JP2004061665A - Method for manufacturing semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the following problem: disconnection or increase and unevenness in wiring resistance due to resist residue remaining in distribution holes, or impossibility of formation of a fine circuit pattern due to swelling by development, in forming a copper wiring circuit by a dual damascene method in which distribution holes are formed prior to formation of distribution grooves using ArF lithography capable of forming a fine pattern. <P>SOLUTION: A dual damascene process in which distribution holes are formed prior to formation of distribution grooves is carried out using ArF lithography with a negative-working resist whose exposed area is made negative by lactonization reaction as a resist in formation of a distribution groove pattern. By this method, the above problem is solved. <P>COPYRIGHT: (C)2004,JPO

Description

【０００８】
【化４】

【０００９】
上記ネガレジストは、化学増幅系レジストであり、ＡｒＦエキシマレーザ露光により、酸発生剤トリフェニルスルホニウムトリフレートから酸が発生し、発生した酸により誘起される酸触媒反応で、構造（Ｉ）中のδ−ヒドロキシ酸構造がδ−ラクトン構造に変化する。この反応は、熱処理により促進されることから、露光後、現像を行う前に、１２０℃で９０秒間の露光後ベークを行った。次に０．２ｗｔ％テトラメチルアンモニウムヒドロキシド水溶液で６０秒間現像を行った。本レジストでは、未露光部では、δ−ヒドロキシ酸構造が存在することからアルカリ可溶性であり、露光部ではδ−ラクトン構造がアルカリ不溶性であることから、現像により未露光部が除去されてネガ型のパタンが形成された。なお上記のクエンチャーは、酸発生剤から露光で発生した酸の拡散を制御して、高解像性を与え、またプロセス安定性を向上させる。
【００１０】
上記のようなネガレジストは、ＡｒＦエキシマレーザ光に感度を持ち、ネガ化反応が極性基から非極性基に変換する極性変換反応であるため現像膨潤がなく、しかもオーバー現像に対する裕度が高いという特長があった。
なおここでは、ベース樹脂として、δ−ヒドロキシ酸と脂環式構造を有する構造（Ｉ）と水酸基を有するα−置換アクリル酸エステルが重合した構造である　（ＩＩ）が含まれるものを用いたが、さらに第三の構造を共重合した構造の樹脂を用いることもできる。また、異なるベース樹脂を混合して用いることもできる。
【００１１】
層間膜１０３上のレジスト塗布膜厚は０．３μｍとしたが、配線孔１０６の深さは０．７μｍあり、配線孔部ではレジストは１．０μｍの膜厚となった。このレジストの未露光部の現像レートは３０ｎｍ／ｓであり、１．０μｍ膜厚のレジストを溶解するのに約３３秒かかる。６０秒現像としたためこの孔の部分では約１．８倍のオーバー現像となった。層間膜上の０．３μｍ膜厚のレジストは約１０秒で溶解するので、約６倍のオーバー現像となった。　このレジストはこのようなオーバー現像を行っても現像膨潤は実用上十分押さえられており、±１０％の寸法精度で配線溝パターン１０７’（図２（ｄ））を形成することができた。かつこの状況で配線孔１０６内にはレジスト残りは発生しなかった。なおここでは溝パターンの最小配線幅は１２０ｎｍとし、このときパタン形成に必要とした露光量は１７ｍＪ／ｃｍ^２であった。架橋型ＡｒＦネガレジストでは現像膨潤によりこのような微細溝パターンを±１０％の精度で形成することはできなかった。
【００１２】
また本発明で用いたネガレジストのベース樹脂として、上記構造（Ｉ）と共重合するコモノマーに水酸基を含まない構造を用いた場合は、０．３μｍという薄膜に現像時間を合わせ短時間現像とすれば配線溝パターンを高精度に形成することは可能であるが、配線孔内のレジストを完全に除去するほどのオーバー現像を行うと配線溝パターンを高精度に形成することは困難であった。なお、本発明で用いる現像液としてはテトラメチルアンモニウムヒドロキサイドの水溶液が挙げられ、その濃度は０．２から０．５ｗｔ％とするとオーバー現像裕度もとれて好ましかった。０．５ｗｔ％以上では濃度が上がるほど現像膨潤が発生し、０．２ｗｔ％以下では現像液のヘタリによる寸法変動が発生した。
【００１３】
その後図２（ｅ）に示すようにネガレジストパターン１０７をマスクに層間膜をエッチングして配線溝１１０を形成する。この方法では配線孔内にレジスト残りがないので従来法と異なりインナークラウンの発生はなかった。その後配線孔や溝に銅を埋め込みＣＭＰを行って図２（ｆ）に示すように銅配線１１１を形成した。この方法によって電気抵抗のバラツキや断線といった問題のない電気的信頼性の高い配線を寸法精度高く形成することができた。なお化学増幅型ポジレジストを用いて配線溝パターンを形成してみたところ、ＳｉＯＣ孔内側壁四方からの酸失活物質の拡散と、孔内の光強度の低下が原因となって、配線溝内にレジスト残りが生じた。
【００１４】
（実施の形態２）
実施の形態２では、半導体装置の製造工程を示した工程図である図３（ａ）から（ｄ）、図４（ａ）から（ｄ）、及び図５（ａ）から（ｄ）を用いて、デュアルダマシン法による銅配線の形成方法を説明する。まず、図３（ａ）に示すように基板１００上に配線３０１、絶縁膜３０２、ストッパ膜３０３、第１の層間膜３０４、中間膜３０５、第２の層間膜３０６、及びキャップ膜３０７が形成された試料を準備した。ここで配線３０１としては銅を用いたがこれに限らずＷ等でも良い。絶縁膜３０２としては酸化膜を用いた。ストッパ膜３０３としてはシリコン窒化膜を用いた。層間膜３０４および３０６としては誘電率の低い有機膜を用いた。ここではダウケミカル社のＳｉＬＫ（登録商標）を用いた。ＳｉＬＫ（登録商標）膜のＡｒＦエキシマレーザ光（波長１９３　ｎｍ）対する屈折率虚部（消衰係数とも呼ばれるもので物質の光吸収量を表す）は０．７１である。中間膜３０５およびキャップ膜３０７としてはＳｉＣ膜あるいはＳｉＯＣ膜を用いた。次に図３（ｂ）に示すように配線孔パターン３０８が開口したポジレジストパターン３０９を形成した。
ここで、このレジストの膜厚は第２の層間膜の膜厚より厚く、第１の層間膜と第２の層間膜の膜厚を足した膜厚より薄い膜厚とした。具体的には第１の層間膜の膜厚を３３０ｎｍ、第２の層間膜の膜厚を２７０ｎｍ、そしてレジストの膜厚を４００ｎｍとした。
その後図３（ｃ）に示すようにエッチングをおこなってキャップ膜３０７に開口３１０を形成し、さらにエッチングを行って図３（ｄ）に示すように第２の層間膜３０６および中間膜３０５にも開口を形成した。
【００１５】
その後図４（ａ）に示すように第１の層間膜３０４のエッチングを行った。膜厚とエッチレートの関係からこのエッチングの際にポジレジストは除去された。エッチングストッパ３０３によって配線３０１は保護される。その後図４（ｂ）に示すようにネガレジスト３１１を塗布し、マスク３１３を介してＡｒＦエキシマレーザ露光光３１４を照射し、配線溝パターンを露光した。図示はしていないがこの露光にはＡｒＦスキャナを用い、レンズを介した露光を行った。ここで用いたネガレジスト３１１は、実施例１と同様にネガ化の機構にラクトン化反応を利用した化学増幅系レジストで、構造（Ｉ）と構造（ＩＩＩ）が７０／３０のモル比で共重合したポリマーをベース樹脂とし、前記ベース樹脂１００重量部、酸発生剤トリフェニルスルホニウムノナフレート２重量部、クエンチャーとして２−ベンジルピリジン０．０１重量部を１−メトキシ−２−プロパノール１０００重量部に溶解して、孔径０．２ミクロンのフッ素樹脂フィルターを用いて、ろ過したものである。
【００１６】
【化５】

【００１７】
【化６】

【００１８】
上記ネガレジストは、化学増幅系レジストであり、ＡｒＦエキシマレーザ露光により、酸発生剤トリフェニルスルホニウムノナフレートから酸が発生し、発生した酸により誘起される酸触媒反応で、構造（Ｉ）中のδ−ヒドロキシ酸構造がδ−ラクトン構造に変化する。この反応は、熱処理により促進されることから、露光後、現像を行う前に、１２０℃で９０秒間の露光後ベークを行った。次に０．２ｗｔ％テトラメチルアンモニウムヒドロキシド水溶液で６０秒間現像を行った。本レジストでは、未露光部では、δ−ヒドロキシ酸構造が存在することからアルカリ可溶性であり、露光部ではδ−ラクトン構造がアルカリ不溶性であることから、現像により未露光部が除去されてネガ型のパタンが形成された。
【００１９】
上記のようなネガレジストは、ＡｒＦエキシマレーザ光に感度を持ち、ネガ化反応が極性基から非極性基に変換する極性変換反応であるため現像膨潤がなく、しかもオーバー現像に対する裕度が高いという特長があった。
【００２０】
層間膜１０３上のレジスト塗布膜厚は第２の層間膜の膜厚より薄い　２００ｎｍとした。これは第２の層間膜エッチングの際のレジスト残りを防止するためである。有機層間膜を用いているため有機層間膜をエッチングするアッシャを用いることができない。そのためレジスト残りは歩留り低下、配線信頼性低下の元凶となる。その後現像を行って図４（ｃ）に示すように配線溝３１５の形成されたネガレジストパターン３１６を形成した。層間膜に用いているＳｉＬＫ（登録商標）は露光光として用いているＡｒＦエキシマレーザ光を良く吸収するので微細孔の中を露光するのにはかなりのオーバー露光が必要になる。しかしこのネガレジストを用いた場合には孔内を露光する必要がなく、露光不足によるレジスト残りなどの不良は発生しない。一方実施例１と同様オーバー現像が必要になるが、上記ネガレジストはオーバー現像裕度が大きく、微細パターンを高精度に形成できた。
【００２１】
その後図４（ｄ）に示すようにネガレジストパターンをマスクにキャップ膜を加工して配線溝キャップパターン３１７を形成し、図５（ａ）に示すようにキャップ膜パターン３１７をエッチングマスクに第２の層間膜をエッチングして配線溝パターンの形成された第２の層間膜３１８を形成する。このエッチングの際、ネガレジストは自動的に除去される。また配線３０１はストッパ３０３によって保護されている。その後図５（ｂ）に示すようにストッパ層をエッチングして開口３１９を形成する。次ぎにＴｉＮバリア膜を埋め込んだ後（図示なし）図５（ｃ）に示すように銅３２０を配線孔や溝に埋め込み、図５（ｄ）に示すようにＣＭＰを行って不要な銅を削り、銅配線３２０’を形成する。このＣＭＰの際キャップ膜３１７はＣＭＰの保護膜としても機能する。このようにして形成された銅配線にはインナークラウンの発生がなかった。そのため電気的信頼性も高く、抵抗のバラツキも少なく、配線のパターン精度も高かった。　なお、本実施例では消衰係数が０．７１のＳｉＬＫ（登録商標）を層間膜に用いた場合を説明したが、層間膜の消衰係数が０．３以上の場合に効果がみられた。
すなわち、ポジレジストで配線溝を形成した場合、層間膜の消衰係数が０．３を越えると孔内にレジストが残り、またインナクラウンが発生したが、本実施例の方法を用いた場合はそのような問題は発生しなかった。
【００２２】
（実施の形態３）
実施の形態３では、実施の形態１で用いたネガレジストの代りに、構造（Ｉ）と構造（ＩＶ）が７５／２５で共重合したポリマーをベース樹脂とし、前記ベース樹脂１００重量部、酸発生剤トリフェニルスルホニウムトリフレート０．５重量部、酸発生剤トリフェニルスルホニウムノナフレート０．５重量部、クエンチャーとしてトリエタノールアミン０．０１重量部を１−メトキシ−２−プロパノール１０００重量部に溶解して、孔径０．２ミクロンのフッ素樹脂フィルターを用いて、ろ過したものを用いた。その結果、実施の形態１と同様に配線溝が形成でき、それをもとに電気抵抗のバラツキや断線といった問題のない電気的信頼性の高い配線を寸法精度高く形成することができた。
【００２３】
【化７】

【００２４】
【化８】

【００２５】
（実施の形態４）
実施の形態４では、実施の形態２で用いたネガレジストの代りに、構造（Ｉ）と構造（Ｖ）が６０／４０で共重合したポリマーをベース樹脂とし、前記ベース樹脂１００重量部、酸発生剤トリフェニルスルホニウムトリフレート０．５重量部、酸発生剤トリフェニルスルホニウムノナフレート０．５重量部、クエンチャーとしてヨウ化テトラメチルアンモニウム０．０１重量部を１−メトキシ−２−プロパノール１０００重量部に溶解して、孔径０．２ミクロンのフッ素樹脂フィルターを用いて、ろ過したものを用いた。その結果、実施の形態２と同様に配線溝が形成でき、それをもとに電気抵抗のバラツキや断線といった問題のない電気的信頼性の高い配線を寸法精度高く形成することができた。
【００２６】
【化９】

【００２７】
【化１０】

【００２８】
（実施の形態５）
実施の形態５では、実施の形態１で用いたネガレジストの代りに、構造（Ｉ）と構造（ＶＩ）と構造（ＶＩＩ）が６０／３０／１０で共重合したポリマーをベース樹脂とし、前記ベース樹脂１００重量部、酸発生剤トリフェニルスルホニウムトリフレート１．５重量部、クエンチャーとして２−ベンジルピリジン０．０１重量部を１−メトキシ−２−プロパノール１０００重量部に溶解して、孔径０．２ミクロンのフッ素樹脂フィルターを用いて、ろ過したものを用いた。その結果、実施の形態１と同様に配線溝が形成でき、それをもとに電気抵抗のバラツキや断線といった問題のない電気的信頼性の高い配線を寸法精度高く形成することができた。
【００２９】
【化１１】

【００３０】
【化１２】

【００３１】
【化１３】

【００３２】
（実施の形態６）
実施の形態５では、実施の形態２で用いたネガレジストの代りに、構造（Ｉ）と構造（ＶＩＩＩ）が２５／７５で共重合したポリマーをベース樹脂とし、前記ベース樹脂１００重量部、酸発生剤トリフェニルスルホニウムトリフレート１．０重量部、クエンチャーとして２−ベンジルピリジン０．０１重量部を１−メトキシ−２−プロパノール１０００重量部に溶解して、孔径０．２ミクロンのフッ素樹脂フィルターを用いて、ろ過したものを用いた。現像液に０．０５ｗｔ％テトラメチルアンモニウムヒドロキシド水溶液を用いることにより、実施の形態１と同様に配線溝が形成でき、それをもとに電気抵抗のバラツキや断線といった問題のない電気的信頼性の高い配線を寸法精度高く形成することができた。
【００３３】
【化１４】

【００３４】
【化１５】

【００３５】
【発明の効果】
本願によって、配線溝内にインナークラウンが形成されることなく、また配線孔内のレジスト残りもなしに微細な銅配線を形成できる。このため配線の高抵抗化や断線のない信頼性の高い微細銅配線回路を持つ半導体装置を製造することが可能となる。
【図面の簡単な説明】
【図１】本発明の第１の実施例による銅配線形成工程を半導体装置の断面図を用いて示した工程図である。
【図２】従来の銅配線形成工程を半導体装置の断面図を用いて示した工程図である。
【図３】本発明の第２の実施例による銅配線形成工程を半導体装置の断面図を用いて示した工程図である。
【図４】本発明の第２の実施例による銅配線形成工程を図３に引き続き示した工程図である。
【図５】本発明の第２の実施例による銅配線形成工程を図４に引き続き示した工程図である。
【符号の説明】
１００…基板、１０１…配線、１０２…バリア膜、１０３…層間膜、１０４…配線孔パターン、１０５…ポジレジスト、１０６…配線孔、１０７…ポジレジスト、１０７’…配線溝パターン、１０８…マスク、１０９…ＡｒＦエキシマレーザ露光光、１１０…配線溝、１１１…銅配線、１２０…ポジレジスト、３０１…配線、３０２…絶縁膜、３０３…ストッパ膜、３０４…第１の層間膜、３０５…中間膜、３０６…第２の層間膜、３０７…キャップ膜、３０７’…キャップ膜、３０８…配線孔パターン、３０９…ポジレジストパターン、３０９’…ポジレジスト、３１０…開口、３１１…共重合ネガレジスト、３１３…マスク、３１４…ＡｒＦエキシマレーザ露光光、３１５…配線溝、３１６…ネガレジストパターン、３１７…配線溝キャップパターン、３１８…配線溝パターンの形成された第２の層間膜、３１９…開口、３２０…銅、３２０’…銅配線。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for manufacturing a semiconductor device, and more particularly to a method for manufacturing a copper wiring having high dimensional accuracy and high electrical reliability.
[0002]
[Prior art]
In recent years, for the purpose of improving the operation speed of a semiconductor device, copper wiring semiconductor devices using copper as a wiring material and having reduced wiring resistance have been mass-produced. As a method of forming a copper wiring, a wiring groove and a wiring hole are formed in an insulating film using a resist pattern as a mask, copper is buried in the groove and the hole, and then CMP (Chemical Mechanical Polishing) is performed to simultaneously form the wiring and the connection wiring with copper. A dual damascene method for forming is used. This method and its problems will be described in detail with reference to FIG. First, as shown in FIG. 2A, a resist having a wiring hole pattern 104 is formed of a positive resist 105 on the substrate 100, the wiring 101, the barrier film 102, and the interlayer film 103. Thereafter, as shown in FIG. 2B, a wiring hole 106 is formed in the interlayer film by an etching method using the positive resist 105 as a mask. Next, as shown in FIG. 2C, a positive resist 120 is applied, and exposure light 109 is irradiated through a mask 108 to expose a wiring groove pattern. Although not shown, this exposure is usually performed by a projection exposure method via a lens.
Here, the result of forming the wiring groove resist pattern by developing the positive resist 120 is shown in FIG. 2D. However, the resist in the wiring hole cannot be completely removed and a part of the resist remains 121. Increasing the amount of exposure so that there is no remaining resist lowers the pattern formation accuracy of the wiring groove. When a wiring groove is formed in the interlayer film by etching, a kind of burr called an inner crown 122 remains as shown in FIG. After that, copper is buried in the wiring holes and grooves to perform a CMP to form a copper wiring.
Here, this method has a problem that the electrical reliability of the wiring is impaired by the insulating etching residue 123 accompanying the inner crown as shown in FIG.
[0003]
As a method of solving this problem, a method of forming a wiring groove pattern using a negative resist has been proposed. This method is described in JP-A-2001-257260. By using a negative resist, the resist in the wiring hole can be removed by development, not by exposure, so that it is possible to accurately form a wiring groove pattern that does not leave the resist in the hole.
[0004]
[Problems to be solved by the invention]
In the dual damascene method using the negative resist, there is a large difference between the resist film thickness on the interlayer film and the resist film thickness including the inside of the hole. For this reason, a thin resist can be accurately formed even during the developing time when the thick resist is removed. In other words, a negative resist having a high development tolerance for over-development is required for a thin portion of the resist.
Conventionally, ArF lithography having a wavelength of 193 nm has no negative resist having such characteristics, and there has been a problem that ArF lithography capable of forming a fine pattern cannot be applied to the dual damascene.
Note that ArF lithography has a higher exposure resolution than KrF lithography with a wavelength of 248 nm, which is currently the mainstream of fine pattern formation, and thus has a higher resolution, thereby enabling finer pattern formation.
[0005]
[Means for Solving the Problems]
In order to solve the above-mentioned conventional problems, a negative resist in which an exposed portion becomes negative by a lactonization reaction is used as a resist for forming a wiring groove pattern, and a wiring hole is formed prior to the wiring groove by using ArF lithography. Perform a dual damascene process. This method solves the above problem.
[0006]
BEST MODE FOR CARRYING OUT THE INVENTION
(Embodiment 1)
In the first embodiment, a method of forming a copper wiring by a dual damascene method will be described with reference to FIGS. 1A to 1F, which are process diagrams showing a manufacturing process of a semiconductor device.
First, as shown in FIG. 1A, a resist having a wiring hole pattern 104 was formed of a positive resist 105 on a substrate 100, a wiring 101, a barrier film 102, and an interlayer film 103. Here, an SiOC film formed by CVD (Chemical Vapor Deposition) was used as the interlayer film. Thereafter, as shown in FIG. 1B, wiring holes 106 were formed in the interlayer film by an etching method using the positive resist 105 as a mask. Next, as shown in FIG. 1C, a negative resist 107 was applied, and an ArF excimer laser exposure light 109 was irradiated through a mask 108 to expose the wiring groove pattern. Although not shown, this exposure was performed through a lens using an ArF scanner. The negative resist 120 used here is a polymer in which the structure (I) and the structure (II) are copolymerized in a molar ratio of 50/50 as a base resin, 100 parts by weight of the base resin, an acid generator triphenylsulfonium triflate. 1 part by weight, 0.01 part by weight of 2-benzylpyridine as a quencher was dissolved in 1000 parts by weight of 1-methoxy-2-propanol, and the solution was filtered using a fluororesin filter having a pore size of 0.2 micron. .
[0007]
Embedded image

[0008]
Embedded image

[0009]
The negative resist is a chemically amplified resist. An acid is generated from an acid generator, triphenylsulfonium triflate, by ArF excimer laser exposure, and an acid catalyzed reaction induced by the generated acid causes a reaction in the structure (I). The δ-hydroxy acid structure changes to a δ-lactone structure. Since this reaction is accelerated by heat treatment, a post-exposure bake was performed at 120 ° C. for 90 seconds after exposure and before development. Next, development was performed with a 0.2 wt% aqueous solution of tetramethylammonium hydroxide for 60 seconds. In the present resist, the unexposed portion is alkali-soluble due to the presence of a δ-hydroxy acid structure, and the exposed portion is alkali-insoluble in the δ-lactone structure in the exposed portion. Was formed. The quencher controls the diffusion of the acid generated by exposure from the acid generator to give high resolution and improve the process stability.
[0010]
Such a negative resist has sensitivity to ArF excimer laser light, and is free from development swelling because the negative reaction is a polarity conversion reaction that converts a polar group into a non-polar group, and has a high tolerance to over-development. There were features.
Here, as the base resin, a resin containing (II) which is a structure obtained by polymerizing a structure (I) having a δ-hydroxy acid and an alicyclic structure and an α-substituted acrylate having a hydroxyl group was used. Alternatively, a resin having a structure obtained by copolymerizing the third structure may be used. Further, different base resins can be used in combination.
[0011]
Although the resist coating thickness on the interlayer film 103 was 0.3 μm, the depth of the wiring hole 106 was 0.7 μm, and the thickness of the resist in the wiring hole portion was 1.0 μm. The development rate of the unexposed portion of the resist is 30 nm / s, and it takes about 33 seconds to dissolve the resist having a thickness of 1.0 μm. Since the development was performed for 60 seconds, over development was performed about 1.8 times in this hole. Since the resist having a thickness of 0.3 μm on the interlayer film was dissolved in about 10 seconds, the over-developing was about 6 times. Even if such resist was subjected to such over-development, the development swelling was sufficiently suppressed for practical use, and the wiring groove pattern 107 ′ (FIG. 2D) could be formed with a dimensional accuracy of ± 10%. In this situation, no resist remains in the wiring hole 106. Here, the minimum wiring width of the groove pattern was 120 nm, and at this time, the exposure required for pattern formation was 17 mJ / cm ² . With a cross-linked ArF negative resist, such a fine groove pattern could not be formed with an accuracy of ± 10% due to development swelling.
[0012]
When a structure not containing a hydroxyl group in the comonomer copolymerized with the above structure (I) is used as the base resin of the negative resist used in the present invention, the development time is adjusted to a thin film of 0.3 μm, and the development is performed in a short time. Although it is possible to form the wiring groove pattern with high precision, it is difficult to form the wiring groove pattern with high precision if over-developing is performed to completely remove the resist in the wiring hole. The developing solution used in the present invention includes an aqueous solution of tetramethylammonium hydroxide. When the concentration is from 0.2 to 0.5 wt%, the over-developing margin is preferable. At 0.5 wt% or more, development swelling occurred as the concentration increased, and at 0.2 wt% or less, dimensional fluctuation due to settling of the developer occurred.
[0013]
Thereafter, as shown in FIG. 2E, the interlayer film is etched using the negative resist pattern 107 as a mask to form a wiring groove 110. In this method, there was no resist remaining in the wiring hole, so that no inner crown was generated unlike the conventional method. Thereafter, copper was buried in the wiring holes and grooves and subjected to CMP to form a copper wiring 111 as shown in FIG. According to this method, a highly reliable wiring having no problem such as variation in electrical resistance and disconnection could be formed with high dimensional accuracy. When a wiring groove pattern was formed using a chemically amplified positive resist, it was found that diffusion of an acid deactivating material from the four side walls of the SiOC hole and a decrease in light intensity in the hole caused the formation of the wiring groove pattern. The resist was left on the surface.
[0014]
(Embodiment 2)
In the second embodiment, FIGS. 3 (a) to 3 (d), FIGS. 4 (a) to 4 (d), and FIGS. 5 (a) to 5 (d), which are process diagrams showing the manufacturing steps of the semiconductor device, are used. Next, a method of forming a copper wiring by a dual damascene method will be described. First, as shown in FIG. 3A, a wiring 301, an insulating film 302, a stopper film 303, a first interlayer film 304, an intermediate film 305, a second interlayer film 306, and a cap film 307 are formed on a substrate 100. The prepared sample was prepared. Here, copper is used as the wiring 301, but it is not limited to this, and W may be used. An oxide film was used as the insulating film 302. As the stopper film 303, a silicon nitride film was used. As the interlayer films 304 and 306, organic films having a low dielectric constant were used. Here, Dow Chemical's SiLK (registered trademark) was used. The imaginary part of the refractive index of the SiLK (registered trademark) film with respect to the ArF excimer laser light (wavelength 193 nm) (also referred to as the extinction coefficient and representing the light absorption amount of the substance) is 0.71. As the intermediate film 305 and the cap film 307, a SiC film or a SiOC film was used. Next, as shown in FIG. 3B, a positive resist pattern 309 having an opening in the wiring hole pattern 308 was formed.
Here, the film thickness of the resist is larger than the film thickness of the second interlayer film and smaller than the film thickness of the first interlayer film and the film thickness of the second interlayer film. Specifically, the thickness of the first interlayer film was 330 nm, the thickness of the second interlayer film was 270 nm, and the thickness of the resist was 400 nm.
Thereafter, etching is performed as shown in FIG. 3C to form an opening 310 in the cap film 307, and further etching is performed to form the second interlayer film 306 and the intermediate film 305 as shown in FIG. An opening was formed.
[0015]
Thereafter, as shown in FIG. 4A, the first interlayer film 304 was etched. Due to the relationship between the film thickness and the etch rate, the positive resist was removed during this etching. The wiring 301 is protected by the etching stopper 303. Thereafter, as shown in FIG. 4B, a negative resist 311 was applied, and ArF excimer laser exposure light 314 was irradiated through a mask 313 to expose the wiring groove pattern. Although not shown, this exposure was performed through a lens using an ArF scanner. The negative resist 311 used here is a chemically amplified resist utilizing a lactonization reaction in the mechanism of negative conversion as in Example 1. The negative resist 311 has a structure (I) and a structure (III) in a molar ratio of 70/30. Using a polymerized polymer as a base resin, 100 parts by weight of the base resin, 2 parts by weight of an acid generator, triphenylsulfonium nonaflate, and 0.01 part by weight of 2-benzylpyridine as a quencher 1000 parts by weight of 1-methoxy-2-propanol And filtered using a fluororesin filter having a pore size of 0.2 micron.
[0016]
Embedded image

[0017]
Embedded image

[0018]
The negative resist is a chemically amplified resist. An acid is generated from an acid generator, triphenylsulfonium nonaflate, by ArF excimer laser exposure, and the acid catalyzed reaction induced by the generated acid causes a reaction in the structure (I). The δ-hydroxy acid structure changes to a δ-lactone structure. Since this reaction is accelerated by heat treatment, a post-exposure bake was performed at 120 ° C. for 90 seconds after exposure and before development. Next, development was performed with a 0.2 wt% aqueous solution of tetramethylammonium hydroxide for 60 seconds. In the present resist, the unexposed portion is alkali-soluble due to the presence of a δ-hydroxy acid structure, and the exposed portion is alkali-insoluble in the δ-lactone structure in the exposed portion. Was formed.
[0019]
Such a negative resist has sensitivity to ArF excimer laser light, and is free from development swelling because the negative reaction is a polarity conversion reaction that converts a polar group into a non-polar group, and has a high tolerance to over-development. There were features.
[0020]
The thickness of the resist coating on the interlayer film 103 was 200 nm, which was smaller than the thickness of the second interlayer film. This is to prevent the resist remaining when the second interlayer film is etched. Since an organic interlayer film is used, an asher for etching the organic interlayer film cannot be used. Therefore, the remaining resist becomes a cause of a decrease in yield and a decrease in wiring reliability. Thereafter, development was performed to form a negative resist pattern 316 in which a wiring groove 315 was formed as shown in FIG. Since SiLK (registered trademark) used for the interlayer film absorbs the ArF excimer laser light used as the exposure light well, a considerable over-exposure is required to expose the inside of the fine hole. However, when this negative resist is used, the inside of the hole does not need to be exposed, and defects such as remaining resist due to insufficient exposure do not occur. On the other hand, over-development was required as in Example 1. However, the above-mentioned negative resist had a large over-development margin, and a fine pattern could be formed with high precision.
[0021]
Thereafter, as shown in FIG. 4D, a cap film is processed using a negative resist pattern as a mask to form a wiring groove cap pattern 317, and as shown in FIG. 5A, a second groove is formed using the cap film pattern 317 as an etching mask. Is etched to form a second interlayer film 318 on which a wiring groove pattern is formed. During this etching, the negative resist is automatically removed. The wiring 301 is protected by a stopper 303. Thereafter, as shown in FIG. 5B, the stopper layer is etched to form an opening 319. Next, after the TiN barrier film is embedded (not shown), copper 320 is embedded in the wiring holes and grooves as shown in FIG. 5C, and unnecessary copper is removed by performing CMP as shown in FIG. 5D. , A copper wiring 320 'is formed. During this CMP, the cap film 317 also functions as a CMP protection film. No inner crown was generated in the copper wiring thus formed. Therefore, electrical reliability was high, resistance variation was small, and wiring pattern accuracy was high. In this embodiment, the case where SiLK (registered trademark) having an extinction coefficient of 0.71 is used for the interlayer film has been described. However, the effect was observed when the extinction coefficient of the interlayer film was 0.3 or more. .
That is, when the wiring groove is formed with a positive resist, when the extinction coefficient of the interlayer film exceeds 0.3, the resist remains in the hole and the inner crown occurs, but when the method of this embodiment is used, No such problem occurred.
[0022]
(Embodiment 3)
In the third embodiment, instead of the negative resist used in the first embodiment, a polymer obtained by copolymerizing the structure (I) and the structure (IV) at 75/25 is used as a base resin, and 100 parts by weight of the base resin is used. 0.5 parts by weight of a generator triphenylsulfonium triflate, 0.5 parts by weight of an acid generator triphenylsulfonium nonaflate, and 0.01 parts by weight of triethanolamine as a quencher in 1000 parts by weight of 1-methoxy-2-propanol It was dissolved and filtered using a fluorine resin filter having a pore size of 0.2 micron. As a result, a wiring groove can be formed in the same manner as in the first embodiment, and based on this, a wiring having high electrical reliability and having no problem such as variation in electric resistance or disconnection can be formed with high dimensional accuracy.
[0023]
Embedded image

[0024]
Embedded image

[0025]
(Embodiment 4)
In the fourth embodiment, instead of the negative resist used in the second embodiment, a polymer obtained by copolymerizing the structure (I) and the structure (V) at 60/40 is used as a base resin, and 100 parts by weight of the base resin is used. 0.5 parts by weight of a generator triphenylsulfonium triflate, 0.5 parts by weight of an acid generator triphenylsulfonium nonaflate, 0.01 parts by weight of tetramethylammonium iodide as a quencher 1000 parts by weight of 1-methoxy-2-propanol And then filtered using a fluororesin filter having a pore size of 0.2 micron. As a result, a wiring groove can be formed in the same manner as in the second embodiment, and based on this, a wiring having high electrical reliability and having no problem such as variation in electric resistance or disconnection can be formed with high dimensional accuracy.
[0026]
Embedded image

[0027]
Embedded image

[0028]
(Embodiment 5)
In the fifth embodiment, instead of the negative resist used in the first embodiment, a polymer obtained by copolymerizing the structure (I), the structure (VI), and the structure (VII) at 60/30/10 is used as a base resin. 100 parts by weight of a base resin, 1.5 parts by weight of an acid generator triphenylsulfonium triflate, 0.01 part by weight of 2-benzylpyridine as a quencher were dissolved in 1000 parts by weight of 1-methoxy-2-propanol, and the pore size was 0%. Filtered using a .2 micron fluororesin filter. As a result, a wiring groove can be formed in the same manner as in the first embodiment, and based on this, a wiring having high electrical reliability and having no problem such as variation in electric resistance or disconnection can be formed with high dimensional accuracy.
[0029]
Embedded image

[0030]
Embedded image

[0031]
Embedded image

[0032]
(Embodiment 6)
In the fifth embodiment, instead of the negative resist used in the second embodiment, a polymer obtained by copolymerizing the structure (I) and the structure (VIII) in a ratio of 25/75 is used as a base resin, and 100 parts by weight of the base resin is used. A fluororesin filter having a pore size of 0.2 micron is prepared by dissolving 1.0 part by weight of triphenylsulfonium triflate and 0.01 part by weight of 2-benzylpyridine as a quencher in 1000 parts by weight of 1-methoxy-2-propanol. And filtered. By using a 0.05 wt% aqueous solution of tetramethylammonium hydroxide for the developer, wiring grooves can be formed in the same manner as in the first embodiment, and based on this, electrical reliability free from problems such as variations in electrical resistance and disconnection. Wiring with high dimensional accuracy could be formed.
[0033]
Embedded image

[0034]
Embedded image

[0035]
【The invention's effect】
According to the present invention, a fine copper wiring can be formed without forming an inner crown in a wiring groove and without a resist remaining in a wiring hole. For this reason, it is possible to manufacture a semiconductor device having a highly reliable fine copper wiring circuit without increasing the wiring resistance or disconnection.
[Brief description of the drawings]
FIG. 1 is a process diagram showing a copper wiring forming process according to a first embodiment of the present invention using a cross-sectional view of a semiconductor device.
FIG. 2 is a process diagram showing a conventional copper wiring forming process using a cross-sectional view of a semiconductor device.
FIG. 3 is a process diagram showing a copper wiring forming process according to a second embodiment of the present invention using a cross-sectional view of a semiconductor device.
FIG. 4 is a process diagram showing a process of forming a copper wiring according to a second embodiment of the present invention, following FIG. 3;
FIG. 5 is a process drawing showing a copper wiring forming process according to a second embodiment of the present invention, which is subsequent to FIG. 4;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 ... board | substrate, 101 ... wiring, 102 ... barrier film, 103 ... interlayer film, 104 ... wiring hole pattern, 105 ... positive resist, 106 ... wiring hole, 107 ... positive resist, 107 '... wiring groove pattern, 108 ... mask, 109: ArF excimer laser exposure light, 110: wiring groove, 111: copper wiring, 120: positive resist, 301: wiring, 302: insulating film, 303: stopper film, 304: first interlayer film, 305: intermediate film, 306: second interlayer film, 307: cap film, 307 ': cap film, 308: wiring hole pattern, 309: positive resist pattern, 309': positive resist, 310: opening, 311: copolymer negative resist, 313 ... Mask, 314: ArF excimer laser exposure light, 315: Wiring groove, 316: Negative resist pattern, 317: Wiring groove cap Turn, 318 ... wiring trench pattern a second interlayer films formed of, 319 ... opening, 320 ... copper, 320 '... copper wiring.

Claims (8)

Forming a through-hole through the film to be processed;
Forming a photosensitive thin film on the film to be processed,
A step of patterning the photosensitive thin film by exposing and developing a desired pattern on the photosensitive thin film,
Etching the film to be processed using the photosensitive thin film as a mask,
Forming a groove in a predetermined region of the film to be processed including the through hole,
The method of manufacturing a semiconductor device, wherein the photosensitive thin film is made of a resist material that undergoes a lactonization reaction by irradiation with the exposure light and becomes negative. Providing an interlayer film on the semiconductor substrate;
Forming a conductive hole for wiring in the interlayer film;
On the photosensitive thin film formed on the interlayer film, a step of irradiating exposure light via a photomask to transfer a wiring groove pattern,
Forming a wiring groove in a predetermined region of the interlayer film using the photosensitive thin film to which the wiring groove pattern has been transferred as a processing mask;
Embedding a metal material in the wiring conductive hole and the wiring groove,
The method of manufacturing a semiconductor device, wherein the photosensitive thin film is made of a resist material that undergoes a lactonization reaction by irradiation with the exposure light and becomes negative. 3. The method according to claim 1, wherein an ArF excimer laser beam is used as the exposure light. 3. The method according to claim 1, wherein a tetramethylammonium hydroxide aqueous solution having a concentration of 0.2 to 0.5 wt% is used for developing the resist. 3. The method according to claim 2, wherein the interlayer film is made of a material having a value of an imaginary part of a refractive index with respect to the exposure light of 0.3 or more during the exposure. 3. The method according to claim 2, wherein the interlayer film is a film including a SiOC film. 7. The method according to claim 1, wherein a base resin of the resist is a copolymer containing the following chemical formula (I).

7. The semiconductor device according to claim 1, wherein the base resin of the resist is a copolymer containing the following chemical formula (I), and is obtained by copolymerizing a monomer having a hydroxyl group. Manufacturing method.

Images