1317348 . 九、發明說明： 【發明所屬之技術領域】 本發明涉及一種奈米碳管陣列製作方法。 【先前技術】 由於奈米碳管獨特之電學性質，其於奈米積體電路、單分子器件等領 域之應用有著不可估量之前景。目前人們已經能夠於實驗室里少量製造基 於奈米碳管之場效應管、或非門等器件，並研究其性質。但大規模之製備 及具有實際思義之應用則必須求助於由下而上（Bottom Up)之製備工藝。 由下而上之製備工藝要求能夠對奈米碳管之生長位置、方向、尺度、 | 甚至奈米碳管之螺旋度進行必要之控制，通過少量而經濟之步驟直接生長 出所需要之器件結構。控制奈米碳管陣裂之生長方向係奈米應用研究中之 一個重要課題。范守善等人於Science 283，512-514(22 Jan 1999)， Self Oriented Regular Arrays of Carbon Nanotubes and Their Field Emission Propertied文中揭露了藉由催化劑圖形(patterne(i Catalyst) 來很好地控制奈米碳管之生長位置，但不能控制其生長方向。 Z_ F. Ren等人於Science 282，1105-1107(6 Nov，1998)’Synthesis of Large Arrays of Well-Aligned Carbon Nanotubes on Glass—文中揭 露了一種藉由奈米碳管陣列之形式使得奈米碳管垂直於基底生長之方法。 B. Q. Wei等人於Nature 416, 495-496(4 Apr，2002)，Organized Assembly > of Carbon Nanotubes—文中揭露了一種藉由設計基底之形狀來實現奈米碳 管垂直於三維基底各表面之方向生長。 惟’上述方法中所獲得之奈米碳管陣列均垂直於生長基底，難以實現 控制奈米碳管陣列向多個方向生長。1317348. EMBODIMENT DESCRIPTION: TECHNICAL FIELD The present invention relates to a method for fabricating a carbon nanotube array. [Prior Art] Due to the unique electrical properties of carbon nanotubes, its application in the field of nano-integrated circuits, single-molecule devices, etc. has an inestimable prospect. At present, it has been possible to manufacture a small number of devices based on carbon nanotubes, NAND gates, etc. in the laboratory, and to study their properties. However, large-scale preparation and practical applications must resort to the Bottom Up preparation process. The bottom-up preparation process requires the necessary control of the growth position, orientation, dimensions, and even the helicity of the carbon nanotubes, and the required device structure is directly grown by a small number of economical steps. Controlling the growth direction of carbon nanotube cracking is an important topic in the application of nanotechnology. Fan Shoushan et al., Science 283, 512-514 (22 Jan 1999), Self Oriented Regular Arrays of Carbon Nanotubes and Their Field Emission Propertied, reveals that carbon nanotubes are well controlled by patterne (i Catalyst). The growth position, but can not control its growth direction. Z_F. Ren et al., Science 282, 1105-1107 (6 Nov, 1998) 'Synthesis of Large Arrays of Well-Aligned Carbon Nanotubes on Glass—disclosed by Nana The carbon nanotube array forms a method in which the carbon nanotubes are grown perpendicular to the substrate. BQ Wei et al., Nature 416, 495-496 (4 Apr, 2002), Organized Assembly > of Carbon Nanotubes - one of which is disclosed The shape of the substrate is designed to realize the growth of the carbon nanotubes perpendicular to the surfaces of the three-dimensional substrate. However, the carbon nanotube arrays obtained in the above method are perpendicular to the growth substrate, and it is difficult to control the carbon nanotube arrays. Growth in one direction.
Yoon-Taek Jang等人於Solid State Communications 2003，126(6): 305-308, Lateral Growth of Aligned Mutilwalled Carbon Nanotubes under· Electric Field —文中揭露了一種藉由電場控制奈米碳管生長方向 之方法，Ki-Hong Lee等人於Applied Physics Letters 2003，82(3): 448-450，Control of Growth Orientation for Carbon Nanotubes—文中 揭露了一種藉由磁場控制奈米碳管生長方向之方法。 1317348 »β. 12.29 惟’上述方法中’由於電場與磁場本身之廣域性， 種生長方向之控制’也加重了奈絲件設計之複雜程度。現對局衫 【發明内容】 下面將以若干實施例說明一種奈米碳管陣列製作方 米碳管陣列至少向-個方向彎曲生長之控制。 -了實現對奈 内容^供—種奈米碳管陣舰作方法，其步驟包括： 如供一基底，於基底上形成一遮擋層； 提供-催侧濺射源，配合遮觀於基底上濺軸— =劑層，所述之催化劑層較薄處之厚度爲Η。奈米，較厚處之厚^ 端 去除上述遮擔層，退火處理所述催化劑層； i Α碳源處理後之催化綱上f曲生長絲碳管陣 ，選的’退火處理催化劑層之前先將所述催化劑較義出至少一催化 劑區塊，該催化劑區塊具有一第一端部和一與該第一端部相對之第 部’確保該催化_塊由該第—端部向第二端部逐漸增厚。 化物==厚膜光娜，崎於—犧牲層上之恤其氧化物或氣 ㈣^催化纖射源包括點催化劑濺射源、線型催化綱射源及面型催 化劑濺射源。 '咏 所述基底材質包括多孔矽、拋光矽片、玻璃及金屬。 所述催化劑層材質爲鐵、鈷、鎳或其合金之一。 較於先前技術，本技術方案所提供之奈米碳管陣列製作方法，不需要 設計基底雜’也不需要外加電場或磁場，藉由普通之麵設備配合一定 厚度之遮擋層，就可以於基底上形成具有梯度之厚度之催化劑層。由於奈 米碳管之生長速度與雜綱厚度侧：催化_上最佳厚度處奈米碳管 生長速度最快’於此厚度基礎上逐漸減薄或增厚，奈米碳管之生長速度都 會逐漸降低。故，可實現㈣奈米碳管_至少向—個方向_生長，所 需設備簡便且工藝簡單易行。所獲得之奈米碳辨列結構可以簡化奈米器 件之設計，且豐富其多樣性。 !317348 θβ. 12.2 9 【實施方式】 下面結合附圖將對本發明實施例作進一步之詳細說明。 , 第一實施例： 本實施例所提供之製作奈米碳管陣列之基本步驟包括： 提供一基底，於基底上形成一遮擋層； 提供一催化劑濺射源，配合遮檔層於基底上濺射出一具有厚度梯度之 催化劑層，並且於所述厚度梯度範圍内，有一處之厚度最接近一最佳厚度； 去除遮擋層’退火處理所述催化劑層； 通入碳源氣，於上述處理後之催化劑層上生長奈米碳管陣列，該奈米 鲁 碳管陣列向背離所述最佳厚度之方向彎曲生長。 針對上述各步驟具體說明如下： 請參閱第一圖，提供一基底10，於該基底10一側（圖中所示爲右侧）上 形成具有一定厚度之遮擋層4〇 ;於遮擋層4〇上方設置一直線型催化劑濺射 源20，以對基底濺射催化劑，催化劑濺射源2〇濺射出之催化劑有一部分被 ' 遮擋層40擋住，從而於基底上形成一具有厚度梯度之催化劑層30。 基底10之材質可以為矽片、玻璃片、金屬片等，本實施例中採用矽片。 催化劑之材質可以為鐵、鈷、鎳或其合金，本實施例中採用鐵催化劑。 遮擋層40可選用厚膜光刻膠、附著於一犧牲層上之金屬或金屬氧化 物、金屬氮化物等，本實施例中採用厚膜光刻膠；遮擋層40之形狀可以係 正方體、長方體或圓柱體等多種形狀，本實施例中採用長方體形；該遮擋 層40具有一定高度之垂直遮擋面42，該高度即爲遮擋層4〇之厚度，該厚度 最好小於預定濺射條件下催化劑之原子平均自由程。催化劑原子平均自由 程*5取決於公式： ⑴Yoon-Taek Jang et al., Solid State Communications 2003, 126(6): 305-308, Lateral Growth of Aligned Mutilwalled Carbon Nanotubes under · Electric Field - a method for controlling the growth direction of carbon nanotubes by electric field is disclosed. Ki-Hong Lee et al., Applied Physics Letters 2003, 82(3): 448-450, Control of Growth Orientation for Carbon Nanotubes, discloses a method for controlling the growth direction of carbon nanotubes by a magnetic field. 1317348 »β. 12.29 However, in the above method, the control of the growth direction of the species due to the wide area of the electric field and the magnetic field itself also increases the complexity of the design of the nanowire. The present invention will be described below with reference to a number of embodiments in which a carbon nanotube array is fabricated to control the growth of the carbon nanotube array in at least one direction. - A method for realizing the nano-carbon nanotubes of the naphthalene content, the steps comprising: forming a shielding layer on the substrate, such as a substrate; providing a side-sputtering source for observing the substrate Splash axis - = agent layer, the thickness of the catalyst layer is Η. Nano, thicker than the thicker ^ end to remove the above-mentioned shielding layer, annealing the catalyst layer; i Α carbon source treatment of the catalyst after the f-growth growth of the carbon tube array, selected 'annealing catalyst layer before Comparing the catalyst to at least one catalyst block, the catalyst block having a first end and a first portion opposite the first end 'ensure that the catalytic block is from the first end to the second The ends are gradually thickened. The compound == thick film light, the sacrificial - sacrificial layer of the oxide or gas (4) ^ catalytic fiber source includes a point catalyst sputtering source, a linear catalytic source and a surface catalyst sputtering source. '咏 The substrate material consists of porous tantalum, polished bakelite, glass and metal. The catalyst layer is made of one of iron, cobalt, nickel or an alloy thereof. Compared with the prior art, the carbon nanotube array manufacturing method provided by the technical solution does not need to design a substrate impurity and does not require an external electric field or a magnetic field, and can be applied to the substrate by using a common surface device with a certain thickness of the shielding layer. A catalyst layer having a gradient thickness is formed thereon. Due to the growth rate of the carbon nanotubes and the thickness side of the catalyst: the fastest growth rate of the carbon nanotubes at the optimum thickness of the catalyst is gradually reduced or thickened on the basis of the thickness, and the growth rate of the carbon nanotubes will be Gradually decreases. Therefore, it is possible to realize (four) carbon nanotubes _ growth in at least one direction, and the required equipment is simple and the process is simple and easy. The obtained nanocarbon discriminating structure can simplify the design of the nano device and enrich its diversity. !317348 θβ. 12.2 9 [Embodiment] Hereinafter, embodiments of the present invention will be further described in detail with reference to the accompanying drawings. The first embodiment: The basic steps of fabricating the carbon nanotube array provided by the embodiment include: providing a substrate to form a shielding layer on the substrate; providing a catalyst sputtering source, and shielding the shielding layer on the substrate Emulating a catalyst layer having a thickness gradient, and having a thickness closest to an optimum thickness within the thickness gradient; removing the shielding layer to anneal the catalyst layer; introducing carbon source gas after the above treatment A carbon nanotube array is grown on the catalyst layer, and the nanolu tube array is bent to grow away from the optimum thickness. The above steps are specifically described as follows: Referring to the first figure, a substrate 10 is provided on the side of the substrate 10 (the right side shown in the figure) to form a shielding layer 4 一定 having a certain thickness; and the shielding layer 4 〇 A linear catalyst sputtering source 20 is disposed above the substrate sputtering catalyst, and a portion of the catalyst sputtered by the catalyst sputtering source is blocked by the 'shielding layer 40, thereby forming a catalyst layer 30 having a thickness gradient on the substrate. The material of the substrate 10 may be a ruthenium sheet, a glass sheet, a metal sheet or the like. In this embodiment, a cymbal sheet is used. The material of the catalyst may be iron, cobalt, nickel or an alloy thereof, and an iron catalyst is used in this embodiment. The mask layer 40 may be a thick film photoresist, a metal or metal oxide adhered to a sacrificial layer, a metal nitride or the like. In this embodiment, a thick film photoresist is used; the shape of the shielding layer 40 may be a square or a rectangular parallelepiped. Or a plurality of shapes, such as a cylinder, in the embodiment, a rectangular parallelepiped shape; the shielding layer 40 has a vertical shielding surface 42 of a certain height, which is the thickness of the shielding layer 4, which is preferably smaller than the catalyst under predetermined sputtering conditions. The mean free path of atoms. The mean free path of the catalyst atom *5 depends on the formula: (1)
kT 4lnd2p 其中，爲催化劑原子直徑；P爲氣體壓強；々爲波爾茨曼常數；^爲 氣體溫度。 本實施例中’於(Tc、IPa氣體壓強之磁控濺射條件下，催化劑鐵之原 子平均自由程約爲5_ 5xl(T3m。該遮擋層4〇也可採用適當厚度之鏤空模板， 1317348 9β. 13.29 但於沈積催化劑時該鏤空模板最好緊密貼合於基底1〇上。kT 4lnd2p where is the atomic diameter of the catalyst; P is the gas pressure; 々 is the Boltzmann constant; ^ is the gas temperature. In the present embodiment, under the condition of magnetron sputtering of Tc and IPa gas pressure, the atomic mean free path of the catalyst iron is about 5 _ 5xl (T3m. The occlusion layer 4 〇 can also adopt a hollow template of appropriate thickness, 1317348 9β 13.29 However, the hollow template is preferably adhered to the substrate 1 when the catalyst is deposited.
直線型催化劑濺射源20到基底10之距離最好大於催化劑之原子平均自 由程，以使濺射出之鐵催化劑能均勻沈積於基底上；催化劑濺射源2〇之尺 寸大小無特別限制，但至少有一部分位於垂直遮擋面42之一侧(本實施例爲 右侧）’以確保催化劑賤射源2〇減射出來催化劑能被遮擋層4〇擋住一部分’， 從而使於基底上沈積之催化綱3〇於—定厚度範圍内具有漸變之梯度。催 化劑層30靠近遮擋面42之一端厚度較薄，優選的，其厚度範圍爲卜仞夺米； 遠離遮擋面42之-稱度較厚，魏的，其厚度範圍爲㈠修米。:射催 化劑時，直線型催化劑濺射源20與基底1〇可以相對靜止，也可以沿平行於 基底10之方向（即垂直於紙面方向）於小範圍内相對移動（請參見第一圖）， 如催化劑職垂直於紙面方向，聽化麵射·與基細可沿平行 於紙面方向於小細_對赫，鱗餅賴硫左纖底上沿一個方 向沈積出較大面積具有厚度梯度之催化劑層3〇(請參見第一圖及第二圖）， 本實施例中，直線型催化劑濺射源20和基底1〇沿垂直於紙面方向於小範圍 内相對移動。 另外，若採用點催化劑濺射源，藉由控制該點催化劑濺射源與遮擋層 之相對位置’也可以於基底上沿一個方向沈積出較大面積具有厚度梯度之 催化劑層30。The distance between the linear catalyst sputtering source 20 and the substrate 10 is preferably greater than the atomic mean free path of the catalyst, so that the sputtered iron catalyst can be uniformly deposited on the substrate; the size of the catalyst sputtering source is not particularly limited, but At least a portion is located on one side of the vertical shielding surface 42 (the right side in this embodiment) to ensure that the catalyst source 2 is reduced and the catalyst can be blocked by the shielding layer 4, thereby catalyzing deposition on the substrate. The outline has a gradient of gradient in the range of thickness. The catalyst layer 30 is thinner near one end of the shielding surface 42. Preferably, the thickness thereof is in the range of dime; the thickness away from the obscuring surface 42 is thicker, and the thickness of the layer 30 is (1) repairing rice. When the catalyst is shot, the linear catalyst sputtering source 20 and the substrate 1〇 may be relatively stationary, or may move relatively in a small range in a direction parallel to the substrate 10 (ie, perpendicular to the paper surface) (see the first figure). If the catalyst is perpendicular to the direction of the paper, the hearing surface and the base can be deposited in a direction parallel to the direction of the paper, and a large area of the catalyst having a thickness gradient is deposited in one direction along the left side of the scale. Layer 3 (see first and second figures), in this embodiment, the linear catalyst sputtering source 20 and the substrate 1 are relatively moved in a small range perpendicular to the plane of the paper. Alternatively, if a spot catalyst sputtering source is used, a catalyst layer 30 having a large thickness gradient may be deposited in one direction on the substrate by controlling the relative position of the spot catalyst sputtering source to the shielding layer.
本實施例中濺射得到之催化劑層3〇之厚度近似滿足公式： Τ{λ) = ⑵ 其中，A爲催化劑層30上某一位置到垂直遮擋面42之距離；y(;l)爲該位 置催化劑層之厚度；I；爲無遮擋層時催化劑層之厚度；λ爲遮擋層4〇之厚度。 根據公式(2)可知，離遮擋面42最近處催化劑層厚度爲7。/2 ;遠離遮擋 面42催化劑層之厚度逐漸增大；若催化劑濺射源2〇位於遮擋面右侧之尺寸 無限長，則當催化劑層30上某一位置離遮擋面42之距離遠大於遮擋層厚度 時，催化劑層之厚度達到7；;其中，催化劑層3〇具有明顯厚度梯度之範圍 近似等於遮擋層厚度λ之2倍。 本實施例中，爲得到具有明顯厚度梯度之催化劑層，應當使遮擋層4〇 1317348 96. 12.29 齐化，由薄到厚之梯*方向上形成之陰影範圍覆蓋需要生長 9位之，域’該陰影範圍於所述方向上之長度最好大致等於遮擋 ?曰η *取遮擋層厚度爲0· 1μιη〜1()刪，則該長度相應爲〇.2叩〜 Ε-私=4〇形成之陰影範圍可這樣確定：假定直_催化劑賤射源20 你一二1;'職位於垂直遮擋面42右侧光騎發出之光會被遮擋層40擋 s從而於垂直遮擋面42左側之基底上形成—定之陰影區域，該陰 衫即爲遮擋層40形成之陰影範圍。 、，二ί㈣—圖’錯遮擋層4°，較―催化劑層3。之最佳厚度線32， ^:二广終奈米石反管生長方向之控制。催化劑層30之厚度由-端向另-端 ®太本’其中兩端之間某一位置之厚度最適合一定化學氣相沈積條件下 之生長’此位置之厚度即爲最佳厚度。該最佳厚度可由這樣之實 疋.於一石夕基底上、配合一遮撐層賤射催化劑，形成一厚度從薄到厚 之催化劑層’要求鱗度變化之範圍足夠大，確雜化劑層上能 ^到-最佳厚度：於預定條件下用化學氣相沈積法生長奈純管陣列；用 ^鏡觀爾奈米碳管生長料德置；触置之雜獅厚度即爲該預 件下之最佳厚度。催化上具有最轉度之位置賴—條線，即最 &度線。例如本實施例中以石夕基底上之鐵作催化劑，於7〇〇攝氏度下用乙 生長，丁、米碳管’催化劑層3〇之最佳厚度約爲5奈米，將厚度約爲5奈米之 位置連成-條線即爲最佳厚度線32，可參閱第二圖與第三圖。催化劑層上 最佳厚度？1置處奈米碳管生長速度最快，於此厚度基礎上逐漸減薄或增 厚’奈米碳管生長速度都會逐漸降低。於實際操作過程中，考慮到最佳厚 度有-定範圍，因此本實施例中之催化劍層3〇厚度分佈可進一步優選爲 3nm~15nm。 請參閱第二圖’於催化劑層30上形成一定圖形。本實施例採用光刻法 形成催化細形：於催化_3G上形成—層光娜，織採用具有預定圖 案之掩模進行曝光，顯影後用酸洗掉未被保護之催化劑膜，並去除催化劑 上之光刻膠’形成與掩模圖案相應之、具有厚度梯度之催化劑區塊33、34。 由於催化劑區塊之位置選擇於最佳厚度之哪一邊將決定奈米碳管陣列之彎 曲方向，因此於形成催化劑圖形進行光刻之時候，應根據需要將掩模位置 96. *2.29 1317348 對準。沈積催化劑層30時若採用適當厚度之鏤空模板作爲遮擋層4〇，則可 將此採用光刻法形成催化劑圖形之步驟提前到沈積催化劑層3〇之前完成， 進而於該步驟中可只通過剝離工藝即可獲取厚度符合要求之催化劑圖形。 請參閱第四圖，將帶有催化劑區塊之基底1〇於空氣中，3〇〇工下退火或 通過類似方法，使催化劑區塊氧化、收縮成爲奈米級之催化劑氧化物顆粒 31。局部之催化劑氧化物顆粒大小與該處催化劑層厚度相關，如厚度小於 ，佳厚度一側之催化劑區塊33被氧化收縮成爲催化劑氧化物顆粒犯，，直徑 較小；厚度大於最佳厚度一侧之催化劑區塊34被氧化收縮成爲催化劑氧化 物顆粒34,，直徑較大。 一明參閱第五圖，將表面形成有催化劑氧化物顆粒之基底置於反應爐（圖 未不）中，通入碳源氣乙烯，利用化學氣相沈積法生長奈米碳管陣列。於生 長過程中碳源、氣分解出氯氣將催化劑氧化物顆粒33,、分別還原成催化劑 顆，33、34，然後於催化劑顆粒33”、34”上生長奈米碳管陣列5〇、51。碳 氣也可轉用其他含碳之氣體，如Μ、乙炔等。但由於不同化學氣相 積條件下’催化騎之最佳厚度不同，因此生長奈米碳管卩車列之化學氣 ^積條件最好與上述確定最佳厚度_生長奈米碳管之條件大致相同。 工丨生長時間則可以控制奈米碳管陣列5〇、51之生長長度。 售Ϊ厚度小於最佳厚度之—側，催化継塊33聽近最佳厚度線一側到 度—侧’厚度騎減薄；於厚度A於最佳厚度之-側，催化劑 =34從錢最佳厚度—則遠離最佳厚度—侧，厚度逐漸增厚。由於催 減簿佳厚度位置處奈米碳管之生長速度最快，於此厚度基礎上逐漸 屮氺1 : 不米碳管生長速度都會逐漸降低，故，催化劑顆粒33”上生長 催化割ϊί碳管陣觸從催化継塊較厚之—侧向較薄之-侧彎曲，而從 較厚之上生長出來之奈米碳管陣列51從催化劑區塊較薄之一側向 佳厚产位i 2办即催化劑顆粒33，’、34”上生長之奈米碳管降列均從靠近最 於最ί厚絲°背離最佳厚度之方向彎曲。如果兩個催化継塊分別分佈 碳管ml兩侧’於該兩讎化舰塊形成之催輔齡上生長奈采 往可得到兩個生長方向相反之奈米碳管陣列。 月參閱第，、圖’利用上述方法獲得之奈米碳管陣列結構】包括一基底 11 95. 1317348 i0相之奈米碳料觸及51，奈米破管陣·抓具有兩 一端向逐漸減薄之-端彎曲；如果催化劑層兩 === :===輪輸嫩裂，分胸物，=ΐ 第二實施例： 化劑型催化舰射源，或線型催化劑滅射源，並使線型催 化劑濺射賴基胁較城_姆鶴，確做麟層關區域之 ^均=出具有厚度梯度之催化劑層。其製作奈米碳管陣列之基本步驟與 第一實施例相同，各步驟具體說明如下： 、 請參閱第七圖，提供-基底7G，於該基底7G上中心處形成一具有一定 逾麟8Q，於麟層8()之上枝置-面難化綱騎6G，、對基底 2催化劑時，有-部分催化麵遮擋細擋住，從而於基底上形成具有 厚度梯度之催化劑層9〇。 — 〃 遮擋層具有-定尚度之垂直遮翻，該高度即爲遮騎⑼之厚度。 面型催化劑濺射源60至少有一部分位於遮擋層8〇之正上方使濺射出 催化劑有4刀此被遮擔層8〇擋住，從而於基底上得到之催化劑層9〇位 於遮擋，80周圍區域内有一漸變梯度，且越靠近遮擋層8〇，催化劑層卵之 厚度越薄，越遠離遮擋層8〇，催化劑層90之厚度逐漸增大。 遮擋層80於基底上沿催化劑層由薄到厚之梯度方向上形成之陰影範圍 覆蓋需要生長奈米碳管陣列之區域，該陰影範圍於所述方向上之長度最好 大致等於賴層厚度之2倍，該陰影細之確定方法與第-實施例相^。 另外也可採用直線型催化劑濺射源，濺射催化劑時， 減射源與基底可以相對靜止;如直線型催化劑嶋平行於== 置，則該催化劑濺射源與基底可沿垂直於紙面方向於較大範圍内相對移 12 1317348 ' 95. 12^2 d 動；如直線型催化劑濺射源垂直於紙面方向放置，則該催化劑濺射源與基 底可沿平行於紙面方向於較大範圍内相對移動，以確保於遮擋層8〇周圍區 域内沈積出大面積具有厚度梯度之催化劑層90。 请參閱第八圖及第九圖，移除遮擋層80，峰定一定化學氣相沈積條件 * 下催化劑層90之最佳厚度，並標記最佳厚度線92。確定該最佳厚度線92之 方法與第一實施例中確定最佳厚度線32之方法相同。 請參閱第十圖，形成催化劑圖形。本實施例中，將催化劑層9〇光刻出 多個催化劑區塊’且該催化劑區塊分別位於最佳厚度線92之兩側，如催化 劑區塊93、94分別處於最佳厚度線92之兩側，催化劑區塊95、96分別處於 φ 最佳厚度線92之兩側。也可於沈積催化劑層90時採用一厚一薄之雙層複合 鏤空遮擋模板取代遮擋層8〇，直接沈積出厚度符合要求之催化劑圖形。 請參閱第十一圖，將帶有催化劑區塊之基底7〇於空氣中、3〇〇°c下退火 或類似方法，使催化劑區塊氧化、收縮成爲奈米級之催化劑氧化物顆粒。 局部之催化劑氧化物顆粒大小與該處催化劑區塊之厚度相關，如厚度小於 - 最佳厚度一側之催化劑區塊94、95被氧化收縮成催化劑氧化物顆粒94,、 95’ ’直徑較小；厚度大於最佳厚度一侧之催化劑區塊⑽、％被氧化收縮成 催化劑氧化物顆粒93,、96,，直徑較大。 請參閱第十二圖’將表面形成有催化劑氧化物顆粒之基底7〇置於反應 爐（圖未示）中，通入碳源氣乙烯，利用化學氣相沈積法生長奈米碳管陣列。 ® 於生長過程中碳源氣分解出氫氣將催化劑氧化物顆粒93,、94,、95,、96,分 別還原成催化劑顆粒93”、94”、95”、96”，然後於催化劑顆粒93”、94”、95”、 96”上生長奈米碳管陣列100、1〇卜1〇2及1〇3。於厚度小於最佳厚度之一側， 催化劑顆粒94”、95”上生長之奈米碳管陣列1〇1、1〇2向催化劑區塊厚度減小 ^方向彎曲，於厚度大於最佳厚度之一侧，催化劑顆粒邪”、96„上生長之奈 米後管陣_G、1G3向催化舰塊厚度增大之方向㈣。相同之處係奈米 - 碳管陣列均向背離最佳厚度線92之一側彎曲。 • 請參閱第十三圖，利用上述方法獲得之奈米碳管陣列結構2包括-基底 7〇、形成於基底70上之多個奈米碳管陣列1〇〇、1〇卜1〇2及1〇3等，多個奈 米碳管陣列具有多個不同之彎曲方向。 ” 13 «1 1317348 器件、電真空器件等微電子學器件、場發射 器等微型功能結構。 ㈣“ *斜爲基之橋梁、電子槍或感測 催化劑區塊之本伽精_職傾化，如適當變更 本發明之技術峨侧縣之形成_，只要其不偏離 惟， 於援依 綜上所述，本發明確已符合發明專利 以上所述者僅為本發明之較佳實 【圖式簡單說ϊγ 應包含於以下之申請專利範圍内 匕=:月第一實施例中催化劑層沈積過程之侧面示意圖。 意圖 第一圖係本發卿—實施财赚賴層麟定最轉麟之立體示 ΐ三發明第—實施例中形成之催化劑圖形立體示意圖。 f四圖係本發明第-實施例中催化劑圖形退火後之示意圖。 第—實施例中生長出奈米碳管陣列之侧面示意圖。 第—實施例製備之奈米碳管陣列結構立體示意圖。 弟圖係本發明第二實施例中催化劑層沈積過程之侧面示意圖。 意圖 第八圖係本㈣第二實施财移除遮擋層並確定最佳厚度^之側面示 第九圖係本發明第二實施例中形成之催化劑層之立體示意圖。 第十圖係轉二實劇t形成之催化麵形之立體示意圖。 ，十一圖係本發明第二實施例中催化劑退火後之側面示意圖。 第十亡圖係本發明第二實施例中生長出奈米碟管陣列之侧面示意圖。 第十二圖係本發明第二實施例製備之奈米碳管陣列結構立體示音 【主要元件符號說明】 〜 10、70 100 、 101 奈米碳管陣列結構1、2 基底 102、103、50、51 奈米碳管陣列 I317348 W· 12.29 % c s > 直線型濺射源 20 催化劑層 30、90 最佳厚度線 32、92 催化劑區塊 33、34、93、94、95、96 催化劑氧化物顆粒 33’ 、 34, 、 93’ 、 94, 、 95, 、 96’ 催化劑顆粒 33”、34”、93”、94”、95”、96 遮擋層 40、80 垂直遮擋面 42 面型濺射源 60 15The thickness of the catalyst layer 3〇 obtained by sputtering in this embodiment approximately satisfies the formula: Τ{λ) = (2) where A is the distance from a position on the catalyst layer 30 to the vertical shielding surface 42; y(;l) is the The thickness of the position catalyst layer; I; the thickness of the catalyst layer when there is no shielding layer; λ is the thickness of the shielding layer 4〇. According to the formula (2), the thickness of the catalyst layer closest to the shielding surface 42 is 7. /2; the thickness of the catalyst layer is gradually increased away from the shielding surface 42; if the size of the catalyst sputtering source 2 〇 on the right side of the shielding surface is infinitely long, when the position on the catalyst layer 30 is farther from the shielding surface 42 than the shielding At the thickness of the layer, the thickness of the catalyst layer reaches 7; wherein the catalyst layer 3 has a significant thickness gradient which is approximately equal to twice the thickness λ of the occlusion layer. In this embodiment, in order to obtain a catalyst layer having a significant thickness gradient, the shielding layer 4〇1317348 96. 12.29 should be homogenized, and the shadow range formed by the thin to thick ladder* direction needs to grow 9 bits, the domain ' The length of the shadow range in the direction is preferably substantially equal to the occlusion? 曰η * The thickness of the occlusion layer is 0·1μιη~1(), and the length corresponds to 〇.2叩~ Ε-私=4〇 formation The shaded range can be determined by assuming that the direct _ catalyst source 20 is one or two; 'the position of the light on the right side of the vertical occlusion surface 42 is blocked by the occlusion layer 40 so as to be on the left side of the vertical occlusion surface 42. A shadow area is formed on the upper surface, which is the shaded range formed by the occlusion layer 40. , ί (4) - Figure 'wrong occlusion layer 4 °, compared to the catalyst layer 3. The optimum thickness line 32, ^: the control of the growth direction of the second-end nano-nose stone. The thickness of the catalyst layer 30 is from - end to the other end - too. The thickness of a position between the ends is most suitable for growth under a certain chemical vapor deposition condition. The thickness of this position is the optimum thickness. The optimum thickness can be achieved by forming a catalyst layer having a thickness ranging from thin to thick on a base of a stone substrate, and forming a catalyst layer having a thickness ranging from a thin layer to a thick layer. The upper layer can reach the optimum thickness: the neat tube array is grown by chemical vapor deposition under predetermined conditions; the growth material of the carbon nanotube is grown by the mirror; the thickness of the contacted lion is the prefab The best thickness below. The position of the most rotational position on the catalysis is the line of the most & For example, in the present embodiment, the iron on the base of the Shixi substrate is used as a catalyst, and is grown by B at 7 ° C. The optimum thickness of the catalyst layer of the butadiene and carbon nanotubes is about 5 nm, and the thickness is about 5 The position of the nanometer is connected to the line - that is the optimum thickness line 32, see the second and third figures. What is the optimum thickness on the catalyst layer? 1 The carbon nanotubes have the fastest growth rate, and the thickness of the carbon nanotubes is gradually thinned or thickened. In the actual operation, the thickness distribution of the catalytic sword layer 3 in the present embodiment may further preferably be 3 nm to 15 nm in consideration of the optimum thickness. Referring to the second figure, a certain pattern is formed on the catalyst layer 30. In this embodiment, a catalytic thin form is formed by photolithography: a layer of light is formed on the catalytic_3G, the woven fabric is exposed by using a mask having a predetermined pattern, and after development, the unprotected catalyst film is washed away with acid, and the catalyst is removed. The photoresist on the 'forms the catalyst blocks 33, 34 having a thickness gradient corresponding to the mask pattern. Since the position of the catalyst block is selected on which side of the optimum thickness will determine the direction of the bend of the carbon nanotube array, the photolithographic pattern should be formed as needed to align the mask position 96. *2.29 1317348 . When the catalyst layer 30 is deposited, if a hollow template of a suitable thickness is used as the shielding layer 4, the step of forming a catalyst pattern by photolithography may be advanced before the deposition of the catalyst layer 3, and in this step, only the stripping may be performed. The process can obtain a catalyst pattern with a thickness that meets the requirements. Referring to the fourth figure, the substrate with the catalyst block is immersed in air, annealed under 3 or the like, and the catalyst block is oxidized and contracted into nanometer-sized catalyst oxide particles 31. The size of the local catalyst oxide particles is related to the thickness of the catalyst layer at the place, for example, the thickness of the catalyst block 33 on the side of the preferred thickness is oxidized and contracted into catalyst oxide particles, and the diameter is smaller; the thickness is larger than the optimum thickness side. The catalyst block 34 is oxidatively shrunk into catalyst oxide particles 34 and has a large diameter. Referring to the fifth figure, the substrate on which the catalyst oxide particles are formed on the surface is placed in a reaction furnace (not shown), and carbon source ethylene is introduced to grow the carbon nanotube array by chemical vapor deposition. During the growth process, the carbon source and the gas decompose the chlorine gas to reduce the catalyst oxide particles 33 to the catalyst particles 33, 34, and then grow the carbon nanotube arrays 5, 51 on the catalyst particles 33", 34". Carbon gas can also be converted to other carbon-containing gases such as hydrazine and acetylene. However, due to the difference in the optimum thickness of the catalytic ride under different chemical vapor deposition conditions, the chemical gas accumulation conditions of the growth carbon nanotubes are better than the above-mentioned conditions for determining the optimum thickness _ growth carbon nanotubes. the same. The growth time of the stack can control the growth length of the carbon nanotube arrays 5, 51. The selling thickness is less than the optimum thickness of the side, the catalytic block 33 is close to the optimum thickness line side to the degree - side 'thickness riding thinning; the thickness A is the best thickness side - the catalyst = 34 from the most money Good thickness - away from the optimum thickness - side, the thickness gradually thickens. Since the growth rate of the carbon nanotubes at the position of the thickness of the book is the fastest, the thickness of the carbon nanotubes is gradually reduced by 1: the growth rate of the carbon nanotubes is gradually reduced, so that the catalytic particles are grown on the catalyst particles 33". The tube array touches the thicker side of the catalytic block - the side is thinner - the side is curved, and the carbon nanotube array 51 grown from the thicker side is from the thinner side of the catalyst block to the side of the thicker layer i 2, that is, the catalyst particles 33, ', 34" on the growth of the carbon nanotubes are deviated from the direction closest to the most thick wire away from the optimum thickness. If two catalytic blocks are respectively distributed on both sides of the carbon tube ml, the two carbon nanotube arrays having opposite growth directions can be obtained by growing the nano-layers formed on the two-staged formation. Referring to the first, the figure 'the carbon nanotube array structure obtained by the above method】 includes a substrate 11 95. 1317348 i0 phase of the nano carbon material touches 51, the nano tube array and the grip have two ends gradually thinning Bending at the end-end; if the catalyst layer is two === :=== round-spinning, splitting the chest, = ΐ second embodiment: a chemical-type catalytic ship source, or a linear catalyst destroying source, and a linear catalyst The sputtering of the base of the base is better than that of the city, and it is true that the layer of the layer is a catalyst layer having a thickness gradient. The basic steps of fabricating the carbon nanotube array are the same as those of the first embodiment, and the steps are specifically described as follows: Referring to the seventh figure, a substrate 7G is provided, and a base 8G is formed at the center of the substrate 7G. On the upper layer 8 (), the branch-face is difficult to ride 6G, and when the base 2 catalyst is applied, the -partial catalytic surface blocks the fine block, thereby forming a catalyst layer 9〇 having a thickness gradient on the substrate. — 〃 The occlusion layer has a vertical offset of a predetermined degree, which is the thickness of the cover (9). At least a portion of the surface-type catalyst sputtering source 60 is located directly above the shielding layer 8〇 so that the sputtering catalyst has 4 knives which are blocked by the shielding layer 8〇, so that the catalyst layer 9〇 obtained on the substrate is located in the occlusion, the area around 80 There is a gradient gradient inside, and the closer to the shielding layer 8〇, the thinner the thickness of the catalyst layer egg, and the further away from the shielding layer 8〇, the thickness of the catalyst layer 90 gradually increases. The shielding layer 80 covers the area of the substrate along the gradient of the thin to thick gradient of the catalyst layer to cover the area where the carbon nanotube array needs to be grown, and the length of the shadow range in the direction is preferably substantially equal to the thickness of the layer. 2 times, the method of determining the shading is compared with the first embodiment. In addition, a linear catalyst sputtering source can also be used. When the catalyst is sputtered, the source of the subtraction can be relatively stationary with the substrate; if the linear catalyst is parallel to the ==, the sputtering source and the substrate can be perpendicular to the plane of the paper. Relatively shifting 12 1317348 '95. 12^2 d in a larger range; if the linear catalyst sputtering source is placed perpendicular to the paper surface, the catalyst sputtering source and the substrate can be in a larger range parallel to the paper surface direction. The relative movement is to ensure that a large area of the catalyst layer 90 having a thickness gradient is deposited in the area around the occlusion layer 8 . Referring to the eighth and ninth figures, the occlusion layer 80 is removed, the optimum thickness of the catalyst layer 90 is determined under certain chemical vapor deposition conditions, and the optimum thickness line 92 is marked. The method of determining the optimum thickness line 92 is the same as the method of determining the optimum thickness line 32 in the first embodiment. Please refer to the tenth figure to form a catalyst pattern. In this embodiment, the catalyst layer 9 is etched out of the plurality of catalyst blocks 'and the catalyst blocks are respectively located on both sides of the optimum thickness line 92, such as the catalyst blocks 93, 94 are respectively at the optimum thickness line 92. On both sides, the catalyst blocks 95, 96 are on either side of the φ optimal thickness line 92, respectively. Alternatively, when the catalyst layer 90 is deposited, a thick and thin double-layer composite hollow masking template is used instead of the shielding layer 8〇 to directly deposit a catalyst pattern having a desired thickness. Referring to Fig. 11, the substrate 7 with the catalyst block is immersed in air, annealed at 3 ° C or the like to oxidize and shrink the catalyst block into nanometer-sized catalyst oxide particles. The local catalyst oxide particle size is related to the thickness of the catalyst block at the site, such as the catalyst block 94, 95 having a thickness less than - the optimum thickness side is oxidatively contracted into catalyst oxide particles 94, and the 95'' diameter is small. The catalyst block (10), % having a thickness greater than the optimum thickness side is oxidized and shrunk into catalyst oxide particles 93, 96, and has a large diameter. Referring to Fig. 12, a substrate 7 having a catalyst oxide particle formed on its surface is placed in a reactor (not shown), a carbon source gas is introduced, and a carbon nanotube array is grown by chemical vapor deposition. The carbon source gas decomposes hydrogen during the growth process to reduce the catalyst oxide particles 93, 94, 95, 96 to catalyst particles 93", 94", 95", 96", respectively, and then to the catalyst particles 93" , 94", 95", 96" grown carbon nanotube arrays 100, 1 〇 1 〇 2 and 1 〇 3. On the side of the thickness less than the optimum thickness, the carbon nanotube arrays 1〇1, 1〇2 grown on the catalyst particles 94", 95" are bent toward the thickness of the catalyst block, and the thickness is greater than the optimum thickness. On one side, the catalyst particles are evil, and the growth of the nano-tube array _G, 1G3 increases the thickness of the catalytic block (4). The same is true for the nano-carbon tube arrays being bent away from one side of the optimum thickness line 92. • Referring to FIG. 13 , the carbon nanotube array structure 2 obtained by the above method includes a substrate 7 , a plurality of carbon nanotube arrays 1 , 1 , 1 , 2 , and 2 formed on the substrate 70 . 1〇3, etc., a plurality of carbon nanotube arrays have a plurality of different bending directions. 13 «1 1317348 Micro-electronic devices such as devices and electric vacuum devices, field emitters and other micro-functional structures. (4) " * oblique-based bridge, electron gun or sensing catalyst block of the gamma _ job dumping, such as Appropriately changing the technology of the present invention 峨 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ It is said that ϊγ should be included in the following patent application 匕 =: a side view of the catalyst layer deposition process in the first embodiment of the month. Intent The first picture is a three-dimensional diagram of the catalyst pattern formed in the embodiment of the present invention. f is a schematic view of the catalyst pattern after annealing in the first embodiment of the present invention. A schematic side view of a carbon nanotube array grown in the first embodiment. A perspective view of a carbon nanotube array structure prepared in the first embodiment. BRIEF DESCRIPTION OF THE DRAWINGS The side view of the catalyst layer deposition process in the second embodiment of the present invention. Intentions Fig. 8 is a perspective view showing the catalyst layer formed in the second embodiment of the present invention. The ninth drawing is a side view showing the catalyst layer formed in the second embodiment of the present invention. The tenth figure is a three-dimensional schematic diagram of the catalytic surface formed by the second real drama t. 11 is a schematic side view of the catalyst after annealing in the second embodiment of the present invention. The tenth death diagram is a schematic side view of the nanodisk array grown in the second embodiment of the present invention. Figure 12 is a perspective view of a carbon nanotube array structure prepared by a second embodiment of the present invention. [Main component symbol description] ~ 10, 70 100 , 101 carbon nanotube array structure 1, 2 substrates 102, 103, 50 , 51 carbon nanotube array I317348 W· 12.29 % cs > linear sputtering source 20 catalyst layer 30, 90 optimal thickness line 32, 92 catalyst block 33, 34, 93, 94, 95, 96 catalyst oxide Particles 33', 34, 93', 94, 95, 96' catalyst particles 33", 34", 93", 94", 95", 96 shielding layer 40, 80 vertical shielding surface 42 surface sputtering source 60 15