JP3907518B2

JP3907518B2 - Shape measuring device

Info

Publication number: JP3907518B2
Application number: JP2002136753A
Authority: JP
Inventors: 勉森本; 弘行高松
Original assignee: Kobe Steel Ltd
Current assignee: Kobe Steel Ltd
Priority date: 2002-05-13
Filing date: 2002-05-13
Publication date: 2007-04-18
Anticipated expiration: 2022-05-13
Also published as: JP2003329414A

Description

【０００１】
【発明の属する技術分野】
本発明は，干渉計を用いて半導体ウェハの形状（厚さ分布）を測定する形状測定装置に関するものである。
【０００２】
【従来の技術】
半導体シリコン等のウェハの厚さ分布を測定する形状測定装置としては，例えば，特開２００１−２４１９２３号公報（公報１）に提案されているものがある。該公報１に示される形状測定装置は，ウェハの面に垂直に平行光を入射させるフィゾー型干渉計を，ウェハの主面側及びその裏面側に対向して配置し，ウェハの厚さ分布を測定する装置である。前記公報１の技術では，ウェハの主面からの反射光とウェハの主面側の基準平面からの反射光とによって形成される干渉縞に基づいて，ウェハの主面と主面側の基準平面との距離Ｌａ（ｘ，ｙ）を，ウェハの裏面からの反射光とウェハの裏面側の基準平面からの反射光とによって形成される干渉縞に基づいて，ウェハの裏面と裏面側の基準平面との距離Ｌｂ（ｘ，ｙ）を，ウェハを２つの基準平面の間から除去した状態で，前記平行光の一方に対するウェハの主面側及び裏面側それぞれの基準平面からの反射光によって形成される干渉縞に基づいて，２つの基準平面の距離Ｌｔ（ｘ，ｙ）をそれぞれ求め，ウェハの厚さＴ（ｘ，ｙ）を次の（１）式により求めるものである。
Ｔ（ｘ，ｙ）＝Ｌｔ（ｘ，ｙ）−Ｌａ（ｘ，ｙ）−Ｌｂ（ｘ，ｙ）…（１）
これにより，前記２つの基準平面それぞれの各位置（ｘ，ｙ）ごとにその変位（凹凸）分が相殺されるので，求められるウェハの測定精度に対して前記２つの基準平面それぞれの平坦度が十分でない場合であっても，これに依存せず精度の高いウェハの形状測定を行うことが可能となる。
即ち，ウェハ表面に垂直な方向から見た所定の位置（ｘ０，ｙ０）において，主面側及び裏面側それぞれの基準平面が，理想平面に対してそれぞれΔａ，Δｂの誤差（凹凸）を有しているとすると，その位置での各測定距離Ｌａ０（＝Ｌａ（ｘ０，ｙ０）），Ｌｂ０（＝Ｌｂ（ｘ０，ｙ０）），Ｌｔ０（＝Ｌｔ（ｘ０，ｙ０））は，それぞれ以下の式で表される。
Ｌａ０＝Ｌａ０’−Δａ
Ｌｂ０＝Ｌｂ０’−Δｂ …（２）
Ｌｔ０＝Ｌｔ０’−Δａ−Δｂ
ここで，Ｌａ０’，Ｌｂ０’，Ｌｔ０’は，２つの基準平面が理想平面であるときのＬａ０，Ｌｂ０，Ｌｔ０に対応する真の距離である。この（２）式を（１）式に適用すると，各誤差Δａ，Δｂが相殺され，理想平面を基準としたウェハの真の厚みを求められる。
【０００３】
【発明が解決しようとする課題】
しかしながら，現在利用されている主な半導体ウェハ（以下，単にウェハという）は，主面は鏡面であるが，裏面はエッチング処理やサンドブラスト処理等が施された散乱面（以下，粗面という）である。このような粗面（裏面）に対して前記公報１に示されるフィゾー干渉計を用いると，照射した平行光の反射率が低くなり，測定に十分な干渉縞が得られないという問題点があった。
これに対し，粗面の干渉測定を行う方法として，例えば特開２０００−８１３２１号公報（公報２）に示されるような斜入射干渉計が知られている。該公報２に示される技術は，ウェハに近接して配置されたプリズムの基準平面（参照面）を介してウェハ表面に平行光を斜めに照射し，ウェハの面及び基準平面それぞれからの反射光で形成される干渉縞画像に基づいて，ウェハの面と基準平面との距離を測定するものである。これを用いれば，粗面であっても反射率を高めることができ，測定に十分な干渉縞を得ることが可能となる。
しかしながら，前記公報２に示される技術では，平行光をウェハ表面に対して斜めに照射するため，プリズムの基準平面の平坦度が測定精度に影響することになる。このことについて，図３を用いて説明する。
【０００４】
図３（ａ）は，従来の斜入射干渉計Ｚにより基準平面とウェハとの距離を測定する際の光路を模式的に表した図である。図３（ａ）に示すように，斜入射干渉計をウェハ１の主面１ａ側，及び裏面１ｂ側それぞれに配置した場合を考える。即ち，主面１ａ側及び裏面１ｂ側それぞれにプリズム４１，４２を，その基準平面４１ａ，４２ａが主面１ａ及び裏面１ｂそれぞれに近接するよう配置する。
今，ウェハ１の裏面１ｂに斜めに照射される平行光Ｂが，プリズム４２の基準平面４２ａにおける所定の出射位置Ｌを通過してウェハ１の裏面１ｂに向かうとする。この光は，ウェハ１の面に垂直な方向から見て（図３（ａ）においては上側又は下側から見て），前記出射位置Ｌとずれた位置Ｍでウェハ１の裏面１ｂで反射した後，さらにずれた位置Ｎにおいて基準平面４２ａに到達する。従って，ウェハ１の裏面１ｂからの反射光Ｓと，基準平面４２ａからの反射光Ｒとの光路差は，（Ｌ〜Ｍ間の距離）＋（Ｍ〜Ｎ間の距離）となり，この光路差による干渉縞が得られる。このため，基準平面４２ａの平坦度が十分でなく，基準平面４２ａが理想平面に対して位置Ｌと位置Ｎとがずれていれば（表面に凹凸があれば），そのずれが測定誤差となってしまう。ここで，基準平面４２ａをウェハ１の裏面１ｂに十分近接させれば，位置Ｌ〜Ｎ間の距離が十分小さくなり，位置Ｌ，Ｍのずれによる測定誤差を許容範囲内に抑えることができる。このことは，ウェハ１の主面１ａ側においても同様である。しかし，このようにして得られる干渉縞に基づいて求められるのは，あくまで基準平面４１ａ，４２ａとウェハ１の表面１ａ，１ｂとの距離であり，求められるウェハの測定精度に対して２つの基準平面４１ａ，４２ａそれぞれの平坦度が十分でない場合には精度のよい形状測定が行えない。そこで，ウェハ１の主面１ａ側及び裏面１ｂ側それぞれの基準平面４１ａ，４２ａの平坦度の誤差の影響を除去する（相殺する）前記公報１の技術を適用することが考えられるが，そのためにはウェハ１を２つの基準平面４１ａ，４２ａの間から除去し，２つの基準平面４１ａ，４２ａの間の距離の分布を測定する必要がある。
【０００５】
しかしながら，一般に，ウェハ１の厚みは０．７〜０．８ｍｍ程度あるため，２つの基準平面４１ａ，４２ａの間隔は少なくとも１ｍｍ程度は必要となる。
図３（ｂ）は従来の斜入射干渉計Ｚにより２つの基準平面４１ａ，４２ａの間の距離を測定する際の光路を模式的に表した図である。図３（ｂ）に示すように，基準平面４２ａにおける所定の入射位置Ｌ’を通過して基準平面４１ａに向かう光は，前記入所位置Ｌ’とずれた位置Ｍ’で基準平面４１ａに反射した後，さらにずれた位置Ｎ’において基準平面４２ａに到達する。このときの位置Ｌ’〜Ｎ’間の距離（ずれ）は，１０ｍｍ程度にもなってしまう。一般に，基準平面には，１０ｍｍ程度を空間周期（間隔）として最大の平坦度誤差（例えば，０．０３μｍ程度）が生じやすいため，前記公報２に示される斜入射干渉計を前記公報１の技術で用いられる干渉計に置き換えると，基準平面の平坦度の最大誤差がそのまま測定誤差となってしまうという問題点があった。これでは，０．０１μｍオーダーの測定精度が要求された場合には適用できない。
従って，本発明は上記事情に鑑みてなされたものであり，その目的とするところは，ウェハの主面が鏡面，その裏面が粗面であっても，干渉計の基準平面が有する平坦度の誤差に影響されずに精度よくウェハの形状（厚み分布）を測定できる形状測定装置及びその方法を提供することにある。
【０００６】
【課題を解決するための手段】
上記目的を達成するために本発明は，ウェハの主面側に配置される基準平面及び前記ウェハの裏面側に近接して配置される基準平面の２つの基準平面を構成する手段と，前記主面側の基準平面を介して前記ウェハの主面に略垂直に平行光を照射する主面側発光手段と，前記主面側発光手段による平行光の前記ウェハの主面及び前記主面側の基準平面それぞれからの反射光で形成される第１の干渉縞画像を取得する主面側干渉縞画像取得手段と，前記裏面側の基準平面を介して前記ウェハの裏面に斜めに照射する裏面側発光手段と，前記裏面側発光手段による平行光の前記ウェハの裏面及び前記裏面側の基準平面それぞれからの反射光で形成される第２の干渉縞画像を取得する裏面側干渉縞画像取得手段と，前記ウェハを前記２つの基準平面の間から除去した状態で，前記主面側発光手段による平行光の前記裏面側の基準平面及び前記主面側の基準平面それぞれからの反射光で形成される第３の干渉縞画像を取得する基準平面干渉縞画像取得手段と，前記第１乃至３の干渉縞画像に基づいて前記ウェハの厚さ分布を算出する形状算出手段と，を具備してなることを特徴とする形状測定装置である。
【０００７】
ここで，前記形状算出手段は，前記ウェハの厚さ分布Ｔ（ｘ，ｙ）を，前記第１の干渉縞画像に基づき求められる前記ウェハの主面と前記主面側の基準平面との距離Ｌａ（ｘ，ｙ）と，前記第２の干渉縞画像に基づき求められる前記ウェハの裏面と前記裏面側の基準平面との距離Ｌｂ（ｘ，ｙ）と，前記第３の干渉縞画像により求められる前記２つの基準平面の距離Ｌｔ（ｘ，ｙ）とに基づいて，次式により算出することが可能である。
Ｔ（ｘ，ｙ）＝Ｌｔ（ｘ，ｙ）−Ｌａ（ｘ，ｙ）−Ｌｂ（ｘ，ｙ）
これにより，ウェハの主面（鏡面）及びその裏面（粗面）それぞれに適した干渉計を配置してウェハの主面及び裏面の形状を測定できるとともに，主面側の干渉計によって求めた主面側及び裏面側の２つの基準平面間の距離に基づいて，２つの基準平面それぞれの各位置ごとにその変位（凹凸）分が相殺されるので，求められるウェハの測定精度に対して２つの基準平面それぞれの平坦度が十分でない場合であっても，これに依存せず精度の高いウェハの形状測定を行うことが可能となる。さらに，垂直入射光が用いられる主面側では，主面側の基準平面とウェハとの間隔を確保できるので，斜入射光を用いる基準平面をウェハ両面に近接させて測定する場合に比べ，ウェハを２つの基準平面の間へ挿入及び除去するハンドリングが容易となる。
【０００８】
また，前記ウェハの少なくとも１箇所の絶対厚さを測定する絶対厚さ測定手段を具備し，前記形状算出手段が，前記絶対厚さ測定手段による前記絶対厚さに基づいて前記ウェハの絶対厚さ分布を算出するものであってもよい。
【０００９】
【発明の実施の形態】
以下添付図面を参照しながら，本発明の実施の形態及び実施例について説明し，本発明の理解に供する。尚，以下の実施の形態及び実施例は，本発明を具体化した一例であって，本発明の技術的範囲を限定する性格のものではない。
ここに，図１は本発明の実施の形態に係るウェハ形状測定装置Ｘの構成図，図２は本発明の実施の形態に係るウェハ形状測定装置Ｘにおける基準平面とウェハとの間及び２つの基準平面間の光路を模式的に表した図，図３は従来の斜入射干渉計Ｚによる基準平面とウェハとの間及び２つの基準平面間の光路を模式的に表した図である。
【００１０】
以下，図１を用いて本発明の実施の形態に係るウェハ形状測定装置Ｘについて説明する。
本ウェハ形状測定装置Ｘは，ウェハ支持部３０により支持されたウェーハ１の主面１ａ及び裏面１ｂの両面側に２つの光学測定系１０，２０が対向配置されている。
前記主面１ａ側の光学測定系１０はフィゾー干渉計であり，測定光１２を出射する主面側発光器１１，その測定光１２を透過させるハーフミラー１３，該ハーフミラー１３を透過後の前記測定光１２を前記主面１ａに垂直な平行ビームとするコリメータレンズ１４，その平行ビームの一部を透過し，一部を反射する主面側基準平面１５ａを有する参照板１５，前記ウェハ１の主面１ａで反射された測定光が前記参照板１５及び前記コリメータレンズ１４を経て，前記ハーフミラー１３に反射した光を受光するＣＣＤカメラ等の主面側受光器１６，及び前記参照板１５を前記ウェハ１の面に垂直な方向に移動させるピエゾアクチュエータ１７を備えている。ここで，前記主面側基準平面１５ａでの反射光と，前記ウェハ１の主面１ａでの反射光とには，前記主面側基準平面１５ａと前記主面１ａとの距離に対応する光路差があるため，この光路差が，前記主面側受光器１６において，両反射光で形成される干渉縞として観測される。
【００１１】
一方，前記裏面１ｂ側の光学測定系２０は，斜入射干渉計であり，測定光２２を出射する裏面側発光器２１，その測定光２２を前記裏面１ｂに対して斜め方向の平行ビームとするコリメータレンズ２４，その平行ビームの一部を透過し，一部を反射する裏面側基準平面２５ａを有する三角プリズム２５，前記ウェハ１の裏面１ｂで反射された測定光が前記裏面側基準平面２５ａ（前記三角プリズム２５）及びコリメータレンズ２４’等を経た光を受光するＣＣＤカメラ等の裏面側受光器２６，及び前記三角プリズム２５を前記ウェハ１の面に略垂直な方向に移動させるピエゾアクチュエータ２７を備えている。そして，前記裏面側基準平面２５ａでの反射光と，前記ウェハ１の裏面１ｂでの反射光とには，前記裏面側基準平面２５ａと前記裏面１ｂとの距離に対応する光路差があるため，この光路差が，前記裏面側受光器２６において，両反射光で形成される干渉縞として観測される。前記三角プリズム２５は，前記裏面側基準平面２５ａが前記ウェハ１の裏面１ｂに近接するよう（例えば，０．１ｍｍ程度の距離で）配置されている。前記測定光１２，２２としては，例えば，ＨｎＮｅレーザ（λ＝６３３ｎｍ）や，半導体レーザ等のコヒーレント光が用いられる。また，前記三角プリズム２５は，フィゾー干渉で用いられるようなウェッジプリズム等の参照板を用いてもかまわない。
【００１２】
前記主面側受光器１６及び前記裏面側受光器２６で得られる干渉縞画像は，画像入力手段を有するパーソナルコンピュータ等である演算器３１（前記形状算出手段の一例）に入力され，入力された干渉縞画像に基づいて前記ウェハ１の主面１ａ及び裏面１ｂの表面形状（高さ分布）が演算される。以下，前記主面側受光器１６で得られたウェハ１の主面１ａと前記主面側基準平面１５ａとの反射光で形成される干渉縞画像をａ，前記裏面側受光器２６で得られたウェハ１の裏面１ｂと前記裏面側基準平面２５ａとの反射光で形成される干渉縞画像をｂとする。
【００１３】
さらに，前記演算器３１には，前記２つの干渉縞画像ａ，ｂの他に，前記主面側基準平面１５ａ及び前記裏面側基準平面２５ａそれぞれからの反射光で形成される干渉縞画像ｃが取り込まれる。前記干渉縞画像ｃは，前記ウェハ１を前記２つの光学測定系１０，２０の間から取り除いた状態で，前記主面１ａ側の発光器１１からの測定光の前記主面側及び裏面側の各基準平面１５ａ，２５ａにおける反射光を前記主面側受光器１６で受光することにより観測する。前記演算器３１は，入力した前記干渉縞画像ｃにより，前記主面側基準平面１５ａと前記裏面側基準平面２５ｂとの距離の分布を演算する。ここで，前記主面側発光器１１及び前記コリメータレンズ１４が前記主面側発光手段の一例，前記裏面側発光器２１及び前記コリメータレンズ２４が前記裏面側発光手段の一例，前記主面側受光器１６が前記主面側干渉縞画像取得手段及び基準平面干渉縞画像取得手段の一例，前記裏面側受光器２６が前記裏面側干渉縞画像取得手段の一例である。
【００１４】
次に，図２を用いて，前記干渉縞画像ａ〜ｃそれぞれを観測時の前記平行光の光路について説明する。
前記干渉縞画像ａ（主面）の観測時においては，図２（ａ）に示すように，前記主面側基準平面１５ａの所定の入射位置Ｑから前記ウェハ１の主面１ａに向かう光は，前記主面１ａに垂直な方向に照射されているため，前記ウェハ１の面に垂直な方向から見て（図２（ａ）においては上側又は下側から見て），前記入射位置Ｑと同じ位置Ｍ’でウェハ１の主面１ａに反射した後，前記主面側基準平面１５ａの前記入射位置Ｑに戻る。このように，前記出射位置Ｑから出た光が元の位置に戻ってくるため，前記主面側基準平面１５ａの平坦度（凹凸）の影響を受けない。
一方，前記干渉縞画像ｂ（裏面）の観測時においては，図２（ａ）に示すように，前記裏面側基準平面２５ａの所定の入射位置Ｌから前記ウェハ１の裏面１ｂに向かう光は，前記裏面１ｂに対して斜めに照射されているため，前記ウェハ１の面に垂直な方向から見て，前記入射位置Ｌとずれた位置Ｍで反射した後，さらにずれた位置Ｎにおいて前記裏面側基準平面２５ａに到達する。ここで，前述したように，前記裏面側基準平面２５ａは前記ウェハ１の裏面１ｂに近接しているので，前記位置ＬとＮのずれは，測定精度に問題が生じない程度に十分小さい。前述したように，一般に，基準平面には１０ｍｍ程度を空間周期（間隔）として最大の平坦度誤差が生じやすいが，例えば，前記裏面側基準平面２５ａを前記ウェハ１の裏面１ｂに対して０．１ｍｍ程度まで近接させれば，前記位置ＬとＮのずれは，１ｍｍ程度となり，基準平面の平坦度誤差の影響を無視できる程度に抑えられる。
また，前記干渉縞画像ｃ（２つの基準平面間）の観測時においては，図２（ｂ）に示すように，前記主面側基準平面１５ａの前記入射位置Ｑから前記ウェハ１の主面１ａに向かう光は，前記ウェハ１の面に垂直な方向から見て（図２（ｂ）においては上側又は下側から見て），前記入射位置Ｑと同じ位置Ｐで前記裏面側基準平面２５ａで反射した後，前記主面側基準平面１５ａの前記入射位置Ｑに戻る。このように，前記出射位置Ｑから出た光が元の位置に戻ってくるため，前記主面側基準平面１５ａの平坦度（凹凸）の影響を受けない。
従って，前記ウェハ１の主面１ａ側における前記位置Ｍ’と前記裏面１ｂ側における前記位置Ｍとが表裏関係の位置であると考えると，前記裏面側基準平面２５ａにおける前記位置Ｌ，Ｐ，Ｎそれぞれのずれは，前記裏面側基準平面２５ａの平坦度誤差の影響を無視できる程度に十分小さく抑えられる。
また，前記ウェハ１の裏面１ｂ側においては，前記裏面側基準平面２５ａと前記ウェハ１とが近接するが，前記ウェハ１の主面１ａ側においては，前記主面側基準平面１５ａと前記ウェハ１との間隔を確保できるため，前記ウェハ１を前記２つの基準平面１５ａ，２５ａの間に挿入したり除去したりするハンドリングが容易となる。
【００１５】
以上の測定により得られた前記干渉縞画像ａ〜ｃを用いて，前記演算器３１では次のような演算を行う。
まず，前記干渉縞画像ａに基づいて，前記主面側基準平面１５ａと前記主面１ａとの間の距離Ｌａ（ｘ，ｙ），前記干渉縞画像ｂに基づいて前記裏面側基準平面２５ａと前記裏面１ｂとの間の距離Ｌｂ（ｘ，ｙ），前記干渉縞画像ｃに基づいて前記主面側基準平面１５ａと前記裏面側基準平面２５ａとの間の距離Ｌｔ（ｘ，ｙ）が求められる。
前記主面１ａ側の光学測定系１０による前記干渉縞画像ａは，前記ピアゾアクチュエータ１７によって前記参照板１５（即ち，前記主面側基準平面１５ａ）を，例えば，０（所定の基準位置），λ／８，２λ／８，３λ／８の各位置（λは測定光の波長）に移動させることにより，それぞれの位置での干渉縞画像の輝度データＩ₀（ｘ，ｙ），Ｉ₉₀（ｘ，ｙ），Ｉ₁₈₀（ｘ，ｙ），Ｉ₂₇₀（ｘ，ｙ）として前記演算器３１に取り込まれる。この輝度データに基づいて，前記干渉縞画像ａの位相φａ（ｘ，ｙ）は，次の（２）式で求められる。
φａ（ｘ，ｙ）＝arctan｛（Ｉ₀−Ｉ₁₈₀）／（Ｉ₉₀／Ｉ₂₇₀）｝…（２）
このようにして得られたφａ（ｘ，ｙ）に所定のアンラップ処理（処理後の位相をφａ’（ｘ，ｙ）とする）を施した場合，前記距離Ｌａ（ｘ，ｙ）は，次の（３）式で求められる。
Ｌａ（ｘ，ｙ）＝（φａ’（ｘ，ｙ）／２π）×（λ／２） …（３）
前記距離Ｌｃ（ｘ，ｙ）も，前記距離Ｌａ（ｘ，ｙ）と同様にして求められる。
また，前記裏面１ｂ側の光学測定系２０による前記干渉縞画像ｂも同様に，前記ピアゾアクチュエータ２７によって前記三角プリズム２５（即ち，前記裏面側基準平面２５ａ）を，前記測定光２２の前記ウェハ１の裏面１ｂへの入射角θを考慮した位置，例えば，０（所定の基準位置），λ／８／ｃｏｎθ，２λ／８／ｃｏｓθ，３λ／８／ｃｏｓθの各位置に移動させて，それぞれの干渉縞画像の輝度データとして前記演算器３１に取り込まれる。さらに，その輝度データから，前記φａ’（ｘ，ｙ）を求めたのと同様にアンラップ処理後の前記干渉縞画像ｂの位相φｂ’（ｘ，ｙ）が求められる。そして，次の（４）式により，前記距離Ｌｂ（ｘ，ｙ）が求められる。
Ｌｂ（ｘ，ｙ）＝（φｂ’（ｘ，ｙ）／２π）×（λ／２×１／ｃｏｓθ）
…（４）
そして，前記３つの距離Ｌａ（ｘ，ｙ），Ｌｂ（ｘ，ｙ），Ｌｔ（ｘ，ｙ）に基づいて，前記（１）式により，前記ウェハ１の厚さ分布Ｔ（ｘ，ｙ）が求められる。
このようにして求められたウェハ１の厚さ分布Ｔ（ｘ，ｙ）は，前記公報１の段落００１４に示されるように，前記２つの基準平面１５ａ，２５ａそれぞれの各位置（ｘ，ｙ）ごとにその変位（凹凸）分が相殺されるので，前記主面側基準平面１５ａ及び前記裏面側基準平面２５ａの平坦度（平面形状）に依存しない値である。即ち，本ウェハ形状測定装置Ｘによれば，基準平面の平坦度に依存せず高精度にウェハの厚さ分布測定を行うことが可能となる。これにより，前記ウェハ１の主面１ａが鏡面，その裏面１ｂが粗面であっても，前記２つの基準平面１５ａ，２５ａの平坦度の誤差に影響されずに精度よく前記ウェハ１の形状（厚み分布）を測定できる。また，前記２つの基準平面１５ａ，２５ａは必ずしも理想平面である必要がないため，装置の低コスト化が実現できる。
【００１６】
【実施例】
前記実施の携帯に係るウェハ形状測定装置Ｘで求められる前記ウェハ１の厚さ分布Ｔ（ｘ，ｙ）は未知のＯｆｆｓｅｔ量を含むものとなる。そこで，前記ウェハ１の任意位置での厚さを実測する厚さ測定器（前記絶対厚さ測定手段に相当）を設け，得られた厚さ実測値ｔを前記演算機３１に入力すれば，前記厚さ実測値ｔに基づいて前記ウェハ１の絶対厚さ分布を求めることが可能である。即ち，前記厚さ分布Ｔ（ｘ，ｙ）のうち，前記厚さ実測値ｔの測定点での値が前記厚さ実測値ｔと一致するように，前記厚さ分布Ｔ（ｘ，ｙ）の各値を平行移動させればよい。
また，前記実施の形態では，前記ピエゾアクチュエータ１７，２７により，前記２つの基準平面１５ａ，２５ａを，９０°の位相ごとに４段階移動させているが，他の位相シフト方法（例えば，波長掃引等）であってもかまわない。さらに，前記２つの基準平面１５ａ，２５ａを移動させるのではなく，前記ウェハ１をアクチュエータで移動させる構成としてもよい。
また，前記実施の形態では，前記ウェハ１の主面１ａ側の光学測定系１０として，フィゾー干渉計を用いているが，前記ウェハ１の面に略垂直に光を照射する他の干渉計（例えば，マイケルソン干渉計等）であってもかまわない。
【００１７】
【発明の効果】
以上説明したように，本発明によれば，ウェハの主面（鏡面）及びその裏面（粗面）それぞれに適した干渉計を配置してウェハの主面及び裏面の形状を測定するとともに，主面側の干渉計によって求めた主面側及び裏面側の２つの基準平面間の距離に基づいて，２つの基準平面それぞれの各位置ごとにその変位（凹凸）分が相殺されるので，求められるウェハの測定精度に対して２つの基準平面それぞれの平坦度が十分でない場合であっても，これに依存せず精度の高いウェハの形状測定を行うことが可能となる。
また，垂直入射光が用いられる主面側の干渉計とウェハとの間隔を確保できるので，斜入射干渉計をウェハ両面に近接させて測定する場合に比べ，ウェハを２つの基準平面の間へ挿入及び除去するハンドリングが容易となる。
【図面の簡単な説明】
【図１】本発明の実施の形態に係るウェハ形状測定装置Ｘの構成図。
【図２】本発明の実施の形態に係るウェハ形状測定装置Ｘにおける基準平面とウェハとの間及び２つの基準平面間の光路を模式的に表した図。
【図３】従来の斜入射干渉計Ｚによる基準平面とウェハとの間及び２つの基準平面間の光路を模式的に表した図。
【符号の説明】
１…ウェハ
１０…主面側の光学測定系（フィゾー干渉計）
１１…主面側発光器
１２…測定光（主面側）
１３…ハーフミラー
１４，２４，２４’…コリメータレンズ
１５…参照板
１５ａ…主面側基準平面
１６…主面側受光器
１７，２７…ピエゾアクチュエータ
２０…裏面側の光学測定系（斜入射干渉計）
２１…裏面側発光器
２２…測定光（裏面側）
２５…三角プリズム
２５ａ…裏面側基準平面
２６…裏面側受光器
３０…ウェハ支持部
３１…演算器[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a shape measuring apparatus that measures the shape (thickness distribution) of a semiconductor wafer using an interferometer.
[0002]
[Prior art]
As a shape measuring apparatus for measuring the thickness distribution of a wafer such as semiconductor silicon, for example, there is one proposed in Japanese Patent Application Laid-Open No. 2001-241923 (Publication 1). In the shape measuring apparatus disclosed in the publication 1, a Fizeau interferometer that allows parallel light to be incident perpendicularly to the surface of a wafer is disposed opposite to the main surface side and the back surface side of the wafer, and the thickness distribution of the wafer is measured. It is a device to measure. In the technique of the publication 1, based on interference fringes formed by the reflected light from the main surface of the wafer and the reflected light from the reference plane on the main surface side of the wafer, the main surface of the wafer and the reference plane on the main surface side. The distance La (x, y) to the reference plane on the back side and the back side of the wafer based on interference fringes formed by the reflected light from the back side of the wafer and the reflected light from the reference plane on the back side of the wafer The distance Lb (x, y) is formed by the reflected light from the reference plane on each of the main surface side and the back surface side of the wafer with respect to one of the parallel lights with the wafer removed from between the two reference planes. The distance Lt (x, y) between the two reference planes is obtained based on the interference fringes, and the wafer thickness T (x, y) is obtained by the following equation (1).
T (x, y) = Lt (x, y) −La (x, y) −Lb (x, y) (1)
As a result, the displacement (unevenness) is canceled at each position (x, y) of each of the two reference planes, so that the flatness of each of the two reference planes can be obtained with respect to the required measurement accuracy of the wafer. Even if it is not sufficient, the wafer shape can be measured with high accuracy without depending on this.
That is, at a predetermined position (x0, y0) viewed from the direction perpendicular to the wafer surface, the reference planes on the main surface side and the back surface side have errors (unevenness) of Δa and Δb, respectively, with respect to the ideal plane. Assuming that the measurement distances La0 (= La (x0, y0)), Lb0 (= Lb (x0, y0)), and Lt0 (= Lt (x0, y0)) at the position are respectively expressed by the following equations: It is represented by
La0 = La0′−Δa
Lb0 = Lb0′−Δb (2)
Lt0 = Lt0′−Δa−Δb
Here, La0 ′, Lb0 ′, and Lt0 ′ are true distances corresponding to La0, Lb0, and Lt0 when the two reference planes are ideal planes. When this equation (2) is applied to the equation (1), the errors Δa and Δb are canceled out, and the true thickness of the wafer based on the ideal plane can be obtained.
[0003]
[Problems to be solved by the invention]
However, the main semiconductor wafers currently used (hereinafter simply referred to as wafers) are mirror surfaces on the main surface, but the back surfaces are scattering surfaces (hereinafter referred to as rough surfaces) that have been subjected to etching or sandblasting. is there. When the Fizeau interferometer disclosed in the above publication 1 is used for such a rough surface (back surface), the reflectance of the irradiated parallel light becomes low, and there is a problem that sufficient interference fringes cannot be obtained for measurement. It was.
On the other hand, an oblique incidence interferometer as disclosed in, for example, Japanese Patent Application Laid-Open No. 2000-81321 (Publication 2) is known as a method for performing interference measurement on a rough surface. The technique disclosed in the publication 2 irradiates parallel light obliquely on the wafer surface via a reference plane (reference surface) of a prism arranged close to the wafer, and reflects light from each of the wafer surface and the reference plane. The distance between the surface of the wafer and the reference plane is measured based on the interference fringe image formed in (1). If this is used, the reflectance can be increased even on a rough surface, and interference fringes sufficient for measurement can be obtained.
However, in the technique disclosed in the publication 2, since the parallel light is irradiated obliquely with respect to the wafer surface, the flatness of the reference plane of the prism affects the measurement accuracy. This will be described with reference to FIG.
[0004]
FIG. 3A is a diagram schematically showing an optical path when the distance between the reference plane and the wafer is measured by the conventional oblique incidence interferometer Z. FIG. As shown in FIG. 3A, consider a case where the oblique incidence interferometers are arranged on the main surface 1a side and the back surface 1b side of the wafer 1, respectively. That is, the prisms 41 and 42 are arranged on the main surface 1a side and the back surface 1b side, respectively, so that the reference planes 41a and 42a are close to the main surface 1a and the back surface 1b, respectively.
Now, it is assumed that the parallel light B obliquely applied to the back surface 1 b of the wafer 1 passes through a predetermined emission position L on the reference plane 42 a of the prism 42 and travels toward the back surface 1 b of the wafer 1. This light is reflected from the back surface 1b of the wafer 1 at a position M that is shifted from the emission position L when viewed from a direction perpendicular to the surface of the wafer 1 (viewed from the upper side or the lower side in FIG. 3A). Thereafter, the reference plane 42a is reached at a position N that is further shifted. Therefore, the optical path difference between the reflected light S from the back surface 1b of the wafer 1 and the reflected light R from the reference plane 42a is (distance between L and M) + (distance between M and N). Interference fringes due to are obtained. For this reason, if the flatness of the reference plane 42a is not sufficient and the position L and the position N of the reference plane 42a are deviated from the ideal plane (if the surface has irregularities), the deviation becomes a measurement error. End up. Here, if the reference plane 42a is sufficiently close to the back surface 1b of the wafer 1, the distance between the positions L to N becomes sufficiently small, and the measurement error due to the deviation of the positions L and M can be suppressed within an allowable range. The same applies to the main surface 1a side of the wafer 1. However, what is obtained based on the interference fringes obtained in this way is only the distance between the reference planes 41a and 42a and the surfaces 1a and 1b of the wafer 1, and there are two criteria for the required measurement accuracy of the wafer. If the flatness of each of the flat surfaces 41a and 42a is not sufficient, accurate shape measurement cannot be performed. Therefore, it is conceivable to apply the technique of the publication 1 for removing (cancelling) the influence of the flatness error of the reference planes 41a and 42a on the main surface 1a side and the back surface 1b side of the wafer 1, respectively. Needs to remove the wafer 1 from between the two reference planes 41a and 42a and measure the distribution of distances between the two reference planes 41a and 42a.
[0005]
However, since the thickness of the wafer 1 is generally about 0.7 to 0.8 mm, the interval between the two reference planes 41a and 42a is required to be at least about 1 mm.
FIG. 3B is a diagram schematically showing an optical path when the distance between the two reference planes 41 a and 42 a is measured by the conventional oblique incidence interferometer Z. As shown in FIG. 3B, light that passes through a predetermined incident position L ′ on the reference plane 42a and travels toward the reference plane 41a is reflected on the reference plane 41a at a position M ′ that deviates from the entrance position L ′. After that, the reference plane 42a is reached at a further shifted position N ′. At this time, the distance (shift) between the positions L ′ to N ′ is about 10 mm. In general, since the maximum flatness error (for example, about 0.03 μm) is likely to occur in the reference plane with a spatial period (interval) of about 10 mm, the oblique incidence interferometer shown in the publication 2 is used in the technique of the publication 1. When using the interferometer used in, the maximum flatness error of the reference plane becomes the measurement error as it is. This is not applicable when a measurement accuracy of the order of 0.01 μm is required.
Accordingly, the present invention has been made in view of the above circumstances, and the object of the present invention is to provide the flatness of the reference plane of the interferometer even if the main surface of the wafer is a mirror surface and the back surface thereof is a rough surface. An object of the present invention is to provide a shape measuring apparatus and method that can accurately measure the shape (thickness distribution) of a wafer without being affected by errors.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, the present invention comprises means for forming two reference planes, a reference plane disposed on the main surface side of a wafer and a reference plane disposed close to the back surface side of the wafer, A main surface side light emitting means for irradiating parallel light substantially perpendicularly to the main surface of the wafer through a reference plane on the surface side; and the main surface of the wafer and the main surface side of the parallel light by the main surface side light emitting means Main surface side interference fringe image acquisition means for acquiring a first interference fringe image formed by reflected light from each of the reference planes, and a back surface side for obliquely irradiating the back surface of the wafer via the back surface reference plane A light-emitting unit, and a back-side interference fringe image acquisition unit that acquires a second interference fringe image formed by reflected light from the back surface of the wafer and the reference plane on the back side of the parallel light by the back-side light-emitting unit. , The wafer of the two reference planes The reference plane for acquiring the third interference fringe image formed by the reflected light from the reference plane on the back surface side and the reference plane on the main surface side of the parallel light by the main surface side light emitting means in a state removed from A shape measuring apparatus comprising: an interference fringe image acquiring unit; and a shape calculating unit for calculating a thickness distribution of the wafer based on the first to third interference fringe images.
[0007]
Here, the shape calculating means calculates the thickness distribution T (x, y) of the wafer based on the first interference fringe image and the distance between the main surface of the wafer and the reference plane on the main surface side. Obtained from La (x, y), the distance Lb (x, y) between the back surface of the wafer and the reference plane on the back surface side obtained based on the second interference fringe image, and the third interference fringe image. On the basis of the distance Lt (x, y) between the two reference planes to be calculated, the following equation can be used.
T (x, y) = Lt (x, y) −La (x, y) −Lb (x, y)
As a result, an interferometer suitable for each of the main surface (mirror surface) and the back surface (rough surface) of the wafer can be arranged to measure the shape of the main surface and back surface of the wafer, and the main surface obtained by the interferometer on the main surface side can be measured. Based on the distance between the two reference planes on the front side and the back side, the displacement (unevenness) is offset at each position of each of the two reference planes. Even when the flatness of each of the reference planes is not sufficient, it is possible to perform highly accurate wafer shape measurement without depending on this. Furthermore, on the main surface side where normal incident light is used, the distance between the reference plane on the main surface side and the wafer can be secured, so that the wafer is compared with the case where the reference plane using oblique incident light is measured close to both surfaces of the wafer. Can be easily inserted and removed between the two reference planes.
[0008]
And an absolute thickness measuring means for measuring the absolute thickness of at least one location of the wafer, wherein the shape calculating means is based on the absolute thickness measured by the absolute thickness measuring means. A distribution may be calculated.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments and examples of the present invention will be described with reference to the accompanying drawings so that the present invention can be understood. It should be noted that the following embodiments and examples are examples embodying the present invention, and do not limit the technical scope of the present invention.
FIG. 1 is a block diagram of a wafer shape measuring apparatus X according to the embodiment of the present invention. FIG. 2 is a block diagram between the reference plane and the wafer in the wafer shape measuring apparatus X according to the embodiment of the present invention. FIG. 3 is a diagram schematically showing an optical path between the reference planes, and FIG. 3 is a diagram schematically showing an optical path between the reference plane and the wafer by the conventional oblique incidence interferometer Z and between the two reference planes.
[0010]
Hereinafter, the wafer shape measuring apparatus X according to the embodiment of the present invention will be described with reference to FIG.
In the wafer shape measuring apparatus X, two optical measuring systems 10 and 20 are arranged to face each other on both the main surface 1a and the back surface 1b of the wafer 1 supported by the wafer support 30.
The optical measurement system 10 on the main surface 1a side is a Fizeau interferometer, and includes a main surface side light emitter 11 that emits measurement light 12, a half mirror 13 that transmits the measurement light 12, and the light after passing through the half mirror 13. A collimator lens 14 for making the measurement light 12 a parallel beam perpendicular to the main surface 1a, a reference plate 15 having a main surface side reference plane 15a that transmits a part of the parallel beam and reflects a part thereof, and the wafer 1 The measurement light reflected by the main surface 1a passes through the reference plate 15 and the collimator lens 14 and receives the light reflected by the half mirror 13 on the main surface side light receiver 16, such as a CCD camera, and the reference plate 15. A piezo actuator 17 that moves in a direction perpendicular to the surface of the wafer 1 is provided. Here, the reflected light from the main surface side reference plane 15a and the reflected light from the main surface 1a of the wafer 1 are optical paths corresponding to the distance between the main surface side reference plane 15a and the main surface 1a. Since there is a difference, this optical path difference is observed as interference fringes formed by both reflected lights in the main surface side light receiver 16.
[0011]
On the other hand, the optical measurement system 20 on the back surface 1b side is a grazing incidence interferometer, and a back-side light emitter 21 that emits measurement light 22 and the measurement light 22 are parallel beams oblique to the back surface 1b. The collimator lens 24, the triangular prism 25 having the back side reference plane 25 a that transmits a part of the parallel beam and reflects a part thereof, and the measurement light reflected by the back side 1 b of the wafer 1 is the back side reference plane 25 a ( A back light-receiving device 26 such as a CCD camera that receives light having passed through the triangular prism 25) and the collimator lens 24 ′, and a piezo actuator 27 that moves the triangular prism 25 in a direction substantially perpendicular to the surface of the wafer 1; I have. The reflected light on the back side reference plane 25a and the reflected light on the back side 1b of the wafer 1 have an optical path difference corresponding to the distance between the back side reference plane 25a and the back side 1b. This optical path difference is observed as interference fringes formed by both reflected lights in the back side light receiver 26. The triangular prism 25 is arranged so that the back side reference plane 25a is close to the back surface 1b of the wafer 1 (for example, at a distance of about 0.1 mm). As the measurement lights 12 and 22, for example, coherent light such as an HnNe laser (λ = 633 nm) or a semiconductor laser is used. The triangular prism 25 may be a reference plate such as a wedge prism used for Fizeau interference.
[0012]
Interference fringe images obtained by the main surface side light receiver 16 and the back surface side light receiver 26 are input and input to a calculator 31 (an example of the shape calculation means) such as a personal computer having image input means. Based on the interference fringe image, the surface shapes (height distribution) of the main surface 1a and the back surface 1b of the wafer 1 are calculated. Hereinafter, the interference fringe image formed by the reflected light of the main surface 1a of the wafer 1 obtained by the main surface side light receiver 16 and the main surface side reference plane 15a is obtained by the a, and the back surface light receiver 26. An interference fringe image formed by reflected light from the back surface 1b of the wafer 1 and the back-side reference plane 25a is defined as b.
[0013]
Furthermore, in addition to the two interference fringe images a and b, the computing unit 31 has an interference fringe image c formed by reflected light from the main surface side reference plane 15a and the back surface side reference plane 25a. It is captured. The interference fringe image c is obtained when the wafer 1 is removed from between the two optical measurement systems 10 and 20 on the main surface side and the back surface side of the measurement light from the light emitter 11 on the main surface 1a side. Observation is performed by receiving the reflected light on each of the reference planes 15 a and 25 a by the main surface side light receiver 16. The calculator 31 calculates the distribution of the distance between the main surface side reference plane 15a and the back side reference plane 25b based on the input interference fringe image c. Here, the main surface side light emitter 11 and the collimator lens 14 are examples of the main surface side light emitting means, the back surface side light emitter 21 and the collimator lens 24 are examples of the back surface side light emitting means, and the main surface side light receiving. The device 16 is an example of the main surface side interference fringe image acquisition unit and the reference plane interference fringe image acquisition unit, and the back surface side light receiver 26 is an example of the back surface side interference fringe image acquisition unit.
[0014]
Next, the optical path of the parallel light when observing each of the interference fringe images a to c will be described with reference to FIG.
At the time of observation of the interference fringe image a (main surface), as shown in FIG. 2A, the light traveling from the predetermined incident position Q of the main surface side reference plane 15a toward the main surface 1a of the wafer 1 is Since the light is irradiated in a direction perpendicular to the main surface 1a, the incident position Q is seen from a direction perpendicular to the surface of the wafer 1 (viewed from the upper side or the lower side in FIG. 2A). After being reflected on the main surface 1a of the wafer 1 at the same position M ′, it returns to the incident position Q on the main surface side reference plane 15a. Thus, since the light emitted from the emission position Q returns to the original position, it is not affected by the flatness (unevenness) of the main surface side reference plane 15a.
On the other hand, at the time of observation of the interference fringe image b (back surface), as shown in FIG. 2A, the light traveling from the predetermined incident position L of the back surface side reference plane 25a toward the back surface 1b of the wafer 1 is Since the back surface 1b is irradiated obliquely, the back surface side is further reflected at a position N after being reflected at a position M shifted from the incident position L when viewed from the direction perpendicular to the surface of the wafer 1. The reference plane 25a is reached. Here, as described above, since the back side reference plane 25a is close to the back side 1b of the wafer 1, the deviation between the positions L and N is sufficiently small so as not to cause a problem in measurement accuracy. As described above, in general, the maximum flatness error is likely to occur in the reference plane with a spatial period (interval) of about 10 mm. For example, the back side reference plane 25a is set to 0. If they are close to about 1 mm, the deviation between the positions L and N is about 1 mm, and the influence of the flatness error of the reference plane can be suppressed to a negligible level.
At the time of observation of the interference fringe image c (between two reference planes), as shown in FIG. 2B, the main surface 1a of the wafer 1 from the incident position Q of the main surface side reference plane 15a. When viewed from a direction perpendicular to the surface of the wafer 1 (viewed from the upper side or the lower side in FIG. 2B), the light traveling toward the surface 1 is at the same position P as the incident position Q on the back side reference plane 25a. After the reflection, the light returns to the incident position Q on the main surface side reference plane 15a. Thus, since the light emitted from the emission position Q returns to the original position, it is not affected by the flatness (unevenness) of the main surface side reference plane 15a.
Accordingly, when the position M ′ on the main surface 1a side of the wafer 1 and the position M on the back surface 1b side are considered to be front-back relation positions, the positions L, P, N on the back-side reference plane 25a are considered. Each shift is suppressed to a sufficiently small extent that the influence of the flatness error of the back side reference plane 25a can be ignored.
Further, the back surface side reference plane 25a and the wafer 1 are close to each other on the back surface 1b side of the wafer 1, but on the main surface 1a side of the wafer 1, the main surface side reference plane 15a and the wafer 1 are disposed. Therefore, handling for inserting and removing the wafer 1 between the two reference planes 15a and 25a is facilitated.
[0015]
Using the interference fringe images a to c obtained by the above measurement, the calculator 31 performs the following calculation.
First, based on the interference fringe image a, the distance La (x, y) between the main surface side reference plane 15a and the main surface 1a and the back side reference plane 25a based on the interference fringe image b A distance Lb (x, y) between the back surface 1b and a distance Lt (x, y) between the main surface side reference plane 15a and the back surface side reference plane 25a is obtained based on the interference fringe image c. It is done.
The interference fringe image a obtained by the optical measurement system 10 on the main surface 1a side is transferred to the reference plate 15 (that is, the main surface side reference plane 15a) by the piazo actuator 17, for example, 0 (predetermined reference position), By moving to each position of λ / 8, 2λ / 8, 3λ / 8 (λ is the wavelength of the measurement light), luminance data I ₀ (x, y), I ₉₀ ( x, y), I ₁₈₀ (x, y), I ₂₇₀ (x, y) are taken into the arithmetic unit 31. Based on the luminance data, the phase φa (x, y) of the interference fringe image a is obtained by the following equation (2).
φa (x, y) = arctan {(I ₀ −I ₁₈₀ ) / (I ₉₀ / I ₂₇₀ )} (2)
When φa (x, y) obtained in this way is subjected to a predetermined unwrap process (the phase after processing is assumed to be φa ′ (x, y)), the distance La (x, y) is It is calculated | required by (3) Formula.
La (x, y) = (φa ′ (x, y) / 2π) × (λ / 2) (3)
The distance Lc (x, y) is also obtained in the same manner as the distance La (x, y).
Similarly, the interference fringe image b by the optical measurement system 20 on the back surface 1 b side is also moved by the piazo actuator 27 so that the triangular prism 25 (that is, the back surface side reference plane 25 a) is moved to the wafer 1 of the measurement light 22. Are moved to positions that take into account the incident angle θ on the back surface 1b, for example, 0 (predetermined reference position), λ / 8 / conθ, 2λ / 8 / cosθ, and 3λ / 8 / cosθ. The calculation unit 31 takes in the luminance data of the interference fringe image. Further, from the luminance data, the phase φb ′ (x, y) of the interference fringe image b after the unwrap processing is obtained in the same manner as obtaining φa ′ (x, y). Then, the distance Lb (x, y) is obtained by the following equation (4).
Lb (x, y) = (φb ′ (x, y) / 2π) × (λ / 2 × 1 / cos θ)
... (4)
Then, based on the three distances La (x, y), Lb (x, y), and Lt (x, y), the thickness distribution T (x, y) of the wafer 1 is expressed by the equation (1). Is required.
The thickness distribution T (x, y) of the wafer 1 thus obtained is the position (x, y) of each of the two reference planes 15a and 25a as shown in paragraph 0014 of the publication 1. Since the displacement (unevenness) is canceled every time, the value does not depend on the flatness (planar shape) of the main surface side reference plane 15a and the back surface side reference plane 25a. That is, according to the wafer shape measuring apparatus X, it is possible to measure the wafer thickness distribution with high accuracy without depending on the flatness of the reference plane. As a result, even if the main surface 1a of the wafer 1 is a mirror surface and the back surface 1b is a rough surface, the shape (of the wafer 1) can be accurately determined without being affected by the flatness error of the two reference planes 15a and 25a. Thickness distribution). Further, since the two reference planes 15a and 25a are not necessarily ideal planes, the cost of the apparatus can be reduced.
[0016]
【Example】
The thickness distribution T (x, y) of the wafer 1 obtained by the wafer shape measuring apparatus X according to the portable embodiment includes an unknown offset amount. Therefore, if a thickness measuring device (corresponding to the absolute thickness measuring means) for actually measuring the thickness of the wafer 1 at an arbitrary position is provided, and the obtained thickness measured value t is input to the calculator 31, It is possible to obtain the absolute thickness distribution of the wafer 1 based on the measured thickness t. That is, in the thickness distribution T (x, y), the thickness distribution T (x, y) is set so that the value at the measurement point of the measured thickness value t coincides with the measured thickness value t. It is sufficient to translate each value of.
In the embodiment, the two reference planes 15a and 25a are moved by the piezoelectric actuators 17 and 27 in four steps for each 90 ° phase, but other phase shift methods (for example, wavelength sweeping) are used. Etc.). Further, the wafer 1 may be moved by an actuator instead of moving the two reference planes 15a and 25a.
In the embodiment, a Fizeau interferometer is used as the optical measurement system 10 on the main surface 1a side of the wafer 1. However, other interferometers that irradiate light substantially perpendicularly on the surface of the wafer 1 (see FIG. For example, a Michelson interferometer may be used.
[0017]
【The invention's effect】
As described above, according to the present invention, an interferometer suitable for each of the main surface (mirror surface) and the back surface (rough surface) of the wafer is arranged to measure the shape of the main surface and back surface of the wafer, and Based on the distance between the two reference planes on the main surface side and the back surface obtained by the interferometer on the surface side, the displacement (unevenness) is canceled for each position of each of the two reference planes. Even if the flatness of each of the two reference planes is not sufficient with respect to the measurement accuracy of the wafer, the wafer shape can be measured with high accuracy without depending on this.
In addition, since the distance between the main surface side interferometer where normal incidence light is used and the wafer can be secured, the wafer is moved between two reference planes as compared with the case where the oblique incidence interferometer is measured close to both surfaces of the wafer. Easy handling for insertion and removal.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a wafer shape measuring apparatus X according to an embodiment of the present invention.
FIG. 2 is a diagram schematically illustrating an optical path between a reference plane and a wafer and between two reference planes in the wafer shape measuring apparatus X according to the embodiment of the present invention.
FIG. 3 is a diagram schematically showing an optical path between a reference plane and a wafer and between two reference planes by a conventional oblique incidence interferometer Z.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Wafer 10 ... Optical measurement system (Fizeau interferometer) on main surface side
11 ... Main surface side light emitter 12 ... Measuring light (main surface side)
DESCRIPTION OF SYMBOLS 13 ... Half mirror 14, 24, 24 '... Collimator lens 15 ... Reference board 15a ... Main surface side reference plane 16 ... Main surface side light receiver 17, 27 ... Piezo actuator 20 ... Optical measurement system on back side (oblique incidence interferometer) )
21 ... Back side light emitter 22 ... Measuring light (back side)
25 ... Triangular prism 25a ... Back side reference plane 26 ... Back side photo detector 30 ... Wafer support 31 ... Calculator

Claims

Means for constituting two reference planes, a reference plane arranged on the main surface side of the wafer and a reference plane arranged close to the back side of the wafer;
A main surface side light emitting means for irradiating parallel light substantially perpendicularly to the main surface of the wafer via a reference plane on the main surface side;
Main surface side interference fringe image acquisition means for acquiring a first interference fringe image formed by reflected light from the main surface of the wafer and a reference plane on the main surface side of the parallel light by the main surface side light emitting means; ,
Backside light emitting means for irradiating the backside of the wafer obliquely through the backside reference plane;
Backside interference fringe image obtaining means for obtaining a second interference fringe image formed by reflected light from the backside of the wafer and the reference plane on the backside of the parallel light by the backside light emitting means;
In a state where the wafer is removed from between the two reference planes, the parallel light by the main surface side light emitting means is formed by reflected light from the reference plane on the back surface side and the reference plane on the main surface side. Reference plane interference fringe image acquisition means for acquiring three interference fringe images;
Shape calculating means for calculating a thickness distribution of the wafer based on the first to third interference fringe images;
A shape measuring apparatus comprising:

The shape calculation means calculates the thickness distribution T (x, y) of the wafer,
The distance La (x, y) between the main surface of the wafer and the reference plane on the main surface side obtained based on the first interference fringe image, and the back surface of the wafer obtained based on the second interference fringe image. And a distance Lb (x, y) between the reference plane on the back surface side and the distance Lt (x, y) between the two reference planes obtained from the third interference fringe image. The shape measuring apparatus according to claim 1.
T (x, y) = Lt (x, y) −La (x, y) −Lb (x, y)

Comprising an absolute thickness measuring means for measuring an absolute thickness of at least one location of the wafer;
The shape measuring apparatus according to claim 1, wherein the shape calculating unit calculates an absolute thickness distribution of the wafer based on the absolute thickness obtained by the absolute thickness measuring unit.