioxane. In step 380, 85°/ is used. Trioxane, 12% dioxane, 2% diborane and 1% dimethane methane were deposited in a fifth stage boron-doped and carbon-doped germanium superlattice. Under a stream of hydrogen, in step 390, the reactor was simmered for 30 seconds. In a further step 400, a sixth stage ruthenium superlattice is deposited with 90% trioxane and dioxane. Under the hydrogen flow, in step 410, the set point is increased to 650 ° C, and the relative energy of the emitter bank is slightly adjusted to maximize the wafer center uniformity (within_wafer uniformity) of the growing stone roof cap layer. . Wafer stabilization is then provided for 30 seconds. The top cover layer is deposited with 100% trioxane in step 420. Finally, in step 430, the wafer is removed from the reactor and the next wafer is processed. Example 40 A ruthenium-containing film having an average thickness of 1,038 angstroms was deposited using trioxane and decane as chemical precursors at a deposition temperature of 650 ° C and a pressure of 40 Torr. The flow injector set point has been empirically adjusted to the usual method for a prior range of operations. The final ruthenium film with a percent non-uniformity of 0.37% (8 angstrom range) was measured by scanning with a 49 point linear dimension excluding the 6 cm edge. Figure 4 shows the film thickness and the measured site. fl.doc/012 diagram. Example 41 (Comparative Edition) A ruthenium-containing film was deposited on a cerium oxide (SiO 2 ) substrate (without a nucleation layer) at 60 (TC temperature) using decane and decane as precursors. The atomic force microscopy (AFM) was measured at a scan area of 10 μm x 10 μm and was 226 angstroms. The sand raft film was revealed to be island-type deposition by scanning electron microscopy (SEM). (Island-type deposition) pyramidal, faceted grain, as shown in Figures 5 and 6. Scanning electron micrographs are shown in Example 41 at 600°. C deposits a ruthenium-containing film, but replaces decane and decane with trioxane and decane as precursors. The surface roughness of the final ruthenium film is measured by atomic force microscopy (AFM) at a scan area of 10 μm×10 μm. The enamel film reveals a much more uniform surface via a scanning electron microscope (SEM), such as the scanning electron micrographs shown in Figures 7 and 8 (respectively with Fig. 5 and Fig. 6, respectively). Have the same put Magnification and tilt angle.) Examples 43 to 63 A series of ruthenium-containing films were deposited on a cerium oxide (SiO 2 ) substrate (without a nucleation layer) using trioxane and decane at a pressure of 40 Torr. The fourth example is fixed at 77 seem (hydrogen carrier, bubble). The decane flow rate (10% decane, 90% hydrogen) and the deposition temperature are variable as shown in Table 4. The final ruthenium concentration of the ruthenium film (atomic %) and thickness were determined by Raspford 38 c/012 1277 ll.d. Backscattering spectroscopy (RBS), and surface roughness was determined by atomic force microscopy (AFM). The results in Table 4 show A film of high uniformity can be prepared over a range of temperatures and flow rates, and particularly within a range of concentrations. Table 4 No. Temperature CC) decane flow (seem) % 锗 thickness (A) Sink rate (A/min. Roughness 0 (A) 43 450 25 5.0 34* 8.5 3.2 44 450 50 7.5 34* 11 4.1 45 450 100 * 11 59* 15 3.7 46 450 100 11 53* 13 nd 47 500 25 6.0 190 63 7.8 48 500 50 10 230 77 9.1 49 500 100 13.5 290 97 8.3 50 500 100 13.5 380* 127 7.2 51 550 25 6.0 630 315 5.2 52 550 50 9 .5 670 335 13.6 53 550 100 14 300 450 12.1 54 550 100 14 1016 508 9.4 55 600 25 7.0 1160 580 8.1 56 600 50 13 1230 615 25.7 57 600 100 19 1685 843 31.8 58 650 25 11 630 630 23.3 59 650 50 17 800 800 31.5 60 650 100 27 1050 1050 50.2 61 700 25 11 680 680 18.1 62 700 50 18 835 835 37.8 63 700 100 31 960 960 , 44.9 > represents the thickness of the nd measured by optical technique: undetermined 39 127713⁄43⁄4 46pifl.doc/012 Examples 64-78 In Table 5, a series of ruthenium is deposited on the primary oxide layer of ruthenium <100> using trioxane and ammonia (Example 64-77) or decane and ammonia (Example 78). membrane. The carrier gas flow rate is 30 slm and the ammonia gas flow rate is 7 slm. Table 5 shows not only the deposition rate and refractive index ("RI") measured by the final tantalum nitride (SiN) film, but also the atomic ratio of silicon to nitrogen. That is, "Si/N") and hydrogen content (ie "%H"). Table 5 No. Pressure (Torr) Temperature (V) Carrier Source/Flow Rate (seem) Deposition Rate (A/min.) Si/N %H RI 64 20 625 n2 Trioxane / 20 124 0.85 4 2.074 65 20 725 n2矽/20 149 0.85 4 2.034 66 20 725 n2 Trioxane / 80 585 0.95 4 2.182 67 20 725 h2 Trioxane / 80 611 1.0 2.2 2.266 68 20 775 n2 Trioxane / 20 158 0.88 4 2.010 69 20 775 h2 Trioxane / 20 117 0.88 3 1.999 70 20 775 n2 Trioxane / 40 308 0.85 4 2.053 71 20 775 n2 Trioxane / 80 582 0.88 4 2.101 72 20 775 h2 Trioxane / 80 600 0.88 3.5 2.146 73 20 775 n2 Trioxane / 160 1050 0.88 4 2.141 74 20 775 h2 trioxane / 160 1283 0.92 3.5 2.281 75 20 775 n2 trioxane / 80 346 nd nd 2.006 76 100 775 n2 trioxane / 160 589 nd nd 2.028 77 100 775 h2 trioxane / 160 244 nd nd 2.012 78 100 775 n2 decane / 40 208 nd nd 2.007 nd : undetermined atomic ratio of ruthenium to nitrogen ("Si/N") and hydrogen content by Russell 40 127713⁄4; 46pifl.doc/012 backscattering Spectroscopic (RBS) determination. Figure 19 shows the Rutherford backscattering spectrum (2 MeV He++) of a tantalum nitride sample deposited with trioxane at 775 °C and 20 T rr. Figure 20 shows the elastic recoil detection spectrum obtained by recoil detection (ERD). These figures show the raw data and the simulation (smuilat1〇n) obtained from the RUMP manufacturing model that can be quantified as enthalpy, nitrogen, and hydrogen concentration. The simulation map represents that the film has a stoichiometric ratio of about Si45N51H4. Figure 20, the Laceford backscattering spectral elastic recoil detection spectrum (RBS ERD spectrum) also reveals that hydrogen is uniformly distributed throughout the membrane. Examples 79 to 82 use a mixture of trioxane and atomic nitrogen on the primary oxide layer of the ruthenium <1〇〇> substrate to deposit a series of ruthenium-containing materials. A commercially available 800 watt microwave radiation generates a radical radical generator (MRG) to produce the atomic nitrogen ' and supplies it to the chemical vapor deposition chamber. Using a nitrogen carrier gas at a flow rate of 5 slm (1 sim in Example 82), the trioxane is supplied to the chemical vapor deposition chamber via a bubbler together with the atomic nitrogen, and its deposition temperature is as shown in Table 6. Shown. The introduction of trioxane to the chamber is not as in the case of Example 79. Continuously, as in the case of 8 to 82, it is pulsed. Pulse introduction was carried out by continuously introducing atomic nitrogen and pulse-introducing trioxane at intervals of about 1 minute and 30 seconds. Under the above flow conditions, each trioxane pulse lasted for about 6 seconds. Each final tantalum nitride film has a stoichiometric ratio ranging from approximately S43N54 56H3 i . Table 6 shows the thickness, refractive index, and hydrogen level in the final tantalum nitride film. The tantalum nitride film of Example 79 is inconsistent because the center is significantly thicker than the edge of the edge 41 oc/012 I277l^46pifl.d and the measured refractive index of the transmembrane surface also changes (at the center above the edge) ). Examples 80 to 82 using pulsed processes improve uniformity. A uniform film can also be obtained by a continuous process that increases the atomic nitrogen flow rate and/or reduces the flow rate of the three chambers. Table 6 Numbering Process Deposition Temperature (°C) Refractive Index %H Center Edge 79 Continuous 650 869 J 510 1.97-2.2 2 80 Pulse 650 324 268 1.98 2 81 Pulse 600 635 655 1.96 3 82 Pulse 650 1115 1174 2.02 0.7 Section 83 In general, as described in Examples 80-82, a continuous tantalum nitride film having a thickness of about 18 angstroms was deposited using a remotely generated atomic nitrogen and a single six second pulsed trioxane at 650 ° C under a pressure of 3 Torr. This film was coated with an epoxy and profiled to take a transmission electron microscopy (TEM) image, such as a transmission electron micrograph shown in Fig. 9. It can be found that the interface between the film and the substrate has no native oxide. Examples 84 to 87 A series of epitaxial ruthenium films were deposited on a cleaned ruthenium <100> substrate using trioxane at a deposition temperature of 40 Torr at a deposition temperature and a deposition rate as shown in Table 7. As shown in Table 7, Z-mm値 obtained from the Laceford backscattering channel spectrum can be used to produce a high-quality epitaxial film. 42 127713⁄4 46pifl.doc/012 Table 7 Numbered deposition temperature fc) Deposition rate (A/min.) χ -min(%) 84 650 47 2.7 85 650 50 3.1 86 600 145 2.9 87 650 460 3.2 Although the invention has been The preferred embodiments are disclosed above, but are not intended to limit the present invention. Any one skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. The scope defined in the appended patent application shall prevail. 43

Claims (1)

丨 8M6pifl.doc/012 拾、申請專利範圍: 1. 一種半導體薄膜之改良式沉積製程,係於一表面沉積 一高度均勻含矽材料,其步驟包括: 提供配置有一基底的一腔體,該基底具有一被控制溫 度用於使用三矽烷蒸氣的沉積,該被控制溫度係被選擇以 確立質量傳輸限定條件; 引導包括三矽烷的一氣體到該腔體中,其中該氣體的 一流量係選擇用以改進相對用矽烷取代三矽烷的沉積之沉 積均勻性;以及 於該基底上沉積一含矽膜。 2. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜是磊晶的。‘ 3. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 1 積製程,其中該含矽膜是多晶質的。 4. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該被控制溫度是在從450°C到750°C的範圍 內。 ❿ 5. 如申請專利範圍第4項所述之半導體薄膜之改良式沉 積製程,其中該被控制溫度是在從550°C到650°C的範圍 內。 6. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中於該基底上沉積該含矽膜的速率約每分鐘50 埃或更高。 7. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 44 &gt;Pifl.doc/012 積製程,其中於該基底上沉積該含矽膜的速率約每分鐘100 埃或更高。 8. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中非晶質的該含矽膜具有橫過該基底約5%或 更低的一厚度不均勻性。 9. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中非晶質的該含矽膜具有橫過該基底約1%或 更低的一厚度不均勻性。 10. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該氣體更包括由一或多個選自於包括矽烷、 鍺院、二鍺院、二鍺院、二氟化氮、一砂甲院、二砍甲院、 三矽甲烷、四矽甲烷與一摻質前驅物的群組之化合物。 11. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該氣體更包括二鍺烷。 12. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該腔體是一單一晶圓水平氣流反應器。 13. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜是選自於包括一微小縮點、一矽鍺 膜、一矽鍺碳膜、一氮化矽膜、一氧化矽膜、一氮氧化矽 膜、一摻雜硼膜、一摻雜砷膜、一摻雜磷膜、一摻雜銦膜、 一摻雜銻膜以及具有介電常數約2.2或更低之膜的群組。 14. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜是矽以及該基底是具有高介電常數 的材料。 45 12771¾ 46pifl.doc/012 15. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中更包括圖案化該含矽膜,以形成一電晶體閘 極。 16. —種於表面沉積含矽材料的製程,包括: 提供配置有一基底的一化學氣相沉積腔體; 引導包括三矽烷的一氣體到該化學氣相沉積腔體中; 以及 _ 在高於525°C的一溫度下於該基底上沉積一含矽膜, 該含矽膜在一較高沉積速率下比用矽烷取代三矽烷製造的 一對照膜具有一較高度的均勻性。 17. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該基底是維持在是約550°C或更高的溫度下。 18. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該基底是維持在是約620°C或更高的溫度下。 19. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該基底是維持在是約▽⑻它或更高的溫度下。 20. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該基底是維持在從450°C到700°C的溫度範 圍內。 21. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該基底是維持在從525°C到650°C的溫度範 圍內。 22. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中沉積該含矽膜的速率約每分鐘50埃或更高。 46 I2771396pifLd〇c/〇i2 23. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中沉積該含矽膜的速率約每分鐘100埃或更高。 24. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該氣體更包括由一或多個選自於包括鍺烷、 二鍺院、三鍺院、三氟化氮、一砂甲院、二砂甲烷、三石夕 甲烷、四矽甲烷與一摻質前驅物的群組之化合物。 25. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該氣體更包括二鍺烷。 26. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該化學氣相沉積腔體是一單一晶圓水平氣流 反應器。 27. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜具有約5%或更低的一厚度不均勻 性。 28. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜具有約1%或更低的一厚度不均勻 性。 29. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜是選自於包括一微小縮點、一矽鍺 膜、一矽鍺碳膜、一氮化矽膜、一矽-氧膜、一矽-氧-氮膜、 一摻雜硼膜、一摻雜砷膜、一摻雜銦膜、一摻雜銻膜、一 摻雜磷膜以及具有介電常數約2.2或更低之膜的群組。 30. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜是矽以及該基底是具有高介電常數 12771¾ 46pifl.doc/012 的材料。 31. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜是磊晶的。 32. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜是多晶質的。 33. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中該含矽膜是非晶質的。丨 8M6pifl.doc/012 Pickup, Patent Application Range: 1. An improved deposition process for a semiconductor film by depositing a highly uniform germanium-containing material on a surface, the steps comprising: providing a cavity with a substrate, the substrate Having a controlled temperature for the deposition using trioxane vapor, the controlled temperature being selected to establish a mass transfer limiting condition; directing a gas comprising trioxane into the chamber, wherein a flow rate of the gas is selected To improve the deposition uniformity of the deposition of the trioxane relative to the decane; and to deposit a ruthenium-containing film on the substrate. 2. The improved deposition process for a semiconductor film according to claim 1, wherein the ruthenium-containing film is epitaxial. </ RTI> 3. The improved sink process of the semiconductor film of claim 1, wherein the ruthenium-containing film is polycrystalline. 4. The improved deposition process for a semiconductor film according to claim 1, wherein the controlled temperature is in the range of from 450 ° C to 750 ° C. ❿ 5. The improved deposition process of the semiconductor film of claim 4, wherein the controlled temperature is in the range of from 550 ° C to 650 ° C. 6. The improved deposition process for a semiconductor film according to claim 1, wherein the ruthenium-containing film is deposited on the substrate at a rate of about 50 angstroms per minute or more. 7. The modified sink film of the semiconductor film of claim 1, wherein the deposition rate of the ruthenium-containing film on the substrate is about 100 angstroms or more per minute. 8. The improved deposition process of a semiconductor film according to claim 1, wherein the amorphous ruthenium-containing film has a thickness non-uniformity of about 5% or less across the substrate. 9. The improved deposition process of a semiconductor film according to claim 1, wherein the amorphous ruthenium-containing film has a thickness non-uniformity of about 1% or less across the substrate. 10. The improved deposition process for a semiconductor film according to claim 1, wherein the gas further comprises one or more selected from the group consisting of decane, brothel, brothel, brothel, difluorination. A compound of the group of nitrogen, sandstone, squash, triterpene, tetramethane and a dopant precursor. 11. The improved deposition process for a semiconductor film according to claim 1, wherein the gas further comprises dioxane. 12. The improved deposition process for a semiconductor film according to claim 1, wherein the cavity is a single wafer horizontal gas flow reactor. 13. The improved deposition process for a semiconductor thin film according to claim 1, wherein the ruthenium-containing film is selected from the group consisting of a minute shrinkage point, a tantalum film, a tantalum carbon film, and a tantalum nitride layer. a film, a hafnium oxide film, a hafnium oxynitride film, a doped boron film, a doped arsenic film, a doped phosphor film, an doped indium film, a doped germanium film, and having a dielectric constant of about 2.2 or A group of lower membranes. 14. The improved deposition process of a semiconductor film according to claim 1, wherein the ruthenium containing film is ruthenium and the substrate is a material having a high dielectric constant. 45 127713⁄4 46 pifl.doc/012 15. The improved deposition process of the semiconductor film of claim 1, further comprising patterning the ruthenium containing film to form a transistor gate. 16. A process for depositing a ruthenium-containing material on a surface, comprising: providing a chemical vapor deposition chamber configured with a substrate; directing a gas comprising trioxane into the chemical vapor deposition chamber; and _ above A ruthenium-containing film was deposited on the substrate at a temperature of 525 ° C. The ruthenium-containing film had a higher degree of uniformity at a higher deposition rate than a control film made with decane instead of trioxane. 17. A process for depositing a ruthenium-containing material as described in claim 16 wherein the substrate is maintained at a temperature of about 550 ° C or higher. 18. A process for depositing a ruthenium-containing material as described in claim 16 wherein the substrate is maintained at a temperature of about 620 ° C or higher. 19. A process for depositing a ruthenium-containing material as described in claim 16 wherein the substrate is maintained at a temperature of about ▽(8) or higher. 20. The process of depositing a ruthenium-containing material as described in claim 16 wherein the substrate is maintained at a temperature ranging from 450 ° C to 700 ° C. 21. A process for depositing a ruthenium-containing material as described in claim 16 wherein the substrate is maintained at a temperature ranging from 525 ° C to 650 ° C. 22. A process for depositing a ruthenium-containing material as described in claim 16 wherein the ruthenium-containing film is deposited at a rate of about 50 angstroms per minute or more. 46 I2771396pifLd〇c/〇i2 23. A process for depositing a ruthenium-containing material as described in claim 16 wherein the ruthenium-containing film is deposited at a rate of about 100 angstroms per minute or more. 24. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the gas further comprises one or more selected from the group consisting of decane, dioxin, sputum, and nitrogen trifluoride. a compound of a group of sand sands, two sand methane, three stone methane, four methane and one dopant precursor. 25. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the gas further comprises dioxane. 26. The process of depositing a germanium-containing material on a surface as described in claim 16 wherein the chemical vapor deposition chamber is a single wafer horizontal gas flow reactor. 27. A process for depositing a ruthenium-containing material as described in claim 16 wherein the ruthenium-containing film has a thickness non-uniformity of about 5% or less. 28. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the ruthenium-containing film has a thickness non-uniformity of about 1% or less. 29. The process for depositing a ruthenium-containing material on a surface as described in claim 16 wherein the ruthenium-containing film is selected from the group consisting of a minute crater, a ruthenium film, a ruthenium carbon film, and a nitridation layer. a ruthenium film, a ruthenium-oxygen film, a ruthenium-oxygen-nitrogen film, a doped boron film, a doped arsenic film, an doped indium film, a doped yttrium film, a doped phosphor film, and a dielectric A group of films having a constant of about 2.2 or less. 30. The process of depositing a ruthenium-containing material as described in claim 16 wherein the ruthenium-containing film is ruthenium and the substrate is a material having a high dielectric constant of 127713⁄4 46 pifl.doc/012. 31. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the ruthenium-containing film is epitaxial. 32. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the ruthenium-containing film is polycrystalline. 33. The process of depositing a ruthenium-containing material on a surface as described in claim 16 wherein the ruthenium-containing film is amorphous. 34. 如申請專利範圍第16項所述之於表面沉積含矽材料 的製程,其中更包括圖案化以形成一電晶體閘極。 35. —種化合的含矽膜,於一積體電路上,該化合的含矽 膜具有約5%或更低的一厚度不均勻性,以及橫過該化合 的含矽膜在任一設定膜約2%或更低的深度之一組成不均 勻性。 36. 如申請專利範圍第35項所述之化合的含矽膜,包括 一電晶體閘極。34. The process of depositing a germanium-containing material as described in claim 16 of the patent application, further comprising patterning to form a transistor gate. 35. A compounded ruthenium-containing film having a thickness unevenness of about 5% or less on an integrated circuit, and a ruthenium-containing film across the compound in any of the set films One of the depths of about 2% or less constitutes unevenness. 36. A ruthenium-containing film as described in claim 35, comprising a transistor gate. 37. 如申請專利範圍第35項所述之化合的含矽膜,具有 約1%或更低的一厚度不均勻性。 38. 如申請專利範圍第35項所述之化合的含矽膜,包括 矽鍺膜。 39. 如申請專利範圍第35項所述之化合的含矽膜,包括 多晶材料。 40. 如申請專利範圍第35項所述之化合的含矽膜,包括 非晶材料。 41. 一種於表面沉積一矽鍺材料的製程,包括: 48 12771涊6 pifl.doc/012 提供配置有一基底的一化學氣相沉積腔體; 引導包括一較高級矽烷與一較高級鍺烷的一氣體到該 化學氣相沉積腔體中;以及 於該基底上沉積一矽鍺膜。 42. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該較高級矽烷是選自於包括二矽烷、三矽 烷與四矽烷的群組。 43. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該較高級鍺烷是選自於包括二鍺烷、三鍺 烷與四鍺烷的群組。 44. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該較高級矽烷是三矽烷以及該較高級鍺烷 是一錯院。 45. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中沉積該矽鍺膜是在從475°C到700°C的溫 度範圍內。 46. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中沉積該矽鍺膜的速率約每分鐘50埃或更 高。 47. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中沉積該矽鍺膜的速率約每分鐘100埃或更 高。 48. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該氣體更包括由一或多個選自於包括一矽 49 &gt;Pifl.doc/012 甲烷、二矽甲烷、三矽甲烷、四矽甲烷與一摻質前驅物的 群組之化合物。 49. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該化學氣相沉積腔體是一單一晶圓水平氣 流反應器。 50. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該矽鍺膜具有5%或更低的一厚度不均勻 性。 51. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該矽鍺膜具有比用矽烷取代該較高級矽烷 製造的一對照膜優良的均勻性。 52. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中該矽鍺膜具有比用鍺烷取代該較高級鍺烷 製造的一對照膜優良的均勻性。 53. 如申請專利範圍第41項所述之於表面沉積一矽鍺材 料的製程,其中更包括圖案化以形成一電晶體閘極。 54. —種矽鍺膜,於一積體電路上,該矽鍺膜具有約5% 或更低的一厚度不均勻性,以及約2%或更低的一組成不 均勻性。 55. 如申請專利範圍第54項所述之矽鍺膜包括一電晶體 閘極。 56. 如申請專利範圍第54項所述之矽鍺膜,該矽鍺膜具 有約1%或更低的一厚度不均勻性,以及約2%或更低的一 組成不均勻性。 50 I2771^46pifld〇c/〇i2 57.—種於表面沉積含矽材料的製程,包括·· 提供配置有一基底的一化學氣相沉積腔體,該化學氣 相沉積腔體裝配一溫度控制器,以於單一方法中能程式化 多個溫度控制變量; 輸入一溫度控制變量乃至該溫度控制器; 引導包括一 XA的第一含矽化學前驅物的一第一氣體 到该化學氣相沉積腔體中,其中Χι是在1χ 1〇·4到約1〇〇的範 圍內; 於該基底上沉積一第一含矽膜; 輸入一溫度控制變量τ2至該溫度控制器; 引導包括一 X2%的第二含矽化學前驅物的一第二氣體 到該化學氣相沉積腔體中,其中Χ2是在lx 10·4到約100 的範圍內’以及該第二含矽化學前驅物與該第一含矽化學 前驅物是相同與不同其中之一;以及 於該第一含矽膜上沉積一第二含矽膜, 藉以形成具有約5%或更低的一厚度不均勻性與約2% 或更低的一組成不均勻性的一多層含矽膜。 58·如申請專利範圍第57項所述之於表面沉積含矽材料 的製程’其中該溫度控制變量1與T2是溫度控制設定讓占。 59·如申請專利範圍第57項所述之於表面沉積含矽树料 的製程,其中更包括: 、 輸入一溫度控制變量Τ3至該溫度控制器; 引導包括一 ΧΑ的第三含矽化學前驅物的一第3氣體 到該化學氣相沉積腔體中;以及 51 丨 pifl.doc/012 於該第二含矽膜上沉積一第三含矽膜。 60. 如申請專利範圍第57項所述之於表面沉積含矽材料 的製程,其中該第一含矽化學前驅物與該第二含矽化學前 驅物其中之一至少是選自於包括矽烷、二矽烷與三矽烷的 群組。 61. 如申請專利範圍第57項所述之於表面沉積含矽材料 的製程,其中該第一氣體與該第二氣體其中之一至少是包 括由一或多個選自於包括鍺烷、二鍺烷、三鍺烷、三氟化 氮、一矽甲烷、二矽甲烷、三矽甲烷、四矽甲烷與一摻質 前驅物的群組之化合物。 62. 如申請專利範圍第57項所述之於表面沉積含矽材料 、 的製程,其中該基底具有約350°C或更高的溫度。 63. 如申請專利範圍第57項所述之於表面沉積含矽材料 的製程,其中該基底具有約475°C到7⑻t的溫度範圍。 64. 如申請專利範圍第57項所述之於表面沉積含矽材料 的製程,其中該化學氣相沉積腔體是一單一晶圓水平氣流 反應器。 · 65. 如申請專利範圍第57項所述之於表面沉積含矽材料 的製程,其中該含多層矽膜是選自於包括一微小縮點、一 矽鍺膜、一矽鍺碳膜、一氮化矽膜、一矽-氧膜、一矽-氧-氮膜、一摻雜硼膜、一摻雜砷膜、一摻雜銦膜、一摻雜銻 膜、一摻雜磷膜、一非晶膜、一多晶膜、一磊晶膜以及具 有介電常數約2.2或更低之膜的群組。 66. —種於表面上沉積一含矽材料的儀器,包括: 52 I2771^896pifid〇c/0i2 一化學氣相沉積腔體; 一導管,含有三矽烷; 一進料管線,用以操作連接該導管到該化學氣相沉積 腔體,使從該導管到該化學氣相沉積腔體之三矽烷通過; 以及 一溫度控制器,配置圍繞著該導管並維持在約10°C到 70°C的一溫度下,藉以控制三矽烷的蒸發率。 67. 如申請專利範圍第66項所述之於表面上沉積一含矽 材料的儀器,更包括一歧管,用以操作連接該進料管線, 以控制該導管到該化學氣相沉積腔體之三矽烷的通過。 68. 如申請專利範圍第66項所述之於表面上沉積一含矽 材料的儀器,其中該溫度控制器包括加熱毯、加熱液與加 熱燈其中之一。 69. 如申請專利範圍第66項所述之於表面上沉積一含矽 材料的儀器,其中該化學氣相沉積腔體是一單一晶圓水平 氣流反應器。 70. 如申請專利範圍第66項所述之於表面上沉積一含矽 材料的儀器,其中該導管是裝配有一載體氣體源的發泡 器,其中該載體氣體是選自於包括氫氣、氦氣、氖氣、氬 氣、氪氣與氮氣的群組。 71. 如申請專利範圍第70項所述之於表面上沉積一含矽 材料的儀器,其中該載體氣體包括氫氣。 72. 如申請專利範圍第70項所述之於表面上沉積一含矽 材料的儀器,其中更包括一熱源圍繞該進料管線,並維持 53 ►pifl.doc/012 35°C到70°C之間的溫度,以降低該進料管線中三矽烷的凝 結。 73. 如申請專利範圍第66項所述之於表面上沉積一含矽 材料的儀器,其中該溫度控制器係維持在15°C到52°C之 間的溫度。 74. 如申請專利範圍第15項所述之半導體薄膜之改良式 沉積製程,其中該含矽膜係沉積於介電常數大於5的一閘 介電材料上。 75. 如申請專利範圍第74項所述之半導體薄膜之改良式 沉積製程,其中該閘介電材料包括氧化鋁、氧化給與氧化 锆其中之一。 ^ 76. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜包括氧化矽。 β 77. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜包括氮氧化矽。 78. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該含矽膜包括氮化矽。 _ 79. 如申請專利範圍第1項所述之半導體薄膜之改良式沉 積製程,其中該氣體更包括一氮源。 80. 如申請專利範圍第79項所述之半導體薄膜之改良式 沉積製程,其中該氮源是選自於包括三氟化氮、三矽胺、 原子氮與氨的群組。 81. 如申請專利範圍第80項所述之半導體薄膜之改良式 沉積製程,其中該氮源包括原子氮。 54 1277 1 ^46pifl.doc/012 82. 如申請專利範圍第80項所述之半導體薄膜之改良式 沉積製程,其中三矽烷是被脈衝式導入的。 83. 如申請專利範圍第80項所述之半導體薄膜之改良式 沉積製程,其中該含矽膜是厚度在10埃到300埃範圍內 的一氮化砂膜。 55 ic/012 柒、指定代表圖: (一) 本案指定代表圖為:第( )圖。 (二) 本代表圖之元件代表符號簡單說明: 捌、本案若有化學式時,請揭示最能顯示發明特徵的化 學式=37. The ruthenium-containing film according to claim 35, which has a thickness unevenness of about 1% or less. 38. A ruthenium-containing film as described in claim 35, including a ruthenium film. 39. A ruthenium-containing film as described in claim 35, comprising a polycrystalline material. 40. A ruthenium-containing film as described in claim 35, comprising an amorphous material. 41. A process for depositing a layer of material on a surface, comprising: 48 12771 涊 6 pifl.doc/012 providing a chemical vapor deposition chamber configured with a substrate; directing a higher decane and a higher decane a gas into the chemical vapor deposition chamber; and depositing a tantalum film on the substrate. 42. The process of depositing a bismuth material as described in claim 41, wherein the higher decane is selected from the group consisting of dioxane, trioxane and tetraoxane. 43. The process of depositing a bismuth material as described in claim 41, wherein the higher decane is selected from the group consisting of dioxane, trioxane and tetraoxane. 44. The process of depositing a bismuth material as described in claim 41, wherein the higher decane is trioxane and the higher decane is a wrong yard. 45. The process of depositing a tantalum material as described in claim 41, wherein depositing the tantalum film is in a temperature range from 475 ° C to 700 ° C. 46. The process of depositing a tantalum material as described in claim 41, wherein the rate of depositing the tantalum film is about 50 angstroms or more per minute. 47. The process of depositing a tantalum material as described in claim 41, wherein the rate of depositing the tantalum film is about 100 angstroms or more per minute. 48. The process of depositing a tantalum material on a surface as described in claim 41, wherein the gas further comprises one or more selected from the group consisting of: 矽49 &gt;Pifl.doc/012 methane, diterpenes a compound of the group of methane, triterpene methane, tetramethane, and a dopant precursor. 49. The process of depositing a tantalum material on a surface as described in claim 41, wherein the chemical vapor deposition chamber is a single wafer horizontal gas flow reactor. 50. A process for depositing a tantalum material as described in claim 41, wherein the tantalum film has a thickness non-uniformity of 5% or less. 51. The process of depositing a bismuth material as described in claim 41, wherein the ruthenium film has superior uniformity to a control film made by substituting the higher decane with decane. 52. The process of depositing a bismuth material as described in claim 41, wherein the ruthenium film has superior uniformity to a control film made by substituting the higher decane with decane. 53. The process of depositing a tantalum material as described in claim 41, further comprising patterning to form a transistor gate. 54. A ruthenium film having a thickness unevenness of about 5% or less and a compositional unevenness of about 2% or less on an integrated circuit. 55. The ruthenium film as described in claim 54 of the patent application includes a transistor gate. 56. The ruthenium film of claim 54, wherein the ruthenium film has a thickness unevenness of about 1% or less, and a composition unevenness of about 2% or less. 50 I2771^46pifld〇c/〇i2 57. A process for depositing a ruthenium-containing material on a surface, comprising: providing a chemical vapor deposition chamber configured with a substrate, the chemical vapor deposition chamber being assembled with a temperature controller In order to program a plurality of temperature control variables in a single method; input a temperature control variable or even the temperature controller; and direct a first gas comprising a first cerium-containing chemical precursor of XA to the chemical vapor deposition chamber In the body, wherein Χι is in the range of 1χ1〇·4 to about 1〇〇; depositing a first ruthenium-containing film on the substrate; inputting a temperature control variable τ2 to the temperature controller; guiding includes an X2% a second gas containing a ruthenium chemical precursor into the chemical vapor deposition chamber, wherein Χ2 is in the range of lx 10.4 to about 100' and the second ruthenium containing chemical precursor and the first a ruthenium containing chemical precursor is one of the same and different; and depositing a second ruthenium containing film on the first ruthenium containing film to form a thickness unevenness of about 5% or less and about 2% Or a lower compositional unevenness A multilayer containing ruthenium film. 58. The process for depositing a ruthenium-containing material on the surface as described in claim 57, wherein the temperature control variables 1 and T2 are temperature control settings. 59. The process for depositing a eucalyptus-containing material on a surface as described in claim 57, further comprising:: inputting a temperature control variable Τ3 to the temperature controller; guiding a third bismuth-containing chemical precursor including a bismuth a third gas of the substance is introduced into the chemical vapor deposition chamber; and 51 丨pifl.doc/012 deposits a third ruthenium-containing film on the second ruthenium-containing film. 60. The process of depositing a ruthenium-containing material as described in claim 57, wherein one of the first ruthenium-containing chemical precursor and the second ruthenium-containing chemical precursor is at least selected from the group consisting of decane, A group of dioxane and trioxane. 61. The process of depositing a ruthenium-containing material as described in claim 57, wherein one of the first gas and the second gas comprises at least one or more selected from the group consisting of decane, A compound of the group of decane, trioxane, nitrogen trifluoride, monomethane, methane, trimethylmethane, tetramethane, and a dopant precursor. 62. A process for depositing a ruthenium-containing material on a surface as described in claim 57, wherein the substrate has a temperature of about 350 ° C or higher. 63. A process for depositing a ruthenium-containing material as described in claim 57, wherein the substrate has a temperature range of from about 475 ° C to 7 (8) t. 64. The process of depositing a germanium-containing material on a surface as described in claim 57, wherein the chemical vapor deposition chamber is a single wafer horizontal gas flow reactor. · 65. The process for depositing a ruthenium-containing material on a surface as described in claim 57, wherein the multi-layer ruthenium film is selected from the group consisting of a minute crater, a ruthenium film, a ruthenium carbon film, and a a tantalum nitride film, a germanium-oxygen film, a germanium-oxygen-nitrogen film, a doped boron film, a doped arsenic film, an doped indium film, a doped germanium film, a doped phosphor film, and a An amorphous film, a polycrystalline film, an epitaxial film, and a group of films having a dielectric constant of about 2.2 or less. 66. An apparatus for depositing a cerium-containing material on a surface, comprising: 52 I2771^896pifid〇c/0i2 a chemical vapor deposition chamber; a conduit containing trioxane; a feed line for operatively connecting the a conduit to the chemical vapor deposition chamber to pass trioxane from the conduit to the chemical vapor deposition chamber; and a temperature controller disposed around the conduit and maintained at about 10 ° C to 70 ° C At a temperature, the evaporation rate of trioxane is controlled. 67. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 66, further comprising a manifold for operatively connecting the feed line to control the conduit to the chemical vapor deposition chamber The passage of trioxane. 68. An apparatus for depositing a cerium-containing material on a surface as described in claim 66, wherein the temperature controller comprises one of a heating blanket, a heating liquid, and a heating lamp. 69. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 66, wherein the chemical vapor deposition chamber is a single wafer horizontal gas flow reactor. 70. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 66, wherein the conduit is a foamer equipped with a carrier gas source, wherein the carrier gas is selected from the group consisting of hydrogen and helium. Groups of helium, argon, helium and nitrogen. 71. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 70, wherein the carrier gas comprises hydrogen. 72. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 70, further comprising a heat source surrounding the feed line and maintaining 53 ►pifl.doc/012 35 ° C to 70 ° C The temperature between them to reduce the condensation of trioxane in the feed line. 73. An apparatus for depositing a ruthenium-containing material on a surface as described in claim 66, wherein the temperature controller is maintained at a temperature between 15 ° C and 52 ° C. 74. The improved deposition process for a semiconductor film according to claim 15, wherein the ruthenium-containing film is deposited on a gate dielectric material having a dielectric constant greater than 5. 75. The improved deposition process for a semiconductor film according to claim 74, wherein the gate dielectric material comprises one of alumina, oxidized, and zirconia. The improved deposition process of the semiconductor film of claim 1, wherein the ruthenium-containing film comprises ruthenium oxide. The modified deposition process of the semiconductor film of claim 1, wherein the ruthenium-containing film comprises ruthenium oxynitride. 78. The improved deposition process of a semiconductor film according to claim 1, wherein the germanium-containing film comprises tantalum nitride. _ 79. The improved deposition process of the semiconductor film of claim 1, wherein the gas further comprises a nitrogen source. 80. The improved deposition process for a semiconductor thin film according to claim 79, wherein the nitrogen source is selected from the group consisting of nitrogen trifluoride, tridecylamine, atomic nitrogen and ammonia. 81. The improved deposition process for a semiconductor thin film according to claim 80, wherein the nitrogen source comprises atomic nitrogen. 54 1277 1 ^46 pifl.doc/012 82. An improved deposition process for a semiconductor film as described in claim 80, wherein the trioxane is pulsed. 83. The improved deposition process for a semiconductor film according to claim 80, wherein the ruthenium-containing film is a nitrid sand film having a thickness in the range of 10 angstroms to 300 angstroms. 55 ic/012 柒, designated representative map: (1) The representative representative of the case is: ( ). (2) The symbolic representation of the symbol of the representative figure is as follows: 捌 If there is a chemical formula in this case, please reveal the chemical formula that best shows the characteristics of the invention.
TW91101956A 2001-02-12 2002-02-05 Improved process for deposition of semiconductor filme TWI277139B (en)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US26833701P 2001-02-12 2001-02-12
US27925601P 2001-03-27 2001-03-27
US31160901P 2001-08-09 2001-08-09
US32364901P 2001-09-19 2001-09-19
US33269601P 2001-11-13 2001-11-13

Publications (1)

Publication Number Publication Date
TWI277139B true TWI277139B (en) 2007-03-21

Family

ID=38646433

Family Applications (1)

Application Number Title Priority Date Filing Date
TW91101956A TWI277139B (en) 2001-02-12 2002-02-05 Improved process for deposition of semiconductor filme

Country Status (1)

Country Link
TW (1) TWI277139B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI775749B (en) * 2016-04-29 2022-09-01 美商蘭姆研究公司 Variable cycle and time rf activation method for film thickness matching in a multi-station deposition system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI775749B (en) * 2016-04-29 2022-09-01 美商蘭姆研究公司 Variable cycle and time rf activation method for film thickness matching in a multi-station deposition system

Similar Documents

Publication Publication Date Title
JP4866534B2 (en) Improved deposition method for semiconductor films.
JP3914647B2 (en) Metal layer forming method using atomic layer deposition process
JP4728953B2 (en) Method for depositing polycrystalline Si-containing film
US7370848B2 (en) Bubbler for substrate processing
US7341907B2 (en) Single wafer thermal CVD processes for hemispherical grained silicon and nano-crystalline grain-sized polysilicon
TWI277139B (en) Improved process for deposition of semiconductor filme

Legal Events

Date Code Title Description
MK4A Expiration of patent term of an invention patent