JP2004096063A - Phase shifting mask for extremely short ultraviolet light, its manufacturing method, and method of manufacturing semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a phase shifting mask for extremely short ultraviolet light that can be constituted actually even when the mask reflects extremely short ultraviolet light by obtaining an appropriate combination of a refractive index and an absorption coefficient. <P>SOLUTION: This phase shifting mask 10 is provided with a multilayered reflecting film substrate 11 which reflects extremely short ultraviolet light and first and second regions 12a and 12b formed on the substrate 11. At the time of constituting the mask 10, the phase and reflectivity of the reflected light of the extremely short ultraviolet light obtained from an arbitrary complex index of refraction to the ultraviolet light and an arbitrary thickness of a film are specified with respect to the complex index of refraction and thickness. Then the thickness and complex index of refraction of each film formed in the first and second regions 12a and 12b are set based on the specified results, so that a prescribed phase difference may be produced between the reflected light of the ultraviolet light in the first region 12a and the reflected light of the ultraviolet light in the second region 12b. <P>COPYRIGHT: (C)2004,JPO

Description

【００２４】
この（１）式において、θは、マスクに入射する光がマスク面上法線に対してなす角度である。λは露光中心波長である。φ_１およびφ_２は、例えば”Ｙａｍａｍｏｔｏ　ａｎｄ　Ｔ．Ｎａｍｉｏｋａ”　”Ｌａｙｅｒ−ｂｙ−ｌａｙｅｒ　ｄｅｓｉｇｎ　ｍｅｔｈｏｄ　ｆｏｒ　ｓｏｆｔ−ｘ−ｒａｙ　ｍｕｌｔｉｌａｙｅｒｓ”，Ａｐｐｌｉｅｄ　Ｏｐｔｉｃｓ，Ｖｏｌ．３１　ｐｐ１６２２−１６３０，（１９９２）で開示されている方法で求めればよい。
【００２５】
また、（１）式を用いて位相差ψ（λ）を求めるためには、第１領域１２ａと第２領域１２ｂの形成材料の複素屈折率が必要である。極短紫外光の露光中心波長が１３．５ｎｍである場合において、形成材料の複素屈折率は、例えばＭｏ：０．９２１０８−０．００６４３５４３ｉ、Ｓｉ：０．９９９３−０．００１８２６４５ｉ、Ｒｕ：０．８８７４９−０．０１７４７２１ｉ、ＴａＮ：０．９４１３６−０．０３１５７３８ｉとなる。形成材料が任意のｍ層の積層構造で構成されている場合には、以下に示す（２）式および（３）式によって得られる合成複素屈折率を用いればよい。
【００２６】
【数２】

【００２７】
【数３】

【００２８】
このようにして特定される位相差ψが第１領域１２ａと第２領域１２ｂとで所定値（例えば１８０°）となるように、位相シフトマスク１０，１０′では、第１領域１２ａおよび第２領域１２ｂにおける形成膜（バッファ膜１３＋吸収膜１４）の構成材料（特に複素屈折率）およびその膜厚が設定されているのである。つまり、本実施形態で説明する位相シフトマスク１０，１０′は、第１領域１２ａの形成膜および第２領域１２ｂの形成膜における各膜厚および各複素屈折率が、第１領域１２ａでの極短紫外光の反射光と第２領域１２ｂでの極短紫外光の反射光とで所定の位相差が生じるように設定されている点に、大きな特徴がある。
【００２９】
〔位相シフトマスクの製造手順の説明〕
次に、以上のような特徴を持つ位相シフトマスク１０，１０′について、その製造手順を説明する。ただし、ここでは、その製造手順を第１の実施の形態と第２の実施の形態とに分けて説明する。
【００３０】
〔第１の実施の形態〕
第１の実施の形態では、本発明を適用してハーフトーン型位相シフトマスクを構成する場合を例に挙げて説明する。図２は、第１の実施の形態における位相シフトマスクの製造手順を示すフローチャートである。
【００３１】
ハーフトーン型位相シフトマスクの製造にあたっては、先ず、各複素屈折率に対する等位相線および等反射率線を求める（ステップ１０１、以下ステップを「Ｓ」と略す）。これら等位相線および等反射率線は、既存の材料に限定することなく、任意の複素屈折率に対して求めておく。つまり、極短紫外光に対する任意の複素屈折率について、その複素屈折率から得られる極短紫外光の反射光の位相および反射率を特定する。なお、複素屈折率に対する位相および反射率は、理論的に一意的に特定することができる。また、任意の複素屈折率には、反射多層膜基板１１の複素屈折率も含むものとする。すなわち、等位相線および等反射率線は、反射多層膜基板１１に対しても求めておく。
【００３２】
図３は、等位相線の一具体例を示す説明図である。図例の等位相線は、複素屈折率の虚部（ｋ）を０．０１００ｉに固定して求めている。また、図４は、等反射率線の一具体例を示す説明図である。図例の等反射率線は、複素屈折率の実部（ｎ）を０．９１００に固定して求めている。なお、これらの等位相線および等反射率線において、（１）式のλで表される露光波長は１３．５ｎｍであり、θで表されるマスク上への斜め入射光の入射角度は４．８４°である。
【００３３】
反射率および位相は複素屈折率の実部および複素屈折率の虚部それぞれに依存するが、位相は主に複素屈折率の実部に依存し、反射率は主に複素屈折率の虚部に依存する。したがって、最終的に第１領域１２ａと第２領域１２ｂとで１８０°の位相差および所望の反射率を得るためには、近似的にこれら図３の等位相線および図４の等反射率線を用いて、各領域１２ａ，１２ｂの構成材料の複素屈折率または膜厚を設定すればよい。あるいは、複素屈折率および膜厚の両方を設定するようにしてもよい。
【００３４】
複素屈折率または膜厚の設定にあたっては、先ず、反射多層膜基板１１上の第１領域１２ａおよび第２領域１２ｂに加工可能な材料および膜構成を求める（Ｓ１０２）。そして、第１領域１２ａの材料構成における複素屈折率と、第２領域１２ｂの材料構成における複素屈折率とを、それぞれ求める（Ｓ１０３，Ｓ１０４）。
【００３５】
ここで、例えば図１（ｂ）に示した構成のハーフトーン型位相シフトマスク１０′、すなわち第２領域１２ｂにのみバッファ膜１３を介して吸収膜１４が形成され、第１領域１２ａと第２領域１２ｂとで１８０°の位相差が生じる位相シフトマスクを構成する場合を考える。このとき、第１領域１２ａと第２領域１２ｂとでは、それぞれ所望の反射率（互いに異なる値）が得られるものとする。また、第２領域１２ｂでは、バッファ膜１３の構成材料としてＲｕを、吸収膜１４の構成材料としてＴａＮを、それぞれ選択しているものとする。
【００３６】
図５は、Ｒｕ膜厚とＴａＮ膜厚に対する複素屈折率の実部の分布を示す説明図である。図例からは、Ｒｕ膜厚が１ｎｍから２０ｎｍおよびＴａＮ膜厚が１ｎｍから５０ｎｍの間において、合成複素屈折率の実部が０．８９０から０．９４５の間の値を取り得ることがわかる。つまり、図５の内容から、第２領域１２ｂの材料構成における複素屈折率がわかることになる。なお、第１領域１２ａについては、バッファ膜１３および吸収膜１４が形成されていないので、反射多層膜基板１１の積層構造から複素屈折率を求めればよい。
【００３７】
それぞれの複素屈折率を求めた後は、その第１領域１２ａにおける屈折率および第２領域１２ｂにおける屈折率と、既に求めている等位相線とから、第１領域１２ａと第２領域１２ｂとの形成膜の段差、すなわち第２領域１２ｂの形成膜の膜厚を求める（Ｓ１０５）。具体的には、図５の複素屈折率分布および図３の等位相線図を基にすれば、反射多層膜基板１１との位相差が１８０°となるための形成膜（バッファ膜１３＋吸収膜１４）のトータル膜厚と合成複素屈折率の実部（ｎ）との関係が、図６に示すように一意的に求まる。また、反射多層膜基板１１との位相差が１８０°となるためのＲｕ層（バッファ膜１３）とＴａＮ層（吸収膜１４）のそれぞれの膜厚は、合成複素屈折率が位相差１８０°を与える条件から、図７に示すように一意的に求まる。これらのようにして、第２領域１２ｂの形成膜の膜厚、すなわちＲｕ層およびＴａＮ層の膜厚を求めればよい。
【００３８】
Ｒｕ層およびＴａＮ層のそれぞれの膜厚が求まると、次いで、第１領域１２ａの形成膜と第２領域１２ｂにおける極短紫外光の反射率を求める（Ｓ１０６）。反射率は、Ｒｕ層およびＴａＮ層の各膜厚からトータル膜厚に対する合成複素屈折率の虚部（ｋ）を求め、その合成複素屈折率の虚部から反射率を求めればよい。図８は、Ｒｕ層およびＴａＮ層の各膜厚と、トータル膜厚に対する合成複素屈折率の虚部（ｋ）との関係を示す説明図である。また、複素屈折率の虚部（ｋ）がわかれば、反射率は、図４に示した等反射率線から一意的に求まる。図６には、合成複素屈折率虚部（ｋ）および反射率とトータル膜厚との関係をも併せて示している。
【００３９】
ここで、例えば、反射率を０．０７５に設定する場合は、図６の内容からトータル膜厚４３ｎｍであり、このとき図７の内容からＲｕ層の膜厚１４ｎｍおよびＴａＮ層の膜厚２９ｎｍが最も近い条件として得られる。この膜厚において、以下に述べるように、さらに合成複素屈折率の正確な値を用いて位相差および反射率を求めることで、位相差１８２．４°および反射率０．０７５のハーフトーン型位相シフトが得られるようになる。
【００４０】
ただし、図６の内容によれば、反射率を０．０７５に設定する場合の膜厚条件は、トータル膜厚が４６．３ｎｍおよび４８．３ｎｍ近傍にも存在していることがわかる。つまり、トータル膜厚は、複素屈折率から単一の条件として定まるものではないので、以下に述べる膜厚調整を含めて様々な条件を総合的に勘案してように定めるようにすればよい。このとき、マスク製造における膜厚ばらつきに対する複素屈折率に対するプロセス余裕度を大きく確保するためには、図中においてトータル膜厚の変化が平坦な部分（例えば５０．５ｎｍ付近）に設定することが望ましい。このようにすれば、±１．５ｎｍ程度の範囲でトータル膜厚がばらついても、複素屈折率は変化しないので、位相差および反射率も変化しないからである。
【００４１】
その後は、第１領域１２ａと第２領域１２ｂとで１８０°の位相および所望の反射率が得られるように、さらに膜厚の調整を行う（Ｓ１０７，Ｓ１０８）。ここで膜厚の調整を行う更なる理由は、図６で考慮していない膜内多重干渉効果も含めて所望の位相および反射率を得るためである。具体的には、Ｒｕ層の膜厚とＴａＮ層の膜厚をそれぞれ適宜変化させた場合の各位相差と各ハーフトーン反射率を、上述したような手順で求める。図９は、Ｒｕ層の膜厚とＴａＮ層の膜厚に対する位相差およびハーフトーン反射率をマトリクス状に配した例を示す説明図である。ここで、ハーフトーン反射率とは、図１（ｂ）の位相シフトマスク１０′において、第１領域１２ａの反射率と第２領域１２ｂの反射率との差である。そして、これらの結果から、所望の位相差とハーフトーン反射率を満たすウインドウ、すなわち調整後の膜厚の設定値を選択すればよい。なお、図例では、位相差１８０．０±６°およびハーフトーン反射率９．０±１％、および位相差１８０．０±６°およびハーフトーン反射率５．０±１％の場合を例示している（図中のシャドウ部参照）。これにより、例えば、Ｒｕ層の膜厚１３ｎｍおよびＴａＮ層の膜厚３０ｎｍと設定すれば、位相差１７９．４°およびハーフトーン反射率９．５％のハーフトーン型位相シフトが得られるようになる。ここで、反射率は、０．０７０で図６で予測されたものと略一致する。図１０および図１１は、Ｒｕ層の膜厚とＣｒ層の膜厚に対する位相差およびハーフトーン反射率をマトリクス状に配した別の例である。これにより、例えば、Ｒｕ層の膜厚９ｎｍおよびＣｒ層の膜厚を３４ｎｍと設定すれば、位相差１７９．２°およびハーフトーン反射率４．１％のハーフトーン型位相シフトマスクが得られるようになる。
【００４２】
このようにして第２領域１２ｂの形成膜の複素屈折率および膜厚を設定した後には、その設定に従いつつ反射多層膜基板１１上に当該形成膜を成膜することで、ハーフトーン型位相シフトマスクを構成すればよい。なお、形成膜の成膜については、周知技術を用いて行えばよいため、ここではその説明を省略する。
【００４３】
つまり、反射多層膜基板１１上に形成する第２領域１２ｂを構成する材料群の膜厚および複素屈折率の実部から当該第２領域１２ｂにおける反射多層膜基板１１との位相差を求め、また第２領域１２ｂを構成する材料群の膜厚および複素屈折率の虚部から当該第２領域１２ｂにおける反射率を求め、これらから第１領域１２ａ（反射多層膜基板１１）と第２領域１２ｂにおける位相差が１８０°異なり、かつ、第２領域１２ｂの反射率が所望の値である構成、すなわち上述した構成のハーフトーン型位相シフトマスクを得るのである。ただし、位相差と反射率は、どちらを先に求めるようにしても構わない。
【００４４】
このような手順で得られるハーフトーン型位相シフトマスクを用いた場合の光強度分布について説明する。図１２は、光学条件がＮＡ＝０．２５、σ＝０．７０である場合に、マスク上３０ｎｍ（ウエハ上座標で表示、４倍マスクでは１２０ｎｍ）の開口部を具備するホールパターンにおける光強度分布を示す説明図である。図例では、比較のため、従来のバイナリマスクに対する光強度分布も併せて示している。図例によれば、ハーフトーン型位相シフトマスクを用いた場合には、従来のバイナリマスクに比べてパターンエッジコントラストが向上する効果が得られることが明らかである。
【００４５】
なお、ここでは、ハーフトーン型位相シフトマスクとして、図１（ｂ）に示した構成の位相シフトマスク１０′、すなわち第２領域１２ｂにのみバッファ膜１３および吸収膜１４が形成されたものを例に挙げたが、例えば図１（ａ）に示すようなハーフトーン型位相シフトマスク１０を構成することも可能である。ただし、その場合であっても、上述したように、第１領域１２ａと第２領域１２ｂとでは、１８０°の位相差が生じるように、それぞれにおける形成膜の膜厚および複素屈折率が設定されているものとする。
【００４６】
図１３は、ハーフトーン型位相シフトマスクの構成例の断面構造を示す模式図である。図例のマスクでは、反射多層膜基板１１上にて、第１領域１２ａにはＲｕ１０ｎｍ、Ｓｉ４７ｎｍが順に積層され、第２領域１２ｂにはＲｕ５ｎｍ、ＴａＮ４７ｎｍ、Ｒｕ５ｎｍが順に積層されている。第１領域１２ａの形成膜の構成材料にＳｉを用いたのは、以下に述べる理由による。Ｓｉの複素屈折率は０．９９９３２−０．００１８２６４５ｉであり、その実部は真空中の屈折率である１に極めて近く、またその虚部は他材料と比較して小さい。したがって、Ｓｉ材料には、その膜内多重干渉効果を利用することにより、位相差および反射率比を調整する役割を担わせることができるからである。膜内多重干渉効果を利用することによって、第１領域１２ａと第２領域１２ｂとで１８０°の位相差が生じ、しかもその上面が完全に平坦なフラット構造のハーフトーン型位相シフトマスクを構成することが可能となる。
【００４７】
〔第２の実施の形態〕
次に、位相シフトマスクの製造手順の第２の実施の形態について説明する。第２の実施の形態では、本発明を適用してレベンソン型位相シフトマスクを構成する場合を例に挙げて説明する。図１４は、第２の実施の形態における位相シフトマスクの製造手順を示すフローチャートである。
【００４８】
図１４に示すように、レベンソン型位相シフトマスクの製造も、上述した第１の実施の形態におけるハーフトーン型位相シフトマスクの場合（図２参照）と略同様にして行う（Ｓ２０１〜Ｓ２０８）。ただし、レベンソン型位相シフトマスクは、第１領域１２ａと第２領域１２ｂにおける位相が１８０°異なるだけではなく、第１領域１２ａと第２領域１２ｂにおける反射率が略等しい点で、ハーフトーン型位相シフトマスクとは異なる。つまり、レベンソン型位相シフトマスクを構成する場合には、▲１▼第１領域１２ａと第２領域１２ｂとで反射率が略等しいこと、▲２▼第１領域１２ａと第２領域１２ｂとの位相差が１８０ーであること、の２条件を満足しなければならない（Ｓ２０８）。
【００４９】
これらの条件を満足するか否かの判定は、次のようにして行えばよい。先ず、第１領域１２ａの反射率をＲ_１、第２領域１２ｂの反射率をＲ_２とした場合に、以下に示す（４）式によって得られる反射率比Ｐを特定する。
【００５０】
【数４】

【００５１】
そして、特定した反射率比Ｐに対して、判定基準１：｜Ｐ｜≦３．０％を適用し、この判定基準１に合致していれば、上記▲１▼の条件を満足していると判定する。
【００５２】
また、上記▲２▼の条件については、第１の実施の形態で説明した（１）式によって得られるψ（λ）に対して、判定基準２：｜ψ（λ）｜≦６°を適用する。そして、この判定基準２に合致していれば、上記▲２▼の条件を満足していると判定する。
【００５３】
このようにして得られるレベンソン型位相シフトマスクとしては、図１（ａ）に示した構成のもの、すなわち第１領域１２ａおよび第２領域１２ｂのいずれにもバッファ膜１３および吸収膜１４が形成された構成のものが考えられる。
【００５４】
ここで、図１（ａ）に示す構成のレベンソン型位相シフトマスクの製造手順について、具体例を挙げてさらに詳しく説明する。レベンソン型位相シフトマスクを構成する場合にも、第１の実施の形態で説明したハーフトーン型位相シフトマスクの場合と同様に、先ず、任意の複素屈折率に対する等位相線および等反射率線を求める。図１５は、等位相線の一具体例を示す説明図である。図例の等位相線は、複素屈折率の虚部（ｋ）を０．０１００ｉに固定して求めている。
【００５５】
ところで、レベンソン型位相シフトマスクでは、第１領域１２ａの形成膜と反射多層膜基板１１との位相差ψ_１（λ）が、以下に示す（５）式によって特定される。
【００５６】
【数５】

【００５７】
また、第２領域１２ｂの形成膜と反射多層膜基板１１との位相差ψ_２（λ）が、以下に示す（６）式によって特定される。
【００５８】
【数６】

【００５９】
したがって、第１領域１２ａと第２領域１２ｂの位相差ψ（λ）は、以下に示す（７）式によって特定されることになる。
【００６０】
【数７】

【００６１】
この（７）式によって特定される関係は、図１５に示したように、等位相線の相対関係で表される。例えば、図中において、複素屈折率の実部が０．９４であるような材料で、第１領域１２ａと第２領域１２ｂの位相差が１８０ーであるためには、第１領域１２ａと第２領域１２ｂの段差は５６ｎｍであれば良いことがわかる（図中［１］参照）。また、複素屈折率の実部が第１領域１２ａにおいて０．９６、第２領域１２ｂにおいて０．９４であるような材料で、第１領域１２ａと第２領域１２ｂの位相差が１８０ーであるためには、第１領域１２ａと第２領域１２ｂの段差は４２ｎｍであれば良いことがわかる（図中［２］参照）。一方、反射率を求める場合は、相対値ではなくトータル膜厚に対する絶対値をそのまま用いる。
【００６２】
レベンソン型位相シフトマスクを構成する場合にも、このような等位相線（図１５参照）および等反射率線（図４参照）を基に、判定基準１および判定基準２を満たすような形成膜の膜厚および複素屈折率を設定すればよい。
【００６３】
つまり、反射多層膜基板１１上に形成する第１領域１２ａを構成する材料群の膜厚および複素屈折率の実部から第１領域１２ａにおける反射多層膜基板との位相差を求め、さらには反射多層膜基板１１上に形成する第２領域１２ｂを構成する材料群の膜厚および複素屈折率の実部から第２領域１２ｂにおける反射多層膜基板との位相差を求める。また、反射多層膜基板１１上に形成する第１領域１２ａを構成する材料群の膜厚および複素屈折率の虚部から第１領域１２ａにおける反射率を求め、さらには反射多層膜基板１１上に形成する第２領域１２ｂを構成する材料群の膜厚および複素屈折率の虚部から第２領域１２ｂにおける反射率を求める。そして、これらの結果から、判定基準１および判定基準２を満たす構成、すなわち第１領域１２ａと第２領域１２ｂにおける位相差が１８０°異なり、かつ、第１領域１２ａと第２領域１２ｂの反射率が略等しい構成を、レベンソン型位相シフトマスクとして得るのである。なお、位相差と反射率は、どちらを先に求めるようにしても構わない。
【００６４】
ところで、判定基準１および判定基準２を同時に満たすためには、異なった材料を適切に積層して第１領域１２ａおよび第２領域１２ｂの形成膜を成膜する必要がある。これは、現実に存在する材料の複素屈折率の制約から、異なった材料を適切に積層しなければ、判定基準１および判定基準２を同時に満たすレベンソン型位相シフトマスクを現実に製造可能な構造として得ることができないからである。
【００６５】
このことから、第１領域１２ａを構成する材料としては、例えばＴａＮ、ＲｕおよびＳｉを用いる。また、第２領域１２ｂを構成する材料としては、例えばＭｏおよびＲｕを用いる。これらの材料を用いるのは、例えばＪｐｎ．Ｊ．Ａｐｐｌ．Ｐｈｙｓ．Ｖｏｌ４０（２００１）ｐｐ６９９８−７００１の”Ａｐｐｒｏａｃｈ　ｔｏ　ｐａｔｔｅｒｎｉｎｇ　ｏｆ　ｅｘｔｒｅｍｅ　ｕｌｔｒａｖｉｏｌｅｔ　ｌｉｔｈｏｇｒａｐｈｙ　ｍａｓｋｓ”で開示されているように、各材料のエッチング選択比を次に述べるように組み合わせにおいて非常に大きく取ることが公知だからである。すなわち、Ｓｉ下地に対するＲｕ層のエッチングをＣｌ_２＋Ｏ_２ガスによるドライエッチングで行う場合、Ｓｉ下地はＲｕエッチングに対しエッチングストッパとして作用する。Ｒｕ下地に対するＴａＮ層のエッチングをＡｒ＋Ｃｌ_２ガスによるドライエッチングで行う場合、Ｒｕ下地はＴａＮ層エッチングに対しエッチングストッパとして作用する。また、Ｒｕ下地に対するＭｏ層のエッチングおよびＳｉ層のエッチングもＡｒ＋Ｃｌ_２ガスによるドライエッチングで行い選択比を大きく取る。なぜならば、Ｒｕの塩化物ＲｕＣｌ_３は比較的安定な物質であり６００℃以上で分解する。一方、Ｓｉの塩化物ＳｉＣｌ_４の沸点は５７．６℃、Ｍｏの塩化物ＭｏＣｌ_５の沸点は２６８℃、Ｔａの塩化物ＴａＣｌ_５の沸点は２４２℃であり、Ｒｕの塩化物に対して真空中でエッチング反応ガスとして除去され易いためである。
【００６６】
図１６は、レベンソン型位相シフトマスクの構成例の断面構造を示す模式図である。図例の構造（以下「構造１」という）は、図１５の等位相線および図４の等反射率線を基に設定された、判定基準１および判定基準２を満たす構成である。第１領域１２ａでは、反射多層膜基板１１上にＲｕ２ｎｍ、ＴａＮ７ｎｍ、Ｒｕ４ｎｍの順に積層されて形成膜を構成している。第１領域１２ａのトータル膜厚は１３ｎｍ、合成複素屈折率の実部は０．９１６５であり、虚部は０．０２５０７ｉである。また、第２領域１２ｂでは、Ｒｕ４ｎｍ、Ｍｏ４９ｎｍ、Ｒｕ２ｎｍの順に積層されて形成膜を構成している。第２領域１２ｂのトータル膜厚は５５ｎｍ、合成複素屈折率の実部は０．９１７４であり、虚部は０．００７６４ｉである。なお、第１領域１２ａと第２領域１２ｂとの境界部分には、膜厚１２０ｎｍのＴａＮ吸収層が４０ｎｍ幅で形成されている。
【００６７】
このような構造１において、図１５の等位相線から第１領域１２ａと第２領域１２ｂの段差を求めると、９０°と２７０°の等位相線の関係から当該段差は４３ｎｍであり、同様に１８０°と０°の等位相線の関係から求めても４３ｎｍである。また、図４の等透過率線によれば、第１領域１２ａの反射率は０．３９であり、第２領域１２ｂの反射率は０．３８であることがわかる。そして、構造１について、詳細に膜内多重干渉効果も含めて第２領域１２ｂのＭｏ膜厚を変化させて位相差と反射率比を求めると、例えば図１７に示す結果が得られる。図例の結果から、図１６に示した構造１は、各材料の膜厚調整範囲内において最適構成であることが確認できる。
【００６８】
また、構造１において、反射率は、第１領域１２ａに対して０．３８８であり、第２領域１２ｂに対して０．３８７であり、これらの間の反射率比は０．２５８％となる。また、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ（ｔｒａｎｓｖｅｒｓｅ　ｅｌｅｃｔｒｉｃ）波に対して１７８．８°、ＴＭ（ｔｒａｎｓｖｅｒｓｅ　ｍａｇｎｅｔｉｃ）波に対して１７８．７°となる。
【００６９】
このような構造１のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造１の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図１８に示すような結果となる。
【００７０】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、以下に示す（８）式の位相差を求めることで確認できる。なお、（８）式では、ウエハ上における座標を用いている。ｘ軸の単位はｎｍである。
【００７１】
【数８】

【００７２】
この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図１９に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【００７３】
これらのことから、構造１のレベンソン型位相シフトマスクは、判定基準１および判定基準２を同時に満たしており、第１領域１２ａと第２領域１２ｂとで反射率が略等しく、その位相差が１８０°であると言える。
【００７４】
次いで、このように複素屈折率および膜厚が設定された第１領域１２ａの形成膜および第２領域１２ｂの形成膜を、反射多層膜基板１１上に成膜する場合の手順について簡単に説明する。図２０〜２３は、構造１の成膜手順の一例を示す説明図である。構造１による形成膜を成膜する場合には、図２０に示すように、先ず、反射多層膜基板１１上にＲｕ層をスパッタ法により成膜する（工程１）。Ｒｕ層は、通常、バイナリマスクでバッファ層として用いられる材料であるので、その通常の場合と同じ製造装置を用いることができる。そして、Ｒｕ層上に、ＴａＮ層をスパッタ法により成膜する（工程２）。ＴａＮ層は、バイナリマスクで吸収層として用いられる材料であるので、その通常の場合と同じ製造装置を用いることができる。
【００７５】
その後は、ＴａＮ層上にレジストを塗布し（工程３）、描画およびレジスト現像工程を経て、第２領域１２ｂの部分のレジストを除去する（工程４）。レジスト除去後は、第２領域１２ｂの部分のＴａＮ層をＡｒ＋Ｃｌ_２ガスによるドライエッチングによって除去する（工程５）。このＴａＮ層の下のＲｕ層はエッチングストッパ層として機能することになる。そして、レジストを剥離し（工程６）、再びＲｕ層をスパッタ成膜した後に（工程７）、今度はＭｏ層をスパッタ成膜する（工程８）。Ｍｏ層は、反射多層膜基板１１の構成材料であるので、多層膜用製造装置を用いることができる。
【００７６】
Ｍｏ層を成膜すると、続いて、図２１に示すように、レジストを塗布した後に（工程９）、描画およびレジスト現像工程を経て、第１領域１２ａの部分のレジストを除去する（工程１０）。そして、第１領域１２ａの部分のＭｏ層をＡｒ＋Ｃｌ_２ガスによるドライエッチングにより除去する（工程１１）。このＭｏ層の下のＲｕ層はエッチングストッパ層として機能することになる。その後は、レジストを剥離し（工程１２）、再びＲｕ層をスパッタ成膜する（工程１３）。
【００７７】
Ｒｕ層を成膜した後は、次いで、図２２に示すように、ＴａＮ吸収層をスパッタ成膜し（工程１４）、さらにレジストを塗布する（工程１５）。そして、描画およびレジスト現像工程を経て、吸収層として残す部分以外のレジストを除去する（工程１６）。
【００７８】
その後、図２３に示すように、ＴａＮ吸収層をＡｒ＋Ｃｌ_２ガスによるドライエッチングにより除去する（工程１７）。このＴａＮ吸収層の下のＲｕ層はエッチングストッパ層として機能することになる。そして、レジストを剥離すれば（工程１８）、Ｒｕ層をエッチングストッパ層として有効に用いつつ、構造１のレベンソン型位相シフトマスクを構成することができる。
【００７９】
図２４は、レベンソン型位相シフトマスクの他の構成例の断面構造を示す模式図である。図例の構造（以下「構造２」という）も、上述した構造１と同様に、図１５の等位相線および図４の等反射率線を基に設定された、判定基準１および判定基準２を満たす構成である。第１領域１２ａでは、反射多層膜基板１１上にＲｕ３ｎｍ、ＴａＮ５ｎｍ、Ｒｕ７ｎｍの順に積層されて形成膜を構成している。第１領域１２ａのトータル膜厚は１５ｎｍ、合成複素屈折率の実部は０．９０５４であり、虚部は０．０２２１７ｉである。また、第２領域１２ｂでは、Ｒｕ３ｎｍ、Ｍｏ４９ｎｍ、Ｒｕ４ｎｍの順に積層されて形成膜を構成している。第２領域１２ｂのトータル膜厚は５６ｎｍ、合成複素屈折率の実部は０．９１６９であり、虚部は０．００７８２ｉである。
【００８０】
このような構造２において、図１５の等位相線から第１領域１２ａと第２領域１２ｂの段差を求めると、９０°と２７０°の等位相線の関係から当該段差は４４ｎｍであり、同様に１８０°と０°の等位相線の関係から求めても４４ｎｍである。また、図４の等透過率線によれば、第１領域１２ａの反射率は０．３８であり、第２領域１２ｂの反射率は０．３６であることがわかる。そして、構造２について、詳細に膜内多重干渉効果も含めて第２領域１２ｂのＭｏ膜厚を変化させて位相差と反射率比を求めると、例えば図２５に示す結果が得られる。図例の結果から、図２４に示した構造２は、各材料の膜厚調整範囲内において最適構成であることが確認できる。
【００８１】
また、構造２において、反射率は、第１領域１２ａに対して０．３９９であり、第２領域１２ｂに対して０．３９６であり、これらの間の反射率比は０．７１０％となる。したがって、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ波に対して１７８．２°、ＴＭ波に対して１７８．３°となる。
【００８２】
このような構造２のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造２の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図２６に示すような結果となる。
【００８３】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、上述した（８）式の位相差を求めることで確認できる。この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図２７に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【００８４】
このように複素屈折率および膜厚が設定された構造２の成膜手順については、上述した構造１の場合と略同様である。構造１の場合とは、工程５と工程６との間に、第２領域１２ｂの部分のＲｕ層をＣｌ_２＋Ｏ_２ガスによるドライエッチングにより除去する工程が追加される点で異なる。
【００８５】
図２８は、レベンソン型位相シフトマスクのさらに他の構成例の断面構造を示す模式図である。図例の構造（以下「構造３」という）も、上述した構造１および構造２と同様に、図１３の等位相線および図４の等反射率線を基に設定された、判定基準１および判定基準２を満たす構成である。第１領域１２ａでは、反射多層膜基板１１上にＲｕ５ｎｍ、ＴａＮ２０ｎｍ、Ｓｉ８ｎｍ、Ｒｕ５ｎｍの順に積層されて形成膜を構成している。また、第２領域１２ｂでは、Ｒｕ５ｎｍ、Ｓｉ８ｎｍ、Ｒｕ４３．５ｎｍの順に積層されて形成膜を構成している。
【００８６】
この構造３では、構造１または構造２の場合とは異なり、第１領域１２ａおよび第２領域１２ｂの形成膜の構成材料にＳｉを用いている。ここで、Ｓｉの複素屈折率は、０．９９９３２−０．００１８２６４５ｉであり、複素屈折率の実部は真空中の屈折率である１に極めて近く、また複素屈折率の虚部は他材料と比較して小さい。そのため、Ｓｉ材料は、膜内多重干渉効果の利用により、位相差および反射率比を調整する役割を担うことになる。
【００８７】
したがって、図１５の等位相線および図４の等反射率線から判定基準１および判定基準２を満たす構成を得る場合には、Ｓｉ層を考慮せずに、位相差を重点的に合わせ込む条件を抽出し、その後にＳｉ層の膜内多重干渉により反射率を合わせ込むとともに、位相差を１８０°近くに調整する手順を採る、といったことが可能となる。すなわち、判定基準１および判定基準２を満たす構成は、第１領域１２ａに対しては反射多層膜基板１１上にＲｕ５ｎｍ、ＴａＮ２０ｎｍ、Ｒｕ５ｎｍの順に積層したものと考え、第２領域１２ｂに対してはＲｕ５ｎｍ、Ｒｕ４３．５ｎｍの順に積層したものと考える。この場合の第１領域１２ａのトータル膜厚は３０ｎｍ、合成複素屈折率の実部は０．９２３４であり、虚部は０．０２６８７ｉである。第２領域１２ｂのトータル膜厚は４８．５ｎｍ、合成複素屈折率の実部は０．８８７５であり、虚部は０．０１７４７ｉである。
【００８８】
このような構造３において、図１５の等位相線から第１領域１２ａと第２領域１２ｂの段差を求めると、９０°と２７０°の等位相線の関係から当該段差は２２ｎｍであり、同様に１８０°と０°の等位相線の関係から求めると１６ｎｍである。また、図４の等透過率線によれば、第１領域１２ａの反射率は０．１４であり、第２領域１２ｂの反射率は０．１８であることがわかる。そして、構造３について、詳細に膜内多重干渉効果も含めて第２領域１２ｂのＭｏ膜厚を変化させて位相差と反射率比を求めると、例えば図２９に示す結果が得られる。図例の結果から、図２８に示した構造３は、各材料の膜厚調整範囲内において最適構成であることが確認できる。この構造３では、第１領域１２ａおよび第２領域１２ｂの形成膜の構成材料にＳｉを用いており、そのＳｉ層による膜内多重干渉を効果的に利用していることから、位相差と反射率比が基準値を満たすことになる条件を実現することができるのである。
【００８９】
また、構造３において、反射率は、第１領域１２ａに対して０．１９５であり、第２領域１２ｂに対して０．２００であり、これらの間の反射率比は２．４１％となる。したがって、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ波に対して１８５．３°、ＴＭ波に対して１８５．１°となる。
【００９０】
このような構造３のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造３の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図３０に示すような結果となる。
【００９１】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、上述した（８）式の位相差を求めることで確認できる。この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図３１に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【００９２】
このように複素屈折率および膜厚が設定された構造３の成膜手順についても、上述した構造２の場合と略同様である。構造２の場合とは、工程８にて、Ｍｏ層に代わってＳｉ層およびＲｕ層をスパッタ成膜する点で異なる。
【００９３】
ところで、構造１〜３では、いずれも、第１領域１２ａの形成膜と第２領域１２ｂの形成膜とに膜厚差があり、それぞれの間に段差が生じている。このような構成であっても、上述したように位相シフトの効果を得ることが可能である。ただし、第１領域１２ａと第２領域１２ｂとの境界部分にＴａＮ吸収層を形成することを考慮すると、その加工の容易さの点で、第１領域１２ａと第２領域１２ｂとの間に段差がなく、平坦に形成されているほうが望ましい。そこで、以下に、第１領域１２ａと第２領域１２ｂとが平坦である構成の具体例について説明する。
【００９４】
図３２は、レベンソン型位相シフトマスクの平坦な構成例の断面構造を示す模式図である。図例の構造（以下「構造４」という）は、図１５の等位相線および図４の等反射率線を基に設定された、判定基準１および判定基準２を満たす構成である。第１領域１２ａでは、反射多層膜基板１１上にＲｕ３ｎｍ、ＴａＮ７ｎｍ、Ｒｕ６ｎｍ、Ｓｉ３７ｎｍ、Ｒｕ５ｎｍの順に積層されて形成膜を構成している。また、第２領域１２ｂでは、Ｒｕ３ｎｍ、Ｍｏ４７ｎｍ、Ｒｕ８ｎｍの順に積層されて形成膜を構成している。
【００９５】
この構造４において、第１領域１２ａの形成膜の構成材料にＳｉを用いているのは、構造３の場合と同様に、膜内多重干渉効果の利用により位相差および反射率比を調整する役割を担わせるためである。したがって、図１５の等位相線および図４の等反射率線から判定基準１および判定基準２を満たす構成を得る場合には、Ｓｉ層を考慮せずに、位相差を重点的に合わせ込む条件を抽出し、その後にＳｉ層の膜内多重干渉により反射率を合わせ込むとともに、位相差を・・・・近くに調整する手順を採る、といったことが可能となる。すなわち、判定基準１および判定基準２を満たす構成は、第１領域１２ａに対しては反射多層膜基板１１上にＲｕ３ｎｍ、ＴａＮ７ｎｍ、Ｒｕ６ｎｍ、Ｒｕ５ｎｍの順に積層したものと考え、第２領域１２ｂに対してはＲｕ３ｎｍ、Ｍｏ４７ｎｍ、Ｒｕ８ｎｍの順に積層したものと考える。この場合の第１領域１２ａのトータル膜厚は２１ｎｍ、合成複素屈折率の実部は０．９０５４であり、虚部は０．０３６３１ｉである。第２領域１２ｂのトータル膜厚は５８ｎｍ、合成複素屈折率の実部は０．９１４７であり、虚部は０．０００９０４ｉである。
【００９６】
このような構造４において、図１５の等位相線から第１領域１２ａと第２領域１２ｂの段差を求めると、９０°と２７０°の等位相線の関係から当該段差は４０ｎｍであり、同様に１８０°と０°の等位相線の関係から求めると４３ｎｍである。また、図４の等透過率線によれば、第１領域１２ａの反射率は０．１５であり、第２領域１２ｂの反射率は０．３０であることがわかる。
【００９７】
このように、構造４では、Ｓｉ層を考慮しないと、構造１および構造２に類似しているものの、第１領域１２ａと第２領域１２ｂとで反射率が一致しない。これを解消するために、構造４では、Ｓｉ層を挿入し、膜内多重干渉を効果的に利用することで、第１領域１２ａと第２領域１２ｂにおける反射率を一致させているのである。
【００９８】
そのために、構造４においては、第１領域１２ａのＴａＮ層の膜厚、第２領域１２ｂのＭｏ層の膜厚、および第１領域１２ａのＳｉ層の膜厚を適宜変化させ、判定基準１および判定基準２を満たす構成を求める。図３３は、ＴａＮ層の膜厚およびＳｉ層の膜厚を一定としＭｏ層の膜厚をパラメータとして位相差および反射率比を求めた結果である。図例の結果から、図３２に示した構造４は、第１領域１２ａと第２領域１２ｂの膜厚差が０ｎｍとなる条件で判定基準を満たす構成であることが確認できる。この構造４では、第１領域１２ａの形成膜の構成材料にＳｉを用いており、そのＳｉ層による膜内多重干渉を効果的に利用していることから、位相差と反射率比が基準値を満たし、かつ、第１領域１２ａと第２領域１２ｂとが平坦になる構成を実現することができるのである。
【００９９】
また、構造４において、反射率は、第１領域１２ａに対して０．２８５であり、第２領域１２ｂに対して０．２８７であり、これらの間の反射率比は０．６５％となる。したがって、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ波に対して１８０．８°、ＴＭ波に対して１８０．５°となる。
【０１００】
このような構造４のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造４の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図３４に示すような結果となる。
【０１０１】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、上述した（８）式の位相差を求めることで確認できる。この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図３５に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【０１０２】
このように複素屈折率および膜厚が設定された構造４の成膜手順については、上述した構造１の場合（図２０〜２３参照）に加えて、構造２の場合と同様に工程５と工程６との間にて第２領域１２ｂの部分のＲｕ層をＣｌ_２＋Ｏ_２ガスによるドライエッチングにより除去する工程を行うとともに、に工程１３と工程１４との間にて図３６，３７に示すような工程を行えばよい。
【０１０３】
すなわち、図３６に示すように、工程１３にてＲｕ層を成膜した後は、次いで、Ｓｉ層をスパッタ成膜し（工程１３−１）、さらにレジストを塗布する（工程１３−２）。そして、描画およびレジスト現像工程を経て、第２領域１２ｂの部分のレジストを除去する（工程１３−３）。
【０１０４】
その後、図３７に示すように、第２領域１２ｂの部分のＳｉ膜をＡｒ＋Ｃｌ_２ガスによるドライエッチングにより除去する（工程１３−４）。このＳｉ層の下のＲｕ層はエッチングストッパ層として機能することになる。そして、レジストを剥離して（工程１３−５）、Ｒｕ層をスパッタ成膜した後は（工程１３−６）、ＴａＮ吸収層をスパッタ成膜し（工程１４）、それ以降、構造１の場合と同様の処理を行う。このような成膜手順によって、構造４のレベンソン型位相シフトマスクを構成することができる。
【０１０５】
以上のように、構造４については、第１領域１２ａのＴａＮ層の膜厚、第２領域１２ｂのＭｏ層の膜厚、および第１領域１２ａのＳｉ層の膜厚を適宜変化させ、判定基準１および判定基準２を満たし、かつ、第１領域１２ａと第２領域１２ｂとが平坦である構成を特定している。ただし、このような構成は、以下に述べるような構造においても得ることができる。
【０１０６】
図３８は、レベンソン型位相シフトマスクの他の平坦な構成例の断面構造を示す模式図である。図例の構造（以下「構造５」という）も、ＴａＮ層の膜厚およびＳｉ層の膜厚を一定にしてＭｏ層の膜厚をパラメータとして位相差および反射率比を求めた結果である。図３９は、ＴａＮ層の膜厚およびＳｉ層の膜厚を一定としＭｏ層の膜厚をパラメータとして位相差および反射率比を求めた結果である。図例の結果から、構造５についても、第１領域１２ａと第２領域１２ｂの膜厚差が０ｎｍとなる条件で判定基準を満たす構成であることが確認できる。この構造５において、反射率は、第１領域１２ａに対して０．２８８であり、第２領域１２ｂに対して０．２８７であり、これらの間の反射率比は０．３６％である。また、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ波に対して１８１．４°、ＴＭ波に対して１８１．１°である。
【０１０７】
このような構造５のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造５の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図４０に示すような結果となる。
【０１０８】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、上述した（８）式の位相差を求めることで確認できる。この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図４１に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【０１０９】
このような構造５の成膜手順については、上述した構造４の場合と同様である。
【０１１０】
図４２は、レベンソン型位相シフトマスクのさらに他の平坦な構成例の断面構造を示す模式図である。図例の構造（以下「構造６」という）も、ＴａＮ層の膜厚およびＳｉ層の膜厚を一定にしてＭｏ層の膜厚をパラメータとして位相差および反射率比を求めた結果である。図４３は、ＴａＮ層の膜厚およびＳｉ層の膜厚を一定としＭｏ層の膜厚をパラメータとして位相差および反射率比を求めた結果である。図例の結果から、構造６についても、第１領域１２ａと第２領域１２ｂの膜厚差が０ｎｍとなる条件で判定基準を満たす構成であることが確認できる。この構造６において、反射率は、第１領域１２ａに対して０．３００であり、第２領域１２ｂに対して０．２８７であり、これらの間の反射率比は０．９０％である。また、第１領域１２ａと第２領域１２ｂの位相差は、ＴＥ波に対して１８０．６°、ＴＭ波に対して１８０．３°である。
【０１１１】
このような構造６のレベンソン型位相シフトマスクを用いた場合の光強度分布は、構造６の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅（ウエハ上単位では１０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件であると、例えば図４４に示すような結果となる。
【０１１２】
また、第１領域１２ａと第２領域１２ｂの位相差が１８０°であることは、４０ｎｍ幅のＴａＮ吸収層による３２０ｎｍピッチ（ウエハ上単位では１０ｎｍ幅で８０ｎｍピッチ）のパターンを用いて、上述した（８）式の位相差を求めることで確認できる。この（８）式をＴＥｙ波、ＴＭｘ波およびＴＭｚ波に対してマスク直上において求めた結果を図４５に示す。図例によれば、ＴＥｙ波およびＴＭｘ波ともに、１８０°の位相差を良好に保持している。一方、ＴＭｚ波は、１８０°からのずれがより顕著ではあるが、転写に及ぼす寄与は０．４５％程度であるため、ほとんど影響を与えない。
【０１１３】
このような構造６の成膜手順についても、上述した構造４または構造５の場合と同様である。
【０１１４】
次に、以上のようなレベンソン型位相シフトマスクにおける効果を、従来の位相シフト効果を用いないバイナリマスクとの比較において説明する。図４６は、上述した構造５のレベンソン型位相シフトマスクについて、構造５の４倍マスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅、３０ｎｍ幅、２０ｎｍ幅、１０ｎｍ幅、０ｎｍ幅（ウエハ上単位では１０ｎｍ、７．５ｎｍ、５ｎｍ、２．５ｎｍ、０ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件で得られた光強度分布を示す説明図である。図例によれば、いずれのＴａＮ吸収層幅においても、良好なパターンコントラストが得られることがわかる。
【０１１５】
これに対して、図４７は、従来のバイナリマスクについて、そのマスク上に膜厚１２０ｎｍのＴａＮ吸収層を４０ｎｍ幅、３０ｎｍ幅、２０ｎｍ幅、１０ｎｍ幅（ウエハ上単位では１０ｎｍ、７．５ｎｍ、５ｎｍ、２．５ｎｍ）で形成してウエハ上に露光した場合に、ＮＡ＝０．２５、σ＝０．７０という光学条件で得られた光強度分布を示す説明図である。図例によれば、ＴａＮ吸収層幅が小さくなるに伴って、パターンコントラストが著しく低下することがわかる。
【０１１６】
これらのことから、上述した構造のレベンソン型位相シフトマスクを用いた場合には、ＴａＮ吸収層１０ｎｍ幅（ウエハ上単位では２．５ｎｍ）においてウエハ上で１５ｎｍまたはこれ以下のサイズの線幅の転写を行うことが可能となる、といった顕著な効果が得られると言える。さらには、例えば図４８に示すように、ＮＡ＝０．３０の光学条件で露光すれば、ＴａＮ吸収層１０ｎｍ幅（ウエハ上単位では２．５ｎｍ）においてウエハ上で１０ｎｍまたはこれ以下のサイズの線幅の転写を行うことが可能になる、といった顕著な効果も得られるようになる。
【０１１７】
以上のように、本実施形態（第１および第２の実施の形態）にて説明した位相シフトマスクでは、第１領域１２ａおよび第２領域１２ｂの形成膜の膜厚および複素屈折率が、それぞれでの極短紫外光の反射光に所定の位相差が生じるように設定されている。さらに詳しくは、位相シフトマスクを構成するのにあたり、先ず、反射多層膜基板１１上の形成膜（特に、その構成材料における複素屈折率）に依存することなく、任意の複素屈折率および膜厚について当該複素屈折率および当該膜厚から得られる反射光の位相および反射率を特定し、その特定した位相および反射率に基づいて、第１領域１２ａと第２領域１２ｂで１８０°の位相差が生じるように、それぞれにおける形成膜の膜厚および複素屈折率が選定されている。
【０１１８】
したがって、本実施形態で説明した位相シフトマスクによれば、極短紫外光に対応する反射型マスクの場合であっても、超解像技術を用いるための位相シフトマスクを構成することが実現可能となる。すなわち、本実施形態で説明した位相シフトマスクの製造方法を用いれば、極短紫外光の位相シフトマスクを構成することができる。
【０１１９】
しかも、本実施形態で説明した位相シフトマスクでは、単に極短紫外光を用いるだけではなく、超解像技術をも利用することで、ウエハ上のパターンコントラストを著しく増大させ、従来のバイナリマスクでは得られなかった解像性を得ることができる。つまり、本実施形態の位相シフトマスクを用いて半導体装置を製造すれば、従来のバイナリマスクを用いた場合よりもより微細なホールパターン、スペースパターンおよびラインパターンを得ることができ、パターンの極小化への対応に非常に好適となる。
【０１２０】
また、本実施形態で説明した位相シフトマスクでは、膜内多重干渉を効果的に利用することで、第１領域１２ａと第２領域１２ｂとで形成膜の膜厚を略等しくして、これら第１領域１２ａと第２領域１２ｂとが平坦になる構成を実現している。したがって、例えばレベンソン型位相シフトマスクにおいて第１領域１２ａと第２領域１２ｂとの境界部分にＴａＮ吸収層を形成する場合であっても、平坦部分にその形成を行えばよいため、加工が非常に容易なものとなり、またその形成精度の確保も容易となる。
【０１２１】
また、本実施形態で説明した位相シフトマスクでは、第１領域１２ａの形成膜および第２領域１２ｂの形成膜の両方またはいずれか一方が、複数の材料からなる積層構造を有している。したがって、任意の複素屈折率および膜厚について当該複素屈折率および当該膜厚から得られる反射光の位相および反射率を特定しても、その任意の複素屈折率および膜厚に対応する形成膜を得ることが可能となる。つまり、複数の材料からなる積層構造を用いることによって、所望する位相シフトマスクを構成することが可能となるのである。
【０１２２】
【発明の効果】
以上に説明したように、本発明に係る極短紫外光の位相シフトマスクおよびその製造方法によれば、適切な屈折率と吸収係数の組み合わせを得ることで、極短紫外光に対応する反射型マスクの場合であっても、超解像技術を用いるための位相シフトマスクを構成することが実現可能となる。また、本発明に係る半導体装置の製造方法によれば、パターンの極小化への対応に非常に好適となる。
【図面の簡単な説明】
【図１】本発明に係る位相シフトマスクの概略構成の一例を示す模式図である。
【図２】本発明の第１実施形態における位相シフトマスクの製造手順を示すフローチャートである。
【図３】等位相線の一具体例を示す説明図である。
【図４】等反射率線の一具体例を示す説明図である。
【図５】ハーフトーン型位相シフトマスクを得るためのＲｕ膜厚およびＴａＮ膜厚に対する合成複素屈折率の実部の分布を示す説明図である。
【図６】ハーフトーン型位相シフトマスクにおけるトータル膜厚と合成複素屈折率の実部（ｎ）および反射率（ｋ）の関係を示す説明図である。
【図７】ハーフトーン型位相シフトマスクにおけるトータル膜厚とＲｕ膜厚およびＴａＮ膜厚との関係を示す説明図である。
【図８】ハーフトーン型位相シフトマスクを得るためのＲｕ膜厚およびＴａＮ膜厚に対する合成複素屈折率の虚部の分布を示す説明図である。
【図９】Ｒｕ層の膜厚とＴａＮ層の膜厚に対する位相差およびハーフトーン反射率をマトリクス状に配した例を示す説明図である。
【図１０】Ｒｕ層の膜厚とＣｒ層の膜厚に対する位相差をマトリクス状に配した例を示す説明図である。
【図１１】Ｒｕ層の膜厚とＣｒ層の膜厚に対するハーフトーン反射率をマトリクス状に配した例を示す説明図である。
【図１２】ハーフトーン型位相シフトマスクのＮＡ＝０．２５の場合におけるマスク上３０ｎｍ（ウエハ上座標で表示）の開口部を具備するホールパターンの光強度分布を示す説明図である。
【図１３】ハーフトーン型位相シフトマスクの構成例の断面構造を示す模式図である。
【図１４】本発明の第２の実施の形態における位相シフトマスクの製造手順を示すフローチャートである。
【図１５】等位相線の一具体例を示す説明図である。
【図１６】レベンソン型位相シフトマスクの構成例（構造１）の断面構造を示す模式図である。
【図１７】構造１のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図１８】構造１のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図１９】構造１のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図２０】構造１のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その１）である。
【図２１】構造１のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その２）である。
【図２２】構造１のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その３）である。
【図２３】構造１のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その４）である。
【図２４】レベンソン型位相シフトマスクの他の構成例（構造２）の断面構造を示す模式図である。
【図２５】構造２のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図２６】構造２のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図２７】構造２のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図２８】レベンソン型位相シフトマスクのさらに他の構成例（構造３）の断面構造を示す模式図である。
【図２９】構造３のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図３０】構造３のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図３１】構造３のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図３２】レベンソン型位相シフトマスクの第１領域と第２領域とが平坦な構成例（構造４）の断面構造を示す模式図である。
【図３３】構造４のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図３４】構造４のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図３５】構造４のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図３６】構造４のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その３）である。
【図３７】構造４のレベンソン型位相シフトマスクの成膜手順の一例を示す説明図（その４）である。
【図３８】レベンソン型位相シフトマスクの他の平坦な構成例（構造５）の断面構造を示す模式図である。
【図３９】構造５のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図４０】構造５のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図４１】構造５のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図４２】レベンソン型位相シフトマスクのさらに他の平坦な構成例（構造６）の断面構造を示す模式図である。
【図４３】構造６のレベンソン型位相シフトマスクにおいて膜内多重干渉を考慮した位相と反射率の最適化（Ｍｏ膜厚をパラメータとしている）の例を示す説明図である。
【図４４】構造６のレベンソン型位相シフトマスクにおける光強度分布を示す説明図である。
【図４５】構造６のレベンソン型位相シフトマスクの位相差（ウエハ上で表示）を示す説明図である。
【図４６】構造５のレベンソン型位相シフトマスクにおけるマスク上ＴａＮ幅（ウエハ上で表示）に対する光強度分布（ＮＡ＝０．２５）を示す説明図である。
【図４７】従来のバイナリマスクにおけるマスク上ＴａＮ幅（ウエハ上で表示）に対する光強度分布（ＮＡ＝０．２５）を示す説明図である。
【図４８】構造５のレベンソン型位相シフトマスクにおけるマスク上ＴａＮ幅（ウエハ上で表示）に対する光強度分布（ＮＡ＝０．３０）を示す説明図である。
【符号の説明】
１０，１０′…位相シフトマスク、１１…反射多層膜基板、１２ａ…第１領域、１２ｂ…第２領域、１３…バッファ膜、１４…吸収膜[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a phase shift mask used in a lithography process for forming a circuit pattern of a semiconductor device, a method for manufacturing the same, and a method for manufacturing a semiconductor device including the lithography process, and particularly to a so-called ultrashort ultraviolet light. The present invention relates to a phase shift mask, a method for manufacturing the same, and a method for manufacturing a semiconductor device.
[0002]
[Prior art]
In recent years, with the miniaturization of semiconductor devices, circuit patterns formed on wafers and resist patterns for forming the circuit patterns have been minimized in pattern width (line width), pitch between patterns, and the like. Is required. Such a requirement for minimization can be met by making the wavelength of ultraviolet light used for exposure of the resist shorter. For example, a semiconductor device having a design rule of 350 nm has a wavelength of 365 nm, a semiconductor device having a design rule of 250 nm and 180 nm has a wavelength of 248 nm, and a semiconductor device having a design rule of 130 nm and 100 nm has a wavelength of 193 nm. As the miniaturization of GaN has progressed, the wavelength of ultraviolet light used for exposure has been shortened, and further, ultraviolet light having a wavelength of 157 nm has been used.
[0003]
In general, it is known that the resolution based on these wavelengths is represented by a Rayleigh equation of w = k1 × (λ / NA). Here, w is the minimum width pattern to be resolved, NA is the numerical aperture of the lens of the projection optical system, and λ is the wavelength of the exposure light. Further, k1 is a process constant determined mainly by the selection of the performance of the resist and the super-resolution technique, and it is known that up to about k1 = 0.35 can be selected by using the optimum resist and the super-resolution technique. Have been. The super-resolution technique is to obtain a pattern smaller than a wavelength by selectively using ± 1st-order diffracted light of light transmitted through a mask and diffracted by a light-shielding pattern on the mask.
[0004]
According to the Rayleigh's formula, for example, the minimum pattern width that can be used when a wavelength of 157 nm is used is w = 61 nm if a lens with NA = 0.9 is used. That is, in order to obtain a pattern width smaller than 61 nm, it is necessary to use ultraviolet light having a wavelength shorter than 157 nm.
[0005]
For this reason, recently, the use of 13.5 nm wavelength light called EUV (Extreme Ultra Violet) as ultraviolet light having a wavelength shorter than 157 nm has been studied. However, up to an ultraviolet light having a wavelength of 157 nm, for example, CaF ₂ (Calcium fluoride) or SiO ₂ Since there is a light-transmitting material such as (silicon dioxide), a mask and an optical system configured to transmit the ultraviolet light can be manufactured. However, as for ultra-short ultraviolet light having a wavelength of 13.5 nm, there is no material that transmits the ultra-short ultraviolet light with a desired thickness. Therefore, when using ultra-short ultraviolet light having a wavelength of 13.5 nm, a mask and an optical system are configured by a reflective mask and a reflective optical system that reflect light instead of a light-transmitting mask and an optical system. There is a need.
[0006]
When a reflective mask is used, the light reflected on the mask surface must be guided to the projection optical system without interfering with the light incident on the mask. Therefore, the light incident on the mask must necessarily be obliquely incident at an angle φ with respect to the normal to the mask surface. This angle is determined by the numerical aperture NA of the lens of the projection optical system, the mask magnification m, and the size σ of the illumination light source. More specifically, for example, when a mask having a reduction magnification of 5 times is used on a wafer, in an exposure apparatus with NA = 0.3 and σ = 0.8, the light is 3.44 ± 2.75 °. It will be incident on the mask with a solid angle. When a mask having a reduction magnification of 4 times is used on a wafer, in an exposure apparatus with NA = 0.25 and σ = 0.7, light has a solid angle of 3.58 ± 2.51 °. Incident on the mask. In consideration of these solid angles, the incident angle of the exposure light incident on the mask is usually set to be around 5 °. Here, the incident angle is defined as an angle formed with a normal to the mask surface.
[0007]
When ultra-short ultraviolet light having a wavelength of 13.5 nm is reflected by such a reflective mask, for example, in an exposure apparatus with NA = 0.25, if k1 ≧ 0.6 from the above-mentioned Rayleigh equation, , 32.4 nm. In other words, by using ultra-short ultraviolet light and a reflective mask that enables pattern transfer with the ultra-short ultraviolet light, minimization of the pattern width and pattern pitch, etc., that could not be achieved with a light transmission type mask and an optical system. It is possible to correspond to.
[0008]
[Problems to be solved by the invention]
By the way, in recent years, demands for miniaturization are rapidly increasing, and it is desired to cope with further minimization of a pattern width, a pattern pitch, and the like. For example, in a particularly small gate line width required, a line width smaller than 32.4 nm, that is, a condition of k1 <0.6 is also required. Specifically, the gate line width after processing is 15 nm, and the line width on the resist is required to be 25 nm. With respect to the resist line width of 25 nm, in the case of an exposure apparatus having a wavelength of 13.5 nm and NA = 0.25, k1 = 0.46 from the Rayleigh equation. In order to form a line width of such a size, it is necessary to use not only a very short ultraviolet light having a wavelength of 13.5 nm and a reflective mask for reflecting the ultraviolet light but also use a super-resolution technique. Is done.
[0009]
As the super-resolution technology, (1) a deformed illumination light source (annular illumination or four-hole illumination, etc.) and (2) a pupil filter (annular filter or four (3) Halftone phase shift mask, (4) Combination of halftone phase shift mask and modified illumination light source, or (5) Levenson type phase shift mask (or "Alternating phase") It is also known to use a “shift mask”. (3), (4) and (5) (the halftone type phase shift mask and the Levenson type phase shift mask are collectively referred to as “phase shift mask”) all use the phase difference of light. Since it is very effective in improving the resolution performance and increasing the pattern contrast, it is used more frequently in the lithography process than in (1) or (2).
[0010]
However, if it is a transmission mask that transmits light, it is easy to form a phase shift mask as is well known, but if it is a reflection mask corresponding to extremely short ultraviolet light, a phase shift mask is formed. It is very difficult to do. For example, in the case of a transmissive mask, a region where the phase difference of light is different from each other by 180 ° can be formed by dug a mask substrate. As a result, the light reflectance also changes at the same time, so that a phase shift mask cannot be configured. Further, in the case of a transmission mask, a region where the phase difference of light is different from each other by 180 ° can be formed by utilizing the phase shift effect of the material. There is no material that does not absorb the light at the exposure wavelength, and a single material cannot provide a configuration exhibiting the desired reflectance and phase shift effect, so that a phase shift mask cannot be configured. Further, a reflective multilayer substrate used for a reflection mask of ultrashort ultraviolet light generally has a structure in which Si (silicon) layers and Mo (molybdenum) layers are alternately laminated (for example, 40 layers are repeated). For this reason, it has been proposed to form regions having a phase difference of 180 ° and the same reflectance by separately forming regions in which the lamination order is opposite to each other. Is extremely difficult to fabricate, and such a phase shift mask having a laminated structure has not yet been put to practical use. From these facts, it has been considered that it is impossible to actually configure a reflection type phase shift mask corresponding to ultrashort ultraviolet light.
[0011]
In view of the above, the present invention considers that a material suitable as a mask material has a refractive index of 0.89 to 1.01 with respect to the wavelength of ultra-short ultraviolet light. It is an object of the present invention to provide a phase shift mask for extremely short ultraviolet light, a method for manufacturing the same, and a method for manufacturing a semiconductor device, which can be actually configured by obtaining a combination of the phase shift mask and the absorption coefficient.
[0012]
[Means for Solving the Problems]
The present invention is an ultra-short ultraviolet light phase shift mask devised to achieve the above object, and is used to transfer a desired pattern onto an object to be exposed by reflecting the ultra-short ultraviolet light. A phase shift mask for light, comprising: a reflective multilayer substrate that reflects ultra-short ultraviolet light; and a first region and a second region formed on the reflective multilayer substrate. Each thickness and each complex refractive index of the formed film and the formed film of the second region are determined by the reflected light of the ultra short ultraviolet light in the first region and the reflected light of the ultra short ultraviolet light in the second region. It is characterized in that it is set so as to generate a predetermined phase difference.
[0013]
Further, the present invention provides a method of manufacturing a phase shift mask for ultra-short ultraviolet light devised to achieve the above object, and a reflective multilayer film substrate that reflects ultra-short ultraviolet light, A method for manufacturing a phase shift mask for ultra-short ultraviolet light, comprising a first region and a second region formed, the method comprising: forming an arbitrary complex refractive index for the ultra-short ultraviolet light; For any film thickness of the film to be formed, specify the complex refractive index and the phase and reflectance of the reflected light of the ultra-short ultraviolet light obtained from the film thickness, based on the specified phase and reflectance, The formation film of the first region and the second region so that a predetermined phase difference occurs between the reflected light of the ultra short ultraviolet light in the first region and the reflected light of the ultra short ultraviolet light in the second region. Select each film thickness and each complex refractive index in the film formed of And it features.
[0014]
Further, the present invention provides a method of manufacturing a semiconductor device devised to achieve the above object, wherein a semiconductor device includes a lithography step of transferring a desired pattern onto an object using a phase shift mask of ultra-short ultraviolet light. A method of manufacturing a device, wherein an arbitrary complex refractive index for the ultrashort ultraviolet light and an arbitrary film thickness of a film formed on the reflective multilayer substrate are obtained from the complex refractive index and the film thickness. Specifying the phase and reflectance of the reflected light of the ultra-short ultraviolet light, based on the specified phase and reflectance, the reflected light of the ultra-short ultraviolet light in the first region on the reflective multilayer substrate and the reflective multilayer Each film thickness and each complex refraction in the first region forming film and the second region forming film so that a predetermined phase difference occurs between the reflected light of the ultrashort ultraviolet light in the second region on the film substrate. Select the refractive index and the selected complex refractive index and Forming a film for forming the first region and a film for forming the second region on the reflective multilayer film substrate, respectively, and forming the first region and the second region on the reflective multilayer film substrate; An ultra-short ultraviolet light phase shift mask is provided, and a desired pattern is transferred onto an object using the ultra-short ultraviolet light phase shift mask.
[0015]
In the ultrashort ultraviolet light phase shift mask having the above-described configuration, the phase shift mask manufacturing method having the above-described procedure, and the semiconductor device manufacturing method having the above-described procedure, the first region and the second region formed on the reflective multilayer substrate are described. The thickness and the complex refractive index of each are set so that a predetermined phase difference is generated in the reflected ultra-short ultraviolet light. That is, the formation film of the first region and the formation film of the second region are formed to have the set film thickness, and the formation film of the first region and the second region are formed to have the set complex refractive index. The material for forming the film is selected. The set complex refractive index can be achieved by forming each formed film into a laminated structure made of a plurality of materials. As described above, by adjusting the respective film thicknesses and the respective complex refractive indices of the formation film of the first region and the formation film of the second region to the set values, the reflected light of the ultra-short ultraviolet light between the first region and the second region is obtained. A predetermined (for example, 180 °) phase difference occurs.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an ultrashort ultraviolet light phase shift mask, a method of manufacturing the same, and a method of manufacturing a semiconductor device according to the present invention will be described with reference to the drawings. Note that, needless to say, the present invention is not limited to the embodiments described below.
[0017]
[Description of Schematic Configuration of Phase Shift Mask]
First, the schematic configuration of the ultrashort ultraviolet light phase shift mask according to the present invention will be described. The phase shift mask described here reflects a very short ultraviolet light and transfers a desired pattern (for example, a circuit pattern) onto an object to be exposed such as a wafer in a lithography step which is one step in a method of manufacturing a semiconductor device. It is used for. Note that the “extremely short ultraviolet light” referred to here has a shorter wavelength (for example, 1 nm to 100 nm) than the ultraviolet light used in the previous lithography process, as represented by, for example, a wavelength of 13.5 nm. The following are applicable.
[0018]
FIG. 1 is a schematic diagram illustrating an example of a schematic configuration of a phase shift mask according to the present invention. As shown in the figure, the phase shift masks 10 and 10 ′ include a reflective multilayer substrate (mask blanks) 11 that reflects ultra-short ultraviolet light, a first region 12 a formed on the reflective multilayer substrate 11, and a And two regions 12b.
[0019]
The reflective multilayer substrate 11 has a structure in which, for example, Si (silicon) layers and Mo (molybdenum) layers are alternately laminated, and the number of repetitions of the lamination is generally 40 layers. It is known that the ratio の between the total thickness of the Si layer and the Mo layer and the thickness of the Mo layer is appropriately Mo layer thickness / (Si layer thickness + Mo layer thickness) = 0.4. Therefore, in the reflective multilayer substrate 11, when the wavelength λ of the ultra-short ultraviolet light used for exposure is 13.5 nm, the total film thickness of the Si layer and the Mo layer is (λ / 2) / (0.9993 × 0. 6 + 0.9211 × 0.4) = 6.973 nm, the thickness of the Si layer is 6.9730 × 0.6 = 4.184 nm, and the thickness of the Mo layer is 6.9730 × 0.4 = 2.789 nm. .
[0020]
Absorbing film 14 is formed on reflective multilayer substrate 11 with buffer film 13 interposed therebetween. The buffer film 13 is provided as an etching stopper when forming the absorption film or for the purpose of avoiding damage when removing defects after the formation of the absorption film. For example, a Ru (ruthenium) layer or a SiO ₂ (Silicon dioxide). The absorption film 14 is made of a material that absorbs ultra-short ultraviolet light, and is made of, for example, a TaN (tantalum nitride) layer. However, the absorbing film 14 may be made of another material as long as it can be used as a material for masking extremely short ultraviolet light. Specifically, other than TaN, Ta (tantalum) or a Ta compound, Cr (chromium) or a Cr compound, W (tungsten), a W compound, or the like can be considered.
[0021]
Incidentally, a first region 12a and a second region 12b are formed on the reflective multilayer substrate 11. In the first region 12a and the second region 12b, the reflected light of the ultrashort ultraviolet light has a predetermined (for example, 180 °) phase difference. Therefore, as described later, the constituent materials of the formed films (buffer film 13 + absorbing film 14) or the film thicknesses thereof are different between the first region 12a and the second region 12b (see FIG. 1A). However, the buffer film 13 and the absorbing film 14 may be formed only in one of the regions so that a phase difference occurs between the regions 12a and 12b (see FIG. 1B).
[0022]
The phase difference の between the first region 12a and the second region 12b is represented by φ which is the phase difference between incident light and reflected light in the first region 12a. ₁ , The phase difference between the incident light and the reflected light in the second region 12b is φ ₂ When the difference in film thickness between the first region 12a and the second region 12b is h, the difference can be specified by the following equation (1).
[0023]
(Equation 1)

[0024]
In the equation (1), θ is an angle formed by light incident on the mask with respect to a normal on the mask surface. λ is the exposure center wavelength. φ ₁ And φ ₂ Are described in, for example, "Yamamoto and T. Namioka", "Layer-by-layer design method for soft-x-ray multilayers", Applied Optics, Vol. 31 pp1622-1630, (1992).
[0025]
Further, in order to obtain the phase difference ψ (λ) using the expression (1), the complex refractive index of the material forming the first region 12a and the second region 12b is required. When the exposure center wavelength of the ultra-short ultraviolet light is 13.5 nm, the complex refractive index of the forming material is, for example, Mo: 0.92108-0.006454343i, Si: 0.9993-0.00182645i, and Ru: 0. 88749-0.0174721i, TaN: 0.94136-0.0315838i. When the forming material has a laminated structure of arbitrary m layers, a composite complex refractive index obtained by the following equations (2) and (3) may be used.
[0026]
(Equation 2)

[0027]
[Equation 3]

[0028]
In the phase shift masks 10 and 10 ', the first region 12a and the second region 12a are arranged such that the phase difference 特定 specified in this manner becomes a predetermined value (for example, 180 °) between the first region 12a and the second region 12b. The constituent material (particularly the complex refractive index) of the formation film (buffer film 13 + absorption film 14) in the region 12b and the thickness thereof are set. That is, in the phase shift masks 10 and 10 ′ described in the present embodiment, the film thickness and the complex refractive index of the film forming the first region 12a and the film forming the second region 12b are extremely different from those of the first region 12a. A major feature is that a predetermined phase difference is generated between the reflected light of the short ultraviolet light and the reflected light of the extremely short ultraviolet light in the second region 12b.
[0029]
[Description of manufacturing procedure of phase shift mask]
Next, the manufacturing procedure of the phase shift masks 10 and 10 'having the above features will be described. However, here, the manufacturing procedure will be described separately for the first embodiment and the second embodiment.
[0030]
[First Embodiment]
In the first embodiment, a case where a halftone phase shift mask is configured by applying the present invention will be described as an example. FIG. 2 is a flowchart illustrating a procedure for manufacturing the phase shift mask according to the first embodiment.
[0031]
In manufacturing a halftone type phase shift mask, first, an equal phase line and an equal reflectance line for each complex refractive index are obtained (Step 101; hereinafter, the step is abbreviated as “S”). These equal phase lines and equal reflectance lines are determined for any complex refractive index without being limited to existing materials. That is, for an arbitrary complex refractive index with respect to the ultrashort ultraviolet light, the phase and the reflectance of the reflected light of the ultrashort ultraviolet light obtained from the complex refractive index are specified. Note that the phase and the reflectance with respect to the complex refractive index can be uniquely specified theoretically. Also, the arbitrary complex refractive index includes the complex refractive index of the reflective multilayer substrate 11. That is, the equal phase line and the equal reflectance line are also obtained for the reflective multilayer substrate 11.
[0032]
FIG. 3 is an explanatory diagram showing a specific example of the equal phase line. The equiphase lines in the example are obtained by fixing the imaginary part (k) of the complex refractive index to 0.0100i. FIG. 4 is an explanatory diagram showing a specific example of an equal reflectance line. The isoreflectance line in the figure is obtained by fixing the real part (n) of the complex refractive index to 0.9100. In these equal phase lines and equal reflectance lines, the exposure wavelength represented by λ in Expression (1) is 13.5 nm, and the incident angle of obliquely incident light on the mask represented by θ is 4 nm. .84 °.
[0033]
The reflectivity and the phase depend on the real part of the complex refractive index and the imaginary part of the complex refractive index, respectively, but the phase mainly depends on the real part of the complex refractive index, and the reflectivity mainly depends on the imaginary part of the complex refractive index. Dependent. Therefore, in order to finally obtain a phase difference of 180 ° and a desired reflectance between the first region 12a and the second region 12b, approximately the equiphase lines of FIG. May be used to set the complex refractive index or the film thickness of the constituent material of each of the regions 12a and 12b. Alternatively, both the complex refractive index and the film thickness may be set.
[0034]
In setting the complex refractive index or the film thickness, first, a material and a film configuration that can be processed in the first region 12a and the second region 12b on the reflective multilayer substrate 11 are obtained (S102). Then, the complex refractive index in the material configuration of the first region 12a and the complex refractive index in the material configuration of the second region 12b are obtained (S103, S104).
[0035]
Here, for example, the absorption film 14 is formed only on the halftone type phase shift mask 10 'having the configuration shown in FIG. 1B, that is, the second region 12b via the buffer film 13, and the first region 12a and the second region A case where a phase shift mask in which a phase difference of 180 ° occurs with the region 12b is considered. At this time, it is assumed that desired reflectances (different values) are obtained in the first region 12a and the second region 12b. In the second region 12b, it is assumed that Ru is selected as a constituent material of the buffer film 13 and TaN is selected as a constituent material of the absorption film.
[0036]
FIG. 5 is an explanatory diagram showing the distribution of the real part of the complex refractive index with respect to the Ru film thickness and the TaN film thickness. From the example in the figure, it can be seen that when the Ru film thickness is 1 nm to 20 nm and the TaN film thickness is 1 nm to 50 nm, the real part of the composite complex refractive index can take a value between 0.890 and 0.945. That is, the complex refractive index in the material configuration of the second region 12b can be understood from the contents of FIG. Since the buffer film 13 and the absorption film 14 are not formed in the first region 12a, the complex refractive index may be obtained from the laminated structure of the reflective multilayer substrate 11.
[0037]
After obtaining the respective complex refractive indices, the first region 12a and the second region 12b are determined based on the refractive index in the first region 12a and the refractive index in the second region 12b and the already obtained isophase line. The step of the formed film, that is, the thickness of the formed film in the second region 12b is obtained (S105). Specifically, based on the complex refractive index distribution of FIG. 5 and the isophase diagram of FIG. 3, the formation film (buffer film 13 + absorption film) for making the phase difference with the reflective multilayer film substrate 11 180 ° The relationship between the total film thickness 14) and the real part (n) of the composite complex refractive index is uniquely determined as shown in FIG. The thickness of each of the Ru layer (buffer film 13) and the TaN layer (absorbing film 14) for achieving a phase difference of 180 ° with the reflective multilayer substrate 11 is such that the composite complex refractive index has a phase difference of 180 °. From the given conditions, it is uniquely determined as shown in FIG. Thus, the thickness of the film forming the second region 12b, that is, the thicknesses of the Ru layer and the TaN layer may be obtained.
[0038]
When the thickness of each of the Ru layer and the TaN layer is determined, the reflectance of the ultra-short ultraviolet light in the formed film of the first region 12a and the second region 12b is determined (S106). For the reflectance, the imaginary part (k) of the composite complex refractive index with respect to the total film thickness may be obtained from the thicknesses of the Ru layer and the TaN layer, and the reflectance may be obtained from the imaginary part of the composite complex refractive index. FIG. 8 is an explanatory diagram showing the relationship between the thicknesses of the Ru layer and the TaN layer and the imaginary part (k) of the composite complex refractive index with respect to the total thickness. In addition, if the imaginary part (k) of the complex refractive index is known, the reflectance is uniquely obtained from the equal reflectance line shown in FIG. FIG. 6 also shows the relationship between the imaginary part (k) of the composite complex refractive index, the reflectance, and the total film thickness.
[0039]
Here, for example, when the reflectance is set to 0.075, the total film thickness is 43 nm from the contents of FIG. 6, and at this time, the thickness of the Ru layer is 14 nm and the thickness of the TaN layer is 29 nm from the contents of FIG. Obtained as the closest condition. At this film thickness, as described below, the phase difference and the reflectivity are further obtained by using the accurate value of the composite complex refractive index to obtain a halftone type phase having a phase difference of 182.4 ° and a reflectivity of 0.075. A shift can be obtained.
[0040]
However, according to the contents of FIG. 6, it is understood that the film thickness conditions when the reflectance is set to 0.075 also exist near the total film thicknesses of 46.3 nm and 48.3 nm. That is, since the total film thickness is not determined as a single condition from the complex refractive index, the total film thickness may be determined by comprehensively considering various conditions including the film thickness adjustment described below. At this time, in order to secure a large process margin for the complex refractive index with respect to the film thickness variation in mask manufacturing, it is desirable to set the change in the total film thickness in the figure to a flat portion (for example, around 50.5 nm). . This is because even if the total film thickness varies in a range of about ± 1.5 nm, the complex refractive index does not change, so that the phase difference and the reflectance do not change.
[0041]
Thereafter, the film thickness is further adjusted so that a phase of 180 ° and a desired reflectance are obtained between the first region 12a and the second region 12b (S107, S108). Here, a further reason for adjusting the film thickness is to obtain a desired phase and reflectance including an intra-film multiple interference effect which is not considered in FIG. Specifically, when the thickness of the Ru layer and the thickness of the TaN layer are appropriately changed, the respective phase differences and the respective halftone reflectivities are obtained by the above-described procedure. FIG. 9 is an explanatory diagram showing an example in which the phase difference and the halftone reflectance with respect to the thickness of the Ru layer and the thickness of the TaN layer are arranged in a matrix. Here, the halftone reflectance is a difference between the reflectance of the first region 12a and the reflectance of the second region 12b in the phase shift mask 10 'of FIG. From these results, a window that satisfies the desired phase difference and halftone reflectance, that is, a set value of the adjusted film thickness may be selected. In the example of the figure, the case where the phase difference is 180.0 ± 6 ° and the halftone reflectance is 9.0 ± 1%, and the case where the phase difference is 180.0 ± 6 ° and the halftone reflectance is 5.0 ± 1% are illustrated. (See the shadow part in the figure). Thus, for example, when the thickness of the Ru layer is set to 13 nm and the thickness of the TaN layer is set to 30 nm, a halftone phase shift with a phase difference of 179.4 ° and a halftone reflectance of 9.5% can be obtained. . Here, the reflectance is 0.070, which is substantially the same as that predicted in FIG. FIGS. 10 and 11 show another example in which the phase difference and the halftone reflectance with respect to the thickness of the Ru layer and the thickness of the Cr layer are arranged in a matrix. Thus, for example, if the thickness of the Ru layer is set to 9 nm and the thickness of the Cr layer is set to 34 nm, a halftone phase shift mask having a phase difference of 179.2 ° and a halftone reflectance of 4.1% can be obtained. become.
[0042]
After setting the complex refractive index and the film thickness of the formation film of the second region 12b in this way, by forming the formation film on the reflective multilayer film substrate 11 while following the settings, the halftone phase shift What is necessary is just to comprise a mask. Note that the formation of the formation film may be performed using a well-known technique, and a description thereof will not be repeated.
[0043]
That is, the phase difference between the second region 12b and the reflective multilayer substrate 11 in the second region 12b is obtained from the thickness of the material group constituting the second region 12b formed on the reflective multilayer substrate 11 and the real part of the complex refractive index. The reflectance of the second region 12b is determined from the imaginary part of the film thickness of the material group constituting the second region 12b and the complex refractive index, and the first region 12a (the reflective multilayer substrate 11) and the second region 12b are obtained therefrom. This is to obtain a configuration in which the phase difference is different by 180 ° and the reflectance of the second region 12b is a desired value, that is, a halftone phase shift mask having the above-described configuration. However, either of the phase difference and the reflectance may be obtained first.
[0044]
The light intensity distribution when using a halftone phase shift mask obtained by such a procedure will be described. FIG. 12 shows the light intensity in a hole pattern having an opening of 30 nm on a mask (indicated by coordinates on a wafer, 120 nm for a 4 × mask) when the optical conditions are NA = 0.25 and σ = 0.70. It is explanatory drawing which shows distribution. In the figure, the light intensity distribution for the conventional binary mask is also shown for comparison. According to the illustrated example, it is clear that the effect of improving the pattern edge contrast is obtained when the halftone phase shift mask is used, as compared with the conventional binary mask.
[0045]
Here, as the halftone type phase shift mask, a phase shift mask 10 'having the configuration shown in FIG. 1B, that is, a halftone type phase shift mask in which the buffer film 13 and the absorption film 14 are formed only in the second region 12b is taken as an example. However, for example, it is also possible to configure a halftone type phase shift mask 10 as shown in FIG. However, even in that case, as described above, the film thickness of the formed film and the complex refractive index in each of the first region 12a and the second region 12b are set so that a phase difference of 180 ° occurs. It is assumed that
[0046]
FIG. 13 is a schematic diagram illustrating a cross-sectional structure of a configuration example of a halftone phase shift mask. In the mask of the illustrated example, on the reflective multilayer substrate 11, Ru 10 nm and Si 47 nm are sequentially stacked in the first region 12a, and Ru 5 nm, TaN 47 nm, and Ru 5 nm are sequentially stacked in the second region 12b. The reason why Si is used as a constituent material of the formation film of the first region 12a is as follows. The complex refractive index of Si is 0.99932-0.00182645i, the real part of which is very close to 1, which is the refractive index in a vacuum, and the imaginary part thereof is smaller than other materials. Therefore, the Si material can play a role of adjusting the phase difference and the reflectance ratio by using the in-film multiple interference effect. By utilizing the multiple interference effect in the film, a phase difference of 180 ° is generated between the first region 12a and the second region 12b, and furthermore, a halftone type phase shift mask having a flat structure whose upper surface is completely flat is formed. It becomes possible.
[0047]
[Second embodiment]
Next, a second embodiment of the manufacturing procedure of the phase shift mask will be described. In the second embodiment, a case where the present invention is applied to form a Levenson-type phase shift mask will be described as an example. FIG. 14 is a flowchart illustrating a procedure for manufacturing the phase shift mask according to the second embodiment.
[0048]
As shown in FIG. 14, the production of the Levenson-type phase shift mask is performed in substantially the same manner as in the case of the halftone phase shift mask in the first embodiment (see FIG. 2) (S201 to S208). However, the Levenson-type phase shift mask is different from the half-tone type phase shift mask in that not only the phase in the first region 12a and the phase in the second region 12b differ from each other by 180 °, but also the reflectance in the first region 12a and the reflectance in the second region 12b are substantially equal. Different from shift mask. That is, when the Levenson-type phase shift mask is formed, (1) the reflectance is substantially equal between the first region 12a and the second region 12b, and (2) the position between the first region 12a and the second region 12b. The two conditions that the phase difference is 180 ° must be satisfied (S208).
[0049]
The determination as to whether or not these conditions are satisfied may be made as follows. First, the reflectance of the first region 12a is represented by R ₁ , The reflectance of the second region 12b is R ₂ Then, the reflectance ratio P obtained by the following equation (4) is specified.
[0050]
(Equation 4)

[0051]
Then, the criterion 1: | P | ≦ 3.0% is applied to the specified reflectance ratio P, and if the criterion 1 is met, the above condition (1) is satisfied. Is determined.
[0052]
As for the condition (2), the criterion 2: | ψ (λ) | ≦ 6 ° is applied to ψ (λ) obtained by the equation (1) described in the first embodiment. I do. Then, if this determination criterion 2 is met, it is determined that the above condition (2) is satisfied.
[0053]
As the Levenson-type phase shift mask obtained in this manner, the buffer film 13 and the absorption film 14 are formed in the configuration shown in FIG. 1A, that is, in both the first region 12a and the second region 12b. A configuration having a different configuration is conceivable.
[0054]
Here, the manufacturing procedure of the Levenson-type phase shift mask having the configuration shown in FIG. 1A will be described in more detail with reference to specific examples. Even when the Levenson type phase shift mask is formed, similarly to the case of the halftone type phase shift mask described in the first embodiment, first, the equal phase line and the equal reflectance line for an arbitrary complex refractive index are set. Ask. FIG. 15 is an explanatory diagram illustrating a specific example of the equal phase line. The equiphase lines in the example are obtained by fixing the imaginary part (k) of the complex refractive index to 0.0100i.
[0055]
By the way, in the Levenson-type phase shift mask, the phase difference ψ between the film forming the first region 12 a and the reflective multilayer substrate 11. ₁ (Λ) is specified by the following equation (5).
[0056]
(Equation 5)

[0057]
In addition, the phase difference ψ between the film forming the second region 12 b and the reflective multilayer substrate 11. ₂ (Λ) is specified by the following equation (6).
[0058]
(Equation 6)

[0059]
Therefore, the phase difference ψ (λ) between the first region 12a and the second region 12b is specified by the following equation (7).
[0060]
(Equation 7)

[0061]
The relationship specified by the equation (7) is represented by the relative relationship of the equiphase lines as shown in FIG. For example, in the figure, in a material in which the real part of the complex refractive index is 0.94 and the phase difference between the first region 12a and the second region 12b is 180 °, the first region 12a and the second region It can be seen that the step between the two regions 12b should be 56 nm (see [1] in the figure). Further, the material is such that the real part of the complex refractive index is 0.96 in the first region 12a and 0.94 in the second region 12b, and the phase difference between the first region 12a and the second region 12b is 180-. To this end, it can be seen that the step between the first region 12a and the second region 12b only needs to be 42 nm (see [2] in the figure). On the other hand, when calculating the reflectance, the absolute value for the total film thickness is used as it is, not the relative value.
[0062]
Even when a Levenson-type phase shift mask is formed, a formed film that satisfies the criterion 1 and the criterion 2 based on such an equal phase line (see FIG. 15) and an equal reflectance line (see FIG. 4). May be set as the film thickness and the complex refractive index.
[0063]
That is, the phase difference between the first region 12a and the reflective multilayer substrate in the first region 12a is determined from the film thickness of the material group constituting the first region 12a formed on the reflective multilayer substrate 11 and the real part of the complex refractive index. The phase difference between the second region 12b and the reflective multilayer substrate in the second region 12b is determined from the film thickness of the material group constituting the second region 12b formed on the multilayer substrate 11 and the real part of the complex refractive index. Further, the reflectance in the first region 12a is obtained from the film thickness of the material group constituting the first region 12a formed on the reflective multilayer substrate 11 and the imaginary part of the complex refractive index. The reflectance in the second region 12b is determined from the thickness of the material group constituting the second region 12b to be formed and the imaginary part of the complex refractive index. From these results, the configuration satisfying the criterion 1 and the criterion 2, that is, the phase difference between the first region 12a and the second region 12b is different by 180 °, and the reflectance of the first region 12a and the second region 12b is different. Are obtained as a Levenson-type phase shift mask. Either the phase difference or the reflectance may be determined first.
[0064]
By the way, in order to satisfy the criterion 1 and the criterion 2 at the same time, it is necessary to appropriately laminate different materials to form the first region 12a and the second region 12b. This is because the Levenson-type phase shift mask that satisfies the criterion 1 and the criterion 2 at the same time unless a different material is appropriately laminated due to the restriction of the complex refractive index of the material that actually exists is a structure that can be actually manufactured. Because you can't get it.
[0065]
For this reason, for example, TaN, Ru, and Si are used as the material forming the first region 12a. Further, as a material forming the second region 12b, for example, Mo and Ru are used. These materials are used, for example, in Jpn. J. Appl. Phys. As disclosed in Vol 40 (2001) pp 6998-7001, "Approach to patterning of extreme ultraviolet lithography masks", it is known that the etching selectivity of each material can be very large in combination as described below. . That is, the etching of the Ru layer with respect to the Si ₂ + O ₂ When dry etching is performed using a gas, the Si underlayer acts as an etching stopper for Ru etching. Ar + Cl is used for etching the TaN layer on the Ru underlayer. ₂ When dry etching is performed using a gas, the Ru underlayer acts as an etching stopper for the TaN layer etching. Further, the etching of the Mo layer and the etching of the Si layer with respect to the Ru underlayer are also performed by Ar + Cl. ₂ The selectivity is increased by performing dry etching with gas. Because Ru chloride, RuCl ₃ Is a relatively stable substance and decomposes at 600 ° C. or higher. On the other hand, Si chloride SiCl ₄ Is 57.6 ° C., Mo chloride MoCl ₅ Is 268 ° C., and a chloride of TaCl ₅ Has a boiling point of 242 ° C., and is easily removed as an etching reaction gas in a vacuum with respect to Ru chloride.
[0066]
FIG. 16 is a schematic diagram illustrating a cross-sectional structure of a configuration example of a Levenson-type phase shift mask. The structure in the figure (hereinafter referred to as “structure 1”) is a configuration that satisfies the criterion 1 and the criterion 2 set based on the isophase line in FIG. 15 and the isoreflectance line in FIG. In the first region 12a, a film is formed by stacking Ru2 nm, TaN7 nm, and Ru4 nm on the reflective multilayer substrate 11 in this order. The total thickness of the first region 12a is 13 nm, the real part of the composite complex refractive index is 0.9165, and the imaginary part is 0.02507i. Further, in the second region 12b, Ru4 nm, Mo49 nm, and Ru2 nm are laminated in this order to form a formed film. The total thickness of the second region 12b is 55 nm, the real part of the combined complex refractive index is 0.9174, and the imaginary part is 0.00764i. At the boundary between the first region 12a and the second region 12b, a TaN absorption layer having a thickness of 120 nm is formed with a width of 40 nm.
[0067]
In such a structure 1, when the step between the first region 12a and the second region 12b is obtained from the isophase line in FIG. 15, the step is 43 nm from the relationship between the 90 ° and 270 ° isophase lines. It is 43 nm even from the relationship between the 180 ° and 0 ° equiphase lines. Further, according to the iso-transmittance line in FIG. 4, it can be seen that the reflectance of the first region 12a is 0.39 and the reflectance of the second region 12b is 0.38. Then, when the phase difference and the reflectance ratio are obtained by changing the Mo film thickness of the second region 12b including the in-film multiple interference effect in detail for the structure 1, the result shown in FIG. 17, for example, is obtained. From the results of the examples, it can be confirmed that the structure 1 shown in FIG. 16 has an optimum configuration within the range of adjusting the thickness of each material.
[0068]
In the structure 1, the reflectance is 0.388 for the first region 12a and 0.387 for the second region 12b, and the reflectance ratio between them is 0.258%. . In addition, the phase difference between the first region 12a and the second region 12b is 178.8 ° with respect to a TE (transverse electric) wave and 178.7 ° with respect to a TM (transverse magnetic) wave.
[0069]
The light intensity distribution when the Levenson-type phase shift mask having such a structure 1 is used is obtained by forming a 120 nm thick TaN absorption layer with a 40 nm width (10 nm in units on a wafer) on a four-fold mask of the structure 1. In the case of exposure on a wafer, under the optical conditions of NA = 0.25 and σ = 0.70, a result as shown in FIG. 18 is obtained, for example.
[0070]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is determined by using a pattern of a 40 nm width TaN absorption layer having a pitch of 320 nm (80 nm pitch in a unit of 10 nm width on a wafer). It can be confirmed by calculating the phase difference of the following equation (8). In equation (8), coordinates on the wafer are used. The unit of the x-axis is nm.
[0071]
(Equation 8)

[0072]
FIG. 19 shows a result obtained by calculating the equation (8) for the TEy wave, the TMx wave, and the TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0073]
From these facts, the Levenson-type phase shift mask of the structure 1 satisfies the criterion 1 and the criterion 2 at the same time, the reflectance is substantially equal between the first region 12a and the second region 12b, and the phase difference is 180. °.
[0074]
Next, a procedure for forming the first region 12a and the second region 12b with the complex refractive index and the thickness set as described above on the reflective multilayer substrate 11 will be briefly described. . 20 to 23 are explanatory diagrams illustrating an example of a film forming procedure for the structure 1. FIG. When forming a film having the structure 1, as shown in FIG. 20, first, a Ru layer is formed on the reflective multilayer substrate 11 by a sputtering method (step 1). Since the Ru layer is usually a material used as a buffer layer in a binary mask, the same manufacturing apparatus as in the normal case can be used. Then, a TaN layer is formed on the Ru layer by a sputtering method (Step 2). Since the TaN layer is a material used as an absorption layer in a binary mask, the same manufacturing apparatus as in the usual case can be used.
[0075]
Thereafter, a resist is applied on the TaN layer (Step 3), and the resist in the second region 12b is removed through a drawing and a resist developing step (Step 4). After removing the resist, the TaN layer in the second region 12b is replaced with Ar + Cl. ₂ It is removed by dry etching with a gas (step 5). The Ru layer below this TaN layer functions as an etching stopper layer. Then, the resist is removed (Step 6), and a Ru layer is formed again by sputtering (Step 7), and then a Mo layer is formed by sputtering (Step 8). Since the Mo layer is a constituent material of the reflective multilayer substrate 11, a multilayer film manufacturing apparatus can be used.
[0076]
After forming the Mo layer, subsequently, as shown in FIG. 21, after applying a resist (Step 9), the resist in the first region 12a is removed through a drawing and a resist developing step (Step 10). . Then, the Mo layer in the first region 12a is changed to Ar + Cl. ₂ It is removed by dry etching with a gas (step 11). The Ru layer below this Mo layer functions as an etching stopper layer. Thereafter, the resist is removed (step 12), and a Ru layer is formed again by sputtering (step 13).
[0077]
After the formation of the Ru layer, as shown in FIG. 22, a TaN absorption layer is formed by sputtering (step 14), and a resist is applied (step 15). Then, after the drawing and the resist developing steps, the resist other than the portion left as the absorbing layer is removed (step 16).
[0078]
After that, as shown in FIG. ₂ It is removed by dry etching with a gas (step 17). The Ru layer below the TaN absorption layer functions as an etching stopper layer. Then, if the resist is stripped (Step 18), the Levenson-type phase shift mask of Structure 1 can be formed while effectively using the Ru layer as an etching stopper layer.
[0079]
FIG. 24 is a schematic diagram showing a cross-sectional structure of another configuration example of the Levenson-type phase shift mask. In the structure of the figure (hereinafter referred to as “structure 2”), the criterion 1 and the criterion 2 set based on the equal phase line of FIG. 15 and the equal reflectance line of FIG. It is the structure which satisfies. In the first region 12a, a formation film is formed by stacking Ru3 nm, TaN5 nm, and Ru7 nm on the reflective multilayer film substrate 11 in this order. The total thickness of the first region 12a is 15 nm, the real part of the composite complex refractive index is 0.9054, and the imaginary part is 0.02217i. In the second region 12b, a film is formed by stacking Ru3 nm, Mo49 nm, and Ru4 nm in this order. The total thickness of the second region 12b is 56 nm, the real part of the composite complex refractive index is 0.9169, and the imaginary part is 0.00782i.
[0080]
In such a structure 2, when the step between the first region 12a and the second region 12b is obtained from the isophase line in FIG. 15, the step is 44 nm from the relationship between the 90 ° and 270 ° isophase lines. It is 44 nm even from the relationship between the 180 ° and 0 ° equiphase lines. According to the iso-transmittance line in FIG. 4, the reflectance of the first region 12a is 0.38, and the reflectance of the second region 12b is 0.36. Then, when the phase difference and the reflectance ratio are obtained by changing the Mo film thickness of the second region 12b including the in-film multiple interference effect in detail for the structure 2, the result shown in FIG. 25 is obtained, for example. From the results shown in the figure, it can be confirmed that the structure 2 shown in FIG. 24 has the optimum configuration within the range of adjusting the film thickness of each material.
[0081]
In the structure 2, the reflectance is 0.399 for the first region 12a and 0.396 for the second region 12b, and the reflectance ratio between them is 0.710%. . Therefore, the phase difference between the first region 12a and the second region 12b is 178.2 ° for the TE wave and 178.3 ° for the TM wave.
[0082]
The light intensity distribution when the Levenson type phase shift mask having the structure 2 is used is such that a TaN absorption layer having a thickness of 120 nm is formed with a 40 nm width (10 nm in units on a wafer) on a four-fold mask of the structure 2. When exposure is performed on a wafer and the optical conditions are NA = 0.25 and σ = 0.70, a result as shown in FIG. 26 is obtained, for example.
[0083]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is described above by using a pattern of a 320 nm pitch (80 nm pitch in a unit of 10 nm on a wafer) with a 40 nm wide TaN absorption layer. This can be confirmed by calculating the phase difference in equation (8). FIG. 27 shows the result obtained by calculating the equation (8) for the TEy wave, the TMx wave, and the TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0084]
The film forming procedure of the structure 2 in which the complex refractive index and the film thickness are set as described above is substantially the same as that of the structure 1 described above. The structure 1 is different from the structure 1 in that the Ru layer in the second region 12b is ₂ + O ₂ The difference is that a step of removing by gas dry etching is added.
[0085]
FIG. 28 is a schematic diagram showing a cross-sectional structure of still another configuration example of the Levenson-type phase shift mask. Similarly to the structures 1 and 2 described above, the structure in the example of the drawing (hereinafter, referred to as “structure 3”) also includes the determination criterion 1 and the determination criterion 1 set based on the isophase line in FIG. 13 and the isoreflectance line in FIG. This is a configuration that satisfies the criterion 2. In the first region 12a, a film is formed on the reflective multilayer substrate 11 in the order of Ru5 nm, TaN20 nm, Si8 nm, and Ru5 nm. In the second region 12b, a formed film is formed by stacking Ru 5 nm, Si 8 nm, and Ru 43.5 nm in this order.
[0086]
In the structure 3, unlike the structure 1 or the structure 2, Si is used as a constituent material of a formation film of the first region 12a and the second region 12b. Here, the complex refractive index of Si is 0.99932-0.00182645i, the real part of the complex refractive index is very close to 1, which is the refractive index in vacuum, and the imaginary part of the complex refractive index is different from other materials. Small in comparison. Therefore, the Si material plays a role of adjusting the phase difference and the reflectance ratio by utilizing the multiple interference effect in the film.
[0087]
Therefore, when obtaining a configuration that satisfies the criterion 1 and the criterion 2 from the isophase line of FIG. 15 and the isoreflectance line of FIG. 4, the condition for emphasizing the phase difference without considering the Si layer. , And then adjust the reflectivity by multiple interference in the Si layer and adjust the phase difference to near 180 °. That is, the configuration satisfying the criterion 1 and the criterion 2 is considered that the first region 12a is laminated on the reflective multilayer substrate 11 in the order of Ru5 nm, TaN20nm, and Ru5nm, and the second region 12b is laminated. It is assumed that Ru5nm and Ru43.5nm are laminated in this order. In this case, the total film thickness of the first region 12a is 30 nm, the real part of the composite complex refractive index is 0.9234, and the imaginary part is 0.02687i. The total thickness of the second region 12b is 48.5 nm, the real part of the combined complex refractive index is 0.8875, and the imaginary part is 0.01747i.
[0088]
In such a structure 3, when the step between the first region 12a and the second region 12b is obtained from the isophase line in FIG. 15, the step is 22 nm from the relationship between the 90 ° and 270 ° isophase lines. It is 16 nm as determined from the relationship between the 180 ° and 0 ° equiphase lines. Further, according to the iso-transmittance line in FIG. 4, it is understood that the reflectance of the first region 12a is 0.14 and the reflectance of the second region 12b is 0.18. Then, for the structure 3, when the Mo film thickness of the second region 12b is changed including the in-film multiple interference effect in detail and the phase difference and the reflectance ratio are obtained, for example, the result shown in FIG. 29 is obtained. From the results of the examples, it can be confirmed that the structure 3 shown in FIG. 28 has an optimum configuration within the range of adjusting the film thickness of each material. In this structure 3, since Si is used as a constituent material of the film for forming the first region 12a and the second region 12b, and the multiple interference in the film by the Si layer is effectively used, the phase difference and the reflection The condition that the rate ratio satisfies the reference value can be realized.
[0089]
In the structure 3, the reflectance is 0.195 for the first region 12a and 0.200 for the second region 12b, and the reflectance ratio therebetween is 2.41%. . Therefore, the phase difference between the first region 12a and the second region 12b is 185.3 ° for the TE wave and 185.1 ° for the TM wave.
[0090]
The light intensity distribution in the case of using the Levenson-type phase shift mask having the structure 3 is such that a TaN absorption layer having a thickness of 120 nm is formed with a 40 nm width (10 nm in units on a wafer) on a four-fold mask of the structure 3. In the case of exposing on a wafer, under the optical conditions of NA = 0.25 and σ = 0.70, a result as shown in FIG. 30, for example, is obtained.
[0091]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is described above by using a pattern of a 320 nm pitch (80 nm pitch in a unit of 10 nm on a wafer) with a 40 nm wide TaN absorption layer. This can be confirmed by calculating the phase difference in equation (8). FIG. 31 shows a result obtained by calculating the equation (8) for the TEy wave, the TMx wave, and the TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0092]
The film forming procedure of the structure 3 in which the complex refractive index and the film thickness are set as described above is substantially the same as that of the structure 2 described above. The difference from the structure 2 is that in step 8, a Si layer and a Ru layer are formed by sputtering instead of the Mo layer.
[0093]
By the way, in each of the structures 1 to 3, there is a difference in film thickness between the formation film of the first region 12a and the formation film of the second region 12b, and there is a step between them. Even with such a configuration, it is possible to obtain the effect of the phase shift as described above. However, considering that the TaN absorption layer is formed at the boundary between the first region 12a and the second region 12b, a step is formed between the first region 12a and the second region 12b in view of the easiness of the processing. It is desirable that they are formed flat without any irregularities. Therefore, a specific example of a configuration in which the first region 12a and the second region 12b are flat will be described below.
[0094]
FIG. 32 is a schematic diagram showing a cross-sectional structure of a flat configuration example of the Levenson-type phase shift mask. The structure in the figure (hereinafter referred to as “structure 4”) is a configuration that satisfies the criterion 1 and the criterion 2 set based on the isophase line in FIG. 15 and the isoreflectance line in FIG. In the first region 12a, a formed film is formed by stacking Ru3 nm, TaN7 nm, Ru6 nm, Si37 nm, and Ru5 nm on the reflective multilayer film substrate 11 in this order. In the second region 12b, a formation film is formed by stacking Ru3 nm, Mo47 nm, and Ru8 nm in this order.
[0095]
In the structure 4, the reason why Si is used as a constituent material of the film for forming the first region 12a is that the role of adjusting the phase difference and the reflectance ratio by utilizing the multiple interference effect in the film is the same as in the case of the structure 3. It is to carry. Therefore, when obtaining a configuration that satisfies the criterion 1 and the criterion 2 from the isophase line of FIG. 15 and the isoreflectance line of FIG. 4, the condition for emphasizing the phase difference without considering the Si layer. , And then adjust the reflectivity by multiple interference in the Si layer, and adjust the phase difference to near. That is, the configuration satisfying the criterion 1 and the criterion 2 is considered that the first region 12a is laminated on the reflective multilayer substrate 11 in the order of Ru3 nm, TaN7nm, Ru6nm, and Ru5nm. In this case, it is assumed that Ru is stacked in the order of 3 nm, 47 nm of Mo, and 8 nm of Ru. In this case, the total thickness of the first region 12a is 21 nm, the real part of the combined complex refractive index is 0.9054, and the imaginary part is 0.03631i. The total thickness of the second region 12b is 58 nm, the real part of the composite complex refractive index is 0.9147, and the imaginary part is 0.0009004i.
[0096]
In such a structure 4, when the step between the first region 12a and the second region 12b is obtained from the isophase line in FIG. 15, the step is 40 nm from the relationship between the 90 ° and 270 ° isophase lines. It is 43 nm when determined from the relationship between the 180 ° and 0 ° equiphase lines. According to the iso-transmittance line in FIG. 4, the reflectance of the first region 12a is 0.15, and the reflectance of the second region 12b is 0.30.
[0097]
As described above, the structure 4 is similar to the structure 1 and the structure 2 when the Si layer is not considered, but does not have the same reflectance in the first region 12a and the second region 12b. In order to solve this, in the structure 4, the reflectance in the first region 12a and the reflectance in the second region 12b are matched by inserting a Si layer and effectively using the multiple interference in the film.
[0098]
Therefore, in the structure 4, the thickness of the TaN layer in the first region 12a, the thickness of the Mo layer in the second region 12b, and the thickness of the Si layer in the first region 12a are appropriately changed, and the determination criteria 1 and A configuration satisfying the criterion 2 is determined. FIG. 33 shows the result of determining the phase difference and the reflectance ratio using the thickness of the Mo layer as a parameter while keeping the thickness of the TaN layer and the thickness of the Si layer constant. From the results in the example, it can be confirmed that the structure 4 shown in FIG. 32 satisfies the criterion under the condition that the thickness difference between the first region 12a and the second region 12b is 0 nm. In this structure 4, Si is used as a constituent material of the film for forming the first region 12a, and the multiple interference in the film by the Si layer is effectively used. Is satisfied, and a configuration in which the first region 12a and the second region 12b are flat can be realized.
[0099]
In the structure 4, the reflectance is 0.285 for the first region 12a and 0.287 for the second region 12b, and the reflectance ratio between them is 0.65%. . Therefore, the phase difference between the first region 12a and the second region 12b is 180.8 ° for the TE wave and 180.5 ° for the TM wave.
[0100]
The light intensity distribution when the Levenson-type phase shift mask having the structure 4 is used is obtained by forming a TaN absorption layer having a thickness of 120 nm with a width of 40 nm (10 nm in units on a wafer) on a four-time mask of the structure 4. When exposure is performed on a wafer and the optical conditions are NA = 0.25 and σ = 0.70, the result is as shown in FIG. 34, for example.
[0101]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is described above by using a pattern of a 320 nm pitch (80 nm pitch in a unit of 10 nm on a wafer) with a 40 nm wide TaN absorption layer. This can be confirmed by calculating the phase difference in equation (8). FIG. 35 shows a result obtained by calculating the equation (8) for the TEy wave, the TMx wave, and the TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0102]
As to the film forming procedure of the structure 4 in which the complex refractive index and the film thickness are set as described above, in addition to the case of the structure 1 described above (see FIGS. 6 between the Ru layer and the second region 12b. ₂ + O ₂ A step as shown in FIGS. 36 and 37 may be performed between the steps 13 and 14 while performing the step of removing by dry etching with a gas.
[0103]
That is, as shown in FIG. 36, after the Ru layer is formed in Step 13, a Si layer is formed by sputtering (Step 13-1), and a resist is applied (Step 13-2). Then, the resist in the portion of the second region 12b is removed through the drawing and the resist developing process (step 13-3).
[0104]
Thereafter, as shown in FIG. 37, the Si film in the second region 12b is changed to Ar + Cl. ₂ It is removed by dry etching with a gas (step 13-4). The Ru layer below this Si layer functions as an etching stopper layer. Then, after removing the resist (Step 13-5) and forming a Ru layer by sputtering (Step 13-6), a TaN absorption layer is formed by sputtering (Step 14). The same processing is performed. By such a film forming procedure, the Levenson-type phase shift mask having the structure 4 can be formed.
[0105]
As described above, for the structure 4, the thickness of the TaN layer in the first region 12a, the thickness of the Mo layer in the second region 12b, and the thickness of the Si layer in the first region 12a are appropriately changed to determine the determination criteria. 1 and the criterion 2 are specified, and the first region 12a and the second region 12b are flat. However, such a configuration can also be obtained in a structure as described below.
[0106]
FIG. 38 is a schematic diagram showing a cross-sectional structure of another flat configuration example of the Levenson-type phase shift mask. The structure in the figure (hereinafter referred to as “structure 5”) is also a result of determining the phase difference and the reflectance ratio using the thickness of the Mo layer as a parameter while keeping the thickness of the TaN layer and the thickness of the Si layer constant. FIG. 39 shows the result of determining the phase difference and the reflectance ratio using the thickness of the Mo layer as a parameter while keeping the thickness of the TaN layer and the thickness of the Si layer constant. From the results shown in the example, it can be confirmed that the structure 5 satisfies the determination criterion under the condition that the film thickness difference between the first region 12a and the second region 12b is 0 nm. In this structure 5, the reflectance is 0.288 for the first region 12a and 0.287 for the second region 12b, and the reflectance ratio therebetween is 0.36%. The phase difference between the first region 12a and the second region 12b is 181.4 ° for the TE wave and 181.1 ° for the TM wave.
[0107]
The light intensity distribution when the Levenson-type phase shift mask having the structure 5 is used is such that a TaN absorption layer having a thickness of 120 nm is formed with a width of 40 nm (10 nm in units on a wafer) on a four-fold mask of the structure 5. In the case of exposure on a wafer, under the optical conditions of NA = 0.25 and σ = 0.70, a result as shown in FIG. 40 is obtained, for example.
[0108]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is described above by using a pattern of a 320 nm pitch (80 nm pitch in a unit of 10 nm on a wafer) with a 40 nm wide TaN absorption layer. This can be confirmed by calculating the phase difference in equation (8). FIG. 41 shows the result obtained by calculating the equation (8) for the TEy wave, TMx wave, and TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0109]
The film forming procedure of the structure 5 is the same as that of the structure 4 described above.
[0110]
FIG. 42 is a schematic diagram showing a cross-sectional structure of still another flat configuration example of the Levenson-type phase shift mask. The structure in the figure (hereinafter referred to as “structure 6”) is also a result of determining the phase difference and the reflectance ratio using the thickness of the Mo layer as a parameter while keeping the thickness of the TaN layer and the thickness of the Si layer constant. FIG. 43 shows the result of obtaining the phase difference and the reflectance ratio using the thickness of the Mo layer as a parameter while keeping the thickness of the TaN layer and the thickness of the Si layer constant. From the results shown in the example, it can be confirmed that the structure 6 also satisfies the criterion under the condition that the film thickness difference between the first region 12a and the second region 12b is 0 nm. In this structure 6, the reflectance is 0.300 for the first region 12a and 0.287 for the second region 12b, and the reflectance ratio between them is 0.90%. The phase difference between the first region 12a and the second region 12b is 180.6 ° for the TE wave and 180.3 ° for the TM wave.
[0111]
The light intensity distribution when the Levenson-type phase shift mask having the structure 6 is used is obtained by forming a TaN absorption layer having a thickness of 120 nm with a width of 40 nm (10 nm in units on a wafer) on a four-fold mask of the structure 6. When exposure is performed on a wafer and the optical conditions are NA = 0.25 and σ = 0.70, a result as shown in FIG. 44 is obtained, for example.
[0112]
Further, the fact that the phase difference between the first region 12a and the second region 12b is 180 ° is described above by using a pattern of a 320 nm pitch (80 nm pitch in a unit of 10 nm on a wafer) with a 40 nm wide TaN absorption layer. This can be confirmed by calculating the phase difference in equation (8). FIG. 45 shows a result obtained by calculating the equation (8) for the TEy wave, the TMx wave, and the TMz wave immediately above the mask. According to the illustrated example, both the TEy wave and the TMx wave favorably maintain a phase difference of 180 °. On the other hand, although the deviation from 180 ° is more remarkable in the TMz wave, the contribution to the transfer is about 0.45%, so that the TMz wave has almost no influence.
[0113]
The procedure for forming the film of the structure 6 is the same as that of the structure 4 or the structure 5 described above.
[0114]
Next, the effects of the above Levenson-type phase shift mask will be described in comparison with a conventional binary mask that does not use the phase shift effect. FIG. 46 shows the Levenson-type phase shift mask having the structure 5 described above, in which a TaN absorption layer having a thickness of 120 nm is 40 nm, 30 nm, 20 nm, 10 nm, and 0 nm wide on a four-fold mask of the structure 5 (unit on a wafer). Shows a light intensity distribution obtained under the optical conditions of NA = 0.25 and σ = 0.70 when formed and exposed on a wafer at 10 nm, 7.5 nm, 5 nm, 2.5 nm, and 0 nm. FIG. According to the example of the figure, it can be seen that good pattern contrast can be obtained at any TaN absorption layer width.
[0115]
On the other hand, FIG. 47 shows a conventional binary mask in which a TaN absorption layer having a thickness of 120 nm is provided on the mask with a width of 40 nm, 30 nm, 20 nm, and 10 nm (in units of 10 nm, 7.5 nm, and 5 nm on a wafer). , 2.5 nm) is an explanatory diagram showing a light intensity distribution obtained under optical conditions of NA = 0.25 and σ = 0.70 when exposed on a wafer. According to the example of the figure, it can be seen that the pattern contrast significantly decreases as the width of the TaN absorption layer decreases.
[0116]
From these facts, when the Levenson-type phase shift mask having the above-described structure is used, the transfer of a line width of 15 nm or less on the wafer in the TaN absorption layer 10 nm width (2.5 nm in units on the wafer). Can be achieved. Further, for example, as shown in FIG. 48, if exposure is performed under an optical condition of NA = 0.30, a line having a size of 10 nm or less on the wafer in a TaN absorption layer 10 nm width (2.5 nm in units on the wafer). A remarkable effect that the width can be transferred can be obtained.
[0117]
As described above, in the phase shift mask described in the present embodiment (first and second embodiments), the film thickness and the complex refractive index of the formation film of the first region 12a and the second region 12b are respectively Is set such that a predetermined phase difference is generated in the reflected light of the ultrashort ultraviolet light at the point (1). More specifically, in forming the phase shift mask, first, an arbitrary complex refractive index and an arbitrary film thickness can be obtained without depending on the film formed on the reflective multilayer substrate 11 (particularly, the complex refractive index of the constituent material). The phase and the reflectance of the reflected light obtained from the complex refractive index and the film thickness are specified, and a phase difference of 180 ° occurs between the first region 12a and the second region 12b based on the specified phase and the reflectance. As described above, the thickness of the formed film and the complex refractive index in each case are selected.
[0118]
Therefore, according to the phase shift mask described in the present embodiment, it is feasible to configure a phase shift mask for using the super-resolution technique even in the case of a reflective mask corresponding to ultrashort ultraviolet light. It becomes. That is, if the method for manufacturing a phase shift mask described in the present embodiment is used, a phase shift mask for extremely short ultraviolet light can be configured.
[0119]
Moreover, in the phase shift mask described in the present embodiment, the pattern contrast on the wafer is remarkably increased by using not only the ultra-short ultraviolet light but also the super-resolution technology. A resolution that could not be obtained can be obtained. In other words, when a semiconductor device is manufactured using the phase shift mask of the present embodiment, finer hole patterns, space patterns, and line patterns can be obtained than when a conventional binary mask is used, and the pattern can be minimized. It becomes very suitable for correspondence to.
[0120]
Further, in the phase shift mask described in the present embodiment, the film thickness of the formed film is made substantially equal between the first region 12a and the second region 12b by effectively utilizing the multiple interference in the film, and A configuration in which the first region 12a and the second region 12b are flat is realized. Therefore, for example, even when the TaN absorption layer is formed at the boundary between the first region 12a and the second region 12b in the Levenson-type phase shift mask, it is only necessary to form the TaN absorption layer on a flat portion. This facilitates the formation, and also facilitates the formation accuracy.
[0121]
Further, in the phase shift mask described in the present embodiment, both or one of the formation film of the first region 12a and the formation film of the second region 12b has a laminated structure made of a plurality of materials. Therefore, even if the phase and the reflectance of the reflected light obtained from the complex refractive index and the film thickness are specified for an arbitrary complex refractive index and the film thickness, the formation film corresponding to the arbitrary complex refractive index and the film thickness is determined. It is possible to obtain. That is, a desired phase shift mask can be formed by using a laminated structure including a plurality of materials.
[0122]
【The invention's effect】
As described above, according to the ultra-short ultraviolet light phase shift mask and the method of manufacturing the same according to the present invention, by obtaining an appropriate combination of the refractive index and the absorption coefficient, the reflection type corresponding to the ultra-short ultraviolet light is obtained. Even in the case of a mask, it becomes feasible to configure a phase shift mask for using the super-resolution technique. Further, according to the method of manufacturing a semiconductor device according to the present invention, the method is very suitable for minimizing a pattern.
[Brief description of the drawings]
FIG. 1 is a schematic diagram illustrating an example of a schematic configuration of a phase shift mask according to the present invention.
FIG. 2 is a flowchart showing a procedure for manufacturing a phase shift mask according to the first embodiment of the present invention.
FIG. 3 is an explanatory diagram showing a specific example of an equal phase line.
FIG. 4 is an explanatory diagram showing a specific example of an equal reflectance line.
FIG. 5 is an explanatory diagram showing a distribution of a real part of a combined complex refractive index with respect to a Ru film thickness and a TaN film thickness for obtaining a halftone type phase shift mask.
FIG. 6 is an explanatory diagram showing the relationship between the total film thickness and the real part (n) and the reflectance (k) of the composite complex refractive index in the halftone type phase shift mask.
FIG. 7 is an explanatory diagram showing a relationship between a total film thickness, a Ru film thickness, and a TaN film thickness in a halftone phase shift mask.
FIG. 8 is an explanatory diagram showing a distribution of an imaginary part of a composite complex refractive index with respect to a Ru film thickness and a TaN film thickness for obtaining a halftone type phase shift mask.
FIG. 9 is an explanatory diagram showing an example in which a phase difference and a halftone reflectance with respect to the thickness of the Ru layer and the thickness of the TaN layer are arranged in a matrix.
FIG. 10 is an explanatory diagram showing an example in which phase differences with respect to the thickness of a Ru layer and the thickness of a Cr layer are arranged in a matrix.
FIG. 11 is an explanatory diagram showing an example in which halftone reflectances with respect to the thickness of a Ru layer and the thickness of a Cr layer are arranged in a matrix.
FIG. 12 is an explanatory diagram showing a light intensity distribution of a hole pattern having an opening of 30 nm above the mask (indicated by coordinates on a wafer) when the NA of the halftone phase shift mask is 0.25;
FIG. 13 is a schematic diagram illustrating a cross-sectional structure of a configuration example of a halftone phase shift mask.
FIG. 14 is a flowchart illustrating a procedure for manufacturing a phase shift mask according to the second embodiment of the present invention.
FIG. 15 is an explanatory diagram showing a specific example of an equal phase line.
FIG. 16 is a schematic diagram showing a cross-sectional structure of a configuration example (structure 1) of a Levenson-type phase shift mask.
FIG. 17 is an explanatory diagram showing an example of optimizing the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 1 in consideration of the multiple interference in the film.
FIG. 18 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 1;
FIG. 19 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 1;
FIG. 20 is an explanatory diagram (part 1) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having structure 1.
FIG. 21 is an explanatory diagram (part 2) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having structure 1.
FIG. 22 is an explanatory view (part 3) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having structure 1.
FIG. 23 is an explanatory view (part 4) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having structure 1.
FIG. 24 is a schematic view showing a cross-sectional structure of another configuration example (structure 2) of the Levenson-type phase shift mask.
FIG. 25 is an explanatory diagram showing an example of optimizing the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 2 in consideration of the intra-film multiple interference.
FIG. 26 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 2;
FIG. 27 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 2;
FIG. 28 is a schematic diagram showing a cross-sectional structure of still another configuration example (structure 3) of the Levenson-type phase shift mask.
FIG. 29 is an explanatory diagram showing an example of optimization of the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 3 in consideration of the multiple interference in the film.
FIG. 30 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 3;
FIG. 31 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 3;
FIG. 32 is a schematic diagram showing a cross-sectional structure of a configuration example (structure 4) in which a first region and a second region of a Levenson-type phase shift mask are flat.
FIG. 33 is an explanatory diagram showing an example of optimization of the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 4 in consideration of the multiple interference in the film.
FIG. 34 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 4;
FIG. 35 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 4;
FIG. 36 is an explanatory view (No. 3) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having structure 4;
FIG. 37 is an explanatory view (No. 4) illustrating an example of a film forming procedure of the Levenson-type phase shift mask having the structure 4;
FIG. 38 is a schematic diagram showing a cross-sectional structure of another flat configuration example (structure 5) of the Levenson-type phase shift mask.
FIG. 39 is an explanatory diagram showing an example of optimization of the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 5 in consideration of the multiple interference in the film.
FIG. 40 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 5;
FIG. 41 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 5;
FIG. 42 is a schematic view showing a cross-sectional structure of still another flat configuration example (structure 6) of the Levenson-type phase shift mask.
FIG. 43 is an explanatory view showing an example of optimizing the phase and the reflectance (using the Mo film thickness as a parameter) in the Levenson-type phase shift mask having the structure 6 in consideration of the multiple interference in the film.
FIG. 44 is an explanatory diagram showing a light intensity distribution in a Levenson-type phase shift mask having a structure 6;
FIG. 45 is an explanatory diagram showing a phase difference (displayed on a wafer) of a Levenson-type phase shift mask having a structure 6;
FIG. 46 is an explanatory diagram showing a light intensity distribution (NA = 0.25) with respect to a TaN width on a mask (displayed on a wafer) in a Levenson-type phase shift mask having a structure 5;
FIG. 47 is an explanatory diagram showing a light intensity distribution (NA = 0.25) with respect to a TaN width on a mask (displayed on a wafer) in a conventional binary mask.
FIG. 48 is an explanatory diagram showing a light intensity distribution (NA = 0.30) with respect to a TaN width on a mask (displayed on a wafer) in a Levenson-type phase shift mask having a structure 5;
[Explanation of symbols]
10, 10 ': phase shift mask, 11: reflective multilayer substrate, 12a: first region, 12b: second region, 13: buffer film, 14: absorption film

Claims (8)

A phase shift mask of ultra-short ultraviolet light used to transfer a desired pattern on an object to be exposed by reflecting the ultra-short ultraviolet light,
A reflective multilayer substrate that reflects ultra-short ultraviolet light, and a first region and a second region formed on the reflective multilayer substrate;
The thickness and the complex refractive index of the film formed in the first region and the film formed in the second region are reflected light of ultra-short ultraviolet light in the first region and ultra-short ultraviolet light in the second region. A phase shift mask for ultra-short ultraviolet light, wherein the phase shift mask is set so as to generate a predetermined phase difference with the reflected light. In addition to the predetermined phase difference, the first region is formed such that the reflectance of the ultra-short ultraviolet light in the first region is substantially equal to the reflectance of the ultra-short ultraviolet light in the second region. 2. The phase shift mask for ultra-short ultraviolet light according to claim 1, wherein each film thickness and each complex refractive index of the film and the film forming the second region are set. 2. The ultra-short ultraviolet light phase shift mask according to claim 1, wherein the first region and the second region are formed so that the film thickness of the second region is substantially equal to the film thickness of the second region. 2. The ultrashort ultraviolet light according to claim 1, wherein at least one of the formation film of the first region and the formation film of the second region has a stacked structure including a plurality of materials. 3. Phase shift mask. A method for manufacturing a phase shift mask for ultra-short ultraviolet light, comprising: a reflective multilayer film substrate that reflects ultra-short ultraviolet light; and a first region and a second region formed on the reflective multilayer film substrate. ,
Any complex refractive index for the ultra-short ultraviolet light, and for any film thickness of the film formed on the reflective multilayer film substrate, the reflected light of the ultra-short ultraviolet light obtained from the complex refractive index and the film thickness Determine the phase and reflectivity of
On the basis of the specified phase and reflectance, the first and second ultra-short ultraviolet light reflected in the first region and the reflected light of ultra-short ultraviolet light in the second region, a predetermined phase difference is generated, the second A method of manufacturing a phase shift mask for ultra-short ultraviolet light, wherein each film thickness and each complex refractive index of a film formed in one region and a film formed in the second region are selected. 2. The method according to claim 1, wherein the thicknesses of the film formed in the first region and the film formed in the second region are selected in consideration of variations in phase difference and reflectance caused by multiple interference in the film. 6. The method for manufacturing a phase shift mask for ultra-short ultraviolet light according to claim 5. 6. The phase shift mask for ultra-short ultraviolet light according to claim 5, wherein the selected complex refractive index and film thickness are obtained by a combined complex refractive index and total film thickness realized by a laminated structure composed of a plurality of materials. Production method. A method of manufacturing a semiconductor device including a lithography step of transferring a desired pattern onto an object to be exposed using a phase shift mask of ultra-short ultraviolet light,
Any complex refractive index for the ultra-short ultraviolet light, and for any film thickness of the film formed on the reflective multilayer film substrate, the reflected light of the ultra-short ultraviolet light obtained from the complex refractive index and the film thickness Determine the phase and reflectivity of
Based on the specified phase and reflectance, reflected light of ultra-short ultraviolet light in the first region on the reflective multilayer substrate and reflected light of ultra-short ultraviolet light in the second region on the reflective multilayer substrate The thickness and the complex refractive index of each of the film formed in the first region and the film formed in the second region are selected so that a predetermined phase difference occurs in
The first region forming film and the second region forming film are formed on the reflective multilayer substrate at the selected complex refractive index and thickness, respectively, and the first region and the first region are formed on the reflective multilayer substrate. Forming a phase shift mask for ultra-short ultraviolet light comprising the second region,
A method for manufacturing a semiconductor device, comprising transferring a desired pattern onto an object to be exposed using the constituted ultrashort ultraviolet light phase shift mask.

Images