JP4208565B2 - Interferometer and measurement method having the same - Google Patents

Description

となる。ここで波面振幅ａは十分小さいとして波面振幅ａの１次の項までの近似で表している。周波数ｆにおけるコントラストをＭ（ｆ）とすると取得される干渉縞画像の強度Ｉ_mean（ｘ，ｔ）は、
I_meas(x,t)= I₀(1+sin(ωt)+M(f)a cos(2πfx)cos(ωt))
となる。フリンジスキャンは干渉縞変化のｃｏｓ変調成分、ｓｉｎ変調成分を摘出して位相を産出するため、計算される位相は、

つまり実波面の波面振幅ａが周波数ｆにおけるコントラストＭ（ｆ）だけ減少して計算されることになる。この為、干渉縞の位相波面を高精度に測定することが困難になってくる。
【００１３】
本発明は被検面の面形状や物体のホモジニティ等を高精度に測定することができる干渉装置及びそれを有する測定方法の提供を目的とする。
【００１４】
【課題を解決するための手段】
請求項１の発明の干渉装置は、被検物の表面形状またはホモジニティを測定する干渉装置において、光源からの光束を前記被検物を介する被検光束と参照面を介する参照光束とに分割し、それらの光束を合成する光学手段と、
前記被検光束と前記参照光束とを合成することで形成される干渉縞を撮像する撮像手段と、
この干渉装置で補正用サンプルを測定することにより予め得た補正値に基づいて前記撮像手段の出力を補正して、前記被検物の表面形状またはホモジニティを算出する処理手段と、を備え、
前記補正用サンプルの表面には、空間周波数の異なる複数の模様または異なる空間周波数を持つ模様が刻まれており、
前記補正値は、この干渉装置で前記補正用サンプルを測定することにより得た前記模様の振幅の測定結果と、この干渉装置よりも前記補正用サンプルの表面形状を正確に測定できる装置で前記補正用サンプルを測定することにより得た前記模様の振幅の測定結果と、を比べることで空間周波数毎に得られるものであり、
前記処理手段は、空間周波数の増加にともなって発生する前記撮像手段で撮像される干渉縞から算出される位相波面の振幅劣化を、前記補正値に基づいて空間周波数毎に補正することを特徴としている。
請求項２の発明は請求項１の発明において、前記補正用サンプルは、前記空間周波数の異なる複数の模様を形成した複数の基板を有することを特徴としている。
請求項３の発明の測定方法は、被検物の表面形状またはホモジニティを干渉装置で測定する測定方法において、
空間周波数の異なる複数の模様または異なる空間周波数を持つ模様が表面に刻まれている補正用サンプルを前記干渉装置で測定することにより補正値を得る工程と、
空間周波数の増加にともなって発生する前記干渉装置の撮像手段で撮像された干渉縞から算出される位相波面の振幅劣化を、前記補正値に基づいて空間周波数毎に補正する工程と、を有し、
前記補正値は、前記干渉装置で前記補正用サンプルを測定することにより得た前記模様の振幅の測定結果と、前記干渉装置よりも前記補正用サンプルの表面形状を正確に測定できる装置で前記補正用サンプルを測定することにより得た前記模様の振幅の測定結果と、を比べることで空間周波数毎に得られるものであることを特徴としている。
【００１５】
【発明の実施の形態】
以下本発明の各実施形態について、添付図面を参照しながら説明する。
【００１６】
（実施形態１）
本発明の実施形態１について説明する。
【００１７】
実施形態１は補正用サンプル（サンプル）を用意して補正用サンプルの被検面の面形状を機械的に測定する形状測定装置として、例えば３次元形状走査型装置（図４参照）と干渉測定装置（図１参照）で求め、双方で得た測定値より干渉測定装置で求めた測定値を補正する為の補正値を算出する。即ち干渉測定装置で得た被測定サンプルの測定値に補正値をかけ、これより被測定サンプルの被検面の面形状情報を求めている。ここで、補正用サンプルの被検面の面形状を機械的に測定する例を示したが、これはこの限りではなく、補正用サンプルの被検面の面形状を正確に測定できる方法であれば、特に機械的な方法である必要は無い。さらに、もし、補正用サンプルの被検面の形状が正確に分かっている場合は、特に測定の必要は無く、そのサンプルの被検面の形状と、本実施形態の干渉測定装置による面形状の測定結果との両者から補正値を算出しても良い。
【００１８】
以下に被測定サンプルの被検面が平面の場合の測定と補正値を算出し、測定値の補正方法の流れを具体的に説明する。
【００１９】
図１は実施形態１で用いる干渉測定装置の要部概略図である。
【００２０】
まず被測定サンプルを図１の干渉測定装置で測定する方法を説明する。光源（光源手段）１を射出した光束はハーフミラー２を透過し、２軸ティルトステージ３上のＴＦ基板５に至る。２軸ティルトステージ３は制御コンピュータ９からの指令により高精度にティルトし、後述する参照光束と被検光束の波面合わせが可能となっている。また２軸ティルトステージ３上には縞走査法用の光路長差変化手段としての圧電素子（ＰＺＴアクチュエータ）４を介し、ＴＦ基板５が設置されている。
【００２１】
ＴＦ基板５は最終面５ａ以外には光源１からの光束の波長に対する反射防止膜を施す或いはクサビ角を設けており、これによって最終面５ａからのみ光束の一部が反射する。最終面５ａを透過した光束は被測定サンプル（サンプル）６の被検面６ａで反射する。以下ＴＦ基板５の最終面５ａで反射される光束を参照光束、最終面５ａを透過する光束を被検光束と称す。
【００２２】
サンプル６の被検面６ａで反射した被検光束はＴＦ基板５で参照光束と合波され、干渉してハーフミラー２で反射され、結像光学系７とピンホールＰＨを介して拡散板１０上で干渉縞を形成する。
【００２３】
拡散板１０は、駆動手段ＭＯによって、回転する事でスペックル等の光学ノイズを平均化するために用いている。拡散板１０で拡散された干渉縞は結像光学系１１によりＣＣＤカメラ（撮像手段）８上に伝達され、撮像された干渉縞画像データは制御コンピュータ（処理系）（処理手段）９に転送される。ここで結像光学系１１と拡散板１０を省略して拡散板１０が位置するところにＣＣＤカメラ８を配置し、直接に干渉縞画像データを検出しても良い。この方法は光学系の影響による振幅低下が少ない。
【００２４】
光源１から被検面６ａを介して、拡散板１０又は撮像手段８に至る系で干渉計を構成している。又、光源１から被検面６ａを介し、撮像手段８に至る光路中に設けている光学部材は光学手段の一要素を構成している。
【００２５】
制御コンピュータ９ではＰＺＴアクチュエータ４を走査した際の複数の干渉縞画像データを取り込み、所謂、縞走査法により干渉縞の位相を算出しており、これによってサンプル６の被検面６ａの面形状を測定している。そして制御コンピュータ９に組み込まれた振幅補正ユニット（補正手段）１２でサンプル６の面形状の測定値を補正値によって補正する。ここで振幅補正ユニット１２は制御コンピュータ９と分離していても良い。この振幅補正ユニット１２で計算される測定値の補正方法について図２を用いて説明する。
【００２６】
図２において、Ｍ１はサンプル６の面形状の測定値（干渉縞データ）の波面分布（位相波面分布）を表す。この測定値Ｍ１に対し、最小二乗法等を用いて多項式フィッティングを行うことにより、多項式成分Ｚ１と多項式残差成分（残差成分）Ｒ１に分離する。ここで多項式としてはＺＥＲＮＩＫＥ多項式等を用いている。
【００２７】
残差成分Ｒ１に対し２次元フーリエ変換を行い周波数分布ＲＦ１を得る。ここで残差成分Ｒ１を用いるのは、波面瞳端部の極端な変化による不要周波数生成物を抑えるためである。
【００２８】
図２において、位相波面の振幅劣化は前記周波数分布ＲＦ１と同一スケールの空間周波数上に、干渉測定装置で得られた位相波面の振幅分布を作成したものである。図２中ＲＦ２は振幅劣化した位相波面の補正後の残差波面の周波数分布を表し、周波数分布ＲＦ１と補正係数分布Ｒｄｃにより、
ＲＦ２＝ＲＦ１／Ｒｄｃ
と表される。
このように本実施例では処理系９は撮像手段８で得られた位相波面の位相劣化ＲＦ１を補正係数分布（補正値）Ｒｄｃに基づいて補正している。
【００２９】
この補正係数分布Ｒｄｃの求め方は後で詳細を説明する。この補正は位相波面の振幅劣化で補正しようとする領域の波面収差の周波数成分振幅が１ｒａｄより十分小さい場合に適用される。補正後の周波数分布ＲＦ２に対し逆フーリエ変換を行い、実空間上の残差波面Ｒ２を得る。これに前記分離した多項式成分Ｚ１を加えることで、波面収差分布Ｍ２を求め、これより測定値Ｍ１に対する振幅劣化の補正が完了し、高周波域まで、測定誤差の少ない波面計測値Ｍ２を得ることを可能としている。
【００３０】
次に前記補正係数分布Ｒｄｃの求め方について図３の流れ（フローチャート）に沿って説明する。
【００３１】
ステップＡ：「補正用サンプルを用意する」
図５、図６、図７は補正用サンプルの概略図である。図５は補正用サンプルＳ１の被検面を上から見た図、Ｃ１ａ、Ｃ１ｂは補正用サンプルＳ１の直線Ｓ１ａ、Ｓ１ｂの断面における振幅分布を表す。図６は補正用サンプルＳ２被検面を上から見た図、Ｃ２ａ、Ｃ２ｂは補正用サンプルＳ２の直線Ｓ２ａ、Ｓ２ｂの断面における振幅分布を表す。図７のＳ３（Ｓ３ａ、Ｓ３ｂ・・・Ｓ３ｘ）は複数の被測定物（基板）に異なる周波数を刻んだサンプルでＣ３ａ、Ｃ３ｂ・・・Ｃ３ｘはサンプルＳ３ａ、Ｓ３ｂ・・・Ｓ３ｘそれぞれの断面図を表す。このような補正用サンプルＳ１，Ｓ２，Ｓ３を複数用意してそれぞれ干渉測定装置と図４に示す３次元形状走査型計測装置で補正用サンプルＳ１，Ｓ２，Ｓ３の被検面の面形状を測定する。
【００３２】
ステップＢ：「補正用サンプルＳ１，Ｓ２，Ｓ３を干渉測定装置で測定する」
干渉測定装置による補正用サンプルＳ１，Ｓ２，Ｓ３の測定は図１を参照した前記説明と同様の方法、所謂縞走査法により補正用サンプルＳ１，Ｓ２，Ｓ３の被検面の面形状を測定する。測定方法の詳細については既に説明したので省略する。ここでは振幅補正ユニット１２による位相波面の振幅補正は行わない。つまり補正係数分布Ｒｄｃ＝１として測定する。
【００３３】
ステップＣ：「補正用サンプルＳ１，Ｓ２，Ｓ３を３次元形状走査型計測装置で測定する」
接触式の３次元形状走査型測定装置による補正サンプルの面形状の測定について図４を参照しながら説明する。
【００３４】
図４は補正用サンプル面形状を測定する接触式の３次元形状走査型測定装置の概略図である。サンプル２２はＸＹステージ２３の上に設置されている。ここでＸＹステージ２３は制御コンピュータ２５により高精度にＸＹ独立の駆動が可能となっている。
【００３５】
プローブ２１はＸＹＺステージ２４に設置されていて、プローブ２１がサンプル２２の被検面２２ａ上を−定の圧力を加えながらスキャンするように制御コンピュータ２５はＸＹＺステージ２４を制御している。このときＸＹＺステージ２４の位置座標を制御コンピュータ２５で計算して求めることでサンプル２２の被検面２２ａの面形状を算出する。ここで非接触式の３次元形状走査型測定装置で補正サンプルＳ１，Ｓ２，Ｓ３を測定しても全く問題ない。
【００３６】
ステップＤ：「ステップＢ，Ｃの測定結果より補正係数ＲｄｃＨ、ＲｄｃＶを求める」
次にステップＢとステップＣの測定結果よりＣＣＤカメラ８の水平方向と垂直方向の補正係数ＲｄｃＨとＲｄｃＶの求め方について図８を参照しながら説明する。
【００３７】
被検面の切断面がＣＣＤカメラ８の水平方向の場合、計測した結果から標本化間隔に対するある空間周波数ｋｘを計算する。以上で被検面上の空間周波数ｋｘにおける振幅Ｖ_ref（ｋｘ）を得る。
【００３８】
以下、ｈが標本化間隔のナイキスト周波数（標本化周期の２倍の逆数）まで計算する。３次元形状走査型測定装置の場合も同様に空間周波数ｋｘの振幅Ｖ_ref（ｋｘ）を求める。
【００３９】
これら一連の計算結果を図８に示した。被検面上の空間周波数の振幅を測定することにより離散データＶ（ｋｘｉ）が得られる。図８の実線は３次元形状走査型計測装置の振幅Ｖ_ref（ｋｘ）である。点線は干渉測定装置の振幅Ｖ（ｋｘ）を表していて点線の振幅は実線の振幅と比べて振幅が劣化している。即ち３次元形状走査型装置の方が干渉装置よりも補正用サンプルの表面形状を正確に測定できる。干渉装置における振幅劣化の原因は先に「解決しようとする課題」で述べたとおりである。
【００４０】
このとき補正係数Ｒｄｃ０は、
Ｒｄｃ０＝Ｖ（ｋｘ）／Ｖ_ref（ｋｘ）
となり、干渉測定装置の測定値を空間周波数毎に補正係数Ｒｄｃ０で割ることによって補正が完了する。このように本実施例では、空間周波数の増加に伴って発生する撮像手段８で撮像される干渉縞の振幅分布を補正係数（補正値）に基づいて補正している。
即ち本実施例では補正係数（補正値）Ｒｄｃ０を干渉測定装置で補正用サンプル（Ｓ１、Ｓ２、Ｓ３）を測定することにより得た模様の振幅の測定結果Ｖ（Ｋｘ）と、干渉測定装置よりも精度の良い３次元形状走査装置で補正用サンプル（Ｓ１、Ｓ２、Ｓ３）を測定することにより得た測定結果Ｖ _ref （ｋｘ）と、を比べることで空間周波数毎に得ている。
【００４１】
３次元形状走査型測定装置と干渉測定装置の標本間隔が異なる場合は振幅Ｖ_ref（ｋｘ）と振幅Ｖ（ｋｘ）に対して最小二乗法等により多項式関数等でフィッティングして補正係数を求めればよい。
【００４２】
ここで本実施形態のように干渉縞を形成する瞳結像系にインコヒーレント結像が含まれる干渉測定装置における補正係数は、
F_InchOpt(x,k₀)=(2 cos^-1(k₀x)- sin(2 cos^-1(k0x)))/π
となり、この式は理想光学系のインコヒーレント光学系の振幅劣化を表している。
【００４３】
また本実施形態では被検レンズの瞳結像系としてインコヒーレント結像を用いているが、コヒーレント結像の干渉計における補正係数には次式で表されるＭｏｆｆａｔ関数を用いればよい。
【００４４】
Moffat(x,k₀,k₁,k₂)=k₀/(1+(x/k₁)²)^k2
ここでｘは空間周波数、ｋ₀，ｋ_l，ｋ₂はパラメータである。
【００４５】
以上でＣＣＤカメラ８の水平方向に対する補正係数ＲｄｃＨの算出が終了する。次に同様の手続きにより垂直方向に対する補正係数ＲｄｃＶを計算すれば、干渉測定装置の水平方向と垂直方向の補正係数ＲｄｃＨ、ＲｄｃＶがそれぞれ求められる。
【００４６】
ステップＥ：「ステップＤの結果より補正係数分布Ｒｄｃを求める」
前記干渉測定装置の水平方向と垂直方向のぞれぞれの補正係数ＲｄｃＨ，ＲｄｃＶにフィッティングを行い、関数化すれば補正係数分布Ｒｄｃの作成が容易になる。フィッティングした水平方向、垂直方向の補正係数をＲｄｃＨ（ｋｘ），ＲｄｃＶ（ｋｙ）とすると、空間周波数（ｋｘ，ｋｙ）上の補正係数分布Ｒｄｃ（ｋｘ，ｋｙ）は、
Rdc(kx,ky)= RdcH(kx)×RdcV(ky)
と表される。
【００４７】
以上で図２中の補正係数分布Ｒｄｃを求めることができる。図１中の振幅補正ユニット１２はこの補正係数分布Ｒｄｃを用いて測定値に振幅補正計算を行う。
【００４８】
また被検出面の形状として、平面の形状について説明してきたがホモジニティ測定（均質測定）への適用も可能である。ホモジニティ測定のとき図１中のサンプル６はＲＦ基板（反射基準板）として、ＲＦ基板６とＴＦ基板５の問に被測定物１３を置き透過波面を測定する。測定法としては非研磨面のまま測定するオイルオンプレート法、或いは研磨面状態で測定する研暦法等が適用できる。尚、本実施例において、被検物６の表面形状またはホモジニティを干渉装置で測定する測定方法としては、空間周波数の異なる複数の模様または異なる空間周波数を持つ模様を有する補正用サンプルＳ１〜Ｓ３を干渉装置で測定することにより補正値（補正係数）を得る工程と、空間周波数の増加にともなって発生する干渉装置の撮像手段９で撮像された干渉縞の振幅劣化を、前記補正値に基づいて補正する工程と、を利用している。
【００４９】
（実施例２）
次に本発明の実施形態２について説明する。
【００５０】
実施形態２は実施形態１の被測定サンプルが並行平板であったのに対して被測定サンプルの形状として球面ミラーを用いている。この球面ミラーの反射面上に図５、図６に示すパターンが設けられている。この場合の干渉測定装置での測定方法について図９を用いて述べる。
【００５１】
図９において、光源１を射出した光束はハーフミラー２を透過し、ＸＹＺステージ３上に設けた集光レンズ５に至る。ＸＹＺステージ３は制御コンピュータ９からの指令により高精度にＸＹＺ方向に独立の駆動が可能となっている。またＸＹＺステージ３上には縞走査法用の圧電素子（ＰＺＴアクチュエータ）４を介し、集光レンズ３５が設置されている。
【００５２】
ここで集光レンズ３５は最終面３５ａの曲率半径と最終面３５ａ−焦点１０間の距離が等しい、所謂ＴＳレンズである。ＴＳレンズ３５は最終面３５ａ以外には光源１の波長に対する反射防止膜を施す事で、最終面３５ａからのみ光束の一部が反射するようにしている。最終面３５ａを透過した光束はサンプル６の被検面６ａで反射する。以下ＴＳレンズ３５の最終面３５ａで反射される光東を参照光束、最終面３５ａを透過する光束を被検光束と称す。
【００５３】
ここでサンプル６は制御コンピュータ９により制御可能なＸＹＺステージ１１上に設けられ、サンプル６の被検面６ａの曲率中心とＴＳレンズ３５による集光点１０が一致するようにＸＹＺ方向の調整がなされている。
【００５４】
サンプル６の被検面６ａで反射した被検光束はＴＳレンズ３５で参照光束と合波し、互いに干渉し、ハーフミラー２で反射され、結像光学系７とピンホールＰＨを介して、ＣＣＤカメラ８上で干渉縞を形成する。ＣＣＤカメラ８で撮像された画像データは制御コンピュータ９に転送される。
【００５５】
制御コンピュータ９ではＣＣＤカメラ８からＰＺＴアクチュエータ４を走査した際の複数の干渉縞画像データを取り込み、所謂、縞走査法により干渉縞の位相を算出し、位相波面（測定値）を求めている。そして制御コンピュータ９に組み込まれた振幅補正ユニット１２で測定値を補正係数で補正する。ここで振幅補正ユニット１２は制御コンピュータ９と分離していても全く問題ない。
【００５６】
ここで用いるサンプル６はＴＳレンズ３５による集光点１０とサンプル６の被検面６ａの曲率中心が一致する球面に図５や図６や図７に示すような模様を刻んだものである。
【００５７】
３次元形状走査型測定装置によるサンプル６の被検面測定と振幅劣化の補正については実施形態１で述べたので省略する。また、ここでは実施形態１のような拡散板１０と結像光学系１１のない光学系を示したがこれらの部材があっても構わない。
【００５８】
（実施例３）
次に本発明の実施形態３について図１０を用いて説明する。実施形態３は被検物として透過物体を用いている。図１０において光源１を射出した光束はハーフミラー２を透過し、ＸＹＺステージ３上に設けた集光レンズ５に至る。ＸＹＺステージ３は制御コンピュータ９からの指令により高精度にＸＹＺ方向に独立の駆動が可能となっている。またＸＹＺステージ３上には縞走査法用の圧電素子（ＰＺＴアクチュエータ）４を介し、集光レンズ５が設置されている。
【００５９】
ここで集光レンズ３５は最終面３５ａの曲率半径と最終面３５ａ−焦点１０間の距離が等しい、所謂ＴＳレンズである。ＴＳレンズ５は最終面３５ａ以外には光源１の波長に対する反射防止膜を施す事で、最終面３５ａからのみ光束の一部が反射するようにしている。以下ＴＳレンズ３５の最終面３５ａで反射される光束を参照光束、最終面３５ａを透過する光束を被検光束と称す。
【００６０】
集光レンズ３５の焦点１０は被検レンズ１４の物体面と一致するようにＺステージの調整がなされており、被検レンズ１４を透過した被検光束は、被検レンズ１４の像面１５上で集光した後、球面のＲＳミラー６により反射される。
【００６１】
ここでＲＳミラー６はＴＳレンズ５と同様に制御コンピュータ９により制御可能なＸＹＺステージ１１上に設けられ、ＲＳミラー６の曲率中心６ａと像側焦点１５が一致するようにＸＹＺステージ１１方向の調整がなされている。
【００６２】
ＴＳレンズ３５の最終面３５ａにて反射された参照光束と、ＲＳミラー６により反射された被検光束は、ＴＳレンズ３５を介して、合波し、互いに干渉して同一光路となりハーフミラー２で反射され、結像光学系７とピンホールＰＨを介してＣＣＤカメラ８上で干渉縞を形成する。ＣＣＤカメラ８で撮像された干渉縞画像データは制御コンピュータ９に転送される。制御コンピュータ９ではＰＺＴアクチュエータ４を走査した際の複数の干渉縞画像データを取り込み、所謂、縞走査法により干渉縞の位相を算出して被検レンズの透過波面を求める。
【００６３】
そして制御コンピュータ９に組み込まれた振幅補正ユニット１２で測定値を補正係数で補正する。ここで振幅補正ユニット１２は制御コンピュータ９と分離していても全く問題ない。
【００６４】
次に図１０の干渉測定装置における補正用サンプルの測定について説明する。
【００６５】
まず図１０中の補正用サンプル１３を被検レンズ１４に取り付ける、或いは被検レンズ１４無しの状態で工具等により同様の位置に配置する。ここで用いる補正用サンプル１３は実施形態２と同様にＴＳレンズ５による集光点１０と補正用サンプル１３の被検面１３ａの曲率中心が一致する球面に図５や図６や図７に示す模様を刻んだものである。このときＸＹＺステージ１１は補正用サンプル１３の被検面１３ａ曲率中心とＴＳレンズ５の集光点１０が一致するように制御コンピュータ９により制御されている。
【００６６】
補正用サンプル１３の被検面１３ａで反射した被検光束はＴＳレンズ３５で参照光束と合波され、互いに干渉し、ハーフミラー２で反射され、結像光学系７とピンホールＰＨを介してＣＣＤカメラ８上で干渉縞を形成する。ＣＣＤカメラ８で撮像された画像データは制御コンピュータ９に転送される。制御コンピュータ９ではＰＺＴアクチュエータ４を走査した際の複数の干渉縞画像データを取り込み、所謂、縞走査法により干渉縞の位相を算出して干渉測定装置で測定したときの振幅劣化を求める。
【００６７】
３次元形状走査型測定装置による補正用サンプル１３の被検面測定と振幅劣化の補正については実施形態１で述べたので省略する。またここでは実施形態１のような拡散板１０と結像光学系１１のない光学系を示したがこれらの部材があっても構わない。
【００６８】
（実施形態４）
次に本発明の実施形態４について説明する。
【００６９】
本実形態４は振幅補正機能を有する面形状測定装置（実施形態１，２）及び透過波面形状測定装置（実施形態３）を露光装置の投影レンズの製造に適用したものである。単レンズ或いはホモジニティ測定を実施形態１，２の測定装置で行い、組立後のレンズの波面を実施形態３の測定装置で測定し調整を行う。これにより高周波まで精度の良いレンズを製造している。
【００７０】
以上、説明した各実施形態をまとめると次のとおりである。第１の干渉装置は、被検物を介した光を用いて干渉縞を形成する干渉装置である。そして補正用サンプルの形状と前記補正用サンプルの形状を前記干渉装置で測定した測定値とを用いて求めた補正値に基づいて、前記干渉縞から得られる位相差分布を補正する補正手段を有する。
【００７１】
前記補正手段が、前記位相差分布の第1周波数成分と前記第1周波数成分より空間周波数の高い第２周波数成分のうち、前記第1周波数成分のゲインを略一定に保ち、前記第２周波数成分のゲインを補正する。
【００７２】
又、前記補正手段が、前記第２周波数成分のゲインを増幅する。
【００７３】
前記位相差分布に基づいて、前記被検物の形状を導く光学手段を有する。
【００７４】
前記干渉縞を形成する２つの光束の光路長差を変化させる光路長差変化手段を有する。
【００７５】
第２の干渉装置は、光源手段から射出された光束より、被検物を介した被検光束と参照面を介した参照光束とを形成し、双方を合波し干渉波面を形成する光学手段と、該被検光束と該参照光束の光路長差を変化させる光路長差変化素子と、該干渉波面に基づく干渉縞を撮像する撮像素子と、該撮像素子で撮影される干渉縞から該干渉波面の位相差分布を計算する処理系とを有する干渉装置である。そして該位相差分布を、補正用サンプルの被検面の凹凸形状を機械的に測定する形状測定装置で得た値と該干渉装置で該補正用サンプルを測定した値とを用いて求めた補正値で補正する補正手段を有する。
【００７６】
前記被検物の被検面、又は被検物がレンズ系であるときはその瞳と前記撮像素子は共役関係である。
【００７７】
前記光学手段は、干渉縞を拡散板上に形成し、更に該拡散板上に形成された干渉縞を撮像素子上に再結像する。
【００７８】
前記補正用サンプルは、被検面に複数の異なる空間周波数の模様を刻んだサンプル、或いは複数の被測定物に異なる空間周波数の模様を刻んだサンプルより成る。
【００７９】
前記補正手段は、周波数空間で補正値を求めている。
【００８０】
前記補正値は干渉縞を形成する結像系がインコヒーレント系の場合、波面の空間周波数をｆ、空間座標をｘ、２πｆ＝ｋ₀とするとき、
F_InchOpt(x,k₀)=(2 cos^-1(k₀x)- sin(2 cos^-1(k₀x)))/π
という補正係数を用いる。
【００８１】
又は、前記補正値は、干渉縞を形成する結像系がコヒーレント系の場合、ｋ₀、ｋ₁、ｋ₂をパラメータｘを空間座標とするとき、
Moffat(x,k₀,k₁,k₂)=k₀/(1+(x/k₁)²)^k2
という補正係数を用いる。
【００８２】
前記被検物のホモジニティを求める。
【００８４】
第３の干渉装置は、光源手段から射出された光束の一部を被検面で反射或いは被検物を透過させた後に反射面で反射させることによって得られる被検光束と、該光源手段から射出された光束の一部を参照面によって反射させて得られる参照光束を形成する光学手段と、前記被検光束と前記参照光束の光路長差を変化させる光路長差変化素子と、前記被検光束と前記参照光束を干渉させて得られる干渉縞を撮像する撮像素子と、前記光学手段は、前記参照光束と前記被検光束により形成される干渉縞を前記撮像素子に導光しており、前記撮像素子で撮影される前記干渉縞から前記被検光束と前記参照光束の位相差分布を計算する処理系とを有する干渉装置である。そして前記干渉縞の空間位相分布の振幅劣化を補正する補正値を算出する補正手段を有し、前記補正手段は、該補正値を被検面の凹凸を接触或いは非接触で測定する形状測定装置と前記干渉装置で補正用サンプルを測定することにより求める。
【００８５】
前記被検面或いは前記被検物の瞳と前記撮像素子は共役関係である。
【００８６】
前記補正用サンプルは、被検面に複数の異なる空間周波数の模様を刻んだサンプル、或いは複数の被測定物に異なる空間周波数の模様を刻んだサンプルより成る。
【００８７】
前記補正手段は、周波数空間で補正値を求めている。
【００８８】
第４の干渉装置は、被検物を介した被検波面と、参照波面とを合波し、位相波面に基づく干渉縞データを撮像手段に形成する干渉計と、該干渉計からの干渉縞データより該被検物を介した位相波面を演算し求める処理系とを有する干渉装置である。そして該処理系は、補正用サンプルを用いたときの該干渉計で得られる干渉情報の位相波面が空間周波数値（ｋｘ）による振幅劣化するときの振幅をＶ（ｋｘ）、該補正用サンプルの凹凸形状を機械的に測定する形状測定装置で得られる振幅をＶ_ref（ｋｘ）としたとき、
【００８９】
【数１】

【００９０】
より求めた補正係数分布Ｒｄｃを用いて該干渉縞データを補正している。
【００９１】
第５の干渉装置は、被検物を介した被検波面と、参照波面とを合波し、位相波面に基づく干渉縞データを撮像手段に形成する干渉計と、該干渉計からの干渉縞データより該被検物を介した位相波面を演算し求める処理系とを有する干渉装置である。そして該処理系は、該干渉計で得られる干渉縞データ（Ｍ１）に基づく周波数分布（ＲＦ１）を補正係数分布（Ｒｄｃ）で補正した周波数分布（ＲＦ２）を用いて、波面収差分布（Ｍ２）を求めており、このとき補正係数分布（Ｒｄｃ₀）を該干渉計で得られる干渉情報の位相波面が空間周波数値（ｋｘ）による振幅劣化するときの振幅をＶ（ｋｘ）、該補正用サンプルの凹凸形状を機械的に測定する形状測定装置で得られる振幅をＶ_ref（ｋｘ）としたとき、
【００９２】
【数２】

【００９３】
より求めている。
【００９４】
【発明の効果】
本発明によれば干渉装置で測定された干渉縞の振幅劣化によって困難であった高周波成分の被検面の透過波面或いは被検面形状誤差を補正し、高精度な測定を行うことができる干渉装置及びそれを有する測定方法を達成することができる。
【図面の簡単な説明】
【図１】 本発明の実施形態１の概略構成図である。
【図２】 本発明に係る振幅劣化の補正手続きを示すフローチャートである。
【図３】 本発明に係る補正係数分布を求める計算を示すフローチャートである。
【図４】 本発明に係る接触式３次元形状走査型計測装置の概略構成図である。
【図５】 本発明に係る被後面測定用のサンプル例である。
【図６】 本発明に係る被検面測定用のサンプル例である。
【図７】 本発明に係る複数枚の被検面測定用のサンプル例である。
【図８】 本発明に係る本発明に係る被検面の断面をフーリエ変換した振幅のグラフである。
【図９】 本発明の実施形態２の概略構成図である。
【図１０】 本発明の実施形態３のを示す概略構成図である。
【図１１】 従来の干渉測定装置の概略構成図である。
【符号の説明】
Ｅ_test（ｘ，ｔ）：被検光束複素振幅
Ｅ_ref（ｘ，ｔ）：参照光束模索振幅
Ｅ０：電場振幅
ａ：波面振幅
ｆ：波面空間周波数
ω：フリンジスキヤン周波数
Ｍ（ｔ）：空間周波数ｆにおけるコントラスト
ｋ：干渉縞空間周波数
Ｖ：被検面振幅
ｋｘ：水平方向空間周波数
ｋｙ：垂直方向空間周波数
Ｆ_inch0pt：インコヒーレント光学系のフィッテイング関数，
Ｒｄｅ０：１次元の補正係数関数
Ｍ１：測定結果干渉縞位相
Ｚ１：測定結果干渉縞位相の多項式成分
Ｒ１：測定結果干渉縞の位相残差成分
ＲＦ１：測定結果干渉の縞位相残差成分の空間周波数分布
Ｒｄｃ：干渉計振幅劣化の空間周波数の補正係数分布
ＲＦ２：補正結果透過波面の収差残差成分の空間周波数分布
Ｒ２：補正結果透過波面の収差残差成分
Ｍ２：補正結果透過波面の収差分布
ＮＡ：測定光束のＮＡ
ｒ：測定光束半径[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an interference device capable of measuring the surface shape of a test surface, the homogeneity of an object, and the like with high accuracy.And measuring method having the sameAbout.
[0002]
In particular, the present invention relates to an interference apparatus suitable for measuring a shape error on a test surface or a transmitted wavefront aberration of a test object, for example, a test lens, from a two-dimensional interference fringe intensity distribution.
[0003]
[Prior art]
Conventionally, an interference measuring device (interference device) using interference of light has been used for measuring the surface shape and measuring the homogeneity of an object.
[0004]
FIG. 11 is a schematic diagram of a main part of a conventional interference measuring apparatus. In FIG. 11, the light beam emitted from the light source 1 passes through the half mirror 2 and reaches the TF substrate (transmission reference plate) 5 provided on the biaxial tilt stage 3. The biaxial tilt stage 3 is tilted with high accuracy by a command from the control computer 9 so that a wavefront of a reference light beam and a test light beam, which will be described later, can be aligned. A TF substrate 5 is installed on the biaxial tilt stage 3 via a piezoelectric element (PZT actuator) 4 for fringe scanning.
[0005]
In addition to the final surface 5 a, the TF substrate 5 is provided with an antireflection film for the wavelength of the light beam from the light source 1, so that a part of the light beam is reflected only from the final surface 5 a, or a wedge angle is provided and the density exceeds the resolution of the CCD camera 8 Interference fringes. The light beam transmitted through the final surface 5 a is reflected by the test surface 6 a of the sample 6. Hereinafter, the light beam reflected by the final surface 5a of the TF substrate 5 is referred to as a reference light beam, and the light beam transmitted through the final surface 5a is referred to as a test light beam.
[0006]
The test light beam reflected by the test surface 6a of the sample 6 is combined with the reference light beam by the TF substrate 5, interferes and is reflected by the half mirror 2, and is reflected on the diffusion plate 10 via the imaging optical system 7 pinhole PH. Interference fringes are formed.
[0007]
The diffusing plate 10 is used to average optical noise such as speckles by being rotated by the driving means M. The interference fringes diffused by the diffusion plate 10 are transmitted to the CCD (CCD camera) 8 by the imaging optical system 11, and the captured interference fringe image data is transferred to the control computer 9.
[0008]
The control computer 9 takes in a plurality of interference fringe image data when the PZT actuator 4 is scanned, calculates the interference fringe phase by a so-called fringe scanning method, and measures the surface shape of the test surface 6 a of the sample 6. .
[0009]
[Problems to be solved by the invention]
The conventional method for obtaining the surface shape of the test surface as the phase wavefront of the interference fringes depends on the spatial frequency of the interference fringes caused by the imaging optical system 7, the imaging optical system 11, or the image sensor (CCD camera) 8. A decrease in amplitude may occur in the phase wavefront of the interference fringes calculated based on the contrast characteristics, which is particularly noticeable at high spatial frequencies.
[0010]
Hereinafter, the cause of the amplitude degradation of the phase wavefront of the interference fringes will be described using mathematical expressions.
[0011]
For simplicity, a wavefront having a single spatial frequency distribution is considered as the wavefront of the test light beam, and the wavefront of the reference light beam is assumed to be completely flat. At this time, the test beam amplitude E_testReference beam amplitude E_retIs
E_test(x, t) = E₀exp (ia cos (2πifx))
E_ref(x, t) = E₀exp (iωt)
It is expressed. Where a is the wavefront amplitude and I₀Is the intensity of the incident light, x is the spatial coordinate, t is the time, the spatial frequency of the f wavefront, and ω represents the frequency of the fringe scan.
[0012]
The interference fringe intensity I (x, t) by these two light beams is

It becomes. Here, the wavefront amplitude a is assumed to be sufficiently small, and is represented by approximation to the first order term of the wavefront amplitude a. The intensity I of the interference fringe image acquired when the contrast at the frequency f is M (f)._mean(X, t) is
I_meas(x, t) = I₀(1 + sin (ωt) + M (f) a cos (2πfx) cos (ωt))
It becomes. Since the fringe scan extracts the cos modulation component and the sin modulation component of the interference fringe change to produce the phase, the calculated phase is

In other words, the wavefront amplitude a of the actual wavefront is calculated by decreasing by the contrast M (f) at the frequency f. For this reason, it becomes difficult to measure the phase wavefront of the interference fringes with high accuracy.
[0013]
The present invention provides an interference device capable of measuring the surface shape of a surface to be examined and the homogeneity of an object with high accuracy.And measuring method having the sameThe purpose is to provide.
[0014]
[Means for Solving the Problems]
  An interference apparatus according to a first aspect of the invention is an interference apparatus for measuring the surface shape or homogeneity of a test object, and divides a light beam from a light source into a test light beam passing through the test object and a reference light beam passing through a reference surface. Optical means for synthesizing those luminous fluxes;
Imaging means for imaging interference fringes formed by combining the test light flux and the reference light flux;
Processing means for correcting the output of the imaging means based on a correction value obtained in advance by measuring a correction sample with this interference device, and calculating the surface shape or homogeneity of the test object,
Sample for correctionOn the surfaceAre patterns with different spatial frequencies or patterns with different spatial frequenciesIs carved,
The correction value is obtained by measuring the amplitude of the pattern obtained by measuring the correction sample with the interference device, and the correction with an apparatus that can more accurately measure the surface shape of the correction sample than the interference device. Is obtained for each spatial frequency by comparing the measurement result of the amplitude of the pattern obtained by measuring the sample for,
The processing means is an interference fringe imaged by the imaging means that is generated as the spatial frequency increases.Phase wavefront calculated fromAmplitude degradation of theFor each spatial frequencyIt is characterized by correction.
  According to a second aspect of the present invention, in the first aspect, the correction sample has a plurality of substrates on which a plurality of patterns having different spatial frequencies are formed.
  The measurement method of the invention of claim 3 is a measurement method for measuring the surface shape or homogeneity of a test object with an interference device.
Patterns with different spatial frequencies or patterns with different spatial frequenciesIs carved on the surfaceObtaining a correction value by measuring a correction sample with the interference device;
Interference fringes picked up by the image pickup means of the interference device generated as the spatial frequency increasesPhase wavefront calculated fromCorrecting the amplitude deterioration of each spatial frequency based on the correction value.And
The correction value is obtained by measuring the amplitude of the pattern obtained by measuring the correction sample with the interference device and the correction device with an apparatus that can more accurately measure the surface shape of the correction sample than the interference device. It is obtained for each spatial frequency by comparing the measurement result of the amplitude of the pattern obtained by measuring the sample forIt is characterized by that.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the accompanying drawings.
[0016]
(Embodiment 1)
A first embodiment of the present invention will be described.
[0017]
Embodiment 1 prepares a correction sample (sample) and mechanically measures the surface shape of the test surface of the correction sample, for example, a three-dimensional shape scanning apparatus (see FIG. 4) and interference measurement. A correction value for correcting the measurement value obtained by the interference measurement apparatus is calculated from the measurement value obtained by the apparatus (see FIG. 1) and obtained by both. That is, the correction value is applied to the measured value of the sample to be measured obtained by the interference measuring apparatus, and the surface shape information of the surface to be measured of the sample to be measured is obtained from this. Here, an example of mechanically measuring the surface shape of the test surface of the correction sample has been shown, but this is not limited to this, and any method that can accurately measure the surface shape of the test surface of the correction sample. For example, there is no need for a mechanical method. Furthermore, if the shape of the test surface of the correction sample is accurately known, there is no particular need for measurement, and the shape of the test surface of the sample and the surface shape by the interference measurement device of the present embodiment are not required. The correction value may be calculated from both of the measurement results.
[0018]
Hereinafter, the measurement and correction values when the test surface of the sample to be measured is a flat surface are calculated, and the flow of the measurement value correction method will be specifically described.
[0019]
FIG. 1 is a schematic diagram of a main part of an interference measuring apparatus used in the first embodiment.
[0020]
First, a method for measuring a sample to be measured with the interference measuring apparatus of FIG. 1 will be described. The light beam emitted from the light source (light source means) 1 passes through the half mirror 2 and reaches the TF substrate 5 on the biaxial tilt stage 3. The biaxial tilt stage 3 is tilted with high accuracy by a command from the control computer 9 so that a wavefront of a reference light beam and a test light beam, which will be described later, can be aligned. A TF substrate 5 is installed on the biaxial tilt stage 3 via a piezoelectric element (PZT actuator) 4 as optical path length difference changing means for the fringe scanning method.
[0021]
In addition to the final surface 5a, the TF substrate 5 is provided with an antireflection film or a wedge angle for the wavelength of the light beam from the light source 1, so that a part of the light beam is reflected only from the final surface 5a. The light beam transmitted through the final surface 5 a is reflected by the test surface 6 a of the sample 6 to be measured. Hereinafter, the light beam reflected by the final surface 5a of the TF substrate 5 is referred to as a reference light beam, and the light beam transmitted through the final surface 5a is referred to as a test light beam.
[0022]
The test light beam reflected by the test surface 6 a of the sample 6 is combined with the reference light beam by the TF substrate 5, interferes and is reflected by the half mirror 2, and the diffusion plate 10 through the imaging optical system 7 and the pinhole PH. Form interference fringes on top.
[0023]
  The diffusion plate 10 is used to average optical noise such as speckles by rotating by the driving means MO. The interference fringes diffused by the diffusion plate 10 are transmitted to the CCD camera (imaging means) 8 by the imaging optical system 11, and the captured interference fringe image data is controlled by a control computer (processing system).(Processing means)9 is transferred. Here, the imaging optical system 11 and the diffusing plate 10 may be omitted, and the CCD camera 8 may be disposed where the diffusing plate 10 is positioned to directly detect the interference fringe image data. This method has a small decrease in amplitude due to the influence of the optical system.
[0024]
An interferometer is configured by a system that extends from the light source 1 to the diffuser plate 10 or the imaging means 8 through the surface 6a to be examined. The optical member provided in the optical path from the light source 1 to the imaging means 8 through the test surface 6a constitutes an element of the optical means.
[0025]
The control computer 9 takes in a plurality of interference fringe image data when the PZT actuator 4 is scanned, and calculates the phase of the interference fringes by a so-called fringe scanning method, whereby the surface shape of the test surface 6a of the sample 6 is determined. Measuring. Then, the measured value of the surface shape of the sample 6 is corrected by the correction value by the amplitude correction unit (correction means) 12 incorporated in the control computer 9. Here, the amplitude correction unit 12 may be separated from the control computer 9. A method of correcting the measurement value calculated by the amplitude correction unit 12 will be described with reference to FIG.
[0026]
In FIG. 2, M1 represents the wavefront distribution (phase wavefront distribution) of the measurement value (interference fringe data) of the surface shape of the sample 6. The measured value M1 is separated into a polynomial component Z1 and a polynomial residual component (residual component) R1 by performing polynomial fitting using a least square method or the like. Here, a ZERNIKE polynomial or the like is used as the polynomial.
[0027]
A two-dimensional Fourier transform is performed on the residual component R1 to obtain a frequency distribution RF1. The reason why the residual component R1 is used here is to suppress unwanted frequency products due to an extreme change of the wavefront pupil end.
[0028]
  In FIG. 2, the amplitude degradation of the phase wavefront is obtained by creating the amplitude distribution of the phase wavefront obtained by the interference measuring device on the same spatial frequency as the frequency distribution RF1. In FIG. 2, RF2 represents the frequency distribution of the residual wavefront after the correction of the phase wavefront whose amplitude has deteriorated, and the frequency distribution RF1 and the correction coefficient distribution Rdc
      RF2 = RF1 / Rdc
It is expressed.
  As described above, in this embodiment, the processing system 9 corrects the phase degradation RF1 of the phase wavefront obtained by the imaging means 8 based on the correction coefficient distribution (correction value) Rdc.
[0029]
Details of how to obtain the correction coefficient distribution Rdc will be described later. This correction is applied when the frequency component amplitude of the wavefront aberration in the region to be corrected by the amplitude deterioration of the phase wavefront is sufficiently smaller than 1 rad. An inverse Fourier transform is performed on the corrected frequency distribution RF2 to obtain a residual wavefront R2 in real space. By adding the separated polynomial component Z1 to this, the wavefront aberration distribution M2 is obtained, and from this, the correction of the amplitude deterioration with respect to the measurement value M1 is completed, and the wavefront measurement value M2 with little measurement error is obtained up to the high frequency region. It is possible.
[0030]
Next, how to obtain the correction coefficient distribution Rdc will be described with reference to the flow (flow chart) of FIG.
[0031]
  Step A: “Preparing correction samples”
  5, 6 and 7 are schematic views of the correction sample. FIG. 5 is a top view of the test surface of the correction sample S1, and C1a and C1b represent amplitude distributions in the cross sections of the straight lines S1a and S1b of the correction sample S1. FIG. 6 is a top view of the correction sample S2 test surface, and C2a and C2b represent amplitude distributions in the cross sections of the straight lines S2a and S2b of the correction sample S2. In FIG. 7, S3 (S3a, S3b... S3x) is a plurality of objects to be measured.(substrate)C3a, C3b,... C3x are cross-sectional views of samples S3a, S3b,. A plurality of such correction samples S1, S2, and S3 are prepared, and the surface shapes of the test surfaces of the correction samples S1, S2, and S3 are measured by the interference measurement device and the three-dimensional shape scanning measurement device shown in FIG. To do.
[0032]
Step B: “Measure the correction samples S1, S2 and S3 with an interference measuring device”
The measurement of the correction samples S1, S2, and S3 by the interference measuring apparatus is performed by measuring the surface shape of the test surfaces of the correction samples S1, S2, and S3 by the same method as described above with reference to FIG. . Since the details of the measurement method have already been described, a description thereof will be omitted. Here, the amplitude correction of the phase wavefront by the amplitude correction unit 12 is not performed. That is, the correction coefficient distribution Rdc = 1 is measured.
[0033]
Step C: “Measure the correction samples S1, S2 and S3 with a three-dimensional shape scanning measurement device”
The measurement of the surface shape of the correction sample by the contact type three-dimensional shape scanning measurement apparatus will be described with reference to FIG.
[0034]
FIG. 4 is a schematic view of a contact-type three-dimensional shape scanning measurement apparatus for measuring the shape of the correction sample surface. The sample 22 is installed on the XY stage 23. Here, the XY stage 23 can be driven XY independently with high precision by the control computer 25.
[0035]
The probe 21 is installed on the XYZ stage 24, and the control computer 25 controls the XYZ stage 24 so that the probe 21 scans the surface 22 a of the sample 22 while applying a constant pressure. At this time, the surface shape of the test surface 22a of the sample 22 is calculated by calculating the position coordinates of the XYZ stage 24 by the control computer 25. Here, there is no problem even if the correction samples S1, S2, and S3 are measured with a non-contact type three-dimensional shape scanning measurement apparatus.
[0036]
Step D: “Determine correction coefficients RdcH and RdcV from the measurement results of steps B and C”
Next, how to obtain correction coefficients RdcH and RdcV in the horizontal direction and vertical direction of the CCD camera 8 from the measurement results of Step B and Step C will be described with reference to FIG.
[0037]
When the cut surface of the test surface is in the horizontal direction of the CCD camera 8, a certain spatial frequency kx with respect to the sampling interval is calculated from the measurement result. Thus, the amplitude V at the spatial frequency kx on the surface to be measured_ref(Kx) is obtained.
[0038]
Hereinafter, h is calculated up to the Nyquist frequency of the sampling interval (the reciprocal of twice the sampling period). Similarly, in the case of a three-dimensional shape scanning measurement apparatus, the amplitude V of the spatial frequency kx_refFind (kx).
[0039]
  The series of calculation results are shown in FIG. Discrete data V (kxi) is obtained by measuring the amplitude of the spatial frequency on the test surface. The solid line in Fig. 8 is a three-dimensional shape.scanningAmplitude V_ref(Kx)It is.The dotted line represents the amplitude V (kx) of the interference measuring apparatus, and the amplitude of the dotted line is degraded compared to the amplitude of the solid line.That is, the three-dimensional shape scanning type device can measure the surface shape of the correction sample more accurately than the interference device. In interference equipmentThe cause of the amplitude deterioration is as described above in “Problems to be solved”.
[0040]
  At this time, the correction coefficient Rdc0 is
      Rdc0 = V (kx) / V_ref(Kx)
Thus, the correction is completed by dividing the measurement value of the interference measuring apparatus by the correction coefficient Rdc0 for each spatial frequency. As described above, in this embodiment, the amplitude distribution of the interference fringes imaged by the imaging unit 8 generated with the increase in the spatial frequency is corrected based on the correction coefficient (correction value).
  That is, in this embodiment, the correction coefficient (correction value) Rdc0 is obtained by measuring the correction sample (S1, S2, S3) with the interference measurement device and the amplitude measurement result V (Kx) of the pattern obtained from the interference measurement device. The measurement results V obtained by measuring the correction samples (S1, S2, S3) with a highly accurate three-dimensional shape scanning device _ref (Kx) is obtained for each spatial frequency.
[0041]
When the sample interval between the three-dimensional shape scanning measurement device and the interference measurement device is different, the amplitude V_refA correction coefficient may be obtained by fitting (kx) and amplitude V (kx) with a polynomial function or the like by the least square method or the like.
[0042]
Here, the correction coefficient in the interference measuring apparatus in which incoherent imaging is included in the pupil imaging system that forms interference fringes as in this embodiment is
F_InchOpt(x, k₀) = (2 cos^-1(k₀x)-sin (2 cos^-1(k0x))) / π
This equation represents the amplitude degradation of the incoherent optical system of the ideal optical system.
[0043]
In this embodiment, incoherent imaging is used as the pupil imaging system of the test lens. However, a Moffat function expressed by the following equation may be used as a correction coefficient in the coherent imaging interferometer.
[0044]
Moffat (x, k₀, k₁, k₂) = k₀/ (1+ (x / k₁)²)^k2
Where x is the spatial frequency and k₀, K_l, K₂Is a parameter.
[0045]
The calculation of the correction coefficient RdcH with respect to the horizontal direction of the CCD camera 8 is thus completed. Next, when the correction coefficient RdcV for the vertical direction is calculated by the same procedure, the correction coefficients RdcH and RdcV for the horizontal direction and the vertical direction of the interference measuring apparatus are obtained.
[0046]
Step E: “Determine the correction coefficient distribution Rdc from the result of Step D”
If fitting is performed to the correction coefficients RdcH and RdcV in the horizontal direction and the vertical direction of the interference measuring apparatus, and a function is formed, the correction coefficient distribution Rdc can be easily created. Assuming that the fitted horizontal and vertical correction coefficients are RdcH (kx) and RdcV (ky), the correction coefficient distribution Rdc (kx, ky) on the spatial frequency (kx, ky) is
Rdc (kx, ky) = RdcH (kx) × RdcV (ky)
It is expressed.
[0047]
As described above, the correction coefficient distribution Rdc in FIG. 2 can be obtained. The amplitude correction unit 12 in FIG. 1 performs an amplitude correction calculation on the measurement value using the correction coefficient distribution Rdc.
[0048]
Moreover, although the planar shape has been described as the shape of the surface to be detected, application to homogeneity measurement (homogeneous measurement) is also possible. In the homogeneity measurement, the sample 6 in FIG. 1 is an RF substrate (reflective reference plate), and the object to be measured 13 is placed between the RF substrate 6 and the TF substrate 5 and the transmitted wavefront is measured. As a measuring method, an oil-on-plate method for measuring the non-polished surface or a calendar method for measuring in the polished surface state can be applied.In the present embodiment, as a measuring method for measuring the surface shape or homogeneity of the test object 6 with an interference device, correction samples S1 to S3 having a plurality of patterns having different spatial frequencies or patterns having different spatial frequencies are used. Based on the correction value, the step of obtaining a correction value (correction coefficient) by measuring with the interference device, and the amplitude degradation of the interference fringes imaged by the imaging means 9 of the interference device that occurs as the spatial frequency increases. And a correcting step.
[0049]
(Example 2)
Next, a second embodiment of the present invention will be described.
[0050]
In the second embodiment, the sample to be measured in the first embodiment is a parallel plate, but a spherical mirror is used as the shape of the sample to be measured. The patterns shown in FIGS. 5 and 6 are provided on the reflection surface of the spherical mirror. A measurement method using the interference measurement apparatus in this case will be described with reference to FIG.
[0051]
In FIG. 9, the light beam emitted from the light source 1 passes through the half mirror 2 and reaches the condenser lens 5 provided on the XYZ stage 3. The XYZ stage 3 can be driven independently in the XYZ directions with high accuracy by a command from the control computer 9. A condensing lens 35 is installed on the XYZ stage 3 through a piezoelectric element (PZT actuator) 4 for fringe scanning.
[0052]
Here, the condenser lens 35 is a so-called TS lens in which the radius of curvature of the final surface 35a is equal to the distance between the final surface 35a and the focal point 10. In addition to the final surface 35a, the TS lens 35 is provided with an antireflection film for the wavelength of the light source 1 so that a part of the light beam is reflected only from the final surface 35a. The light beam transmitted through the final surface 35 a is reflected by the test surface 6 a of the sample 6. Hereinafter, the light east reflected by the final surface 35a of the TS lens 35 is referred to as a reference light beam, and the light beam transmitted through the final surface 35a is referred to as a test light beam.
[0053]
Here, the sample 6 is provided on an XYZ stage 11 that can be controlled by the control computer 9 and is adjusted in the XYZ directions so that the center of curvature of the test surface 6a of the sample 6 coincides with the condensing point 10 by the TS lens 35. ing.
[0054]
The test light beam reflected by the test surface 6 a of the sample 6 is combined with the reference light beam by the TS lens 35, interferes with each other, is reflected by the half mirror 2, and passes through the imaging optical system 7 and the pinhole PH. Interference fringes are formed on the camera 8. Image data picked up by the CCD camera 8 is transferred to the control computer 9.
[0055]
The control computer 9 takes in a plurality of interference fringe image data when the PZT actuator 4 is scanned from the CCD camera 8, calculates the phase of the interference fringes by the so-called fringe scanning method, and obtains the phase wavefront (measured value). The measured value is corrected with a correction coefficient by an amplitude correction unit 12 incorporated in the control computer 9. Here, there is no problem even if the amplitude correction unit 12 is separated from the control computer 9.
[0056]
  Sample 6 used here is a TS lens35A pattern as shown in FIG. 5, FIG. 6 or FIG. 7 is engraved on a spherical surface where the center of curvature of the condensing point 10 and the test surface 6a of the sample 6 coincide.
[0057]
The measurement of the surface to be tested and the correction of the amplitude deterioration of the sample 6 by the three-dimensional shape scanning measurement apparatus have been described in the first embodiment, and will be omitted. Although the optical system without the diffusing plate 10 and the imaging optical system 11 as in the first embodiment is shown here, these members may be provided.
[0058]
(Example 3)
Next, Embodiment 3 of the present invention will be described with reference to FIG. In the third embodiment, a transmission object is used as the test object. In FIG. 10, the light beam emitted from the light source 1 passes through the half mirror 2 and reaches the condenser lens 5 provided on the XYZ stage 3. The XYZ stage 3 can be driven independently in the XYZ directions with high accuracy by a command from the control computer 9. A condensing lens 5 is installed on the XYZ stage 3 via a piezoelectric element (PZT actuator) 4 for a fringe scanning method.
[0059]
Here, the condenser lens 35 is a so-called TS lens in which the radius of curvature of the final surface 35a is equal to the distance between the final surface 35a and the focal point 10. In addition to the final surface 35a, the TS lens 5 is provided with an antireflection film for the wavelength of the light source 1 so that a part of the light beam is reflected only from the final surface 35a. Hereinafter, the light beam reflected by the final surface 35a of the TS lens 35 is referred to as a reference light beam, and the light beam transmitted through the final surface 35a is referred to as a test light beam.
[0060]
The Z stage is adjusted so that the focal point 10 of the condenser lens 35 coincides with the object plane of the test lens 14, and the test light beam transmitted through the test lens 14 is on the image plane 15 of the test lens 14. And then reflected by the spherical RS mirror 6.
[0061]
Here, the RS mirror 6 is provided on the XYZ stage 11 that can be controlled by the control computer 9 in the same manner as the TS lens 5 and is adjusted in the XYZ stage 11 direction so that the curvature center 6a of the RS mirror 6 coincides with the image side focal point 15. Has been made.
[0062]
The reference light beam reflected by the final surface 35a of the TS lens 35 and the test light beam reflected by the RS mirror 6 are combined through the TS lens 35 and interfere with each other to form the same optical path. Reflected to form interference fringes on the CCD camera 8 via the imaging optical system 7 and the pinhole PH. Interference fringe image data captured by the CCD camera 8 is transferred to the control computer 9. The control computer 9 takes in a plurality of interference fringe image data when the PZT actuator 4 is scanned, calculates the interference fringe phase by the so-called fringe scanning method, and obtains the transmitted wavefront of the test lens.
[0063]
The measured value is corrected with a correction coefficient by an amplitude correction unit 12 incorporated in the control computer 9. Amplitude correction hereunitThere is no problem even if 12 is separated from the control computer 9.
[0064]
Next, measurement of the correction sample in the interference measurement apparatus of FIG. 10 will be described.
[0065]
First, the correction sample 13 shown in FIG. 10 is attached to the test lens 14 or is arranged at the same position by a tool or the like without the test lens 14. Similar to the second embodiment, the correction sample 13 used here is shown in FIGS. 5, 6, and 7 as a spherical surface where the focal point 10 of the TS lens 5 coincides with the center of curvature of the test surface 13 a of the correction sample 13. It is a carved pattern. At this time, the XYZ stage 11 is controlled by the control computer 9 so that the center of curvature of the test surface 13a of the correction sample 13 and the condensing point 10 of the TS lens 5 coincide.
[0066]
The test light beam reflected by the test surface 13a of the correction sample 13 is combined with the reference light beam by the TS lens 35, interferes with each other, is reflected by the half mirror 2, and passes through the imaging optical system 7 and the pinhole PH. Interference fringes are formed on the CCD camera 8. Image data picked up by the CCD camera 8 is transferred to the control computer 9. The control computer 9 takes in a plurality of interference fringe image data when the PZT actuator 4 is scanned, calculates the interference fringe phase by the so-called fringe scanning method, and obtains the amplitude deterioration when measured by the interference measuring device.
[0067]
Since the measurement of the test surface of the correction sample 13 and the correction of the amplitude deterioration by the three-dimensional shape scanning measurement apparatus have been described in the first embodiment, they will be omitted. Although an optical system without the diffusing plate 10 and the imaging optical system 11 as in the first embodiment is shown here, these members may be provided.
[0068]
(Embodiment 4)
Next, a fourth embodiment of the present invention will be described.
[0069]
In the fourth embodiment, a surface shape measuring device (Embodiments 1 and 2) and a transmitted wavefront shape measuring device (Embodiment 3) having an amplitude correction function are applied to manufacture of a projection lens of an exposure apparatus. A single lens or homogeneity measurement is performed by the measurement apparatus of the first and second embodiments, and the wavefront of the assembled lens is measured and adjusted by the measurement apparatus of the third embodiment. As a result, lenses with high accuracy up to high frequencies are manufactured.
[0070]
  The embodiments described above are summarized as follows. The first interfering device isAn interference device that forms interference fringes using light that passes through the test object.The AndCorrection means for correcting a phase difference distribution obtained from the interference fringes based on a correction value obtained by using a shape of the correction sample and a measurement value obtained by measuring the shape of the correction sample with the interference device;The
[0071]
in frontThe correction means keeps the gain of the first frequency component substantially constant among the first frequency component of the phase difference distribution and the second frequency component having a spatial frequency higher than the first frequency component, and the second frequency component Correct the gain ofThe
[0072]
  Also beforeThe correction means amplifies the gain of the second frequency component.The
[0073]
in frontThe shape of the specimen is derived based on the phase difference distribution.OpticalHave meansThe
[0074]
in frontOptical path length difference change that changes the optical path length difference between the two light beams that form the interference fringesmeansHaveThe
[0075]
  The second interfering device isOptical means for forming a test light beam through the test object and a reference light beam through the reference surface from the light beam emitted from the light source means, and combining both to form an interference wave front; and the test light beam; An optical path length change element that changes the optical path length difference of the reference light beam, an image sensor that captures an interference fringe based on the interference wavefront, and a phase difference distribution of the interference wavefront from the interference fringe imaged by the image sensor Apparatus having a processing systemIt is. AndA correction value obtained by using the value obtained by the shape measuring device that mechanically measures the uneven shape of the test surface of the correction sample and the value obtained by measuring the correction sample by the interference device. Has correction means to correct withThe
[0076]
in frontWhen the test surface of the test object or the test object is a lens system, the pupil and the imaging device have a conjugate relationship.The
[0077]
in frontThe optical means forms interference fringes on the diffusion plate, and re-images the interference fringes formed on the diffusion plate on the image sensor.The
[0078]
in frontThe correction sample consists of a sample in which a plurality of different spatial frequency patterns are engraved on the surface to be measured or a sample in which different spatial frequency patterns are engraved on a plurality of objects to be measured.The
[0079]
in frontThe correction means finds the correction value in the frequency space.The
[0080]
in frontWhen the imaging system that forms interference fringes is an incoherent system, the correction value is f, the spatial frequency of the wavefront is x, and the spatial coordinates are x, 2πf = k.₀And when
F_InchOpt(x, k₀) = (2 cos^-1(k₀x)-sin (2 cos^-1(k₀x))) / π
Using the correction factorThe
[0081]
  Or beforeThe correction value is k when the imaging system that forms the interference fringes is a coherent system.₀, K₁, K₂Where x is the spatial coordinate,
      Moffat (x, k₀, k₁, k₂) = k₀/ (1+ (x / k₁)²)^k2
Using the correction factorThe
[0082]
in frontDetermine the homogeneity of the specimenThe
[0084]
  The third interfering device isA test light beam obtained by reflecting a part of the light beam emitted from the light source means on the test surface or reflecting it on the test surface and then reflecting on the reflection surface, and a part of the light beam emitted from the light source means An optical means for forming a reference light beam obtained by reflecting the light beam by a reference surface, an optical path length difference changing element for changing an optical path length difference between the test light beam and the reference light beam, and interfering the test light beam and the reference light beam The imaging device that images the interference fringes obtained by the imaging and the optical means guide the interference fringes formed by the reference light beam and the test light beam to the imaging device, and are photographed by the imaging device. An interference device having a processing system for calculating a phase difference distribution between the test light beam and the reference light beam from the interference fringesIt is. AndA correction unit that calculates a correction value that corrects amplitude degradation of the spatial phase distribution of the interference fringes, and the correction unit is configured to measure the correction value in contact or non-contact with the unevenness of the test surface; Obtained by measuring the correction sample with the interference deviceThe
[0085]
in frontThe test surface or the pupil of the test object and the image sensor have a conjugate relationship.The
[0086]
in frontThe correction sample consists of a sample in which a plurality of different spatial frequency patterns are engraved on the surface to be measured or a sample in which different spatial frequency patterns are engraved on a plurality of objects to be measured.The
[0087]
in frontThe correction means finds the correction value in the frequency space.The
[0088]
  The fourth interfering device isAn interferometer that combines the test wavefront through the test object and the reference wavefront to form interference fringe data based on the phase wavefront in the imaging means, and the test object from the interference fringe data from the interferometer Interferometer having a processing system for calculating and obtaining a phase wavefront viaIt is. AndThe processing system uses the amplitude V (kx) when the phase wavefront of the interference information obtained by the interferometer when the correction sample is used is deteriorated by the spatial frequency value (kx), and the unevenness of the correction sample. The amplitude obtained by a shape measuring device that mechanically measures the shape is expressed as V_ref(Kx)
[0089]
[Expression 1]

[0090]
The interference fringe data is corrected using the correction coefficient distribution Rdc obtained fromThe
[0091]
  The fifth interfering device isAn interferometer that combines the test wavefront through the test object and the reference wavefront to form interference fringe data based on the phase wavefront in the imaging means, and the test object from the interference fringe data from the interferometer Interferometer having a processing system for calculating and obtaining a phase wavefront viaIt is. AndThe processing system uses the frequency distribution (RF2) obtained by correcting the frequency distribution (RF1) based on the interference fringe data (M1) obtained by the interferometer with the correction coefficient distribution (Rdc), to thereby calculate the wavefront aberration distribution (M2). At this time, the correction coefficient distribution (Rdc₀) Is a shape measuring device that mechanically measures the amplitude when the phase wavefront of the interference information obtained by the interferometer deteriorates in amplitude due to the spatial frequency value (kx), and the uneven shape of the correction sample. The resulting amplitude is V_ref(Kx)
[0092]
[Expression 2]

[0093]
Seeking moreThe
[0094]
【The invention's effect】
  According to the present invention, it is possible to correct a transmitted wavefront or a test surface shape error of a test surface of a high-frequency component that has been difficult due to amplitude degradation of interference fringes measured by an interference device, and to perform high-accuracy measurement. apparatusAnd measuring method having the sameCan be achieved.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of Embodiment 1 of the present invention.
FIG. 2 is a flowchart showing a procedure for correcting amplitude deterioration according to the present invention.
FIG. 3 is a flowchart showing a calculation for obtaining a correction coefficient distribution according to the present invention.
FIG. 4 is a schematic configuration diagram of a contact-type three-dimensional shape scanning measurement apparatus according to the present invention.
FIG. 5 is an example of a sample for back surface measurement according to the present invention.
FIG. 6 is an example of a sample for measuring a test surface according to the present invention.
FIG. 7 is an example of a sample for measuring a plurality of test surfaces according to the present invention.
FIG. 8 is a graph of amplitude obtained by Fourier transform of a cross section of a test surface according to the present invention according to the present invention.
FIG. 9 is a schematic configuration diagram of Embodiment 2 of the present invention.
FIG. 10 is a schematic configuration diagram showing Embodiment 3 of the present invention.
FIG. 11 is a schematic configuration diagram of a conventional interference measuring apparatus.
[Explanation of symbols]
E_test(X, t): Test light flux complex amplitude
E_ref(X, t): Reference beam search amplitude
E0: Electric field amplitude
a: Wavefront amplitude
f: wavefront spatial frequency
ω: fringe scanning frequency
M (t): Contrast at spatial frequency f
k: interference fringe spatial frequency
V: Amplitude of test surface
kx: horizontal spatial frequency
ky: vertical spatial frequency
F_inch0pt: Fitting function of incoherent optical system,
Rde0: one-dimensional correction coefficient function
M1: Measurement result interference fringe phase
Z1: Measurement result Interference fringe phase polynomial component
R1: Measurement result Interference fringe phase residual component
RF1: Spatial frequency distribution of fringe phase residual component of interference of measurement results
Rdc: correction coefficient distribution of spatial frequency of interferometer amplitude degradation
RF2: Correction result Spatial frequency distribution of residual aberration component of transmitted wavefront
R2: Aberration residual component of transmission wavefront of correction result
M2: Correction result transmitted wavefront aberration distribution
NA: NA of the measured light beam
r: Measurement beam radius

Claims (3)

In an interference device for measuring the surface shape or homogeneity of a test object, optical means for dividing a light beam from a light source into a test light beam passing through the test object and a reference light beam passing through a reference surface, and combining the light beams ,
Imaging means for imaging interference fringes formed by combining the test light beam and the reference light beam;
Processing means for correcting the output of the imaging means based on a correction value obtained in advance by measuring a correction sample with this interference device, and calculating the surface shape or homogeneity of the test object,
A plurality of patterns having different spatial frequencies or patterns having different spatial frequencies are engraved on the surface of the correction sample,
The correction value is obtained by measuring the amplitude of the pattern obtained by measuring the correction sample with the interference device, and the correction with an apparatus that can more accurately measure the surface shape of the correction sample than the interference device. Is obtained for each spatial frequency by comparing the measurement result of the amplitude of the pattern obtained by measuring the sample for,
The processing means corrects the amplitude deterioration of the phase wavefront calculated from the interference fringes picked up by the image pickup means that occurs as the spatial frequency increases for each spatial frequency based on the correction value. Interfering device.   The interference apparatus according to claim 1, wherein the correction sample includes a plurality of substrates on which a plurality of patterns having different spatial frequencies are formed. In a measurement method for measuring the surface shape or homogeneity of a test object with an interference device,
Obtaining a correction value by measuring a correction sample in which a plurality of patterns having different spatial frequencies or patterns having different spatial frequencies are engraved on the surface with the interference device;
The amplitude deterioration of the phase front being calculated from the interference fringe captured by the imaging means of the interference device generating with increasing spatial frequency, have a, a step of correcting each spatial frequency on the basis of the correction value ,
The correction value is obtained by measuring the amplitude of the pattern obtained by measuring the correction sample with the interference device and the correction device with an apparatus that can more accurately measure the surface shape of the correction sample than the interference device. measurement wherein the der Rukoto those obtained each spatial frequency by comparing the measurement results of the amplitude of the pattern obtained by measuring use samples.

Images