JP3986801B2 - Surface state observation apparatus and method, and semiconductor device and microdevice manufacturing method - Google Patents

Description

試料４に形成されているパターンをＯ（ｘ，ｙ）とすると、このパターンＯ（ｘ，ｙ）は、上記（１）式からＸＹ座標系を用いて表すことができ、更にｘｙステージ４０のｙ方向の移動速度をｖとすると以下の（２）式で表される。
【数２】

【００３７】
次に、撮像素子３５から出力される画素信号について考える。まず、撮像素子３５が備える複数の画素の内の１つの画素から出力される信号について考察する。図７は、撮像素子３５が備える画素の配列の一例を示す図である。いま、図７に示すように、撮像素子３５のＸ方向の画素を第１番目から第ｍ番目のｍ個とし、Ｙ方向の画素を第１番目から第ｎ番目のｎ個とし、Ｘ方向についてｐ（１≦ｐ≦ｍ）番目であって、Ｙ方向についてｑ（１≦ｑ≦ｎ）番目の画素（図７中においては斜線を付した画素）からの出力を考える。また、各画素の感度特性は全て等しいと仮定し、この感度特性を関数ｇ（Ｘ，Ｙ）で表す。図８は、Ｙ方向の位置が画素の中心に設定された感度特性関数ｇ（Ｘ）の一例を示す図である。図８に示したように、この感度特性関数ｇ（Ｘ）は画素の開口内のみで値を有する関数である。尚、感度特性関数ｇ（Ｙ）も、図８と同様に開口内でのみ値を有する関数である。
【００３８】
ここで、時刻ｔにおいて試料４に形成されているパターンの像の一部が着目画素上にあるとすると、着目画素から出力される信号ｇ′（Ｘ_p，Ｙ_q，ｔ）は、感度特性関数ｇ（Ｘ，Ｙ）を用いて以下の（３）式で表される。尚、本明細書において、数式中の積分範囲に積分範囲を明示していない場合には、−∞から∞まで積分するものとする。
【数３】

ところで、撮像素子３５は各画素毎にシャッタが設けられており、シャッタが開状態の時のみパターンの光学像を電気信号に変換する。そこで、シャッタが開状態である時間をΔｔとし、以下の（４）式に示す関数Ω（ｔ）を考える。
【数４】

上記（４）式で表される関数Ω（ｔ）は、シャッタが開状態にあるΔｔの間のみ値「１」となり、それ以外のシャッタが閉状態にある時間は値が「０」となる関数である。よって、シャッタを考慮した上で時刻ｔ_kに着目画素から出力される信号ｇ″（Ｘ_p，Ｙ_q，ｔ_k）は以下の（５）式で表される。
【数５】

【００３９】
次に、撮像素子３５の全画素を考慮した信号を考える。前述したように、撮像素子３５はＹ軸方向に並んだ画素各々が変換した信号を積算して出力する。ここでは、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している場合を考えているので、Ｙ座標を時刻で表すことができる。いま、試料４上の任意の点Ｐの像が撮像素子３５の撮像面３５ａをＹ方向に移動している場合を考える。尚、この場合において、点Ｐの像は撮像素子３５の同一列上を移動する点に注意されたい。
【００４０】
この点Ｐの像が撮像面３５ａの中心に配置された行を横切る（通過する）時刻をＴとすると、この行からＹ方向へｑ個の画素数分だけずれた位置の画素を点Ｐの像が通り過ぎる時刻はＴ＋ｔ_qとなる。尚、ｔ_q＝（Ｙ_q／β）／ｖである。このため、時刻Ｔにおいて、撮像素子３５から出力される信号ｓ（Ｘ_p，Ｔ）は、以下の（６）式で表される。
【数６】

上記（６）式に示した撮像素子３５から出力される信号ｓ（Ｘ_p，Ｔ）を示す式はＸ方向の値が離散的であり、後述するＯＴＦを解析する上で取り扱いが不便である。このため、連続的な変数Ｘを用いた式Ｓ（Ｘ，Ｔ）に変形すると、以下の（７）式となる。
【数７】

この（７）式中の右辺のδ（Ｘ−Ｘ_p）は、Ｄｉｒａｃのδ関数である。上記の（７）式が写像光学系２０に歪曲収差がないとしたときの撮像素子３５から出力される信号を示す式である。
【００４１】
〔２〕写像光学系２０の歪曲収差を考慮したときに撮像素子３５から得られる信号
上記〔１〕においては、写像光学系２０の倍率βを定数として歪曲収差がない場合を考えた。この場合には試料４に形成されたパターンを拡大又は縮小した像が撮像素子３５の撮像面３５ａに形成される。しかしながら、写像光学系２０に歪曲収差がある場合には撮像面３５ａ上の像は歪む。この像の歪みは、写像光学系２０の局所的な倍率の変動と解釈できる。そこで、ｘｙ座標内の位置に応じて写像光学系２０の倍率が異なるものとして歪曲収差を捕らえ、前述した（１）式を以下の（８）式で表す。
【数８】

尚、上述したように、写像光学系２０の倍率がｘｙ座標内の位置に応じて異なるものであるならば、上記（８）式中の倍率β_x，β_yがｘ，ｙの関数とするのが自然であるが、倍率β_x，β_yをＸ，Ｙの関数としても結果は同じになるため、以下の取り扱いを容易にすることを考慮して、上記（８）式では、倍率β_x，β_yをＸ，Ｙの関数としている。
【００４２】
ｘｙ座標系とＸＹ座標系との関係が上記（８）式に示した関係にある場合に、撮像素子３５の撮像面上に形成される像は、（８）式をｘ，ｙ各々について解いた式を、試料４上のパターンを示す式Ｏ（ｘ，ｙ）に代入し、更にｘｙステージ４０のｙ方向の移動速度ｖを考慮すると以下の（９）式となる。
【数９】

写像光学系２０の歪曲収差を考慮したときの、撮像素子３５から出力される信号ｓ_d（Ｘ_p，Ｔ）は、（５）式中の式Ｏ（Ｘ／β，Ｙ／β−ｖ・ｔ）を上記（９）式に換えた式を（６）に代入することにより以下の（１０）式の通りに求まる。
【数１０】

最後に、上記〔１〕の場合と同様に、後述するＯＴＦを解析する上で取り扱いを容易とするため、連続的な変数Ｘを用いた式に変形すると、以下の（１１）式となる。
【数１１】

【００４３】
〔３〕撮像素子３５から得られる信号のＯＴＦの計算
以下、上記（１０）式及び（１１）式で示される撮像素子３５から出力される信号をフーリエ変換することにより撮像素子３５から得られる信号のＯＴＦを求めることができる。しかしながら、通常は写像光学系２０の歪曲収差は極力小さくなるように設計されている。従って、以下においては、歪曲収差が小さいとして式の簡略化を図り、その後にＯＴＦを計算する。
【００４４】
（ａ）式の簡略化
写像光学系２０の歪曲収差が小さいため、（８）式に示した倍率β_x（Ｘ，Ｙ），β_y（Ｘ，Ｙ）は〔１〕で用いた定数の倍率βに極めて近いと考えられる。このため、倍率β_x（Ｘ，Ｙ），β_y（Ｘ，Ｙ）をそれぞれ以下の（１２）式で表す。
【数１２】

ここで、上記（１２）式中のＤ_x（Ｘ，Ｙ），Ｄ_y（Ｘ，Ｙ）は、それぞれ、Ｄ_x（Ｘ，Ｙ）≪１、Ｄ_y（Ｘ，Ｙ）≪１である。
【００４５】
次に、上記（１２）式に示した倍率、β_x（Ｘ，Ｙ），β_y（Ｘ，Ｙ）を（１０）式に適用するわけであるが、ここで、以下の（１３）式に示す変数変換を施す。尚、以下の式では、式の簡略化のために、倍率β_x（Ｘ，Ｙ），β_y（Ｘ，Ｙ）をβ_x，β_yとそれぞれ表記する。
【数１３】

この（１３）式に示した変数変換では、（９）式に示した撮像面３５ａ上の像の位置を規定するＸ座標の値及びＹ座標の値をそれぞれ変数Ｘ′，Ｙ′に変換していることから、撮像素子３５の撮像面３５ａ上で移動する像の移動とともに移動する座標系（Ｘ′，Ｙ′）を設定している。尚、（１３）式中のＬはＸ′Ｙ′座標系の原点とｑ番目の画素とのＹ方向についてのオフセット量（距離）を示している。上記（１０）式に示した変数変換を行うときに、（１０）式中の積分要素ｄＸｄＹｄｔは、ｄＸｄＹｄｔ＝（１／｜Ａ｜）ｄＸ′ｄＹ′ｄＬとなる。ここで、Ａは、以下の（１４）式で表される。
【数１４】

【００４６】
ここで、｜Ａ｜を計算すると、以下の（１５）式となる。
【数１５】

尚、上記（１５）式を導出するにあたっては、（１２）式を用いるとともにＤ_x（Ｘ，Ｙ），Ｄ_y（Ｘ，Ｙ）の２次以上の項を無視している。上記（１３）式中の各式を、ｔ，Ｘ，Ｙについて解いて（１０）式に代入するとともに、（１２）式を（１０）式に代入し、更にｄＸｄＹｄｔ＝（１／｜Ａ｜）ｄＸ′ｄＹ′ｄＬなる関係を用いて（１０）式を表すと以下の（１６）式となる。
【数１６】

【００４７】
ところで、前述しように、撮像素子３５が備える各画素の感度特性を示す関数ｇは、画素の開口内でのみ値を有する関数である。従って、上記（１６）式においても、感度関数ｇはＤ_x(Ｘ，Ｙ)，Ｄ_y(Ｘ，Ｙ)が極めて小さいとすると、Ｘ方向のｐ番目の画素、及び、Ｙ方向のｑ番目の画素に対して、以下の（１７）式が成り立つ極めて近接する領域のみで値を有することになる。
【数１７】

また、上記（１６）式中の感度関数ｇのＸ座標の値におけるＤ_x（Ｘ，Ｙ）・Ｘ′についても、上記（１７）式が成り立つ座標（Ｘ′，Ｙ′）の周りで展開したときに、第２項目以降は無視することができるので、以下の（１８）式のように近似することができる。
【数１８】

【００４８】
同様に、上記（１６）式中の感度関数のＸ座標の値におけるＤ_y（Ｘ，Ｙ）・（Ｙ′＋Ｌ−ｖ・ｔ_q）についても、以下の（１９）式のように近似することができる。
【数１９】

更に、同様の理由から｜Ａ（Ｘ，Ｙ）｜についても以下の（２０）式のように近似することができる。
【数２０】

【００４９】
以上の近似式（１８）式〜（２０）式を（１６）式に代入すると、以下の（２１）式が得られる。
【数２１】

この（２１）式が写像光学系２０に歪曲収差があるときに、撮像素子３５から出力される信号の近似値を示している。
【００５０】
（ｂ）信号のフーリエ成分
次に、写像光学系２０に残存収差があるときに、撮像素子３５から出力される信号Ｓ_d（Ｘ、Ｔ）のフーリエ成分について考察する。いま、信号Ｓ_d（Ｘ，Ｔ）のフーリエ変換をＦＳ_d（μ，ν）とすると、この式は以下の（２２）式で表される。
【数２２】

ここで、（１１）式を参照すると分かるように、式Ｓ_d(Ｘ，Ｔ)は、くし形関数（comb関数）と式ｓ_d（Ｘ，Ｔ）との積であるので、そのフーリエ変換は式ｓ_d（Ｘ，Ｔ）のフーリエ変換、comb関数、及び式ＦＳ_d（μ，ν）のコンボリューションとなる。よって、式ｓ_d（Ｘ，Ｔ）のフーリエ変換をＦｓ_d（μ，ν）とすると、式ＦＳ_d（μ，ν）は以下の（２３）式のように変形することができる。
【数２３】

【００５１】
上記（２３）式を参照すると、式ＦＳ_d（μ，ν）は式ｓ_dのフーリエ変換から得られることが分かる。従って、次に式ｓ_dについて考察する。ここで、上記（２１）式に着目すると、この式中に含まれる感度関数ｇが値を有する領域は、前述したように、極めて狭い領域である。また、試料４に形成されたパターンを示す式Ｏについても、分解能を議論するときには極めて微細なパターンを考えている。従って、関数Ｏが値を有する領域も極めて狭い領域となる。それ故、上記（２１）式も限られた領域でのみ値を持つ式である。かかる限られた領域内においては、Ｘ方向に関して歪曲収差の変化量も小さいと考えられる。つまり、上記（２１）式中のＤ_x（Ｘ_p，Ｙ_q）・（Ｘ_p／β）及びＤ_y（Ｘ_p，Ｙ_q）・（Ｘ_p／β）の変化は小さいものとし、Ｘ方向に関しては、第ｐ列に配列された全ての画素について同一の値をとるものとし、以下の（２４）式に示すように、Ｙ方向の座標にのみ依存するとする。尚、（２４）式中のδ_xp，δ_ypはδ関数ではない点に注意されたい。
【数２４】

また、上記（２２）式中の｜Ａ（Ｘ_p，Ｘ_q）｜についても同様の理由により、Ｘ方向に関して定数と見なして、（２５）式のように近似する。
【数２５】

【００５２】
以上の（２４）式及び（２５）式に示した近似を用いると、式ｓ_d（Ｘ_p，Ｔ）のフーリエ変換は簡単になる。いま、以下の（２６）式に示す関数Ｇ_q（Ｘ，Ｙ）を定義する。
【数２６】

この関数Ｇ_q（Ｘ，Ｙ）を用いて（２１）式を表記すると以下の（２７）式になる。尚、（２７）式中のｄは画素のピッチを示している。
【数２７】

ここで、上記（２７）式に示したように、式ｓ_d（Ｘ，Ｔ）は関数Ｇ_q（Ｘ，Ｙ）と関数Ｏ（Ｘ′，Ｙ′）とのコンボリューション、更にはΩ（Ｔ）とのコンボリューションの形で表されている。ここで、関数Ｇ_qはいわば撮像素子３５に設けられた各画素の感度を示す式であり、関数Ｏ（Ｘ′，Ｙ′）は撮像面に結像している試料４のパターンの像を示す式であり、Ω（Ｔ）は各画素のシャッタ速度を示す式である。
【００５３】
上記（２７）式のフーリエ変換を求めると以下の（２８）式になる。尚、（２８）式中のＦΩは、式Ωのフーリエ変換であり、ＦＯは式Ｏのフーリエ変換である。
【数２８】

【００５４】
更に、感度関数ｇのフーリエ変換をＦｇとすると、（２６）式に示した関数Ｇ_q（Ｘ，Ｙ）のフーリエ変換は、以下の（２９）式となる。
【数２９】

そこで、上記（２９）式を上記（２８）式に代入し、点像のフーリエ変換をＯＴＦと考えると、ＯＴＦは以下の（３０）式で表される。
【数３０】

【００５５】
上記（３０）中において、写像光学系２０の歪曲収差に関係するのは、以下の（３１）式に示す部分である。
【数３１】

上記（３１）式が写像光学系２０に歪曲収差が存在するときのＯＴＦを示している。
【００５６】
〔４〕ＯＴＦを最大にするｘｙステージ４０の移動速度
以上の〔２〕及び〔３〕の議論では、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している状態において、写像光学系２０の残存収差により生ずるＯＴＦの劣化について議論した。以下では、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期として、ＯＴＦを最大にするための非同期量を求める方法について説明する。
【００５７】
本実施形態では、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングを固定とし、ｘｙステージ４０の移動速度を変えることにより、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にしている。同期している状態（ｘｙステージ４０の移動速度がｖのとき）に得られるＯＴＦは上記の（３１）式から求められる。そこで、ｘｙステージ４０の移動速度を、以下の（３２）式で示したように、ｖからδｖだけ変化させた速度ｖ′に変えたときのＯＴＦについて考察する
【数３２】

【００５８】
上記（３２）式を（３０）式に代入し、ｘｙステージ４０の速度をｖ′としたときに得られるＯＴＦ′（μ，ν）は以下の（３３）式で表される。尚、このとき、Ｙ_q＝β・ｖ・ｔ_qを用いた。
【数３３】

ここで、上記（３３）式は、非走査方向（Ｘ方向）及び走査方向（Ｙ方向）のＯＴＦをともに含んでおり、走査方向（Ｙ方向）のみのＯＴＦ_Y′を抜き出すと以下の（３４）式で表される。
【数３４】

【００５９】
上記の（３４）式において、δｖを最適に選ぶことによりＯＴＦ_Y′をより１（最大）に近い値にすることができる。以上から本実施形態では、以下の（３５）式の値が最大となるように、式中のδｖを求めている。
【数３５】

上記（３５）式は、歪曲収差がないとしたときの写像光学系２０の倍率β、画素のＹ方向の位置に応じたＹ方向の歪曲収差の量δ_yq、撮像素子３５の各画素に設けられたシャッタが開状態にある時間Δｔ、画素のＹ方向の位置ｑ、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にするときのステージ４０の速度の変化分δｖ、及び試料４に形成されたパターンの空間周波数ν、更にはＡ_pqを含んでなる式である。この式に於いて、倍率β、歪曲収差の量δ_yq、及びシャッタが開状態にある時間Δｔは装置によって決まってしまう。よって、Ｙ方向の歪曲収差の影響による解像度の低下を防止するためには、試料４に形成されたパターンのピッチに応じてδｖを決定する必要がある。
【００６０】
以上、本発明の一実施形態による表面状態観察装置において、ＯＴＦを最大にする方法、つまり解像度の低下を防止する方法について説明したが、次に、上記方法を用いて求めた方法により求めたステージの移動速度でステージを移動させて、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にしたときに得られるＯＴＦと、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にしたときに得られるＯＴＦとのシミュレーション結果を説明する。
【００６１】
シミュレーション結果を説明する準備として、本実施形態で想定している歪曲収差について説明する。写像光学系２０により歪曲収差の発生する様子は異なるが、本実施形態の表面状態観察装置のように狭い視野を観察する写像光学系２０は、一般的に低次の成分が支配的になる。このため、ここでは、三次以下の歪曲収差を考える。また、写像光学系２０が光軸に対して回転対称に構成されており、このために歪曲収差も光軸に回転対称であるとする。いま、図９に示すように、試料４の表面において、写像光学系２０の光軸との交点から以下の（３６）式で示した距離ｒにある点が写像光学系２０の歪曲収差により撮像面３５ａでは光軸からＲの位置に結像するとする。図９は、光軸からｒにある距離の点と歪曲収差の影響により、この点が結像する位置Ｒとの関係を示す図である。
【数３６】

【００６２】
ここで、三次までの歪曲収差を考えると、上記ｒとＲとの関係は以下の（３７）式で表される。
【数３７】

また、上記ｒ及びＲをベクトル成分で表し、その関係を表記すると以下の（３８）式になる。尚、以下の（３８）式では、Ｒの座標を（Ｘ，Ｙ）、ｒの座標を（ｘ、ｙ）としている。
【数３８】

更に、上記（３７）式の両辺を二乗して、ｒ²について解くと以下の（３９）式になる
【数３９】

上記（３９）式を（３８）式に代入し、（β₃／β₁）≪１を考慮して、（β₃／β₁）の二次以上を無視すると、以下の（４０）式となる。
【数４０】

【００６３】
上記（４０）式と（１２）式とを比較すると、三次までの歪曲収差を考慮した場合には、（１２）式中のβ、Ｄ_x（Ｘ，Ｙ）、Ｄ_y（Ｘ，Ｙ）は以下の（４１）式で表される。
【数４１】

また、｜Ａ｜についても、上記（４１）式を（１５）式に代入すると、以下の（４２）式となる。
【数４２】

以下に示すシミュレーションでは、（４１）式中のβ、Ｄ_x（Ｘ，Ｙ）、Ｄ_y（Ｘ，Ｙ）及び（４２）式中の｜Ａ｜を用いて、ＯＴＦの評価を行っている。尚、撮像素子３５の走査方向（Ｙ方向）の大きさを８．２ｍｍ、非走査方向（Ｘ方向）の大きさを３８．５ｍｍ、撮像素子３５が備える画素のＸ方向及びＹ方向のピッチを共に１６μｍとしている。
【００６４】
まず、比較のために撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している状態をシミュレートする。図１０は、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している状態において得られたシミュレーション結果である。尚、このシミュレーションは、撮像素子３５の最周辺において０．４％の歪曲収差が生じているとしている。図１０中の符号Ｃ１０を付した曲線は撮像面３５ａの周辺において得られるＸ方向のＭＴＦ曲線であり、符号Ｃ１１を付した曲線は撮像面３５ａの中心位置において得られるＹ方向のＭＴＦ曲線であり、符号Ｃ１２を付した曲線は撮像面３５ａの周辺において得られるＹ方向のＭＴＦ曲線である。
【００６５】
また、図１１は、写像光学系２０の歪曲収差により生ずる撮像素子３５の撮像面３５ａ上における像の画素に対するずれ量を示す図である。図１１において、符号Ｄ１０を付した曲線は撮像面３５ａの周辺におけるＸ方向の像のずれ量を示しており、符号Ｄ１１を付した曲線は撮像面３５ａの中心におけるＹ方向の像のずれ量を示しており、符号Ｄ１２を付した曲線は撮像面３５ａの周辺におけるＹ方向の像のずれ量を示している。図１０及び図１１から分かるように、走査方向（Ｙ方向）に直交する非走査方向（Ｘ方向）の歪曲収差の影響よりも、走査方向（Ｙ方向）のＭＴＦが悪化している。ここで、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している状態においては、前述した（３１）式において、ｔ_q＝Ｙ_q／ｖなる関係があるため、Ｘ方向のＯＴＦ_X（μ，ν）及びＹ方向のＯＴＦ_Y（Ｘ，Ｙ）は以下の（４３）式となる。
【数４３】

【００６６】
上記（４３）式を参照すると、何れもｑについて和をとっているが、これは撮像素子３５が備える各画素の信号をＹ方向に加算することを意味する。しかしながら、（４３）式中のδ_xp,δ_yqの値が大きくとも、これらの値がｑに関して一定ならばＯＴＦの劣化は生じないことになる。ＯＴＦの劣化を引き起こすのは、（４３）式中のδ_xp,δ_yqの位置による差である。ここで、図１１中の符号Ｄ１０を付した曲線がδ_xpを示しており、符号Ｄ１１又は符号Ｄ１２を付した曲線がδ_yqを示している。
【００６７】
ここで、図１０と図１１との関連について考えてみると、図１１に示したように、δ_xpの値（曲線Ｄ１０の値）はδ_yqの値（曲線Ｄ１１又は曲線Ｄ１２の値）よりも大きい。しかしながら、図１０をみると撮像面３５ａの中心位置において得られるＸ方向のＭＴＦ曲線を示す曲線Ｃ１０の劣化はほとんど無い。一方、図１１中に示したようにδ_yqの値（曲線Ｄ１１又は曲線Ｄ１２の値）そのものは小さい。しかしながら、δ_xpの最大値と最小値との差（曲線Ｄ１０の差）は３μｍ程度であるのに対し、例えば符号Ｄ１２を付した曲線の最大値と最小値との差は３２μｍ程度と極めて大きい。その差の結果が図１０に反映されており、撮像面３５ａの周辺において得られるＹ方向のＭＴＦ曲線である曲線Ｃ１２が悪化している。
【００６８】
次に、本実施形態のように、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にした場合についてシミュレートする。図１２は、撮像面に結像した像を撮像素子３５が電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にした状態において得られたシミュレーション結果である。尚、このシミュレーションは、撮像素子３５の最周辺において０．４％の歪曲収差が生じているとしており、また、（３４）式において、δｖを最適に選んでいる。図１２中の符号Ｃ２０を付した曲線は撮像面３５ａの周辺において得られるＸ方向のＭＴＦ曲線であり、符号Ｃ２１を付した曲線は撮像面３５ａの中心位置において得られるＹ方向のＭＴＦ曲線であり、符号Ｃ２２を付した曲線は撮像面３５ａの周辺において得られるＹ方向のＭＴＦ曲線である。
【００６９】
また、図１３は、撮像面に結像した像を撮像素子３５が電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にした状態において、写像光学系２０の歪曲収差により生ずる撮像素子３５の撮像面３５ａ上における像の画素に対するずれ量を示す図である。図１３において、符号Ｄ２０を付した曲線は撮像面３５ａ上におけるＸ方向の像のずれ量を示しており、符号Ｄ２１を付した曲線は撮像面３５ａの中心におけるＹ方向の像のずれ量を示しており、符号Ｄ２２を付した曲線は撮像面３５ａの周辺におけるＹ方向の像のずれ量を示している。
【００７０】
図１３を参照すると分かるように、試料４に形成されたパターンの空間周波数に応じて、（３５）式に示される値が最大となるδｖを求めてステージ４０の速度を設定することにより、撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度との非同期量を制御すると、符号Ｄ２１を付した曲線で示されるように、撮像面３５ａの中心におけるＹ方向の像のずれ量の差は大きくなる。しかしながら、符号Ｄ２２を付した曲線で示されるように撮像面３５ａの周辺におけるＹ方向の像のずれ量の差は図１１中に示した曲線Ｄ１２と比較すると小さくなる。また、図１３において符号Ｄ２２を付して示す撮像面３５ａの周辺におけるＹ方向の像のずれ量を示す曲線の差が小さくなったことに伴って、撮像面３５ａの周辺において得られるＹ方向のＭＴＦ曲線（図１２中のＭＴＦ曲線Ｃ２２）の悪化が改善されていることがわかる。
【００７１】
以上、本発明の一実施形態による表面状態観察装置の構成及び本発明の特徴部分である、非同期量の制御方法について説明したが、次に上記構成における本発明の一実施形態による表面状態観察装置の観察時における動作について説明する。まず、観察対象の試料４をローダ（図示省略）上に載置すると、試料４はローダによって図１に示したチャンバ３内に搬送される。試料４がチャンバ３内に搬入されるとｘｙステージ４０上に載置され、主制御系３７は駆動装置４３を介してｘｙステージ４０を駆動して、試料４を計測範囲に移動する。次に、ユーザは、入力装置４５を操作して、試料４に形成されているパターンにのピッチを入力する。
【００７２】
入力装置４５からパターンのピッチが入力されると、主制御系３７は記憶装置４４から写像光学系２０に歪曲収差量δ_yp、写像光学系２０の倍率β、撮像素子３５に設けられているシャッタが開状態にある時間Δｔ、及び撮像素子３５の行方向及び列方向の画素数等を読み出し、前述した（３５）式を最大にするようなδｖ（これは、本発明にいう非同期量に相当する）を算出する。ここで、図１４に示した試料４の表面状態を観察する場合を考える。図１４は、表面状態の観察を行う際の動作を説明するための図である。ここで、試料４の斜線を付した観察対象領域４ａを観察する場合について考える。
【００７３】
観察にあたり、主制御系３７はまず図示せぬＺセンサの計測結果に基づいて制御信号を写像光学系制御部３９へ出力して第２光学系２０の倍率を調整し、試料４から放出される二次電子ビームＢ２の焦点位置を電子ビーム検出器３０の検出面上に設定する。そして、主制御系３７は熱電子放出型電子銃１０に対して電子の放出を開始させ、照明光学系１１、イー・クロス・ビー２３、開口絞りＡＳ、第１アライナ２２、及びカソードレンズ２１を介して照明用ビームＢ１を試料４に照射するとともに、駆動装置４３を介してｘｙステージ４０を上述した処理で求めたδｖを考慮した一定速度ｖ＋δｖで図１４中Ｙ軸方向と平行な方向（図１４中符号ｄを付して示した方向）へ移動させる。試料４に照明用ビームＢ１を照射することにより、試料４の物体面から発生する二次電子ビームＢ２は写像光学系２０で集光され、その像は電子ビーム検出器３０の検出面に結像する。電子ビーム検出器３０から放出された光、即ち試料４の光学像はリレーレンズ３４を透過して画素が写像光学系２０の歪曲収差に応じて配列された撮像素子３５の撮像面に結像する。撮像素子３５の結像面に結像した光学像は、撮像素子３５によって画素信号に光電変換される。
【００７４】
図１４において、符号５０を付した領域が撮像素子３５で撮像される領域であるとし、領域５０中の符号Ｒ１を付した線状の領域を図２に示したＣＣＤ画素列Ｒ１が撮像し、符号Ｒ２５６を付した線状の領域をＣＣＤ画素列Ｒ２５６が撮像するとする。図１４に示した状態から観察を開始したとすると、観察対象領域４ａの座標（Ｘ１，Ｙ１）から（Ｘ１０２４，Ｙ１）までの光学像が、撮像素子３５をなすＴＤＩアレイＣＣＤのＣＣＤ画素列Ｒ１において光電変換されて、電荷が蓄積される。試料４が−Ｙ軸方向に移動することにより、撮像素子３５によって撮像される領域５０が符号ｄを付した方向に移動し、ｘｙステージ４０が速度ｖで移動していると仮定したときに領域５０が１ＣＣＤ画素列分だけ移動した時間が経過すると、主制御系３７から撮像素子３５へ垂直クロック信号が１つ送出される。
【００７５】
垂直クロック信号が送出されてくると、撮像素子３５のＣＣＤ画素列Ｒ１に蓄積されていた電荷は、ＣＣＤ画素列Ｒ２に転送される。尚垂直クロック信号が送出されてくる前にＣＣＤ画素列Ｒ１で撮像されていた光学像は、写像光学系２０の歪曲収差及びステージ４０の移動速度ｖ＋δｖに応じてＣＣＤ画素列Ｒ２の位置からずれた位置に移動する。写像光学系２０に歪曲収差がなく、且つｘｙステージ４０が速度ｖで移動している場合には、ＣＣＤ画素列Ｒ２は、垂直クロック信号が送出されてくる前にＣＣＤ画素列Ｒ１で撮像されていた光学像と同様の光学像を、次の垂直クロック信号が送出されてくるまで撮像することになる。しかしながら、実際には写像光学系２０の歪曲収差のために光学像の位置がずれてしまう。本実施形態ではこの位置ずれを緩和するために、ｘｙステージ４０の移動速度をｖ＋δｖとして各ＣＣＤ画素が光学像を電気信号に変換するタイミングと、ｘｙステージ４０の移動速度（ひいては、撮像素子３５の撮像面上における光学像の移動速度）とを非同期にしている。
【００７６】
以下同様にして、ＣＣＤ画素列Ｒ１によって撮像される領域５０の中のＲ１を付した線状の部分が（Ｘ１，Ｙ２５６）から（Ｘ１０２４，Ｙ２５６）までの位置に達すると撮像は終了し、（Ｘ１，Ｙ１）から（Ｘ１０２４，Ｙ１）までの試料画像が、コントロールユニット３６を介して初めて主制御系３７へ出力され始める。次に送出されてくる垂直クロック信号からは、（Ｘ１，Ｙ２）から（Ｘ１０２４，Ｙ２）までの試料画像が主制御系３７へ出力され、以下順々に画像が主制御系３７に取得されて観察が行われる。
【００７７】
以上説明したように、本発明の一実施形態による表面状態観察装置によれば、照明用ビームＢ１を試料４に照射して得られる二次電子ビームＢ２が写像光学系２０の光学特性に起因して撮像素子３５の撮像面３５ａ上において位置ずれを生じていても、撮像素子３５の各画素が撮像面に結像した光学像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にしているため、撮像面上の光学像の走査方向（Ｙ方向）の位置ずれ量を緩和することによりＯＴＦの悪化を改善することができ、その結果として分解能を向上させることができる。また、本実施形態によれば、撮像素子３５として用いているのは検出感度の高いＴＤＩアレイＣＣＤであり、ＴＤＩアレイＣＣＤを用いて上述の方法により分解能を向上させることができれば、検出信号の強度不足なく短時間で高分解能の観察を行うことができるという効果がある。
【００７８】
以上、本発明の一実施形態による表面状態観察装置について説明したが、本発明は上記実施形態に制限されることなく本発明の範囲内で自由に変更が可能である。例えば、上述した一実施形態及び他の実施形態では、照明用ビームとして照明用ビームＢ１を用い、照明用ビームＢ１を試料４に照射して得られる二次電子、反射電子、及び後方散乱電子の少なくとも１つを含む二次電子ビームＢ２を蛍光板３２で光電変換し、変換された光学像を撮像素子３５で撮像する場合を例に挙げたが、例えば電子ビームの代わりにイオンビームを用いても良く、可視領域の光を試料４に照射して得られる光学像を直接撮像素子３５で撮像するようにしても良い。この場合においては、写像光学系２０が光学レンズを含んで構成されるが、写像光学系２０の歪曲収差を完全に補正するために写像光学系２０を構成する光学レンズの数を増加させる必要がないため、装置の大型化を招かずに高分解能で観察することができる。照明用ビームとして光を用いた場合には、二次ビームは光を物体表面に照射して得られる反射光である。
【００７９】
また、上記実施形態においては、イー・クロス・ビー２３を用いて照明用ビームＢ１を偏向させて試料４に照射し、試料４から発生した二次電子ビームＢ２を直進させる場合について説明した。しかしながら、本発明はこれに限られず、照明用ビームＢ１を直進させて、二次電子ビームＢ２を偏向させる電磁プリズムを用いても良い。更に、本実施形態は線源からの電子ビームにて試料の物体面を照明し像面ヘ結像する、いわゆる面から面への写像投影光学系を備えているが、観察装置及び検査装置の単体装置としてではなく、半導体露光装置等にも簡単に応用することができる。
【００８０】
更に、上記実施形態では、入力手段として入力装置４５を備え、試料４に形成されたパターンのピッチを入力するようにしていたが、予め測定する試料４に形成されているパターンをハードディスク等の外部記憶装置に入力しておき、その内容を読み出してステージ４０の移動速度を算出するようにしても良い。また、上記実施形態では、ｘｙステージ４０を移動させていたが、試料４を載置するステージを固定にし、照明用ビームＢ１を試料４に照射して得られる二次ビームを、偏向させる偏向装置を設けて検出面又は撮像面に結像した像を検出面又は撮像面に対して相対的に移動させるようにしてもよい。更にまた、上記実施形態では試料４のｘｙ座標系の位置に関わらずｘｙステージ４０を一定の速度ｖ＋δｖで移動させていたが、写像光学系２０の歪曲収差に応じてｘｙステージ４０の移動速度を可変する制御を行っても良い。
【００８１】
次に、本発明の一実施形態によるマイクロデバイスの製造方法について説明する。図１５は、マイクロデバイス（ＩＣやＬＳＩ等の半導体チップ、液晶パネル、ＣＣＤ、薄膜磁気ヘッド、マイクロマシン等）の製造例のフローチャートを示す図である。図１５に示すように、まず、ステップＳ１０（設計ステップ）において、マイクロデバイスの機能・性能設計（例えば、半導体デバイスの回路設計等）を行い、その機能を実現するためのパターン設計を行う。引き続き、ステップＳ１１（マスク製作ステップ）において、設計した回路パターンを形成したマスク（レチクル）を製作する。一方、ステップＳ１２（ウェハ製造ステップ）において、シリコン等の材料を用いてウェハを製造する。
【００８２】
次に、ステップＳ１３（ウェハ処理ステップ）において、ステップＳ１０〜ステップＳ１２で用意したマスクとウェハを使用して、後述するように、リソグラフィ技術等によってウェハ上に実際の回路等を形成する。次いで、ステップＳ１４（デバイス組立ステップ）において、ステップＳ１３で処理されたウェハを用いてデバイス組立を行う。このステップＳ１４には、ダイシング工程、ボンティング工程、及びパッケージング工程（チップ封入）等の工程が必要に応じて含まれる。最後に、ステップＳ１５（検査ステップ）において、ステップＳ１４で作製されたマイクロデバイスの動作確認テスト、耐久性テスト等の検査を行う。このステップＳ１５において、前述した本発明の一実施形態による表面状態観察装置によって、マイクロデバイスの表面状態が観察される。こうした工程を経た後にマイクロデバイスが完成し、これが出荷される。尚、図１５に示したフローチャートでは、ステップＳ１４を経た後に、本発明の一実施形態による表面状態観察装置によってマイクロデバイスの表面状態を観察している場合を例示しているが、ステップＳ１４のデバイスを組み立てる前に、ステップＳ１３を経ることにより処理が終了したウェハの表面状態を適宜観察してもよい。
【００８３】
【発明の効果】
以上、説明したように、本発明によれば、写像光学系の光学特性に応じて生ずる検出面上での光学像の位置ずれ量に応じて、撮像手段が検出面に結像した像を電気信号に変換するタイミングと検出面上における像の移動速度とを非同期にするのみならず、その非同期の量を制御するようにしているため、写像光学系の歪曲収差による解像度の低下を防止することができ、その結果として写像光学系に歪曲収差が残存していても高分解能で観察することができる。
【図面の簡単な説明】
【図１】 本発明の一実施形態による表面状態観察装置の構成を示す図である。
【図２】 撮像素子３５としてのＴＤＩアレイＣＣＤの構成例を示す機能ブロック図である。
【図３】 本発明の一実施形態による表面状態観察装置の照明用ビームＢ１の軌道を示す図である。
【図４】 本発明の一実施形態による表面状態観察装置の二次電子ビームＢ２の軌道を示す図である。
【図５】 本発明の一実施形態による表面状態観察装置が備えるイー・クロス・ビー２３の構成及び動作原理を説明するための図である。
【図６】 試料４の表面に設定されたｘｙ座標系を撮像素子３５の撮像面に設定されたＸＹ座標系を示す図である。
【図７】 撮像素子３５が備える画素の配列の一例を示す図である。
【図８】 Ｙ方向の位置が画素の中心に設定された感度特性関数ｇ（Ｘ）の一例を示す図である。
【図９】 光軸からｒにある距離の点と歪曲収差の影響により、この点が結像する位置Ｒとの関係を示す図である。
【図１０】 撮像素子３５が撮像面に結像した像を電気信号に変換するタイミングとｘｙステージ４０の移動速度とが同期している状態において得られたシミュレーション結果である。
【図１１】 写像光学系２０の歪曲収差により生ずる撮像素子３５の撮像面３５ａ上における像の画素に対するずれ量を示す図である。
【図１２】 撮像面に結像した像を撮像素子３５が電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にした状態において得られたシミュレーション結果である。
【図１３】 撮像面に結像した像を撮像素子３５が電気信号に変換するタイミングとｘｙステージ４０の移動速度とを非同期にした状態において、写像光学系２０の歪曲収差により生ずる撮像素子３５の撮像面３５ａ上における像の画素に対するずれ量を示す図である。
【図１４】 表面状態の観察を行う際の動作を説明するための図である。
【図１５】 マイクロデバイスの製造工程の一例示すフローチャートである。
【図１６】 歪曲収差の一例を示す図である。
【図１７】 歪曲収差が生じている像をＴＤＩアレイＣＣＤで検出した場合の検出精度の悪化を説明する図である。
【図１８】 歪曲収差が生じている像をＴＤＩアレイＣＣＤで検出した場合のスキャン方向の検出精度の悪化を説明する図である。
【符号の説明】
４ 試料（物体）
１１ 照明光学系
２０ 写像光学系
３０ 電子ビーム検出器（撮像手段）
３２ 蛍光板（変換手段）
３４ リレーレンズ（撮像手段）
３５ 撮像素子（撮像手段）
３７ 主制御系（制御手段、算出装置）
４０ ｘｙステージ（ステージ）
４４ 記憶装置（記憶手段）
４５ 入力装置（入力手段）
Ｂ１ 照明用ビーム
Ｂ２ 二次電子ビーム（二次ビーム）[0001]
BACKGROUND OF THE INVENTION
The present invention irradiates an object with a charged particle beam such as an electron beam or an ion beam or an electromagnetic wave such as light or X-ray as an illumination beam, and a secondary beam (secondary electron) obtained by irradiating the object with the illumination beam. , Reflected electron, backscattered light, reflected light) to observe and inspect the surface state of an object, and a surface state observation apparatus and method, and a semiconductor element observed using the surface state observation apparatus and method The present invention also relates to a method of manufacturing a microdevice, which includes a step of observing the surface of the microdevice using the surface state observation apparatus and method.
[0002]
[Prior art]
A semiconductor element is manufactured by forming a fine pattern on the surface of a semiconductor substrate by using a planar technology, but the pattern is miniaturized and highly integrated due to a demand for miniaturization of the semiconductor element. A charged particle beam microscope using charged particles such as an electron beam (electron beam) is used for observing the surface state of the semiconductor element and performing defect inspection. A microscope that has been generally known as a charged particle microscope and is frequently used is a scanning electron microscope (SEM). The scanning electron microscope irradiates a point on the surface of an object with an electron beam, detects secondary electrons, reflected electrons, and backscattered electrons generated from the point irradiated with the electron beam, and irradiates the object with such an operation. An image of the object surface is formed by performing scanning while relatively scanning the electron beam. The scanning electron microscope has a problem that it takes a relatively long time for observation because it is necessary to scan the electron beam over the entire region to be observed.
[0003]
In recent years, a mapping electron microscope has been devised as an apparatus for solving this problem. This mapping electron microscope irradiates a charged particle beam such as an electron beam or an ion beam or an electromagnetic wave such as light or X-ray in a planar manner on the observation surface of the object, secondary electrons generated from the surface of the object, reflected electrons, This is a device that observes the surface of an object by accelerating and focusing backscattered electrons, enlarging and projecting the image by an electron optical system, forming it on the imaging surface, and converting the electron intensity distribution into a light intensity distribution. . A mapping electron microscope for irradiating an object with an electron beam is disclosed in, for example, Japanese Patent Application Laid-Open Nos. 10-197462 and 11-64256.
[0004]
In the mapping type electron microscope, a control method at the time of observation differs depending on whether a sensor such as a CCD (Charge Coupled Device) provided in the electron optical system is a one-dimensional sensor or a two-dimensional sensor. A mapping electron microscope equipped with a one-dimensional sensor has a control method for obtaining a light intensity distribution sequentially while moving the stage on which the object to be observed is placed at a constant speed and scanning the image of the object with respect to the sensor. Basic. On the other hand, a mapping electron microscope equipped with a two-dimensional sensor is an operation in which a stage on which an object is placed is moved by a certain distance and imaged by the sensor, and then the stage is further moved by a certain distance and imaged by the sensor. That is, a control method for obtaining the light intensity distribution by step-and-repeat of the stage is fundamental.
[0005]
A mapping electron microscope provided with a one-dimensional sensor has the following advantages compared to a mapping electron microscope provided with a two-dimensional sensor. First, since the field of view of an electron optical system that expands secondary electrons, reflected electrons, and backscattered electrons generated from the surface of an object may be narrower than that provided with a two-dimensional sensor, electrons that take aberration correction into consideration There is an advantage that the design / manufacturing cost of the optical system can be reduced. Second, even when observing a wide range, there is an advantage that stage control can be simplified because it is only necessary to move the stage at a constant speed. For these reasons, the mapping electron microscope often includes a one-dimensional sensor.
[0006]
Incidentally, a mapping electron microscope disclosed in Japanese Patent Application Laid-Open No. 10-197462 has been devised as a microscope developed from a mapping electron microscope provided with a one-dimensional sensor having the above characteristics. In this mapping type electron microscope, an object to be observed is placed on a stage, and a two-dimensional image formed on an imaging surface while moving the stage is converted into a TDI (Time-Delay-Integration) array CCD ( It is detected by Charge Coupled Device). Here, the configuration of the TDI array CCD will be briefly described as follows. The TDI array CCD has a plurality of rows (for example, 256 rows) and a plurality of columns (for example, 1024 columns). Pixels are arranged at intersections between the rows and columns, and are stored in the pixels arranged in the rows. In this configuration, the generated charges are transferred to adjacent rows for each column. In other words, the TDI array CCD can be considered as a line sensor for the number of rows having a length for the number of columns. Therefore, if the charge transfer timing and the stage movement timing in the TDI array CCD are synchronized, the same signal is sequentially integrated every time the charge is transferred, so that the detection sensitivity can be improved.
[0007]
[Problems to be solved by the invention]
By the way, the charge transfer timing in the TDI array CCD is fixed, and in order to obtain a high-resolution observation result, the stage movement timing (moving speed) is controlled with high accuracy to match the charge transfer timing in the TDI array CCD. There is a need. In addition, in order to accurately inspect the surface state of an object, aberrations of the mapping optical system that focuses secondary electrons, reflected electrons, and backscattered electrons from the object surface to form an image on the imaging surface, in particular, distortion aberration. It is necessary to suppress as much as possible. FIG. 16 is a diagram illustrating an example of distortion. Now, as shown in FIG. 16A, when the lattice shape is arranged on the object surface (for example, the surface of the object to be observed), when the mapping optical system has a negative distortion, the image is displayed. An image formed on a surface (for example, an imaging surface) is deformed into a barrel shape as shown in FIG. When such a deformed image is picked up by the CCD, the surface of the object is observed in a state where an error corresponding to the deformation has occurred.
[0008]
Further, when the image shown in FIG. 16B is picked up by the TDI array CCD, the detection accuracy further deteriorates due to the configuration of the TDI array CCD. The cause will be described below. FIG. 17 is a diagram for explaining deterioration in detection accuracy when an image in which distortion is generated is detected by a TDI array CCD. For the sake of easy understanding, it is assumed that there is a point light source on the observation surface on the object. When observing with a TDI array CCD, since the stage is moved as described above, the trajectory tr1 of the point light source on the object surface (observation surface) is linear as shown in FIG. When there is no distortion in the mapping optical system, the locus of the point light source image on the image plane is also linear.
[0009]
However, when distortion occurs in the mapping optical system, the image formed on the image plane is deformed as shown in FIG. 16B, so that when the point light source moves on the object plane, the image plane In FIG. 17B, the locus of the image of the point light source is not a straight line, but becomes an arc-shaped curved locus tr2, as shown in FIG. When such a trajectory tr2 of the arcuate curve is detected by the TDI array CCD, in the rows r1 and r3 located at the end of the TDI array CCD, it is observed by pixels in the jth column (j is a natural number), and the center of the TDI array CCD is observed. In row r2 located in the part, detection is performed with pixels in the i-th column (i is a natural number satisfying i <j) (this relationship is reversed on the right side in the figure). As described above, since the TDI array CCD sequentially transfers the accumulated charges to adjacent rows, the image of the point light source on the object plane is the width of the interval Δx between the i-th column and the j-th column on the image plane. Obsessed and observed. As described above, when an optical image having distortion is detected by the TDI array CCD, the image is deteriorated.
[0010]
Assuming that there is no distortion in the scanning direction (Y direction in FIG. 17), the image of the point light source on the object surface is sequentially observed by different pixels arranged in the same column at regular time intervals. Since the time interval at which the light source is observed and the charge transfer timing of the TDI array CCD are synchronized, the point light source in which the signal intensity is finally integrated from the TDI array CCD by the number of pixels arranged in the same column Is output. However, if distortion occurs in the mapping optical system, the moving speed of the point light source image on the TDI array CCD changes according to the distortion even if the moving speed of the point light source on the object surface is constant. The moving speed of the image of the point light source and the charge transfer timing of the TDI array CCD are different, and as a result, the shape of the point light source extending in the Y direction according to distortion is observed (exactly X Since there is also distortion in the direction, a circular or elliptical image is observed).
[0011]
As described above, when an image through the mapping optical system in which distortion is generated is to be observed with the TDI array CCD, a blurred image is obtained in both the X direction and the Y direction, and the resolution is lowered. As shown in FIG. 17B, an image blurred by Δx in the X direction is obtained, but an image blurred by 2 · Δy at the maximum is obtained in the scanning direction as shown in FIG. FIG. 18 is a diagram for explaining deterioration in detection accuracy in the scanning direction when an image with distortion is detected by a TDI array CCD. As shown in FIG. 18, with respect to the scanning direction (Y direction), pixels arranged in the same column are an image distorted in the + Y direction (an image protruding in the + Y direction) and an image distorted in the −Y direction. Since both (an image protruding in the −Y direction) are detected, the resolution in the scan direction is lower than the resolution in the X direction shown in FIG. Therefore, in order to improve the overall resolution, it is first necessary to prevent a decrease in resolution in the Y direction (scan direction) as much as possible.
[0012]
The present invention has been made in view of the above circumstances, and a surface state observation apparatus and method capable of observing with high resolution even when distortion aberration remains in the mapping optical system, and the surface state observation An object of the present invention is to provide a semiconductor device observed using the apparatus and method, and a method of manufacturing a microdevice including a step of observing the surface of the microdevice using the surface state observation apparatus and method.
[0013]
[Means for Solving the Problems]
  In order to solve the above problems, a surface state observation apparatus according to a first aspect of the present invention includes an illumination optical system that irradiates an illumination beam on the surface of an object to be observed, and the illumination beam on the object surface. A mapping optical system that forms an image of a secondary beam obtained by irradiation on a predetermined detection surface, an image formed on the detection surface is converted into an electric signal, and the electric signal is converted to an image of the image on the detection surface. An imaging unit that outputs an accumulated signal while being transferred in the moving direction, and a conversion timing of the imaging unit and a detection surface on the detection surface according to an amount of positional deviation of the image caused by an optical characteristic of the mapping optical system Control means for controlling an asynchronous amount with respect to the moving speed of the image, the control means having an optical transfer function obtained from a signal output from the imaging means at a pitch of a pattern formed on the object. According to the sky of the statue It is characterized by controlling the asynchronous amount so as to maximize the frequency.
  According to the present invention, the timing at which the imaging means converts the image formed on the detection surface into an electric signal and detection according to the positional deviation amount of the optical image on the detection surface that occurs according to the optical characteristics of the mapping optical system. Not only is the movement speed of the image on the surface asynchronous, but the amount of the asynchronous movement is controlled, so that it is possible to prevent a reduction in resolution due to distortion in the mapping optical system. Even if distortion remains in the optical system, it can be observed with high resolution.
  The surface state observation apparatus according to the second aspect of the present invention is the surface state observation apparatus according to the first aspect, further comprising a stage configured to be movable with the object placed thereon, wherein the control means The asynchronous amount between the conversion timing of the imaging means and the moving speed of the image on the detection surface is controlled by making the moving speed of the stage variable according to the amount of positional deviation of the image. Yes.
  The surface state observation apparatus according to a third aspect of the present invention is the surface state observation apparatus according to the first aspect, wherein the control means is configured to perform the imaging in a state where the moving speed of the image on the detection surface is constant. The asynchronous amount is controlled so that an optical transfer function obtained from a signal output from the means becomes maximum at a spatial frequency of the image corresponding to a pitch of a pattern formed on the object.
  The surface state observation device according to a fourth aspect of the present invention is the surface state observation device according to the third aspect, wherein the spatial frequency is a spatial frequency in the moving direction of the image on the detection surface. It is said.
  The surface state observation apparatus according to the fifth aspect of the present invention is the surface state observation apparatus according to the third aspect or the fourth aspect, wherein the mapping optics obtained in advance from a design value of the mapping optical system or from an actual measurement value. Storage means for storing the optical characteristics of the system, input means for inputting the pitch of the pattern formed on the object, optical characteristics of the mapping optical system stored in the storage means, and the input from the input means Calculating means for calculating an asynchronous amount of the image with respect to the conversion timing based on a pattern pitch.
  The surface state observation apparatus according to the sixth aspect of the present invention is the surface state observation apparatus according to any one of the first to fifth aspects, wherein the imaging means is two-dimensionally equidistant on the detection surface. And the conversion timing of each pixel is the same.
  Further, a surface state observation apparatus according to a seventh aspect of the present invention is obtained by irradiating an illumination optical system for irradiating an illumination beam onto the surface of an object to be observed, and irradiating the object beam with the illumination beam. A mapping optical system that forms an image of the next beam on a predetermined detection surface, and an image formed on the detection surface is converted into an electric signal, and the electric signal is transferred in the moving direction of the image on the detection surface. An imaging unit that outputs an integrated signal, and a conversion timing of the imaging unit and a moving speed of the image on the detection surface according to an amount of positional deviation of the image caused by optical characteristics of the mapping optical system. Control means for controlling the asynchronous amount of the image, and the positional deviation of the image caused by the optical characteristics of the mapping optical system is a positional deviation due to distortion aberration remaining in the mapping optical system. .
  The surface state observation apparatus according to an eighth aspect of the present invention is the surface state observation apparatus according to any one of the first to seventh aspects, wherein the illumination beam is a charged particle beam, and the secondary beam Is characterized in that it is at least one of secondary electrons, reflected electrons, and backscattered electrons generated by irradiating the object with the charged particle beam.
  A surface state observation apparatus according to a ninth aspect of the present invention is the surface state observation apparatus according to the eighth aspect, wherein the imaging means converts an image formed on the detection surface by the mapping optical system into an optical image. Conversion means for converting is provided, and an optical image converted by the conversion means is photoelectrically converted into the electric signal.
  In order to solve the above-described problems, the surface state observation method of the present invention provides a secondary beam obtained by irradiating an illumination beam on the surface of an object to be observed on a predetermined detection surface via a mapping optical system. The image formed and converted into an electric signal while moving the image formed on the detection surface, and the signal obtained by integrating the electric signal while transferring the electric signal in the moving direction of the image on the detection surface. In the surface state observation method for observing the surface state of an object, an image is formed on the detection surface in accordance with a displacement amount caused by distortion aberration remaining in the mapping optical system caused by the optical characteristics of the mapping optical system. Controlling an asynchronous amount between the conversion timing of an image into an electric signal and the moving speed of the image on the detection surface.
  Furthermore, in order to solve the said subject, the manufacturing method of the microdevice of this invention uses the surface state observation apparatus or the said surface state observation method in any one of the said 1st viewpoint to the 9th viewpoint. It includes the step of observing the surface of the device.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an apparatus and method for observing a surface state and a method for manufacturing a semiconductor element and a micro device according to an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing a configuration of a surface state observation apparatus according to an embodiment of the present invention. The surface state observation apparatus according to the present embodiment mainly detects a primary column 1 for accelerating an electron beam to a sample and a secondary electron beam generated when the sample is irradiated with an electron beam by an electron beam detector 30. It is composed of a secondary column 2 for focusing on a surface and a chamber 3 for accommodating a sample (object) 4 to be observed. The optical axis of the primary column 1 is set obliquely with respect to the vertical direction, and the optical axis of the secondary column 2 is set substantially vertical. Therefore, the illumination beam B1 is incident on the secondary column 2 from the oblique direction from the primary column 1. The primary column 1, the secondary column 2, and the chamber 3 are connected to an evacuation system (not shown), and are evacuated by a vacuum pump such as a turbo pump provided in the evacuation system. Maintained.
[0015]
A thermionic emission electron gun 10 is provided inside the primary column 1, and an illumination optical system 11 is disposed on the optical axis of the illumination beam B 1 emitted from the thermionic emission electron gun 10. Here, as a chip of the thermoelectron emission type electron gun 10, it is preferable that a large current can be taken out by a rectangular cathode, for example. The illumination optical system 11 includes a field stop FS1, irradiation lenses 12, 13, and 14, aligners 15 and 16, a scan aligner 17, an aperture 18, and the like. Here, the irradiation lenses 12, 13, and 14 are electronic lenses, and for example, a circular lens, a quadrupole lens, an octupole lens, or the like is used. The convergence characteristics of the irradiation lenses 12, 13, and 14 included in the illumination optical system 11 with respect to the illumination beam B1 are changed by changing the applied voltage. The irradiation lenses 12, 13, and 14 may be rotationally symmetric lenses called unipotential lenses or Einzel lenses.
[0016]
A mapping optical system 20 is disposed in the secondary column 2. The mapping optical system 20 is for converging the secondary electron beam B2 as a secondary beam generated when the illumination beam B1 is irradiated on the sample 4 to form an image on the detection surface of the electron beam detector 30. Cathode lens 21, first aligner 22, aperture stop AS, e-cross bee 23, stigmator 24, imaging lens front group 25, second aligner 26, stigmator 27, field of view in order from the sample 4 side vertically upward A stop FS2, an imaging lens rear group 28, and a third aligner 29 are arranged.
[0017]
The field stop FS2 provided in the mapping optical system 20 is set in a positional relationship conjugate with the object plane of the sample 4 with respect to the cathode lens 21 and the imaging lens front group 25. The imaging lens front group 25 and the imaging lens rear group 28 of the mapping optical system 20 are electronic lenses, and for example, a circular lens, a quadrupole lens, an octupole lens, or the like is used. The cathode lens 21, the imaging lens front group 25, and the imaging lens rear group 28 may be rotationally symmetric lenses called unipotential lenses or Einzel lenses. Further, the convergence characteristics of the cathode lens 21, the imaging lens front group 25, and the imaging lens rear group 28 included in the mapping optical system 20 with respect to the secondary electron beam B2, that is, the focal position of the secondary electron beam B2, is determined by the applied voltage. It changes by changing. Further, the deflection characteristic and the convergence characteristic of the e-cross bee 23 with respect to the illumination beam B1 and the secondary electron beam B2 are changed by changing the applied voltage or current.
[0018]
Further, an electron beam detector 30 is disposed in the vertical upward direction of the third aligner 29 provided in the mapping optical system 20. A secondary electron beam B2 emitted when the sample 4 is irradiated with the electron beam B1 is formed on the detection surface of the electron beam detector 30 by the mapping optical system 20. Here, the electron beam detector 30 includes an MCP (Micro Channel Plate) 31 for amplifying electrons, and a fluorescent plate as conversion means for converting an image formed on the detection surface of the electron beam detector 30 into an optical image. 32 and a vacuum window 33 for emitting light converted by the fluorescent screen to the outside of the secondary column 2 kept in a vacuum state. The light emitted from the electron beam detector 30, that is, the optical image of the sample 4 is transmitted through the relay lens 34, and the imaging device 35 of a TDI (Time-Delay-Integration) array CCD (Charge Coupled Device). Is incident on the imaging surface. The electron beam detector 30, the relay lens 34, and the image pickup device 35 correspond to the image pickup means in the present invention. In addition, since the detection surface of the electron beam detector 30 is optically conjugate with the detection surface of the electron beam detector 30 by the relay lens 34, hereinafter, the detection surface of the electron beam detector 30 is simply referred to as the detection surface. Both the surface and the imaging surface of the imaging device 35 are meant.
[0019]
Here, the configuration of the image sensor 35 will be described. FIG. 2 is a functional block diagram showing a configuration example of a TDI array CCD as the image sensor 35. The TDI array CCD shown in FIG. 2 has a line-shaped CCD pixel array by arranging 1024 pixels C1 to C1024 in the horizontal direction (X-axis direction in FIG. 2), and further in the vertical direction (FIG. 2 in the Y-axis direction), and 256 CCD pixel rows R1 to R256 are arranged. The CCD pixels are arranged at equal intervals in the row direction and the column direction, and the timing for converting the image formed on the imaging surface by each CCD pixel is set to be the same. Note that the interval in which the pixels are arranged in the row direction may be different from the interval in which the pixels are arranged in the column direction. The accumulated charges on the pixels arranged in each of the CCD pixel rows R1 to R256 are transferred by one CCD pixel row in the vertical direction at a time by one vertical clock signal supplied from the outside.
[0020]
Among images formed on the imaging surface of the image sensor 35, an image captured by the CCD pixel array R1 at a certain time moves in the vertical direction by one CCD pixel array, and a vertical clock signal is given in synchronization therewith. The pixel signals accumulated in each pixel of the CCD pixel row R1 are transferred to the CCD pixel row R2. In the CCD pixel row R2, since the same image as the image taken in the CCD pixel row R1 is picked up before the vertical clock signal is given, the charge of the image accumulated in the CCD pixel row R2 is doubled. Subsequently, when the image further moves in the vertical direction by one CCD pixel column and a synchronization clock signal is given, the accumulated image is transferred to the CCD pixel column R3, where three times the image charge is accumulated. When the transfer and imaging of the charge to the CCD pixel row R256 are repeated following the movement of the image in order, the result of accumulating 256 times the charge first accumulated in the CCD pixel row R1 is horizontal. The pixel signal is output serially from the output register HR. Here, the pixels included in the imaging device 35 are arranged at equal intervals in the horizontal direction (X-axis direction in FIG. 2) and the vertical direction (Y-axis direction in FIG. 2).
[0021]
A control unit 36 is connected to the image sensor 35. The control unit 36 serially reads out pixel signals output from the image sensor 35 and sequentially outputs them to the main control system 37. The main control system 37 performs image processing such as template matching on the pixel signal output from the control unit 36 to determine the presence or absence of a defect in the sample 4. The main control system 37 outputs control signals to the illumination optical system control unit 38 and the mapping optical system control unit 39 to control the optical characteristics of the illumination optical system 11 and the mapping optical system 20 and to control the electromagnetic wave of the e-cross bee 23. Boundary control is performed. If the image signal output from the control unit 36 to the main control system 37 is displayed on a display device such as a CRT (Cathod Ray Tube), the image of the sample 4 is displayed on the display device.
[0022]
Next, the configuration in the chamber 3 will be described. Inside the chamber 3, an xy stage 40 configured to be movable in an xy plane set substantially parallel to a horizontal plane in a state where the sample 4 is placed is disposed. An L-shaped movable mirror 41 is attached to one end on the xy stage 40, and a laser interferometer 42 is disposed at a position facing the mirror surface of the movable mirror 41. Although shown in a simplified manner in FIG. 1, the movable mirror 41 includes a plane mirror having a reflecting surface perpendicular to the x axis and a plane mirror having a reflecting surface perpendicular to the y axis. The laser interferometer 42 includes two x-axis laser interferometers that irradiate the moving mirror 41 with a laser beam along the x-axis and a y-axis laser that irradiates the movable mirror 41 with a laser beam along the y-axis. The x coordinate and the y coordinate of the xy stage 40 are measured by one laser interferometer for the x axis and one laser interferometer for the y axis. Further, the rotation angle of the xy stage 40 in the xy plane is measured based on the difference between the measurement values of the two x-axis laser interferometers.
[0023]
The measurement result of the laser interferometer 42 is output to the main control system 37. The main control system 37 outputs a control signal to the driving device 43 based on the measurement result, and controls the position of the xy stage 40 in the xy plane. To do. Although not shown in the drawing, in addition to the xy stage 40, a z stage that can change the position of the sample 4 in the z-axis direction (vertical direction) and a tilt that controls the inclination of the object surface of the sample 4 with respect to the xy plane. It is preferable to provide a stage.
[0024]
In addition to the above-described configuration, the surface state observation apparatus according to the present embodiment includes a storage device 44 as a storage unit that stores a distortion aberration amount of the mapping optical system 20 obtained in advance from a design value of the mapping optical system 20 or an actual measurement value, and a sample. 4 is provided with an input device 45 as input means for inputting the pitch of the pattern formed in the pattern 4. The storage device 45 is realized by an external storage device such as a hard disk, and the input device 45 is realized by a keyboard and a mouse. Note that the storage device 44 does not store only the amount of distortion of the mapping optical system 20, but also information about the magnification of the mapping optical system 20 and the image sensor 35 (for example, the shutter provided in the image sensor 35 is in an open state). A certain time and the number of pixels in the row direction and the column direction) are stored.
[0025]
The pitch of the pattern input from the input device 45 is output to the main control system 37, and the main control system 37 detects the detection surface (and thus imaging) according to the distortion amount of the mapping optical system 20 based on the value from the input device 45. The moving speed of the stage 20 is calculated so as not to reduce the resolution reduction caused by the image shift amount generated on the imaging surface of the element 35 as much as possible. Details of this calculation method will be described later. The main control system 37 corresponds to the control means and calculation means referred to in the present invention.
[0026]
The schematic configuration of the surface state observation apparatus according to the embodiment of the present invention has been described above. Next, the trajectories of the illumination beam B1 and the secondary electron beam B2 of the surface state observation apparatus according to the embodiment of the present invention will be described in detail. explain. FIG. 3 is a diagram showing the trajectory of the illumination beam B1 of the surface state observation apparatus according to the embodiment of the present invention. In FIG. 3, in order to facilitate understanding, illustration of some of the members included in the illumination optical system 11 is omitted. The illumination beam B1 emitted from the thermionic emission electron gun 10 is focused and diverged by the influence of the electric field formed by the irradiation lenses 12, 13, and 14, as shown in FIG. Here, when the major axis direction of the rectangular chip of the thermionic emission electron gun 10 is set to the a-axis direction and the minor axis direction is set to the b-axis direction, the rectangular cathode is emitted to the cross section in the a-axis direction. The electron trajectory is denoted by the symbol P in FIG._aThe trajectory of electrons emitted in the cross section in the b-axis direction of the rectangular cathode is denoted by the symbol P in FIG._bIt becomes the orbit shown with.
[0027]
After being affected by the electric field by the irradiation lenses 12, 13, and 14, the illumination beam B1 enters the e-cross bee 23 from an oblique direction. When the illumination beam B1 enters the e-cross bee 23, its optical path is deflected in a direction substantially parallel to the vertical direction. The illumination beam B1 deflected by the e-cross beam 23 reaches the aperture stop AS, and forms a crossover image of the thermionic emission electron gun 10 at this position. The illumination beam B1 that has passed through the aperture stop AS passes through the first aligner 22 and is then subjected to the lens action by the cathode lens 21 to cause Koehler illumination of the sample 4.
[0028]
When the sample 4 is irradiated with the illumination beam B1, the sample 4 includes secondary electrons, reflected electrons, and backscattered electrons having a distribution according to the surface shape, material distribution, potential change, and the like of the sample 4. An electron beam B2 is generated. Among these, the secondary electron beam B2 mainly composed of secondary electrons becomes an observation electron beam. The initial energy of the secondary electron beam B2 is low, about 0.5 to 2 eV. Next, the trajectory of the secondary electron beam B2 generated from the sample 4 will be described. FIG. 4 is a diagram showing the trajectory of the secondary electron beam B2 of the surface state observation apparatus according to the embodiment of the present invention. In FIG. 4, some members of the mapping optical system 20 are not shown for easy understanding.
[0029]
The secondary electron beam B2 emitted from the sample 4 sequentially passes through the cathode lens 21, the first aligner 22, the aperture stop AS, and the e-cross bee 23 included in the mapping optical system 20. When the secondary electron beam B2 passes through the e-cross bee 23, it is converged by the imaging lens front group 25 and forms an image of the sample 4 at the position of the field stop FS2. The secondary electron beam B2 that has passed through the field stop FS2 is converged again by the imaging lens rear group 28, and an enlarged image of the object surface of the sample 4 is formed on the detection surface of the electron beam detector 30. At this time, when distortion aberration remains in the mapping optical system 20, the image formed on the detection surface reflects the distortion aberration. When the secondary electron beam B2 enters the electron beam detector 30, the number of electrons is first amplified by the MCP 31, and then enters the fluorescent plate 32 to be converted into photons. The optical image converted by the fluorescent plate 32 forms an image on the imaging surface of the image sensor 35 via the relay lens 34. At this time, since the size of the image of the sample 4 on the fluorescent plate 32 is larger than the image pickup surface of the image pickup device 35, the relay lens 34 is set to reduce the optical image converted by the fluorescent plate 32 at a predetermined magnification. .
[0030]
Next, the e-cross bee 23 provided in the surface state observation apparatus according to the embodiment of the present invention will be described in detail. FIG. 5 is a diagram for explaining the configuration and operation principle of the e-cross bee 23 provided in the surface state observation device according to the embodiment of the present invention. FIG. 5A is a perspective view showing the configuration of the e-cross bee 23. As shown in FIG. 5A, after the illumination beam B 1 emitted from the thermionic emission electron gun 10 is converged by the lens action by the illumination optical system 11 and enters the e-cross bee 23. The orbit (optical path) is bent by the deflection action of the e-cross bee 23. This is because, as shown in FIG. 5B, when an electron of charge q (illumination beam B1) travels in the + z direction at a velocity v in an electric field E and a magnetic field B orthogonal to each other, the −x direction. Force F by electric field acting on_E(= QE) and force F by magnetic field_BThis is to receive the resultant force with (= qvB). Thereby, the trajectory of the illumination beam B1 is bent in the zx plane.
[0031]
On the other hand, the secondary electron beam B2 generated from the sample 4 irradiated with the illumination beam B1 is subjected to the lens action by the cathode lens 21, passes through the aperture stop AS arranged at the focal position of the cathode lens 21, and is -After entering the cross bee 23, go straight through the e-cross bee 23. This is due to the following reason. As shown in FIG. 5C, when an electron of charge q (secondary electron beam B2) travels in a −z direction at a velocity v in an electric field E and a magnetic field B orthogonal to each other, Force F by working electric field_EAnd the force F caused by the magnetic field acting in the + x direction_BReceive the combined power of. At this time, the force F by the electric field_EAnd force F by magnetic field_BAre set to be equal (E = vB), that is, to satisfy the Vienna condition. Therefore, the force F due to the electric field_EAnd force F by magnetic field_BAnd the apparent force received by the secondary electron beam B2 is zero, and the secondary electron beam B2 travels straight through the e-cross bee 23. As described above, the e-cross bee 23 has a function as a so-called electromagnetic prism that selects the optical path of the passing electron beam.
[0032]
The configuration of the surface state observation device according to the embodiment of the present invention has been described above. The conventional general surface state observation apparatus moves the xy stage 40 in a certain direction (this book) in synchronization with the timing for converting the image formed on the imaging surface of the image sensor 35 into an electrical signal and the timing for transferring the converted electrical signal. In the embodiment, the scanning is basically performed at a constant speed in the y direction). However, in the surface state observation apparatus according to the embodiment of the present invention, in order to improve the resolution, the timing at which the imaging element 35 converts the image formed on the imaging surface into an electrical signal, and the moving speed of the xy stage 40 (and eventually) The moving speed of the xy stage 40 is controlled so that the moving speed of the image on the detection surface becomes asynchronous.
[0033]
More specifically, in the present embodiment, the moving speed of the xy stage 40 is controlled so that the OTF (Optical Transfer Function) is maximized. Here, OTF is a concept for evaluating the image transmission performance of an optical system such as the imaging characteristics of a lens. In short, an image of an object having a spatial frequency f on the object surface is formed on the image plane by the optical system. It is the spatial frequency characteristic of the image obtained when it is made to do.
[0034]
The OTF is a function of the spatial frequency on the imaging surface of the image sensor 35 and has a distribution. In this embodiment, the OTF has a spatial frequency on the imaging surface corresponding to the pitch of the pattern formed on the sample 4. The moving speed of the xy stage 40 is controlled so that the OTF is maximized. Here, the decrease in resolution due to the distortion of the mapping optical system 20 is more remarkable in the scanning direction (y direction) than in the non-scanning direction (x direction). Therefore, even when OTF is considered, the xy stage 40 is set so that the OTF at the spatial frequency in the direction corresponding to the pitch in the y direction (Y direction on the imaging surface of the imaging device 35 (see FIG. 2)) is maximized. The movement speed is controlled. Hereinafter, an example of a method for controlling the moving speed of the xy stage 40 will be described in detail.
[0035]
Now, the coordinate system shown in FIG. 6 is set. FIG. 6 is a diagram illustrating an XY coordinate system in which the xy coordinate system set on the surface of the sample 4 is set on the imaging surface of the image sensor 35. In FIG. 6, reference numeral 35a denotes an imaging surface. In the following description, when the timing at which the image sensor 35 converts an image formed on the imaging surface 35a into an electrical signal and the moving speed of the xy stage 40 are synchronized, [1] the mapping optical system 20 is distorted. A signal obtained from the image sensor 35 when there is no aberration is obtained, [2] a signal obtained from the image sensor 35 when the distortion of the mapping optical system 20 is taken into account, and [3] from the image sensor 35. Obtain the OTF of the resulting signal. Finally, [4] the moving speed of the xy stage 40 that maximizes the OTF is obtained. Note that in the case of [4], the timing at which the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous.
[0036]
[1] A signal obtained from the image sensor 35 when it is assumed that the mapping optical system 20 has no distortion.
Assuming that the mapping optical system 20 has no distortion, assuming that the magnification of the mapping optical system 20 is β (here, generally β> 1), the xy coordinate system and the XY coordinate system shown in FIG. Is expressed by the following equation (1).
[Expression 1]

Assuming that the pattern formed on the sample 4 is O (x, y), this pattern O (x, y) can be expressed using the XY coordinate system from the above equation (1). If the moving speed in the y direction is v, it is expressed by the following equation (2).
[Expression 2]

[0037]
Next, a pixel signal output from the image sensor 35 will be considered. First, a signal output from one pixel among a plurality of pixels included in the image sensor 35 will be considered. FIG. 7 is a diagram illustrating an example of an array of pixels included in the image sensor 35. Now, as shown in FIG. 7, the X-direction pixels of the image sensor 35 are the first to m-th m pixels, the Y-direction pixels are the first to n-th n pixels, and the X-direction pixels Consider the output from the p (1 ≦ p ≦ m) th pixel and the q (1 ≦ q ≦ n) th pixel in the Y direction (the hatched pixel in FIG. 7). Further, it is assumed that the sensitivity characteristics of the respective pixels are all equal, and this sensitivity characteristic is represented by a function g (X, Y). FIG. 8 is a diagram illustrating an example of the sensitivity characteristic function g (X) in which the position in the Y direction is set at the center of the pixel. As shown in FIG. 8, the sensitivity characteristic function g (X) is a function having a value only in the pixel aperture. Note that the sensitivity characteristic function g (Y) is also a function having a value only in the opening as in FIG.
[0038]
Here, if a part of the pattern image formed on the sample 4 at the time t is on the target pixel, the signal g ′ (X_p, Y_q, T) is expressed by the following equation (3) using the sensitivity characteristic function g (X, Y). In the present specification, when the integration range is not explicitly specified in the integration range in the mathematical formula, the integration is performed from −∞ to ∞.
[Equation 3]

By the way, the image pickup device 35 is provided with a shutter for each pixel, and converts an optical image of a pattern into an electric signal only when the shutter is open. Therefore, let Δt be the time during which the shutter is open, and consider the function Ω (t) shown in the following equation (4).
[Expression 4]

The function Ω (t) expressed by the above equation (4) has a value “1” only during Δt when the shutter is in the open state, and has a value “0” during the time when the other shutters are in the closed state. It is a function. Therefore, the time t after considering the shutter_kThe signal g ″ (X_p, Y_q, T_k) Is expressed by the following equation (5).
[Equation 5]

[0039]
Next, consider a signal that considers all the pixels of the image sensor 35. As described above, the image sensor 35 integrates and outputs the signals converted by the pixels arranged in the Y-axis direction. Here, since the timing when the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are synchronized, the Y coordinate can be expressed by time. . Consider a case where an image of an arbitrary point P on the sample 4 is moving in the Y direction on the imaging surface 35a of the imaging device 35. Note that in this case, the image of the point P moves on the same column of the image sensor 35.
[0040]
If the time at which the image of the point P crosses (passes through) the row arranged at the center of the imaging surface 35a is T, the pixel at the position shifted by q pixels from the row in the Y direction is the point P. The time when the image passes is T + t_qIt becomes. T_q= (Y_q/ Β) / v. For this reason, at time T, the signal s (X_p, T) is expressed by the following equation (6).
[Formula 6]

Signal s (X) output from the image sensor 35 shown in the above equation (6)_p, T) has a discrete value in the X direction, which is inconvenient to handle in analyzing OTF described later. For this reason, when transformed into the equation S (X, T) using the continuous variable X, the following equation (7) is obtained.
[Expression 7]

Δ (X−X) on the right side in the equation (7)_p) Is a Dirac δ function. The above expression (7) is an expression showing a signal output from the image sensor 35 when the mapping optical system 20 has no distortion.
[0041]
[2] Signal obtained from the image sensor 35 when the distortion of the mapping optical system 20 is taken into consideration
In the above [1], the case where there is no distortion with the magnification β of the mapping optical system 20 as a constant was considered. In this case, an image obtained by enlarging or reducing the pattern formed on the sample 4 is formed on the imaging surface 35 a of the imaging element 35. However, when the mapping optical system 20 has distortion, the image on the imaging surface 35a is distorted. This image distortion can be interpreted as a local magnification variation of the mapping optical system 20. Therefore, the distortion is captured assuming that the magnification of the mapping optical system 20 differs according to the position in the xy coordinates, and the above-described equation (1) is expressed by the following equation (8).
[Equation 8]

As described above, if the magnification of the mapping optical system 20 differs depending on the position in the xy coordinates, the magnification β in the above equation (8)._x, Β_yIs a function of x and y, but the magnification β_x, Β_yAs a function of X and Y, the result is the same. Therefore, in consideration of facilitating the following handling, in the above equation (8), the magnification β_x, Β_yIs a function of X and Y.
[0042]
When the relationship between the xy coordinate system and the XY coordinate system is the relationship shown in the above equation (8), the image formed on the imaging surface of the image sensor 35 solves the equation (8) for each of x and y. Substituting this equation into the equation O (x, y) indicating the pattern on the sample 4, and further considering the moving speed v in the y direction of the xy stage 40, the following equation (9) is obtained.
[Equation 9]

A signal s output from the image sensor 35 when the distortion of the mapping optical system 20 is taken into account._d(X_p, T) is obtained by substituting the formula O (X / β, Y / β−v · t) in the formula (5) for the formula (9) into the formula (6) below. It is obtained according to the formula.
[Expression 10]

Finally, as in the case of [1] above, the following equation (11) is obtained by transforming into an equation using a continuous variable X in order to facilitate handling in analyzing the OTF described later.
## EQU11 ##

[0043]
[3] Calculation of OTF of signal obtained from image sensor 35
Hereinafter, the OTF of the signal obtained from the image sensor 35 can be obtained by performing Fourier transform on the signal output from the image sensor 35 represented by the above equations (10) and (11). However, it is normally designed so that the distortion aberration of the mapping optical system 20 is minimized. Therefore, in the following, the equation is simplified assuming that the distortion is small, and then the OTF is calculated.
[0044]
Simplification of formula (a)
Since the distortion of the mapping optical system 20 is small, the magnification β shown in the equation (8)_x(X, Y), β_y(X, Y) is considered to be very close to the constant magnification β used in [1]. For this reason, the magnification β_x(X, Y), β_y(X, Y) is represented by the following equation (12).
[Expression 12]

Here, D in the above equation (12)_x(X, Y), D_y(X, Y) is D_x(X, Y) << 1, D_y(X, Y) << 1.
[0045]
Next, the magnification shown in the above equation (12), β_x(X, Y), β_y(X, Y) is applied to the equation (10). Here, the variable conversion shown in the following equation (13) is performed. In the following formula, the magnification β is used for simplification of the formula._x(X, Y), β_y(X, Y) to β_x, Β_yRespectively.
[Formula 13]

In the variable conversion shown in the equation (13), the value of the X coordinate and the value of the Y coordinate that define the position of the image on the imaging surface 35a shown in the equation (9) are converted into variables X ′ and Y ′, respectively. Therefore, a coordinate system (X ′, Y ′) that moves with the movement of the image that moves on the imaging surface 35a of the image sensor 35 is set. Note that L in the equation (13) indicates an offset amount (distance) in the Y direction between the origin of the X′Y ′ coordinate system and the qth pixel. When the variable conversion shown in the above equation (10) is performed, the integral element dXdYdt in the equation (10) becomes dXdYdt = (1 / | A |) dX′dY′dL. Here, A is represented by the following equation (14).
[Expression 14]

[0046]
Here, when | A | is calculated, the following equation (15) is obtained.
[Expression 15]

In deriving equation (15), equation (12) is used and D_x(X, Y), D_yThe second and higher order terms of (X, Y) are ignored. Each equation in equation (13) is solved for t, X, and Y and substituted into equation (10), equation (12) is substituted into equation (10), and dXdYdt = (1 / | A | ) When the expression (10) is expressed using the relationship dX′dY′dL, the following expression (16) is obtained.
[Expression 16]

[0047]
By the way, as described above, the function g indicating the sensitivity characteristic of each pixel included in the image sensor 35 is a function having a value only within the aperture of the pixel. Therefore, also in the above equation (16), the sensitivity function g is D_x(X, Y), D_yAssuming that (X, Y) is extremely small, it has a value only in a very close region where the following expression (17) holds for the p-th pixel in the X direction and the q-th pixel in the Y direction. become.
[Expression 17]

Further, D in the value of the X coordinate of the sensitivity function g in the above equation (16)_xAs for (X, Y) · X ′, the second and subsequent items can be ignored when expanded around the coordinates (X ′, Y ′) where the above equation (17) is established. ).
[Expression 18]

[0048]
Similarly, D in the value of the X coordinate of the sensitivity function in the above equation (16)_y(X, Y) · (Y ′ + L−v · t_q) Can also be approximated by the following equation (19).
[Equation 19]

Further, for the same reason, | A (X, Y) | can be approximated as the following equation (20).
[Expression 20]

[0049]
Substituting the above approximate expressions (18) to (20) into expression (16) yields the following expression (21).
[Expression 21]

This equation (21) represents an approximate value of a signal output from the image sensor 35 when the mapping optical system 20 has distortion.
[0050]
(B) Fourier component of signal
Next, when there is residual aberration in the mapping optical system 20, the signal S output from the image sensor 35._dConsider the Fourier component of (X, T). Now, signal S_dFS Fourier transform of (X, T)_dAssuming (μ, ν), this equation is expressed by the following equation (22).
[Expression 22]

Here, as can be seen by referring to the equation (11), the equation S_d(X, T) is a comb function (comb function) and an expression s_dSince it is a product of (X, T), its Fourier transform is_dFourier transform of (X, T), comb function, and formula FS_dThis is a convolution of (μ, ν). Thus, the expression s_dThe Fourier transform of (X, T) is Fs_dIf (μ, ν), then the formula FS_d(Μ, ν) can be modified as shown in the following equation (23).
[Expression 23]

[0051]
Referring to equation (23) above, equation FS_d(Μ, ν) is the expression s_dAs can be seen from the Fourier transform of Therefore, the expression s_dConsider. Here, paying attention to the above equation (21), the region where the sensitivity function g included in the equation has a value is a very narrow region as described above. In addition, regarding the expression O indicating the pattern formed on the sample 4, a very fine pattern is considered when discussing the resolution. Therefore, the region where the function O has a value is also a very narrow region. Therefore, the above equation (21) is also an equation having a value only in a limited region. In such a limited region, it is considered that the amount of change in distortion in the X direction is small. That is, D in the above equation (21)_x(X_p, Y_q) ・ (X_p/ Β) and D_y(X_p, Y_q) ・ (X_p/ Β) is assumed to be small, and with respect to the X direction, the same value is assumed for all the pixels arranged in the p-th column, and the coordinates in the Y direction are Depends only on. In the equation (24), δ_xp, Δ_ypNote that is not a δ function.
[Expression 24]

Also, | A (X_p, X_qFor the same reason,) | is regarded as a constant in the X direction, and is approximated as in equation (25).
[Expression 25]

[0052]
Using the approximation shown in the above equations (24) and (25), the equation s_d(X_p, T) is simplified. Now, the function G shown in the following equation (26)_qDefine (X, Y).
[Equation 26]

This function G_qWhen (21) is expressed using (X, Y), the following (27) is obtained. In the equation (27), d indicates the pixel pitch.
[Expression 27]

Here, as shown in the equation (27), the equation s_d(X, T) is the function G_qIt is expressed in the form of a convolution of (X, Y) and a function O (X ′, Y ′), and further a convolution with Ω (T). Where function G_qIn other words, it is an expression indicating the sensitivity of each pixel provided in the image sensor 35, and the function O (X ′, Y ′) is an expression indicating the pattern image of the sample 4 formed on the imaging surface, and Ω (T) is an expression indicating the shutter speed of each pixel.
[0053]
When the Fourier transform of the above equation (27) is obtained, the following equation (28) is obtained. In the equation (28), FΩ is the Fourier transform of the equation Ω, and FO is the Fourier transform of the equation O.
[Expression 28]

[0054]
Furthermore, if the Fourier transform of the sensitivity function g is Fg, the function G shown in the equation (26)_qThe Fourier transform of (X, Y) is given by the following equation (29).
[Expression 29]

Therefore, when the above equation (29) is substituted into the above equation (28) and the Fourier transform of the point image is considered as OTF, the OTF is expressed by the following equation (30).
[30]

[0055]
In (30) above, the part shown in the following equation (31) relates to the distortion of the mapping optical system 20.
[31]

The above equation (31) represents the OTF when distortion aberration exists in the mapping optical system 20.
[0056]
[4] Movement speed of xy stage 40 that maximizes OTF
In the above discussions [2] and [3], in the state where the timing at which the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are synchronized, the mapping optical system The degradation of OTF caused by 20 residual aberrations was discussed. Hereinafter, a method of obtaining an asynchronous amount for maximizing the OTF with the timing at which the image formed by the imaging element 35 is converted into an electrical signal and the moving speed of the xy stage 40 being asynchronous will be described.
[0057]
In the present embodiment, the timing at which the image sensor 35 converts an image formed on the image pickup surface into an electrical signal is fixed, and the moving speed of the xy stage 40 is changed to change the image formed on the image pickup surface 35. The timing for converting to an electrical signal and the moving speed of the xy stage 40 are asynchronous. The OTF obtained in the synchronized state (when the movement speed of the xy stage 40 is v) can be obtained from the above equation (31). Therefore, consider the OTF when the moving speed of the xy stage 40 is changed from v to a speed v ′ changed by δv as shown in the following equation (32).
[Expression 32]

[0058]
Substituting the above equation (32) into the equation (30) and setting the velocity of the xy stage 40 to v ′, OTF ′ (μ, ν) obtained by the following equation (33). At this time, Y_q= Β ・ v ・ t_qWas used.
[Expression 33]

Here, the above expression (33) includes both OTFs in the non-scanning direction (X direction) and the scanning direction (Y direction), and only in the scanning direction (Y direction)._YWhen ′ is extracted, it is expressed by the following equation (34).
[Expression 34]

[0059]
In the above equation (34), by selecting δv optimally, OTF_Y′ Can be made closer to 1 (maximum). From the above, in the present embodiment, δv in the equation is obtained so that the value of the following equation (35) is maximized.
[Expression 35]

The above equation (35) is the amount of distortion β in the Y direction according to the magnification β of the mapping optical system 20 and the position of the pixel in the Y direction when there is no distortion._yqThe time Δt during which the shutter provided in each pixel of the image sensor 35 is open, the position q of the pixel in the Y direction, the timing at which the image formed on the imaging surface by the image sensor 35 is converted into an electrical signal, and the xy stage 40 A change δv of the speed of the stage 40 when the movement speed of the pattern is made asynchronous, the spatial frequency ν of the pattern formed on the sample 4, and A_pqIs an expression comprising In this equation, the magnification β, the amount of distortion δ_yqAnd the time Δt during which the shutter is open is determined by the apparatus. Therefore, in order to prevent a decrease in resolution due to the influence of distortion in the Y direction, it is necessary to determine δv according to the pitch of the pattern formed on the sample 4.
[0060]
As described above, in the surface state observation apparatus according to the embodiment of the present invention, the method for maximizing the OTF, that is, the method for preventing the decrease in the resolution has been described. Next, the stage obtained by the method obtained by using the above method The OTF obtained when the stage is moved at a moving speed and the timing at which the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous, and the image sensor 35. A simulation result of OTF obtained when the image formed on the imaging surface is converted into an electrical signal and the movement speed of the xy stage 40 is asynchronous will be described.
[0061]
As preparation for explaining the simulation result, the distortion assumed in this embodiment will be described. Although how the distortion aberration is generated varies depending on the mapping optical system 20, generally, in the mapping optical system 20 that observes a narrow field of view as in the surface state observation apparatus of the present embodiment, lower order components are dominant. For this reason, distortion of the third order or less is considered here. Further, it is assumed that the mapping optical system 20 is configured to be rotationally symmetric with respect to the optical axis, and therefore distortion is also rotationally symmetric with respect to the optical axis. Now, as shown in FIG. 9, on the surface of the sample 4, a point at a distance r represented by the following expression (36) from the intersection with the optical axis of the mapping optical system 20 is imaged by the distortion aberration of the mapping optical system 20. It is assumed that an image is formed on the surface 35a at a position R from the optical axis. FIG. 9 is a diagram showing a relationship between a point at a distance r from the optical axis and a position R at which this point forms an image due to the influence of distortion.
[Expression 36]

[0062]
Here, considering the distortion aberration up to the third order, the relationship between r and R is expressed by the following equation (37).
[Expression 37]

Further, when r and R are represented by vector components and their relationship is expressed, the following equation (38) is obtained. In the following equation (38), the coordinates of R are (X, Y) and the coordinates of r are (x, y).
[Formula 38]

Furthermore, the both sides of the above equation (37) are squared and r²Solve for (39)
[39]

Substituting the above equation (39) into equation (38), (β_Three/ Β₁) ≪ Considering 1, (β_Three/ Β₁If the second or higher order of) is ignored, the following equation (40) is obtained.
[Formula 40]

[0063]
Comparing the above equation (40) and equation (12), when distortion up to the third order is considered, β, D in equation (12)_x(X, Y), D_y(X, Y) is expressed by the following equation (41).
[Expression 41]

Also for | A |, if the above equation (41) is substituted into equation (15), the following equation (42) is obtained.
[Expression 42]

In the simulation shown below, β and D in the equation (41)_x(X, Y), D_yOTF is evaluated using | A | in the formulas (X, Y) and (42). The size of the image sensor 35 in the scanning direction (Y direction) is 8.2 mm, the size in the non-scanning direction (X direction) is 38.5 mm, and the pitches of the pixels included in the image sensor 35 in the X direction and Y direction are set. Both are 16 μm.
[0064]
First, for comparison, a state in which the timing at which the image sensor 35 converts an image formed on the imaging surface into an electric signal and the moving speed of the xy stage 40 are synchronized is simulated. FIG. 10 is a simulation result obtained in a state where the timing at which the image sensor 35 converts an image formed on the imaging surface into an electric signal and the moving speed of the xy stage 40 are synchronized. In this simulation, it is assumed that a distortion of 0.4% occurs in the outermost periphery of the image sensor 35. 10 is an MTF curve in the X direction obtained at the periphery of the imaging surface 35a, and the curve denoted by C11 is an MTF curve in the Y direction obtained at the center position of the imaging surface 35a. A curve denoted by reference numeral C12 is an MTF curve in the Y direction obtained around the imaging surface 35a.
[0065]
FIG. 11 is a diagram showing a shift amount of the image on the image pickup surface 35a of the image pickup element 35 caused by distortion of the mapping optical system 20 with respect to the pixels. In FIG. 11, the curve denoted by reference numeral D10 indicates the amount of deviation of the image in the X direction around the imaging surface 35a, and the curve denoted by reference numeral D11 indicates the amount of displacement of the image in the Y direction at the center of the imaging surface 35a. The curve with the symbol D12 indicates the amount of image shift in the Y direction around the imaging surface 35a. As can be seen from FIGS. 10 and 11, the MTF in the scanning direction (Y direction) is worse than the influence of distortion in the non-scanning direction (X direction) orthogonal to the scanning direction (Y direction). Here, in the state where the timing at which the image sensor 35 converts the image formed on the imaging surface into an electric signal and the moving speed of the xy stage 40 are synchronized, in the above-described equation (31), t_q= Y_q/ V relationship, so OTF in the X direction_X(Μ, ν) and Y direction OTF_Y(X, Y) becomes the following equation (43).
[Expression 43]

[0066]
Referring to the above equation (43), the sum is taken for q, which means that the signals of the respective pixels included in the image sensor 35 are added in the Y direction. However, δ in equation (43)_xp, δ_yqEven if the value of is large, if these values are constant with respect to q, the OTF will not deteriorate. It is δ in the equation (43) that causes the OTF to deteriorate._xp, δ_yqIt is the difference by the position of. Here, the curve with the symbol D10 in FIG._xp, And the curve with reference D11 or D12 is δ_yqIs shown.
[0067]
Now, considering the relationship between FIG. 10 and FIG. 11, as shown in FIG._xpThe value of (curve D10) is δ_yq(Value of curve D11 or curve D12). However, when FIG. 10 is seen, there is almost no deterioration of the curve C10 which shows the MTF curve of the X direction obtained in the center position of the imaging surface 35a. On the other hand, as shown in FIG._yq(The value of the curve D11 or the curve D12) itself is small. However, δ_xpThe difference between the maximum value and the minimum value (difference between the curve D10) is about 3 μm, whereas the difference between the maximum value and the minimum value of the curve denoted by reference sign D12 is as large as about 32 μm. The result of the difference is reflected in FIG. 10, and the curve C12 that is the MTF curve in the Y direction obtained around the imaging surface 35a is deteriorated.
[0068]
Next, as in this embodiment, a case is simulated in which the timing at which the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the movement speed of the xy stage 40 are asynchronous. FIG. 12 is a simulation result obtained in a state where the timing at which the image sensor 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous. In this simulation, it is assumed that a distortion of 0.4% occurs in the outermost periphery of the image sensor 35, and δv is optimally selected in the equation (34). In FIG. 12, the curve with reference C20 is an MTF curve in the X direction obtained around the imaging surface 35a, and the curve with reference C21 is an MTF curve in the Y direction obtained at the center position of the imaging surface 35a. A curve with a symbol C22 is an MTF curve in the Y direction obtained around the imaging surface 35a.
[0069]
FIG. 13 shows an image pickup device that is caused by distortion of the mapping optical system 20 in a state where the timing at which the image pickup device 35 converts the image formed on the image pickup surface into an electric signal and the moving speed of the xy stage 40 are asynchronous. It is a figure which shows the deviation | shift amount with respect to the pixel of the image on the imaging surface 35a of 35. In FIG. 13, the curve denoted by reference sign D20 indicates the amount of image shift in the X direction on the image pickup surface 35a, and the curve indicated by reference sign D21 indicates the amount of image shift in the Y direction at the center of the image pickup surface 35a. The curve with the symbol D22 indicates the amount of image shift in the Y direction around the imaging surface 35a.
[0070]
As can be seen from FIG. 13, imaging is performed by obtaining δv that maximizes the value shown in the equation (35) according to the spatial frequency of the pattern formed on the sample 4 and setting the speed of the stage 40. When the asynchronous amount between the timing at which the element 35 converts the image formed on the imaging surface into an electric signal and the moving speed of the xy stage 40 is controlled, as shown by the curve with the reference symbol D21, the center of the imaging surface 35a is obtained. The difference in image shift amount in the Y direction becomes large. However, as shown by the curve with the symbol D22, the difference in the image shift amount in the Y direction around the imaging surface 35a is smaller than the curve D12 shown in FIG. In addition, the Y-direction obtained in the periphery of the imaging surface 35a is reduced in accordance with the decrease in the difference in the curve indicating the shift amount of the image in the Y-direction in the periphery of the imaging surface 35a indicated by reference numeral D22 in FIG. It can be seen that the deterioration of the MTF curve (MTF curve C22 in FIG. 12) is improved.
[0071]
As described above, the configuration of the surface state observation apparatus according to the embodiment of the present invention and the asynchronous amount control method which is a characteristic part of the present invention have been described. Next, the surface state observation apparatus according to the embodiment of the present invention having the above configuration will be described. The operation during observation will be described. First, when the sample 4 to be observed is placed on a loader (not shown), the sample 4 is transported into the chamber 3 shown in FIG. 1 by the loader. When the sample 4 is carried into the chamber 3, it is placed on the xy stage 40, and the main control system 37 drives the xy stage 40 via the driving device 43 to move the sample 4 to the measurement range. Next, the user operates the input device 45 to input the pitch to the pattern formed on the sample 4.
[0072]
When the pattern pitch is input from the input device 45, the main control system 37 sends the distortion aberration amount δ from the storage device 44 to the mapping optical system 20._ypThen, the magnification β of the mapping optical system 20, the time Δt during which the shutter provided in the image sensor 35 is open, the number of pixels in the row direction and the column direction of the image sensor 35, etc. are read out, and the above-described equation (35) is obtained. Δv that maximizes (this corresponds to the asynchronous amount in the present invention) is calculated. Here, the case where the surface state of the sample 4 shown in FIG. 14 is observed is considered. FIG. 14 is a diagram for explaining the operation when observing the surface state. Here, consider the case of observing the observation target region 4a of the sample 4 with diagonal lines.
[0073]
In the observation, the main control system 37 first outputs a control signal to the mapping optical system control unit 39 based on the measurement result of a Z sensor (not shown) to adjust the magnification of the second optical system 20 and is emitted from the sample 4. The focal position of the secondary electron beam B2 is set on the detection surface of the electron beam detector 30. Then, the main control system 37 causes the thermoelectron emission electron gun 10 to start emitting electrons, and the illumination optical system 11, the e-cross bee 23, the aperture stop AS, the first aligner 22, and the cathode lens 21. 14 irradiates the sample 4 with the illumination beam B1, and the xy stage 40 through the driving device 43 at a constant speed v + δv in consideration of δv obtained by the above-described processing, in a direction parallel to the Y-axis direction in FIG. 14 in the direction indicated by reference numeral d. By irradiating the specimen 4 with the illumination beam B1, the secondary electron beam B2 generated from the object plane of the specimen 4 is condensed by the mapping optical system 20, and the image is formed on the detection plane of the electron beam detector 30. To do. The light emitted from the electron beam detector 30, that is, the optical image of the sample 4 passes through the relay lens 34 and forms an image on the imaging surface of the imaging device 35 in which the pixels are arranged according to the distortion aberration of the mapping optical system 20. . The optical image formed on the imaging surface of the image sensor 35 is photoelectrically converted into a pixel signal by the image sensor 35.
[0074]
In FIG. 14, it is assumed that an area denoted by reference numeral 50 is an area imaged by the image sensor 35, and a linear area denoted by reference numeral R <b> 1 in the area 50 is imaged by the CCD pixel row R <b> 1 illustrated in FIG. It is assumed that the CCD pixel row R256 images a linear region denoted by reference numeral R256. If the observation is started from the state shown in FIG. 14, the optical image from the coordinates (X1, Y1) to (X1024, Y1) of the observation target region 4a is the CCD pixel row R1 of the TDI array CCD that forms the image sensor 35. In the case of photoelectric conversion, charges are accumulated. When the sample 4 moves in the −Y-axis direction, the region 50 imaged by the image sensor 35 moves in the direction indicated by the symbol d, and the region when the xy stage 40 is moved at the speed v. When the time that 50 has moved by one CCD pixel row has elapsed, one vertical clock signal is sent from the main control system 37 to the image sensor 35.
[0075]
When the vertical clock signal is transmitted, the charges accumulated in the CCD pixel row R1 of the image sensor 35 are transferred to the CCD pixel row R2. The optical image picked up by the CCD pixel row R1 before the vertical clock signal is sent is shifted from the position of the CCD pixel row R2 according to the distortion aberration of the mapping optical system 20 and the moving speed v + δv of the stage 40. Move to position. When there is no distortion in the mapping optical system 20 and the xy stage 40 moves at the speed v, the CCD pixel row R2 is imaged by the CCD pixel row R1 before the vertical clock signal is sent out. An optical image similar to the optical image is captured until the next vertical clock signal is sent. However, the position of the optical image is actually shifted due to the distortion of the mapping optical system 20. In this embodiment, in order to alleviate this positional shift, the moving speed of the xy stage 40 is set to v + δv, the timing at which each CCD pixel converts an optical image into an electrical signal, and the moving speed of the xy stage 40 (and thus the image sensor 35). The movement speed of the optical image on the imaging surface is asynchronous.
[0076]
Similarly, when the linear portion with R1 in the region 50 imaged by the CCD pixel row R1 reaches the position from (X1, Y256) to (X1024, Y256), the imaging is terminated. Sample images from (X1, Y1) to (X1024, Y1) begin to be output to the main control system 37 through the control unit 36 for the first time. Next, from the vertical clock signal sent out, sample images from (X1, Y2) to (X1024, Y2) are output to the main control system 37, and the images are sequentially acquired by the main control system 37. Observations are made.
[0077]
As described above, according to the surface state observation apparatus according to the embodiment of the present invention, the secondary electron beam B2 obtained by irradiating the sample 4 with the illumination beam B1 is caused by the optical characteristics of the mapping optical system 20. Thus, even when a positional deviation occurs on the imaging surface 35a of the imaging device 35, the timing at which each pixel of the imaging device 35 converts the optical image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous. Therefore, the deterioration of the OTF can be improved by reducing the amount of positional deviation in the scanning direction (Y direction) of the optical image on the imaging surface, and as a result, the resolution can be improved. Further, according to the present embodiment, a TDI array CCD with high detection sensitivity is used as the image sensor 35. If the resolution can be improved by the above-described method using the TDI array CCD, the intensity of the detection signal There is an effect that high-resolution observation can be performed in a short time without lack.
[0078]
The surface state observation apparatus according to one embodiment of the present invention has been described above. However, the present invention is not limited to the above embodiment, and can be freely changed within the scope of the present invention. For example, in the above-described embodiment and other embodiments, the illumination beam B1 is used as the illumination beam, and secondary electrons, reflected electrons, and backscattered electrons obtained by irradiating the sample 4 with the illumination beam B1 are used. Although the case where the secondary electron beam B2 including at least one is photoelectrically converted by the fluorescent plate 32 and the converted optical image is picked up by the image pickup device 35 has been described as an example, for example, an ion beam may be used instead of the electron beam Alternatively, an optical image obtained by irradiating the sample 4 with light in the visible region may be directly captured by the image sensor 35. In this case, the mapping optical system 20 includes an optical lens. However, in order to completely correct the distortion of the mapping optical system 20, it is necessary to increase the number of optical lenses constituting the mapping optical system 20. Therefore, it is possible to observe with high resolution without increasing the size of the apparatus. When light is used as the illumination beam, the secondary beam is reflected light obtained by irradiating the object surface with light.
[0079]
In the above-described embodiment, the case where the illumination beam B1 is deflected using the e-cross bee 23 to irradiate the sample 4 and the secondary electron beam B2 generated from the sample 4 is caused to travel straight has been described. However, the present invention is not limited to this, and an electromagnetic prism that linearly moves the illumination beam B1 and deflects the secondary electron beam B2 may be used. Furthermore, this embodiment includes a so-called plane-to-plane projection optical system that illuminates the object surface of the sample with an electron beam from a radiation source and forms an image on the image plane. It can be easily applied not only as a single device but also to a semiconductor exposure device.
[0080]
Furthermore, in the above embodiment, the input device 45 is provided as an input means, and the pitch of the pattern formed on the sample 4 is input. However, the pattern formed on the sample 4 to be measured in advance is externally connected to a hard disk or the like. The movement speed of the stage 40 may be calculated by inputting the data into a storage device and reading the contents. In the above-described embodiment, the xy stage 40 is moved. However, a deflection apparatus that fixes the stage on which the sample 4 is placed and deflects the secondary beam obtained by irradiating the sample 4 with the illumination beam B1. The image formed on the detection surface or the imaging surface may be moved relative to the detection surface or the imaging surface. Furthermore, in the above embodiment, the xy stage 40 is moved at a constant speed v + δv regardless of the position of the sample 4 in the xy coordinate system, but the moving speed of the xy stage 40 is set according to the distortion aberration of the mapping optical system 20. Variable control may be performed.
[0081]
Next, a method for manufacturing a microdevice according to an embodiment of the present invention will be described. FIG. 15 is a flowchart illustrating a manufacturing example of a micro device (a semiconductor chip such as an IC or LSI, a liquid crystal panel, a CCD, a thin film magnetic head, a micro machine, or the like). As shown in FIG. 15, first, in step S10 (design step), a function / performance design (for example, circuit design of a semiconductor device) of a micro device is performed, and a pattern design for realizing the function is performed. Subsequently, in step S11 (mask manufacturing step), a mask (reticle) on which the designed circuit pattern is formed is manufactured. On the other hand, in step S12 (wafer manufacturing step), a wafer is manufactured using a material such as silicon.
[0082]
Next, in step S13 (wafer processing step), as will be described later, an actual circuit or the like is formed on the wafer using lithography and the like using the mask and wafer prepared in steps S10 to S12. Next, in step S14 (device assembly step), device assembly is performed using the wafer processed in step S13. This step S14 includes processes such as a dicing process, a bonding process, and a packaging process (chip encapsulation) as necessary. Finally, in step S15 (inspection step), inspections such as an operation confirmation test and a durability test of the microdevice manufactured in step S14 are performed. In step S15, the surface state of the microdevice is observed by the surface state observation apparatus according to the embodiment of the present invention described above. After these steps, the microdevice is completed and shipped. The flow chart shown in FIG. 15 illustrates the case where the surface state of the microdevice is observed by the surface state observation apparatus according to the embodiment of the present invention after step S14. Before assembling, the surface state of the wafer that has been processed through step S13 may be appropriately observed.
[0083]
【The invention's effect】
As described above, according to the present invention, the image formed by the imaging unit on the detection surface is electrically converted according to the amount of positional deviation of the optical image on the detection surface that occurs according to the optical characteristics of the mapping optical system. Not only makes the timing of signal conversion and the moving speed of the image on the detection surface asynchronous, but also controls the amount of that asynchronous, thus preventing resolution degradation due to distortion in the mapping optical system. As a result, even if distortion aberration remains in the mapping optical system, it can be observed with high resolution.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a surface state observation apparatus according to an embodiment of the present invention.
FIG. 2 is a functional block diagram illustrating a configuration example of a TDI array CCD as an image sensor 35. FIG.
FIG. 3 is a diagram showing a trajectory of an illumination beam B1 of the surface state observation device according to the embodiment of the present invention.
FIG. 4 is a diagram showing a trajectory of a secondary electron beam B2 of the surface state observation apparatus according to the embodiment of the present invention.
FIG. 5 is a diagram for explaining the configuration and operation principle of an e-cross bee 23 provided in the surface state observation device according to the embodiment of the present invention.
6 is a diagram illustrating an XY coordinate system in which an xy coordinate system set on the surface of a sample 4 is set on the imaging surface of the imaging device 35. FIG.
7 is a diagram illustrating an example of an array of pixels provided in the image sensor 35. FIG.
FIG. 8 is a diagram illustrating an example of a sensitivity characteristic function g (X) in which the position in the Y direction is set at the center of a pixel.
FIG. 9 is a diagram showing a relationship between a point at a distance r from the optical axis and a position R at which this point forms an image due to the influence of distortion.
FIG. 10 is a simulation result obtained in a state where the timing at which the image sensor 35 converts an image formed on the imaging surface into an electric signal and the moving speed of the xy stage 40 are synchronized.
FIG. 11 is a diagram showing a shift amount of an image with respect to a pixel on an imaging surface 35a of an imaging element 35 caused by distortion of the mapping optical system 20.
12 is a simulation result obtained in a state where the timing at which the image sensor 35 converts an image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous. FIG.
FIG. 13 shows a state in which the imaging device 35 is caused by distortion of the mapping optical system 20 in a state where the timing at which the imaging device 35 converts the image formed on the imaging surface into an electrical signal and the moving speed of the xy stage 40 are asynchronous. It is a figure which shows the deviation | shift amount with respect to the pixel of the image on the imaging surface 35a.
FIG. 14 is a diagram for explaining an operation when observing a surface state;
FIG. 15 is a flowchart showing an example of a microdevice manufacturing process.
FIG. 16 is a diagram illustrating an example of distortion.
FIG. 17 is a diagram for explaining deterioration in detection accuracy when an image in which distortion is generated is detected by a TDI array CCD.
FIG. 18 is a diagram for explaining deterioration in detection accuracy in the scanning direction when an image in which distortion is generated is detected by a TDI array CCD.
[Explanation of symbols]
4 Sample (object)
11 Illumination optical system
20 Mapping optics
30 Electron beam detector (imaging means)
32 Fluorescent screen (conversion means)
34 Relay lens (imaging means)
35 Imaging device (imaging means)
37 Main control system (control means, calculation device)
40 xy stage (stage)
44 Storage device (storage means)
45 Input device (input means)
B1 Beam for illumination
B2 Secondary electron beam (secondary beam)

Claims (11)

An illumination optical system for irradiating an illumination beam on the surface of the object to be observed;
A mapping optical system that forms an image of a secondary beam obtained by irradiating the object surface with the illumination beam on a predetermined detection surface;
Imaging means for converting an image formed on the detection surface into an electrical signal, and outputting an integrated signal while transferring the electrical signal in the moving direction of the image on the detection surface;
Control means for controlling an asynchronous amount between the conversion timing of the imaging means and the moving speed of the image on the detection surface in accordance with the amount of positional deviation of the image caused by the optical characteristics of the mapping optical system. Prepared,
The control unit controls the asynchronous amount so that an optical transfer function obtained from a signal output from the imaging unit is maximized at a spatial frequency of the image corresponding to a pitch of a pattern formed on the object. A surface state observation device. A stage configured to be movable with the object placed thereon;
The control unit controls an asynchronous amount between the conversion timing of the imaging unit and the moving speed of the image on the detection surface by making the moving speed of the stage variable according to the amount of positional deviation of the image. The surface state observation apparatus according to claim 1.   In the state where the moving speed of the image on the detection surface is constant, the control means has an optical transfer function obtained from a signal output from the imaging means in accordance with a pitch of a pattern formed on the object. 2. The surface state observation apparatus according to claim 1, wherein the asynchronous amount is controlled so as to be maximum at a spatial frequency of the image.   The surface state observation apparatus according to claim 3, wherein the spatial frequency is a spatial frequency in a moving direction of the image on the detection surface. Storage means for storing optical characteristics of the mapping optical system obtained in advance from design values of the mapping optical system or from measured values;
Input means for inputting a pitch of a pattern formed on the object;
Calculation means for calculating an asynchronous amount of the image with respect to the conversion timing based on the optical characteristics of the mapping optical system stored in the storage means and the pitch of the pattern input from the input means. The surface state observation apparatus of Claim 3 or Claim 4 to do.   6. The imaging device according to claim 1, wherein the imaging unit has a plurality of pixels arranged at equal intervals two-dimensionally on the detection surface, and the conversion timing in each pixel is the same. The surface state observation apparatus according to claim 1. An illumination optical system for irradiating an illumination beam on the surface of the object to be observed;
A mapping optical system that forms an image of a secondary beam obtained by irradiating the object surface with the illumination beam on a predetermined detection surface;
Imaging means for converting an image formed on the detection surface into an electrical signal, and outputting an integrated signal while transferring the electrical signal in the moving direction of the image on the detection surface;
Control means for controlling an asynchronous amount between the conversion timing of the imaging means and the moving speed of the image on the detection surface in accordance with the amount of positional deviation of the image caused by the optical characteristics of the mapping optical system. Prepared,
The surface state observing apparatus according to claim 1, wherein the positional deviation of the image caused by the optical characteristics of the mapping optical system is a positional deviation due to distortion aberration remaining in the mapping optical system. The illumination beam is a charged particle beam;
8. The secondary beam according to claim 1, wherein the secondary beam is at least one of secondary electrons, reflected electrons, and backscattered electrons generated by irradiating the object with the charged particle beam. The surface state observation apparatus according to one item.   The imaging means includes conversion means for converting an image formed on the detection surface by the mapping optical system into an optical image, and photoelectrically converts the optical image converted by the conversion means into the electrical signal. The surface state observation apparatus according to claim 8. A secondary beam obtained by irradiating the surface of the object to be observed with the illumination beam is imaged on a predetermined detection surface via the mapping optical system, and the image formed on the detection surface is moved while the image is moved. In the surface state observation method for observing the state of the surface of the object using a signal obtained by converting the signal and integrating the electric signal while transferring the electric signal in the moving direction of the image on the detection surface,
According to the amount of displacement caused by distortion aberration remaining in the mapping optical system caused by the optical characteristics of the mapping optical system, the conversion timing of the image formed on the detection surface to an electrical signal and the detection surface A method for observing a surface state, characterized by controlling an asynchronous amount with respect to a moving speed of the image. A step of observing the surface of a microdevice using the surface state observation apparatus according to any one of claims 1 to 9 or the surface state observation method according to claim 10. Production method.

Images