JP4284675B2

JP4284675B2 - Substrate inspection device with height measurement

Info

Publication number: JP4284675B2
Application number: JP2003038993A
Authority: JP
Inventors: 克田代
Original assignee: オプトウエア株式会社; 有限会社アイシスウエア
Priority date: 2002-01-11
Filing date: 2003-01-10
Publication date: 2009-06-24
Anticipated expiration: 2023-01-10
Also published as: JP2004045372A

Description

【０００１】
【発明の属する技術分野】
本発明は、表面に電極等の材質の異なる部分をもつ基板の基板面上にレーザ光を走査したときの反射光を受光して表面上の位置や高さ検査を行う高さ測定装置および基板検査装置に関する。
【０００２】
【従来の技術】
従来技術としては例えば特開平６−１６７３２２に示されたものがあげられる。
図７はレーザ走査を用いた基板検査装置の一例の構成図である。レーザ光源１は、例えば半導体レーザとコリメーションレンズの組み合わせであり、このレーザ光源１とポリゴンスキャナ２と走査レンズ３の組み合わせによる投光光学系によって搬送ステージ４上に載置されている測定対象としてたとえば電極５が形成された基板６面上に一次元的に走査されるものとなっている。
【０００３】
又、搬送ステージ４は、制御／測定部７の制御によって投光光学系によるレーザ光の一次元的な走査方向に対して直交する方向に移動するものとなっている。
【０００４】
従って、投光光学系によるレーザ光の一次元的な走査と搬送ステージ４の移動とにより、レーザ光は、基板６の電極５の配置されている領域に対して二次元走査される。
【０００５】
一方、基板６の法線について投光光学系と対称となる斜め上方には、結像光学系８が配置され、基板６からの反射光が結像光学系８を通して高さ測定手段としてのＰＳＤ（光位置検出器）９に結像されるものとなっている。
【０００６】
このＰＳＤ９に結像されるレーザ光は、基板６上のレーザ光の照射される位置の高さに応じて移動し、これによりＰＳＤ９からは、レーザ光の位置と基板６からの反射光量に応じて電流による出力信号Ａ１、Ｂ１が流れるものとなっている。
【０００７】
このＰＳＤ９の各出力Ａ１、Ｂ１の出力端子には、それぞれＩ−Ｖ変換器１０が接続され、さらにこれらＩ−Ｖ変換器１０の出力がＡ／Ｄ変換機１１によりデジタルデータに変換されたあと制御／測定部７に送られ各ＰＳＤ出力ごとに（Ａ１−Ｂ１）／（Ａ１＋Ｂ１）の計算が行われ、これが高さ測定信号として出力される。
【０００８】
高さ測定信号についてはデータ処理が行われる。データ処理の従来技術としては例えば特開平１１−２８７６２２などがある。たとえば測定対象領域が基板上の電極部であったとすれば電極部の高さを出力するため制御／測定部７においてまずどこが電極５であるかについて判別する。この判別方法については（Ａ１＋Ｂ１）、つまり光量と比例する信号の値が一定値以上となるものが電極であるという判別を行い電極部の高さ測定信号を抽出し測定データを分析して異常検査をしていた。
【０００９】
【発明が解決しようとする課題】
上記装置では、電極等の測定対象領域５の形状や基板６表面の反射率によって正確な高さ測定信号が得られない問題がある。すなわち、測定対象領域５が電極等の金属である場合、投光光学系により走査されたレーザ光が測定対象領域５の頂上や平面部分で反射して結像光学系８に入射する場合きわめて光量が大きくなり、このときＡ／Ｄ変換器の出力信号が飽和しないためには回路ゲインを小さくしておかなければならない。
【００１０】
一方基板６の基板表面と測定対象領域５の高さを測定しようとする場合、基板６の表面が樹脂であると金属面に比べ光学的性能は一定せず吸収が大きく平坦度も悪く反射が拡散的でもある。反射が拡散的の場合には反射光が全空間に広がってしまうため結像光学系８に入射する光量は測定対象領域５上の反射に比べてきわめて小さくなる。
【００１１】
このため、測定対象領域５の表面での反射光量に合わせてレーザ光源１からのレーザ出射光量及びＰＳＤ９からＩ−Ｖ変換器１０での検出ゲインを設定すると、基板６表面での反射光量による信号が極めて小さくなるために分解能が悪くなりまたノイズの影響を受けやすくなり、正確な高さ測定信号（Ａ１−Ｂ１）／（Ａ１＋Ｂ１）が得られなくなる。
【００１２】
図８は基板６の基板と測定対象領域５の高さ測定を行った場合の反射光量と高さ測定信号（Ａ１−Ｂ１）／（Ａ１＋Ｂ１）とを示している。上述したように基板表面の反射光量が測定対象領域５の表面での反射光量よりも非常に小さい場合、基板表面における高さ測定信号（Ａ１−Ｂ１）／（Ａ１＋Ｂ１）はノイズの影響を受けて乱れる。そのうえ、測定対象領域５が電極の場合その側面における反射光量は殆どゼロであり、さらに大きくノイズの影響を受けたものとなり、高さ測定信号としては全く信用できないものである。
【００１３】
またどこが測定対象領域５であるかの判別認識をＡ１＋Ｂ１信号と一定値の比較で行った場合、たとえば基板の光学特性が正反射性で反射率も大きかった場合には基板からの反射光量が測定対象領域からの反射光量と近くなるので測定対象領域かどうかの判別認識ができなくなるという問題点があった。また、測定対象領域を決定する段階でＡ１＋Ｂ１等の演算をすべての画素において行わなければならず、計算時間がかかる問題があった。
【００１４】
そこで本発明は、ノイズの影響のない安定した高さ測定信号を得て正確な高さ測定ができる基板検査装置を提供することを目的とする。また、本発明は、測定対象領域と基板表面との反射光量が近くても特別な演算を行わずに判別認識が可能となる基板検査装置を提供することを目的とする。
【００１５】
【課題を解決するための手段】
本発明は以上の課題を解決するため以下のような構成としたものである。
【００１６】
レーザ光源と光走査機と走査レンズとを備えた投光光学系と、測定対象基板の法線に対して投光光学系と対称な角度に設けられた結像光学系と複数の高さ測定手段とを備え、投光光学系のレーザ光の収束角が結像光学系の開口角より小であって、結像光学系においてレーザ光の収束角に相当する開口で走査される部分を含む範囲の光と残る範囲の光をミラーにより分離して一方を第１の高さ測定手段に導き、他方を第２の高さ測定手段に導くこととした。
【００１７】
あるいはレーザ光源と光走査機と走査レンズとを備えた投光光学系と、測定対象基板の法線に対して投光光学系と対称な角度に設けられた結像光学系と複数の高さ測定手段とを備え、投光光学系のレーザ光の収束角が結像光学系の開口角より小であって、結像光学系の光量の一部をビームスプリッタにより分離して第１の高さ測定手段に導き、残るレーザ光について投光光学系の収束角に相当する開口で走査される部分を含む範囲を遮光し、残る範囲の光を第２の高さ測定手段に導くこととした。
【００１８】
あるいはレーザ光源と光走査機と走査レンズとを備えた投光光学系と、測定対象基板の法線に対して投光光学系と対称な角度に設けられた結像光学系と複数の高さ測定手段とを備え、投光光学系のレーザ光の収束角が結像光学系の開口角より小であって、結像光学系の光量の一部をビームスプリッタにより分離して第１の高さ測定手段に導き、残るレーザ光について投光光学系の収束角に相当する開口で走査される部分を含む範囲の光量を光量調節フィルタにより減光したうえで第２の高さ測定手段に導くこととした。
【００１９】
また前記第１の高さ測定手段に導かれる光路中に光量減少フィルタを挿入することとした。
【００２０】
また前記第1の高さ測定手段の第1出力（Ａ１）、第2出力（Ｂ１）、および前記第2の高さ測定手段の第1出力（Ａ２）、第2出力（Ｂ２）のいずれかひとつの信号を基板と測定対象領域の状態によって選択し、選択した出力信号に反映される基板表面の出力と測定対象領域の出力との差に基づき、基板上の測定対象領域を特定することとした。
【００２１】
【発明の実施の形態】
以下、本発明の実施の形態について図面を参照して説明する。なお、図７と同一部分には同一符号を付す。
【００２２】
図１は本発明の第１の実施例の構成図、図２はその原理図である。レーザ光源１は例えば半導体レーザとコリメーションレンズの組み合わせでありこのレーザ光源１、光走査機の一種であるポリゴンスキャナ２と走査レンズ３の組み合わせによる投光光学系によって搬送ステージ４上に載置されている凹凸、例えば電極５が形成された基板６面上に一次元的に走査されるものとなっている。
【００２３】
又、搬送ステージ４は、制御／測定部７の制御によって投光光学系によるレーザ光の一次元的な走査方向に対して直交する方向に移動するものとなっている。
【００２４】
従って、投光光学系によるレーザ光の一次元的な走査と搬送ステージ４の移動とにより、レーザ光は、基板６の電極５の配置されている領域に対して二次元走査される。
【００２５】
一方、基板６の法線について投光光学系３と対称となる方向の斜め上方には、従来例と同様に結像光学系８が配置され、基板６からの反射光が結像光学系８を通してミラーにより分離され、第1の高さ測定手段であるところのＰＳＤ（光位置検出器）９および第２のＰＳＤ１８に結像されるものとなっている。
【００２６】
このＰＳＤ９、１８に結像されるレーザ光は、基板６上のレーザ光の照射される位置の高さに応じて一次元的に移動し、これによりＰＳＤ９、１８からは、レーザ光の位置と基板６からの反射光量に応じて電流Ａ１、Ｂ１、Ａ２，Ｂ２が流れるものとなっている。
【００２７】
これらの各ＰＳＤ９、１８の各電流Ａ１、Ｂ１、Ａ２，Ｂ２の出力端子には、それぞれＩ−Ｖ変換器１０が接続され、さらにこれらＩ−Ｖ変換器１０の出力はＡ／Ｄ変換器１１に入力されデジタル化されたあと制御／測定部７に送られるようになっている。
【００２８】
次にこの構成および作用を図２の原理図に基づき詳細に説明する。
レーザ光はポリゴンスキャナ２で反射、走査され、走査レンズ３に入射する。走査レンズ３は対象基板６上にレーザ光を収束させる。このときレーザ光の収束角のFナンバーF1は、走査レンズ３の焦点距離をｆ１とし、走査レンズ前のレーザ光の径をｄ１とするとF１＝ｆ１／ｄ１となる。
【００２９】
レーザ光は対象基板上で反射し結像光学系８に入射する。結像光学系８の開口数はF2であり、F2<F1となっている。結像光学系内の光束径はd２で、前側距離をｓ１とするとｄ２＝ｓ１／Ｆ２である。しかし対象基板６上でレーザ光が正反射した場合、レーザの反射光の開口数は反射前と変わらずＦ１であるので結像光学系内のレーザビーム径ｄ３はｄ３＝ｓ１／Ｆ１と計算され、Ｆ２＜Ｆ１であるためｄ３＜ｄ２である。
【００３０】
そこでｄ３部分が走査により移動する範囲内にミラー１９を設置し、光量減少フィルタ２０を介して第１の高さ測定手段、この場合は走査方向に長く測定方向が走査方向と直交している１次元ＰＳＤ９に入射させる。
【００３１】
一方ミラー１９外を通過した光は第２のＰＳＤ１８に入射させる。このような構成とした場合の作用を次に説明する。対象基板６上で正反射が起こると反射光はきわめて強く、また指向性が高いため上記ｄ３部分に集中的に入射する。この光はミラー１９によって第１のＰＳＤ９に導くことができ、光量減少フィルタ２０により光量を適切に下げ、また第１のＰＳＤ９のゲインを適切に下げて測定することにより安定した高速高精度の測定ができる。
【００３２】
一方対象基板６上で拡散反射が起きると反射光は広範囲に分散するため弱くなり、結像光学系８内においても結像光学系の開口数いっぱい、ｄ２部分まで広がって入射する。よってｄ３部分以外の部分が第２のＰＳＤ１８に集光されるような構成であっても拡散光は第２のＰＳＤ１８に入射することができる。
【００３３】
そのためＰＳＤ１８のゲインを適正にあげることによって対象基板上の拡散反射部分についても安定的に測定することができる。第２のＰＳＤ１８のゲインをあげると正反射光が入射した場合にＰＳＤが飽和する等の障害がでるが、正反射光はミラー１９により第１のＰＳＤ９の側に導かれているので飽和の問題はでない。ただし電極表面が斜め部分の正反射光はｄ３部分のミラー１９をはずれて第２のＰＳＤ１８に入射してきてしまうが短時間であるので致命的な障害とはならない。
【００３４】
図３は本発明の第２の実施例の構成図、図４はその原理図である。結像光学系８は前側８ａと後ろ側８ｂ、８ｃに別れ、その間は平行光となっている。前側結像光学系８ａの直後にシリンドリカルレンズ２１が設置されている。その後にビームスプリッタ２２がおかれ５％程度の光量を反射し、残りを透過する。反射された光は第１の後ろ側結像光学系８ｂを介して第１のＰＳＤ９に入射する。
【００３５】
一方透過した光のうち第２の実施例で説明したｄ３対応部分に遮光板２３が設置され直接反射光のみ遮光される。このとき遮光板でなく５％程度の光量減少フィルタを設置してもよい。この場合には直接反射光は拡散反射光と同等の光量レベルとなって透過する。その後第２の後ろ側結像光学系８ｃにより集光され第２のＰＳＤ１８に結像される。
【００３６】
このような構成とした場合の作用を次に説明する。対象基板上で正反射がおこると反射光はきわめて強く、またｄ３部分に集中的に入射する。この際前側結像光学系８ａ直後のシリンドリカルレンズ２１に入射するとシリンドリカルレンズ２１の作用によりＰＳＤ９，１８上のスポットが走査方向に広がる。ＰＳＤには光パワー密度が大きいと応答が遅くなる性質がある。よってシリンドリカルレンズ２１を挿入することによってＰＳＤ９，１８上の光パワー密度を下げ、ＰＳＤの応答を高速のまま保つ効果がある。
【００３７】
しかしながら実施例１の場合などの収束光中ではシリンドリカルレンズ挿入により収差が悪化し、精度に悪影響を及ぼす。そこで第２の実施例においては結像光学系８を前側８ａと後ろ側８ｂ、８ｃに分け間を平行光とすることによりシリンドリカルレンズ２１による収差の発生を抑えている。
【００３８】
正反射光はビームスプリッタ２２で反射後第１の後ろ側結像光学系８ｂに入射する。正反射光は強く、拡散反射光は弱いため、拡散反射光を検知しようとするとレーザーパワーをあげざるを得ないがそうすると正反射光をそのままＰＳＤ９に入射させるのはパワーが大きすぎる。そのため５％程度のビームスプリッタ２２によって光学的にパワーを落とすことによりＰＳＤ９に適正光量が入射するような構成となっている。
【００３９】
また平行光部分は機械的ずれに対する収差の変化量がきわめて少なく、組み立て上、調整上も有利である。
【００４０】
一方ビームスプリッタ２２を透過した光束は第２の後ろ側結像光学系８ｃに入射し、第２のＰＳＤ１８上に結像する。しかし拡散光をみるためには正反射光が混じると光量が大きすぎ応答速度が下がるなどの不具合を生じる。そこでｄ３部分に遮光板、あるいは５％程度の光量減少フィルタ２３をおき正反射光の光量を減衰させることにより安定的に拡散光を測定することができる。第２の後ろ側結像光学系８ｃに入射する以前は平行光なのでビームスプリッタ２２透過による収差の発生も抑えられている。
【００４１】
しかし電極斜面の正反射光は遮光板または光量減少フィルタ２３でさえぎられることなく第２のＰＳＤ１８に入射してしまう。正反射光が入射するとＰＳＤ１８は飽和してしまうがシリンドリカルレンズ２１によりパワー密度が抑えられていればＰＳＤ１８自身の応答速度の低下を避けることができ、電気的な飽和のみに抑えることができ、測定への影響を最小限にすることができる。
【００４２】
なお光量減少フィルタを５％程度とした実施例を説明したが、金属面、基板表面の反射率の比により適正な値にするべきことはいうまでもない。またビームスプリッタを５％反射としたが９５％反射としてもよく、この場合は第１と第２の後ろ側結像光学系、ＰＳＤが入れ替わることになる。
【００４３】
なお第２の実施例ではビームスプリッタ２２をもちいたがｄ３部のみ反射するミラーまたはｄ３部のみ穴のあいているミラーを用いてもよい。この場合には第１の結像光学系前に光量減少フィルタ２０をいれることによりＰＳＤ９に対して適切な光量とすることができる。
【００４４】
またｄ３部のみ光量減少フィルタにした平行平面ガラス板を用いてもよく、この場合にはｄ３部は光量減少フィルタで光量を落とされて拡散光用ＰＳＤ１８に入射させることができ、拡散光と正反射光を同様な条件で測定することが可能になる。
【００４５】
図５は本発明の第３の実施例の構成図である。本実施例では光走査は共振型ミラースキャナ２ｂで行っている。共振型ミラースキャナ２ｂは走査速度が等速でないというデメリットがあるが、小型で走査効率がよく、高速化でき、寿命が長いメリットがある。
【００４６】
結像光学系８の前にミラー１９がおかれ第１の実施例で説明したｄ３対応部分を反射し、残りを透過する。反射された光は第１の結像光学系８ｄ、光量減少フィルタ２０を介して第１のＰＳＤ９に入射する。透過した光は第２の結像光学系８により集光され第２のＰＳＤ１８に結像される。
【００４７】
このような構成とした場合の作用を次に説明する。対象基板６上で正反射がおこると反射光はきわめて強く、またｄ３部分に集中的に入射する。このためミラー１９の反射光を第１の結像光学系８ｄを介して第１のＰＳＤ９に入射させることにより直接反射光を安定的に測定することができる。なおこの場合第１の結像光学系８ｄとしては開口が投光光学系３と同程度に狭くてよいため安価なレンズが使用できる。
【００４８】
一方ミラー１９に反射しない部分の光束は第２の結像光学系８に入射し、第２のＰＳＤ１８上に結像する。拡散光をみるためには正反射光があると光量が大きすぎ応答速度が下がるなどの不具合を生じるがｄ３部分がミラーによって遮光されているため正反射光は第２のＰＳＤ１８には入射せず安定的に拡散光を測定することができる。
【００４９】
なお第３の実施例で説明したように第１、第２の結像光学系内にシリンドリカルレンズを挿入することがＰＳＤの応答速度を確保する上で望ましい。
【００５０】
なおいずれの例でも走査レンズ３、結像光学系８、８ａ、８ｄの測定対象物側はテレセントリックである。これは正反射光を確実に取り込むために必要なことである。
【００５１】
以下、第１のＰＳＤからの信号および第２のＰＳＤからの信号の特性と使用方法について説明する。
まず第１の実施の形態で説明した信号の利用方法を説明する。基板上の電極の基板からの高さを測定する場合、第１のＰＳＤ９には主として電極の正反射光が入射する。一般的には電極５が金属であるために、投光光学系３により走査されたレーザ光が電極５の頂上や平面部分で反射して結像光学系８に入射する場合きわめて光量が大きくなるので、このときＡ／Ｄ変換器１１の出力信号が飽和しないようにＩ−Ｖ変換器１０の回路ゲインを小さくしておく。
【００５２】
一方第２のＰＳＤ１８には主として基板６上の拡散反射光の一部が入射する。一般的には基板６表面の表面は樹脂であるために金属面に比べ吸収が大きく平坦度も悪い場合がある。しかも拡散反射光の一部しか取り込まないため基板６上の反射光量は電極５上の反射に比べてきわめて小さくなる。しかし一般的には電極５上の金属反射面は平滑面であり拡散反射光はきわめて小さいので第２のＰＳＤ１８には主として基板上の拡散反射光が入射する事となる。
【００５３】
このため、第１のＰＳＤ１８以降のＩ−Ｖ変換器１０等のゲインを基板上の拡散反射光にあわせて大きくとる事により第２のＰＳＤ１８により基板６上の高さが正確に得られる事となる。
【００５４】
図６は基板６の基板と電極５の高さ測定を行った場合の第１のＰＳＤ９による反射光量と高さ測定信号および第２のＰＳＤ１８による反射光量と高さ測定信号とを示している。
【００５５】
第１のＰＳＤ９による高さ測定信号は電極５上の反射光量にあわせてゲイン設定しているため電極５上の高さは明確にでるが上述したように一般に基板６表面の反射光量が電極５の表面での反射光量よりも小さいために、基板６表面における高さ測定信号はノイズの影響を受けて乱れている。そのうえ、電極５の側面における反射光量は殆どゼロであり、さらに大きくノイズの影響を受けたものとなり、高さ測定信号としては全く信用できないものである。
【００５６】
一方第２のＰＳＤ１８による高さ測定信号は基板６の拡散反射光にあわせてゲイン設定しているため基板６上の高さが明確に出ている。しかし電極５上、電極５側面の拡散反射光は少ないためノイズの影響を受けている。
【００５７】
このように第１のＰＳＤ９により電極５上の高さが、第２のＰＳＤ１８により基板上の高さが明確に得られるためこれらのデータより基板６表面から電極５頂点までの高さを求める事ができる。
【００５８】
データの選択方法としては各画素ごとに各ＰＳＤ９，１８から得られた光量データ（Ａ１＋Ｂ１）および（Ａ２＋Ｂ２）が飽和しておらずかつ高いほうのデータを採用することにより適切なデータ選択が可能となる。
【００５９】
ただし電極５側面についてはどちらの受光系でも光量が足りないのでデータは使用できない場合が多く、その場合にはデータを削除する。
【００６０】
以上のデータ選択例ではデータ処理でどちらのＰＳＤからのデータを選択したかにより測定点が基板上であるか電極であるかどうかの判別が同時になされることになる。すなわちこの場合にはＰＳＤ９からのデータ採用部分は電極５、ＰＳＤ１８からのデータ採用部分は基板６上と判別される。第１の結像光学系８において電極５からの反射光量が基板６からの反射光量と同様であったとしても第２の結像光学系においては基板６からの反射光量が電極５の反射光量より大きくなるので上記データ選択処理を行うことにより確実に電極部を判別認識することができる。
【００６１】
上記データ選択方法は基板状態が次のようである場合に効果的である。すなわち「測定対象が電極である」「電極部の反射率が基板表面の反射率よりも高い」「電極部表面は正反射性で基板表面は拡散反射性である」ことである。なお基板表面より電極部が位置が高いことも想定している。
【００６２】
しかしすべての測定対象がこれらの条件にあてはまるわけではなく、基板によっては測定対象よりも正反射性が高く反射率が大きい場合もありうる。条件としては「反射率」「拡散性」「位置高さ」の３条件で「高い」「同じ」「低い」などの状態があり後述するように全部で３６通りの組み合わせがあることになる。実際に基板表面はロットごとに変わることが多く３６通りのうちの一つにしか対応しないのは基板検査装置としては望ましくない。
【００６３】
上記データ選択例や従来例では測定対象領域確定のためには一般的にＰＳＤの２出力を合計した「輝度」が用いられている。しかしこの「輝度」算出動作は「高さ」情報を消去する動作であり測定対象位置の高低差がある場合にはわざわざ高さ情報を消去することは必ずしも適切とはいえない。
【００６４】
また、実際にはすべての画素で輝度や高さのデータが必要なわけではないため、計算時間短縮のためには輝度や高さのデータを算出する前にそれより簡単な手段で算出すべきエリアを絞り込むことが望ましい。
そこでＰＳＤの出力のうち片側のみをとりだして２値化すると以下に説明するように光量と高さの両方の情報を一度に利用でき、なおかつ計算時間もすくなくすることができる。
【００６５】
ＰＳＤの電極間幅をＬとし光がその中をｘ：ｙで分割する位置に入射して電流Ｉが発生したとすると片側電極にはＩｘ／Ｌ、もう片側にはＩｙ／Ｌの電流が出力される。２出力を合計すればＩが算出されるが、１出力のみを利用するとＩｘつまり光量と位置の積の数値が得られることになる。
【００６６】
よってＩｘがもっとも高くなる（あるいは低くなる）出力を選んで（もしくは反転させて）一定のしきい値により2値化して測定対象領域を決めることによって単にＩのみによって測定対象領域を決めるよりもより正確に領域を確定することもでき、また光量が同じ場合などＩのみによっては区別できない場合にも対応できることになる。
【００６７】
具体的には位置が高いものが結像する側のＰＳＤ電極をＡ、逆をＢとすると、測定対象部の反射率が高く測定対象部の高さが高い場合にはＡを、測定対象部の高さが低い場合にはＢを、測定対象部の反射率が低く測定対象部の高さが高い場合にはＢの反転を、測定対象部の高さが低い場合にはＡの反転を選ぶことにより測定対象部をより区別しやすくなる。
【００６８】
さらに第１のＰＳＤ、第２のＰＳＤの出力を選択することにより各部分の正反射性、拡散性によっても最適な出力をえらぶことができる。具体的には正反射光取得側を１、拡散反射光取得側を２とすると、測定対象部の反射率が高く正反射性の場合には１を、拡散反射性の場合には２を、測定対象部の反射率が低く正反射性の場合には２の反転を、拡散反射性の場合には１の反転を選ぶことにより測定対象部をより区別しやすくなる。
【００６９】
望ましい選択をまとめると次のようになる。正反射光取得側ＰＳＤを１、拡散反射光取得側ＰＳＤを２とし、位置が高い側の電極をＡ、低い側をＢとする。なお小文字ｔは反転を示す。
【００７０】
測定対象位置が高く基板表面が低い
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性Ａ１
測定対象が拡散性、基板表面が拡散性Ａ２
測定対象が正反射性、基板表面が正反射性Ａ１
測定対象が拡散性、基板表面が正反射性Ａ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ａ１
測定対象が拡散性、基板表面が拡散性Ａ２
測定対象が正反射性、基板表面が正反射性Ａ１
測定対象が拡散性、基板表面が正反射性Ａ２
測定対象の反射率が低く基板表面が高い
測定対象が正反射性、基板表面が拡散性ｔＢ２
測定対象が拡散性、基板表面が拡散性ｔＢ１
測定対象が正反射性、基板表面が正反射性ｔＢ２
測定対象が拡散性、基板表面が正反射性ｔＢ１
測定対象位置と基板表面の高さが同じ
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性Ａ１ｏｒＢ１
測定対象が拡散性、基板表面が拡散性Ａ２ｏｒＢ２
測定対象が正反射性、基板表面が正反射性Ａ１ｏｒＢ１
測定対象が拡散性、基板表面が正反射性Ａ２ｏｒＢ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ａ１ｏｒＢ１
測定対象が拡散性、基板表面が拡散性 ×
測定対象が正反射性、基板表面が正反射性 ×
測定対象が拡散性、基板表面が正反射性Ａ２ｏr Ｂ２
測定対象の反射率が低く基板表面のそれが高い
測定対象が正反射性、基板表面が拡散性ｔＡ２ｏｒｔＢ２
測定対象が拡散性、基板表面が拡散性ｔＡ１ｏｒｔＢ２
測定対象が正反射性、基板表面が正反射性ｔＡ２ｏｒｔＢ１
測定対象が拡散性、基板表面が正反射性ｔＡ１ or ｔＢ１
測定対象位置がひくく基板表面が高い
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性Ｂ１
測定対象が拡散性、基板表面が拡散性Ｂ２
測定対象が正反射性、基板表面が正反射性Ｂ１
測定対象が拡散性、基板表面が正反射性Ｂ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ｂ１
測定対象が拡散性、基板表面が拡散性Ｂ２
測定対象が正反射性、基板表面が正反射性Ｂ１
測定対象が拡散性、基板表面が正反射性Ｂ２
測定対象の反射率が低く基板表面のそれが高い
測定対象が正反射性、基板表面が拡散性ｔＡ２
測定対象が拡散性、基板表面が拡散性ｔＡ１
測定対象が正反射性、基板表面が正反射性ｔＡ２
測定対象が拡散性、基板表面が正反射性ｔＡ１
【００７１】
以上のように二つのＰＳＤの各２出力のうち１出力のみを選択することにより３６通りのうちすべての条件が同じもの以外はすべて測定対象領域を区別できることになり格段に対応測定対象および基板の種類が増え、また区別もしやすくなることがわかる。
【００７２】
さらに測定対象領域を確定するにあたりＰＳＤの２出力を合計する輝度計算が不要となるので処理速度も向上することができる。
【００７３】
上記は測定対象部を判別しやすい選択例であるが同様に基板表面部を判別しやすい選択例も考えることができる。具体的には以下のようになる。
測定対象位置が高く基板表面が低い
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性ｔＡ１
測定対象が拡散性、基板表面が拡散性ｔＡ１
測定対象が正反射性、基板表面が正反射性ｔＡ２
測定対象が拡散性、基板表面が正反射性ｔＡ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ｂ２
測定対象が拡散性、基板表面が拡散性Ｂ２
測定対象が正反射性、基板表面が正反射性Ｂ１
測定対象が拡散性、基板表面が正反射性Ｂ１
測定対象の反射率が低く基板表面が高い
測定対象が正反射性、基板表面が拡散性Ｂ２
測定対象が拡散性、基板表面が拡散性Ｂ２
測定対象が正反射性、基板表面が正反射性Ｂ１
測定対象が拡散性、基板表面が正反射性Ｂ１
測定対象位置と基板表面の高さが同じ
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性ｔＡ１ｏｒｔＢ１
測定対象が拡散性、基板表面が拡散性ｔＡ１ｏｒｔＢ１
測定対象が正反射性、基板表面が正反射性ｔＡ２ｏｒｔＢ２
測定対象が拡散性、基板表面が正反射性ｔＡ２ｏｒｔＢ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ａ２ｏｒＢ２
測定対象が拡散性、基板表面が拡散性 ×
測定対象が正反射性、基板表面が正反射性 ×
測定対象が拡散性、基板表面が正反射性Ａ１ｏr Ｂ１
測定対象の反射率が低く基板表面のそれが高い
測定対象が正反射性、基板表面が拡散性Ａ２ｏｒＢ２
測定対象が拡散性、基板表面が拡散性Ａ２ｏｒＢ２
測定対象が正反射性、基板表面が正反射性Ａ１ｏｒＢ１
測定対象が拡散性、基板表面が正反射性Ａ１ or Ｂ１
測定対象位置がひくく基板表面が高い
測定対象が反射率が高く基板表面が低い
測定対象が正反射性、基板表面が拡散性ｔＢ１
測定対象が拡散性、基板表面が拡散性ｔＢ１
測定対象が正反射性、基板表面が正反射性ｔＢ２
測定対象が拡散性、基板表面が正反射性ｔＢ２
測定対象の反射率と基板表面のそれが同じ
測定対象が正反射性、基板表面が拡散性Ａ２
測定対象が拡散性、基板表面が拡散性Ａ２
測定対象が正反射性、基板表面が正反射性Ａ１
測定対象が拡散性、基板表面が正反射性Ａ１
測定対象の反射率が低く基板表面のそれが高い
測定対象が正反射性、基板表面が拡散性Ａ２
測定対象が拡散性、基板表面が拡散性Ａ２
測定対象が正反射性、基板表面が正反射性Ａ１
測定対象が拡散性、基板表面が正反射性Ａ１
【００７４】
なお、本発明は、上記実施の形態に限定されるものでなく次の通り変形してもよい。たとえば上記実施の形態では、基板６として基板上に形成された電極５の高さ測定に適用した場合について説明したが、これに限らず各種基板上に形成された凸状部や凹状部の高さ測定に適用できる。
【００７５】
また材質の判別例として電極と樹脂基板を例に説明したがこれに限らず各種基板上に形成された異種物質の判別に適用できる。例えば蛍光体物質とガラスの場合においても蛍光体物質が乱反射性でありガラスが正反射性であるのでガラスの反射光は第１のＰＳＤ９で、蛍光体の反射光は第２のＰＳＤ１８でとらえることができる。
【００７６】
【発明の効果】
以上詳記したように本発明によれば、簡単な構成で、ノイズの影響のない安定した高さ測定信号を得て正確な高さ測定ができる基板検査装置を提供できる。
【００７７】
又、本発明によれば、電極等の測定対象部と基板表面との双方から適正な反射光量信号を得て正確な高さ測定ができる基板検査装置を提供できる。
【００７８】
また、本発明によれば測定対象と基板表面等の条件の差がいかなるものであってもその領域を高速に区別できる基板検査装置を提供できる。
【図面の簡単な説明】
【図１】本発明の第１の実施例の構成図
【図２】本発明の第１の実施例の原理図
【図３】本発明の第２の実施例の構成図
【図４】本発明の第２の実施例の原理図
【図５】本発明の第３の実施例の構成図
【図６】本発明の実施例による受光信号の図
【図７】従来例の構成図
【図８】従来例の受光信号の図
【符号の説明】
１はレーザ光源
２はポリゴンスキャナ
２ｂは共振型ミラースキャナ
３は走査レンズ
４はステージ
５は電極等の測定対象領域
６は基板
７は制御／測定部
８は結像光学系
８ａは前側結像光学系
８ｂ、８ｃは後ろ側結像光学系
８ｄは結像光学系
９は第1のＰＳＤ
１０はＩ−Ｖ変換器
１１はＡ／Ｄ変換器
１８は第2のＰＳＤ
１９はミラー
２０は光量減少フィルタ
２１はシリンドリカルレンズ
２２はビームスプリッタ
２３は遮光板または光量減少フィルタ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a height measuring device and a substrate for receiving a reflected light when a laser beam is scanned on a substrate surface of a substrate having a different part of material such as an electrode on the surface and inspecting the position and height on the surface. It relates to an inspection device.
[0002]
[Prior art]
Examples of the prior art include those disclosed in JP-A-6-167322.
FIG. 7 is a configuration diagram of an example of a substrate inspection apparatus using laser scanning. The laser light source 1 is, for example, a combination of a semiconductor laser and a collimation lens. As a measurement target placed on the transport stage 4 by a light projecting optical system using a combination of the laser light source 1, the polygon scanner 2, and the scanning lens 3, for example, One-dimensional scanning is performed on the surface of the substrate 6 on which the electrode 5 is formed.
[0003]
Further, the transfer stage 4 is moved in a direction orthogonal to the one-dimensional scanning direction of the laser light by the light projecting optical system under the control of the control / measurement unit 7.
[0004]
Therefore, the laser light is two-dimensionally scanned with respect to the region where the electrode 5 of the substrate 6 is arranged by one-dimensional scanning of the laser light by the light projecting optical system and the movement of the transport stage 4.
[0005]
On the other hand, an imaging optical system 8 is disposed obliquely above and symmetrical with the projection optical system with respect to the normal line of the substrate 6, and the reflected light from the substrate 6 passes through the imaging optical system 8 as a PSD as height measuring means. (Optical position detector) 9 is imaged.
[0006]
The laser light imaged on the PSD 9 moves in accordance with the height of the position on the substrate 6 where the laser light is irradiated. As a result, the PSD 9 responds to the position of the laser light and the amount of reflected light from the substrate 6. Thus, output signals A1 and B1 due to current flow.
[0007]
An IV converter 10 is connected to the output terminals of the outputs A1 and B1 of the PSD 9, and the output of the IV converter 10 is converted into digital data by the A / D converter 11. A calculation of (A1-B1) / (A1 + B1) is performed for each PSD output sent to the control / measurement unit 7, and this is output as a height measurement signal.
[0008]
Data processing is performed on the height measurement signal. As a prior art of data processing, for example, there is JP-A-11-287622. For example, if the measurement target region is an electrode part on the substrate, the control / measurement unit 7 first determines where the electrode 5 is in order to output the height of the electrode part. About this discriminating method, (A1 + B1), that is, discriminating that the electrode whose signal value proportional to the light quantity exceeds a certain value is an electrode, extracting the height measurement signal of the electrode part, analyzing the measurement data, and checking the abnormality I was doing.
[0009]
[Problems to be solved by the invention]
In the above apparatus, there is a problem that an accurate height measurement signal cannot be obtained due to the shape of the measurement target region 5 such as an electrode or the reflectance of the surface of the substrate 6. That is, when the measurement target region 5 is a metal such as an electrode, the amount of light is extremely high when the laser beam scanned by the light projecting optical system is reflected by the top or plane portion of the measurement target region 5 and enters the imaging optical system 8. In this case, in order for the output signal of the A / D converter not to be saturated, the circuit gain must be reduced.
[0010]
On the other hand, when the height of the substrate surface of the substrate 6 and the measurement target region 5 is to be measured, if the surface of the substrate 6 is a resin, the optical performance is not constant compared to the metal surface, the absorption is large, the flatness is poor, and the reflection is poor. It is also diffuse. When the reflection is diffusive, the reflected light spreads over the entire space, so that the amount of light incident on the imaging optical system 8 is extremely smaller than the reflection on the measurement target region 5.
[0011]
For this reason, if the laser output light quantity from the laser light source 1 and the detection gain from the PSD 9 to the IV converter 10 are set in accordance with the reflection light quantity on the surface of the measurement target region 5, a signal based on the reflection light quantity on the surface of the substrate 6 is set. Is extremely small, the resolution is deteriorated and it is easily affected by noise, and an accurate height measurement signal (A1-B1) / (A1 + B1) cannot be obtained.
[0012]
FIG. 8 shows the amount of reflected light and the height measurement signal (A1-B1) / (A1 + B1) when the height of the substrate 6 and the measurement target region 5 is measured. As described above, when the amount of reflected light on the substrate surface is much smaller than the amount of reflected light on the surface of the measurement target region 5, the height measurement signal (A1-B1) / (A1 + B1) on the substrate surface is affected by noise. Disturbed. In addition, when the measurement target region 5 is an electrode, the amount of reflected light on the side surface is almost zero, and is greatly affected by noise, and the height measurement signal cannot be trusted at all.
[0013]
Further, when the recognition of the measurement target region 5 is performed by comparing the A1 + B1 signal with a constant value, for example, when the optical characteristics of the substrate are specular and the reflectance is large, the amount of light reflected from the substrate is measured. Since the amount of light reflected from the target area is close, there is a problem that it is impossible to discriminate and recognize whether the target area is a measurement target area. In addition, calculation such as A1 + B1 has to be performed on all the pixels at the stage of determining the measurement target region, and there is a problem that it takes a long calculation time.
[0014]
SUMMARY OF THE INVENTION An object of the present invention is to provide a substrate inspection apparatus capable of obtaining a stable height measurement signal free from noise and performing accurate height measurement. It is another object of the present invention to provide a substrate inspection apparatus capable of discriminating and recognizing without performing a special calculation even when the amount of reflected light between the measurement target region and the substrate surface is close.
[0015]
[Means for Solving the Problems]
In order to solve the above problems, the present invention has the following configuration.
[0016]
Projection optical system including a laser light source, an optical scanner, and a scanning lens, an imaging optical system provided at an angle symmetrical to the projection optical system with respect to the normal line of the measurement target substrate, and a plurality of height measurements Including a portion that is scanned by an aperture corresponding to the convergence angle of the laser beam in the imaging optical system, the convergence angle of the laser beam of the light projecting optical system being smaller than the aperture angle of the imaging optical system. The light in the range and the light in the remaining range are separated by a mirror, and one is guided to the first height measuring means, and the other is guided to the second height measuring means.
[0017]
Alternatively, a projection optical system including a laser light source, an optical scanner, and a scanning lens, an imaging optical system provided at an angle symmetrical to the projection optical system with respect to the normal line of the measurement target substrate, and a plurality of heights Measuring means, the convergence angle of the laser light of the light projecting optical system is smaller than the aperture angle of the imaging optical system, and a part of the light quantity of the imaging optical system is separated by a beam splitter. The remaining laser light is shielded from the range including the portion scanned with the aperture corresponding to the convergence angle of the projection optical system, and the remaining light is guided to the second height measuring means. .
[0018]
Alternatively, a projection optical system including a laser light source, an optical scanner, and a scanning lens, an imaging optical system provided at an angle symmetrical to the projection optical system with respect to the normal line of the measurement target substrate, and a plurality of heights Measuring means, the convergence angle of the laser light of the light projecting optical system is smaller than the aperture angle of the imaging optical system, and a part of the light quantity of the imaging optical system is separated by a beam splitter. The remaining laser light is guided to the second height measuring means after the amount of light in the range including the portion scanned with the aperture corresponding to the convergence angle of the projection optical system is dimmed by the light quantity adjusting filter. It was decided.
[0019]
Further, a light quantity reduction filter is inserted in the optical path guided to the first height measuring means.
[0020]
One of the first output (A1) and the second output (B1) of the first height measuring means, and the first output (A2) and the second output (B2) of the second height measuring means. Selecting one signal according to the state of the substrate and the measurement target region, and specifying the measurement target region on the substrate based on the difference between the output of the substrate surface reflected in the selected output signal and the output of the measurement target region; did.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. The same parts as those in FIG.
[0022]
FIG. 1 is a block diagram of a first embodiment of the present invention, and FIG. 2 is a principle diagram thereof. The laser light source 1 is, for example, a combination of a semiconductor laser and a collimation lens. The laser light source 1 is placed on the transport stage 4 by a light projection optical system including a combination of the laser light source 1 and a polygon scanner 2 which is a kind of optical scanner and a scanning lens 3. It is one-dimensionally scanned on the surface of the substrate 6 on which the unevenness, for example, the electrode 5 is formed.
[0023]
Further, the transfer stage 4 is moved in a direction orthogonal to the one-dimensional scanning direction of the laser light by the light projecting optical system under the control of the control / measurement unit 7.
[0024]
Therefore, the laser light is two-dimensionally scanned with respect to the region where the electrode 5 of the substrate 6 is arranged by one-dimensional scanning of the laser light by the light projecting optical system and the movement of the transport stage 4.
[0025]
On the other hand, an imaging optical system 8 is arranged obliquely above in a direction symmetric with the light projecting optical system 3 with respect to the normal line of the substrate 6, and the reflected light from the substrate 6 is reflected by the imaging optical system 8. And is imaged on a PSD (optical position detector) 9 and a second PSD 18 which are first height measuring means.
[0026]
The laser light imaged on the PSDs 9 and 18 moves one-dimensionally in accordance with the height of the position on the substrate 6 where the laser light is irradiated. Currents A1, B1, A2, and B2 flow in accordance with the amount of light reflected from the substrate 6.
[0027]
The IV converters 10 are connected to the output terminals of the currents A1, B1, A2 and B2 of the PSDs 9 and 18, respectively. Further, the outputs of the IV converters 10 are converted into A / D converters 11 respectively. And is digitized and sent to the control / measurement unit 7.
[0028]
Next, this configuration and operation will be described in detail based on the principle diagram of FIG.
The laser light is reflected and scanned by the polygon scanner 2 and enters the scanning lens 3. The scanning lens 3 converges the laser beam on the target substrate 6. At this time, the F number F1 of the convergence angle of the laser beam is F1 = f1 / d1 where the focal length of the scanning lens 3 is f1 and the diameter of the laser beam before the scanning lens is d1.
[0029]
The laser light is reflected on the target substrate and enters the imaging optical system 8. The numerical aperture of the imaging optical system 8 is F2, and F2 <F1. The beam diameter in the imaging optical system is d2, and d2 = s1 / F2 where the front distance is s1. However, when the laser light is specularly reflected on the target substrate 6, the numerical aperture of the reflected light of the laser is F1, which is the same as before reflection, so the laser beam diameter d3 in the imaging optical system is calculated as d3 = s1 / F1. , F2 <F1, so d3 <d2.
[0030]
Therefore, the mirror 19 is installed within the range in which the d3 portion moves by scanning, and the first height measuring means, in this case, is long in the scanning direction and the measuring direction is orthogonal to the scanning direction via the light quantity reduction filter 20. Incident on the dimension PSD9.
[0031]
On the other hand, the light that has passed through the mirror 19 is incident on the second PSD 18. The operation of such a configuration will be described next. When regular reflection occurs on the target substrate 6, the reflected light is extremely strong and has high directivity, so that it is incident on the d3 portion in a concentrated manner. This light can be guided to the first PSD 9 by the mirror 19, and can be measured stably with a high speed and high accuracy by measuring the light amount by appropriately reducing the light amount by the light amount reducing filter 20 and also by appropriately reducing the gain of the first PSD 9. Can do.
[0032]
On the other hand, when diffuse reflection occurs on the target substrate 6, the reflected light is weakened because it is dispersed over a wide range, and also enters the imaging optical system 8 with the full numerical aperture of the imaging optical system, extending to the portion d 2. Therefore, the diffused light can be incident on the second PSD 18 even when the portion other than the d3 portion is condensed on the second PSD 18.
[0033]
Therefore, the diffuse reflection portion on the target substrate can be stably measured by appropriately increasing the gain of the PSD 18. Increasing the gain of the second PSD 18 causes obstacles such as saturation of the PSD when specularly reflected light is incident. However, since the specularly reflected light is guided to the first PSD 9 side by the mirror 19, there is a problem of saturation. Is not. However, the specularly reflected light whose electrode surface is slanted is off the mirror 19 at the d3 portion and enters the second PSD 18, but it is not a fatal obstacle because of the short time.
[0034]
FIG. 3 is a block diagram of a second embodiment of the present invention, and FIG. 4 is a principle diagram thereof. The imaging optical system 8 is divided into a front side 8a and rear sides 8b, 8c, and the light is parallel light between them. A cylindrical lens 21 is installed immediately after the front imaging optical system 8a. Thereafter, the beam splitter 22 is placed to reflect a light amount of about 5% and transmit the remaining light. The reflected light is incident on the first PSD 9 via the first rear imaging optical system 8b.
[0035]
On the other hand, a light shielding plate 23 is installed at the portion corresponding to d3 described in the second embodiment of the transmitted light, and only the directly reflected light is shielded. At this time, a light quantity reduction filter of about 5% may be installed instead of the light shielding plate. In this case, the directly reflected light is transmitted at the same light level as the diffusely reflected light. Thereafter, the light is condensed by the second rear imaging optical system 8 c and imaged on the second PSD 18.
[0036]
The operation of such a configuration will be described next. When regular reflection occurs on the target substrate, the reflected light is extremely strong and is intensively incident on the d3 portion. At this time, when the light enters the cylindrical lens 21 immediately after the front imaging optical system 8a, the spots on the PSDs 9 and 18 spread in the scanning direction by the action of the cylindrical lens 21. PSD has the property of slowing down the response when the optical power density is high. Therefore, the insertion of the cylindrical lens 21 has the effect of reducing the optical power density on the PSDs 9 and 18 and keeping the PSD response at a high speed.
[0037]
However, in the convergent light as in the case of Example 1, the aberration is deteriorated by the insertion of the cylindrical lens, and the accuracy is adversely affected. Therefore, in the second embodiment, the imaging optical system 8 is divided into the parallel light between the front side 8a and the back side 8b, 8c, thereby suppressing the occurrence of aberration by the cylindrical lens 21.
[0038]
The specularly reflected light is reflected by the beam splitter 22 and then enters the first rear imaging optical system 8b. Since the specular reflection light is strong and the diffuse reflection light is weak, the laser power has to be increased when trying to detect the diffuse reflection light, but then it is too powerful to cause the regular reflection light to enter the PSD 9 as it is. For this reason, an appropriate amount of light enters the PSD 9 by optically reducing the power by the beam splitter 22 of about 5%.
[0039]
Further, the parallel light portion has a very small amount of change in aberration due to mechanical deviation, and is advantageous in terms of assembly and adjustment.
[0040]
On the other hand, the light beam that has passed through the beam splitter 22 enters the second rear imaging optical system 8 c and forms an image on the second PSD 18. However, in order to see the diffused light, if regular reflection light is mixed, the light quantity is too large and the response speed is lowered. Therefore, the diffused light can be stably measured by placing a light shielding plate or a light amount reducing filter 23 of about 5% in the d3 portion to attenuate the light amount of the regular reflection light. Occurrence of aberration due to transmission through the beam splitter 22 is suppressed because it is parallel light before entering the second rear imaging optical system 8c.
[0041]
However, the specularly reflected light on the electrode slope is incident on the second PSD 18 without being blocked by the light shielding plate or the light amount reducing filter 23. When the regular reflection light is incident, the PSD 18 is saturated, but if the power density is suppressed by the cylindrical lens 21, a decrease in the response speed of the PSD 18 itself can be avoided, and only the electric saturation can be suppressed. The impact on can be minimized.
[0042]
Although the embodiment in which the light quantity reduction filter is set to about 5% has been described, it goes without saying that an appropriate value should be set according to the ratio of the reflectance of the metal surface to the substrate surface. Further, the beam splitter may be 5% reflection, but may be 95% reflection. In this case, the first and second rear imaging optical systems and PSDs are interchanged.
[0043]
Although the beam splitter 22 is used in the second embodiment, a mirror that reflects only the d3 portion or a mirror that has a hole only in the d3 portion may be used. In this case, an appropriate light quantity can be obtained with respect to the PSD 9 by inserting a light quantity reduction filter 20 before the first imaging optical system.
[0044]
In addition, a plane parallel glass plate in which only the d3 part is a light quantity reducing filter may be used. In this case, the d3 part can be reduced in the light quantity by the light quantity reducing filter and incident on the diffused light PSD 18, so The reflected light can be measured under similar conditions.
[0045]
FIG. 5 is a block diagram of the third embodiment of the present invention. In this embodiment, the optical scanning is performed by the resonance type mirror scanner 2b. The resonant mirror scanner 2b has a demerit that the scanning speed is not constant, but has a merit that it is small in size, has a high scanning efficiency, can be increased in speed, and has a long life.
[0046]
A mirror 19 is placed in front of the imaging optical system 8 to reflect the portion corresponding to d3 described in the first embodiment and transmit the rest. The reflected light is incident on the first PSD 9 via the first imaging optical system 8d and the light quantity reduction filter 20. The transmitted light is condensed by the second imaging optical system 8 and imaged on the second PSD 18.
[0047]
The operation of such a configuration will be described next. When regular reflection occurs on the target substrate 6, the reflected light is extremely strong and is intensively incident on the d3 portion. Therefore, the direct reflected light can be stably measured by making the reflected light of the mirror 19 incident on the first PSD 9 via the first imaging optical system 8d. In this case, an inexpensive lens can be used as the first imaging optical system 8d because the aperture may be as narrow as the light projecting optical system 3.
[0048]
On the other hand, the portion of the light beam that is not reflected by the mirror 19 enters the second imaging optical system 8 and forms an image on the second PSD 18. In order to see the diffused light, if there is specularly reflected light, there is a problem that the light amount is too large and the response speed is lowered. However, since the d3 portion is shielded by the mirror, the specularly reflected light does not enter the second PSD 18. Diffuse light can be measured stably.
[0049]
As described in the third embodiment, it is desirable to insert cylindrical lenses in the first and second imaging optical systems in order to secure the response speed of PSD.
[0050]
In any example, the measurement object side of the scanning lens 3 and the imaging optical systems 8, 8a, 8d is telecentric. This is necessary for reliably capturing the specularly reflected light.
[0051]
Hereinafter, characteristics and usage methods of the signal from the first PSD and the signal from the second PSD will be described.
First, a method of using the signal described in the first embodiment will be described. When measuring the height of the electrode on the substrate from the substrate, the regular reflected light of the electrode is mainly incident on the first PSD 9. In general, since the electrode 5 is made of metal, the amount of light becomes extremely large when the laser light scanned by the light projecting optical system 3 is reflected by the top or a plane portion of the electrode 5 and enters the imaging optical system 8. Therefore, at this time, the circuit gain of the I-V converter 10 is reduced so that the output signal of the A / D converter 11 is not saturated.
[0052]
On the other hand, a part of the diffusely reflected light on the substrate 6 is incident on the second PSD 18. In general, since the surface of the substrate 6 is made of resin, it may absorb more than the metal surface and may have poor flatness. In addition, since only a part of the diffuse reflected light is captured, the amount of light reflected on the substrate 6 is extremely small compared to the reflection on the electrode 5. However, in general, the metal reflection surface on the electrode 5 is a smooth surface and the diffuse reflection light is extremely small. Therefore, the diffuse reflection light on the substrate is mainly incident on the second PSD 18.
[0053]
For this reason, when the gain of the IV converter 10 and the like after the first PSD 18 is increased according to the diffuse reflection light on the substrate, the height on the substrate 6 can be accurately obtained by the second PSD 18. Become.
[0054]
FIG. 6 shows the amount of reflected light and height measurement signal by the first PSD 9 and the amount of reflected light and height measurement signal by the second PSD 18 when the height of the substrate 6 and the electrode 5 is measured.
[0055]
Since the height measurement signal by the first PSD 9 is set in accordance with the amount of light reflected on the electrode 5, the height on the electrode 5 is clear, but as described above, the amount of reflected light on the surface of the substrate 6 is generally the electrode 5. Therefore, the height measurement signal on the surface of the substrate 6 is disturbed by the influence of noise. In addition, the amount of reflected light on the side surface of the electrode 5 is almost zero, and is greatly influenced by noise, which is not reliable as a height measurement signal.
[0056]
On the other hand, the height measurement signal by the second PSD 18 is set in accordance with the diffuse reflection light of the substrate 6, so that the height on the substrate 6 clearly appears. However, since there is little diffused reflected light on the electrode 5 and on the side surface of the electrode 5, it is affected by noise.
[0057]
Thus, since the height on the electrode 5 can be clearly obtained by the first PSD 9 and the height on the substrate by the second PSD 18, the height from the surface of the substrate 6 to the top of the electrode 5 can be obtained from these data. Can do.
[0058]
As a data selection method, the light quantity data (A1 + B1) and (A2 + B2) obtained from the PSDs 9 and 18 for each pixel are not saturated, and appropriate data selection can be performed by adopting the higher data. Become.
[0059]
However, the data on the side surface of the electrode 5 cannot be used because there is not enough light in either light receiving system, and in this case, the data is deleted.
[0060]
In the above data selection example, it is simultaneously determined whether the measurement point is on the substrate or the electrode depending on which PSD data is selected in the data processing. That is, in this case, the data adoption part from the PSD 9 is discriminated from the electrode 5, and the data adoption part from the PSD 18 is discriminated from the substrate 6. Even if the amount of reflected light from the electrode 5 is the same as the amount of reflected light from the substrate 6 in the first imaging optical system 8, the amount of reflected light from the substrate 6 is reflected by the electrode 5 in the second imaging optical system. Since it becomes larger, the electrode portion can be reliably identified and recognized by performing the data selection process.
[0061]
The above data selection method is effective when the substrate state is as follows. That is, “the object to be measured is an electrode”, “the reflectance of the electrode part is higher than the reflectance of the substrate surface”, and “the electrode part surface is specular and the substrate surface is diffusely reflective”. It is assumed that the position of the electrode part is higher than the surface of the substrate.
[0062]
However, not all measurement objects are applicable to these conditions, and depending on the substrate, specular reflection and reflectance may be higher than the measurement object. As conditions, there are three conditions of “reflectance”, “diffusivity”, and “position height”, such as “high”, “same”, and “low”, and there will be 36 combinations in total as will be described later. Actually, the substrate surface often changes from lot to lot, and it is not desirable for a substrate inspection apparatus to correspond to only one of 36 patterns.
[0063]
In the above data selection example and the conventional example, “luminance” obtained by adding two PSD outputs is generally used to determine the measurement target region. However, this “brightness” calculation operation is an operation for deleting the “height” information, and it is not always appropriate to delete the height information when there is a difference in height of the measurement target position.
[0064]
In addition, since luminance and height data are not necessarily required for all pixels, it should be calculated by a simpler means before calculating luminance and height data in order to reduce calculation time. It is desirable to narrow down the area.
Therefore, if only one side of the PSD output is taken out and binarized, information on both the light amount and the height can be used at a time as described below, and the calculation time can be shortened.
[0065]
If the interelectrode width of the PSD is L and light is incident on a position where x: y divides the light and current I is generated, current Ix / L is output to one electrode and current Iy / L is output to the other side. Is done. If two outputs are summed, I is calculated, but if only one output is used, Ix, that is, the numerical value of the product of the light amount and the position is obtained.
[0066]
Therefore, by selecting (or inverting) the output at which Ix is the highest (or lowering) and binarizing with a certain threshold value to determine the measurement target region, it is more than simply determining the measurement target region only by I. It is possible to accurately determine the region, and it is possible to cope with a case where the region cannot be distinguished only by I, such as when the light amount is the same.
[0067]
Specifically, assuming that the PSD electrode on the side where an image with a high position forms an image is A and the reverse is B, when the reflectance of the measurement target portion is high and the height of the measurement target portion is high, A is determined as the measurement target portion. When the height of the measurement target is low, B is inverted. When the measurement target is low and the measurement target is high, B is inverted. When the measurement target is low, A is inverted. By selecting, it becomes easier to distinguish the measurement target part.
[0068]
Further, by selecting the output of the first PSD and the second PSD, the optimum output can be selected depending on the specular reflectivity and diffusivity of each part. Specifically, assuming that the specular reflection light acquisition side is 1, and the diffuse reflection light acquisition side is 2, the measurement target portion has a high reflectivity and is specular, and 1 is diffuse reflection. By selecting the inversion of 2 when the reflectivity of the measurement target portion is low and regular reflection, and selecting the inversion of 1 when the reflectivity is diffuse reflection, it becomes easier to distinguish the measurement target portion.
[0069]
To summarize the preferred choices: The regular reflection light acquisition side PSD is 1, the diffuse reflection light acquisition side PSD is 2, and the electrode on the higher position side is A, and the low side is B. The lowercase letter t indicates inversion.
[0070]
High measurement target position and low substrate surface
Measurement object has high reflectivity and low substrate surface
Measurement target is specular reflection, substrate surface is diffusive A1
Measurement object is diffusive, substrate surface is diffusive A2
Measurement target is specular, substrate surface is specular A1
Measurement object is diffusive, substrate surface is specularly reflective A2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular reflection, substrate surface is diffusive A1
Measurement object is diffusive, substrate surface is diffusive A2
Measurement target is specular, substrate surface is specular A1
Measurement object is diffusive, substrate surface is specularly reflective A2
Low reflectivity of measurement object and high substrate surface
Measurement object is specular reflection, substrate surface is diffusive tB2
Measurement object is diffusive, substrate surface is diffusive tB1
The object to be measured is specular, and the substrate surface is specular. TB2
Measurement object is diffusive, substrate surface is specularly reflective tB1
The measurement target position and the substrate surface height are the same
Measurement object has high reflectivity and low substrate surface
Measurement target is specular reflection, substrate surface is diffusive A1 or B1
Measurement object is diffusive, substrate surface is diffusive A2 or B2
Measurement target is specular, substrate surface is specular A1 or B1
The object to be measured is diffusive and the substrate surface is specularly reflected. A2 or B2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular reflection, substrate surface is diffusive A1 or B1
Measurement object is diffusive, substrate surface is diffusive ×
Measurement target is specular, substrate surface is specular ×
The object to be measured is diffusive and the substrate surface is specularly reflected. A2 or B2
Low reflectivity of measurement object and high on substrate surface
Measurement object is specular reflection, substrate surface is diffusive tA2 or tB2
Measurement target is diffusive, substrate surface is diffusive tA1 or tB2
The object to be measured is specular, and the substrate surface is specular. TA2 or tB1
The object to be measured is diffusive and the substrate surface is specularly reflected. TA1 or tB1
The measurement target position is low and the substrate surface is high.
Measurement object has high reflectivity and low substrate surface
Measurement target is specular, substrate surface is diffusive B1
Measurement object is diffusive, substrate surface is diffusive B2
Measurement target is specular, substrate surface is specular. B1
Measurement object is diffusive, substrate surface is specularly reflected B2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular, substrate surface is diffusive B1
Measurement object is diffusive, substrate surface is diffusive B2
Measurement target is specular, substrate surface is specular. B1
Measurement object is diffusive, substrate surface is specularly reflected B2
Low reflectivity of measurement object and high on substrate surface
Measurement target is specular reflection, substrate surface is diffusive tA2
Measurement target is diffusive, substrate surface is diffusive tA1
Measurement target is specular, substrate surface is specular. TA2
The object to be measured is diffusive and the substrate surface is specularly reflective tA1
[0071]
As described above, by selecting only one output from each of two outputs of two PSDs, it is possible to distinguish all measurement target areas except for those in which all conditions are the same among 36 patterns. It can be seen that the number of types increases and it is easy to distinguish them.
[0072]
Further, since the luminance calculation for summing the two PSD outputs is not required when determining the measurement target region, the processing speed can be improved.
[0073]
The above is an example of selection that makes it easy to discriminate the measurement target part. Similarly, a selection example that makes it easy to discriminate the substrate surface part can be considered. Specifically:
High measurement target position and low substrate surface
Measurement object has high reflectivity and low substrate surface
Measurement target is specular, substrate surface is diffusive tA1
Measurement target is diffusive, substrate surface is diffusive tA1
Measurement target is specular, substrate surface is specular. TA2
The object to be measured is diffusive and the substrate surface is specularly reflective tA2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular reflection, substrate surface is diffusive B2
Measurement object is diffusive, substrate surface is diffusive B2
Measurement target is specular, substrate surface is specular. B1
Measurement target is diffusive, substrate surface is specularly reflected B1
Low reflectivity of measurement object and high substrate surface
Measurement target is specular reflection, substrate surface is diffusive B2
Measurement object is diffusive, substrate surface is diffusive B2
Measurement target is specular, substrate surface is specular. B1
Measurement target is diffusive, substrate surface is specularly reflected B1
The measurement target position and the substrate surface height are the same
Measurement object has high reflectivity and low substrate surface
Measurement target is specular reflection, substrate surface is diffusive tA1 or tB1
Measurement target is diffusive, substrate surface is diffusive tA1 or tB1
The object to be measured is specular, and the substrate surface is specular. TA2 or tB2
The object to be measured is diffusive and the substrate surface is specularly reflected. TA2 or tB2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular reflection, substrate surface is diffusive A2 or B2
Measurement object is diffusive, substrate surface is diffusive ×
Measurement target is specular, substrate surface is specular ×
The object to be measured is diffusive and the substrate surface is specularly reflected. A1 or B1
Low reflectivity of measurement object and high on substrate surface
Measurement target is specular reflection, substrate surface is diffusive A2 or B2
Measurement object is diffusive, substrate surface is diffusive A2 or B2
Measurement target is specular, substrate surface is specular A1 or B1
The object to be measured is diffusive and the substrate surface is specularly reflected A1 or B1
The measurement target position is low and the substrate surface is high.
Measurement object has high reflectivity and low substrate surface
Measurement target is specular reflection, substrate surface is diffusive tB1
Measurement object is diffusive, substrate surface is diffusive tB1
The object to be measured is specular, and the substrate surface is specular. TB2
The object to be measured is diffusive and the substrate surface is specular. TB2
The reflectance of the measurement target is the same as that of the substrate surface
Measurement target is specular reflection, substrate surface is diffusive A2
Measurement object is diffusive, substrate surface is diffusive A2
Measurement target is specular, substrate surface is specular A1
Measurement target is diffusive, substrate surface is specularly reflective A1
Low reflectivity of measurement object and high on substrate surface
Measurement target is specular reflection, substrate surface is diffusive A2
Measurement object is diffusive, substrate surface is diffusive A2
Measurement target is specular, substrate surface is specular A1
Measurement target is diffusive, substrate surface is specularly reflective A1
[0074]
In addition, this invention is not limited to the said embodiment, You may deform | transform as follows. For example, in the above embodiment, the case where the substrate 6 is applied to the height measurement of the electrode 5 formed on the substrate has been described. However, the present invention is not limited to this, and the height of the convex portion and the concave portion formed on various substrates is described. Applicable to the measurement.
[0075]
Moreover, although the electrode and the resin substrate have been described as examples of the material discrimination, the present invention is not limited to this and can be applied to discrimination of different types of substances formed on various substrates. For example, even in the case of a phosphor material and glass, since the phosphor material is irregularly reflective and the glass is specularly reflected, the reflected light of the glass can be captured by the first PSD 9 and the reflected light of the phosphor can be captured by the second PSD 18. Can do.
[0076]
【The invention's effect】
As described above in detail, according to the present invention, it is possible to provide a substrate inspection apparatus that can obtain a stable height measurement signal that is not affected by noise and that can perform accurate height measurement with a simple configuration.
[0077]
In addition, according to the present invention, it is possible to provide a substrate inspection apparatus capable of obtaining an accurate reflected light amount signal from both a measurement target portion such as an electrode and the substrate surface and accurately measuring the height.
[0078]
In addition, according to the present invention, it is possible to provide a substrate inspection apparatus that can distinguish the region at high speed regardless of the difference in conditions between the measurement object and the substrate surface.
[Brief description of the drawings]
FIG. 1 is a block diagram of a first embodiment of the present invention.
FIG. 2 is a principle diagram of the first embodiment of the present invention.
FIG. 3 is a block diagram of a second embodiment of the present invention.
FIG. 4 is a principle diagram of a second embodiment of the present invention.
FIG. 5 is a block diagram of a third embodiment of the present invention.
FIG. 6 is a diagram of a received light signal according to an embodiment of the present invention.
FIG. 7 is a configuration diagram of a conventional example.
FIG. 8 is a diagram of a received light signal in a conventional example.
[Explanation of symbols]
1 is a laser light source
2 is a polygon scanner
2b is a resonant mirror scanner
3 is a scanning lens
4 is the stage
5 is an area to be measured such as an electrode
6 is the substrate
7 is a control / measurement unit
8 is an imaging optical system
8a is a front imaging optical system
8b and 8c are rear imaging optical systems
8d is an imaging optical system
9 is the first PSD
10 is an IV converter
11 is an A / D converter
18 is the second PSD
19 is a mirror
20 is a light reduction filter
21 is a cylindrical lens
22 is a beam splitter
23 is a light shielding plate or a light quantity reduction filter

Claims

A light projecting optical system including a laser light source (1), an optical scanner (2), and a scanning lens (3), and provided at an angle symmetrical to the light projecting optical system with respect to the normal line of the measurement target substrate (6) The imaging optical system (8) and a plurality of height measuring means (9) and (18), and the convergence angle of the laser light of the projection optical system is smaller than the aperture angle of the imaging optical system, In the imaging optical system, the light in the range including the portion scanned by the aperture corresponding to the convergence angle of the laser beam and the light in the remaining range are separated by the mirror (19), and one of the first height measuring means (9 ), And the other is guided to the second height measuring means (18).

A light projecting optical system including a laser light source (1), an optical scanner (2), and a scanning lens (3), and provided at an angle symmetrical to the light projecting optical system with respect to the normal line of the measurement target substrate (6) The imaging optical system (8) and a plurality of height measuring means (9) and (18), and the convergence angle of the laser light of the projection optical system is smaller than the aperture angle of the imaging optical system, A part of the light quantity of the imaging optical system is separated by the beam splitter (22), one is guided to the first height measuring means (9), and the other is scanned by the aperture corresponding to the convergence angle of the light projecting optical system. A substrate inspection apparatus characterized by shielding a range including the portion to be guided to the second height measuring means (18).

A light projecting optical system including a laser light source (1), an optical scanner (2), and a scanning lens (3), and provided at an angle symmetrical to the light projecting optical system with respect to the normal line of the measurement target substrate (6) The imaging optical system (8) and a plurality of height measuring means (9) and (18), and the convergence angle of the laser light of the projection optical system is smaller than the aperture angle of the imaging optical system, A part of the light quantity of the imaging optical system is separated by the beam splitter (22), one is guided to the first height measuring means (9), and the other is scanned by the aperture corresponding to the convergence angle of the light projecting optical system. A substrate inspection apparatus characterized in that the amount of light in a range including the portion to be reduced is reduced by the light amount reduction filter (23) and led to the second height measuring means (18).

4. The substrate inspection apparatus according to claim 1, wherein a light quantity reduction filter (20) is inserted in an optical path guided to the first height measuring means (9).

One of the first output (A1) and the second output (B1) of the first height measuring means, and the first output (A2) and the second output (B2) of the second height measuring means. The measurement target specified after the measurement target region on the substrate is specified based on the difference between the output of the substrate surface reflected in the selected output signal and the output of the measurement target region on the substrate. 5. The substrate inspection apparatus according to claim 1, wherein the height and the luminance are calculated for each area.