JPH06147827A - Positional shift detecting method - Google Patents

Abstract

PURPOSE:To enhance resolving power in detection while enhancing the intensity of diffracted light to be detected by projecting monochromatic light of double wavelength having slightly different frequencies to first and second objects from two directions symmetric to the optical axis. CONSTITUTION:Monochromatic light of double wavelength having slightly different frequencies is projected to diffraction gratings 36a, 36b of first and second objects from two directions symmetrical to the optical axis at an angle of thetai/2, where thetai is the angle between the optical axis and + or - 1st order direction of diffracted light being taken out upon impingement in the direction of the optical axis. A -1st order diffracted light, i.e., a light diffracted reversely to the direction of impingement, and a zero order diffracted light being diffracted in same direction are taken out from both objects M, W and subjected to interference with each other thus producing first beat signals on the opposite sides of the optical axis from the objects M, W. On the other hand, -1st order diffracted lights taken out on the opposite sides of the optical axis are subjected to interference with each other to produce a second beat signal. Displacement amounts of the objects M, W are then determined by measuring phase differences based on the first and second beat signals.

Description

【発明の詳細な説明】Detailed Description of the Invention

【０００１】[0001]

【産業上の利用分野】この発明は、半導体超微細加工装
置（ＳＯＲアライナ・ステッパ、液晶ステッパ等のプロ
キシミティ露光装置）や感光基板に露光されたパターン
の重ね合せ精度を測定するレジストレーション超精密測
定等において光ヘテロダイン干渉光を利用する位置ずれ
検出方法に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor ultra-fine processing apparatus (proximity exposure apparatus such as SOR aligner / stepper, liquid crystal stepper) and a registration ultra-precision registration measuring superposition accuracy of a pattern exposed on a photosensitive substrate. The present invention relates to a position shift detection method using optical heterodyne interference light in measurement or the like.

【０００２】[0002]

【従来の技術】シンクロトロン放射光リソグラフィ用ア
ライナやフォトステッパ等の超精密位置合せにあっては
位置合せすべき２つの物体の位置ずれ検出を高精度に行
なう必要があり、そのために例えば特開昭６２−２６１
００３号や特開昭６４−８９３２３号等では、光ヘテロ
ダイン干渉光を利用した位置ずれ検出方法が提案されて
いる。2. Description of the Related Art In ultra-precision alignment of aligners for synchrotron radiation lithography, photosteppers and the like, it is necessary to detect the positional deviation between two objects to be aligned with high accuracy. 62-261
No. 003 and Japanese Patent Laid-Open No. 64-89323 propose a position shift detection method using optical heterodyne interference light.

【０００３】これらはいずれも２つの物体の各回折格子
から取り出された回折光を干渉させ、これらの両物体の
夫々において生成されたビート信号の位相差を検出する
ことで両物体の位置ずれ量を測るものであり、位置ずれ
検出分解能力が飛躍的に向上するとして期待されてい
る。Both of these interfere with the diffracted light extracted from the diffraction gratings of the two objects, and detect the phase difference of the beat signals generated in each of these two objects to detect the amount of positional deviation between the two objects. It is expected to dramatically improve the ability to detect and resolve misalignment.

【０００４】[0004]

【発明が解決しようとする課題】ところがこれらの方法
では、位相ずれが１周期以上になると位置合せにおける
ピッチ跳び・ずれ等が起こり不正確なものとなるため、
その位置ずれが予めある範囲内に納まっていなければな
らないという検出範囲の限定がある。特に回折格子の格
子ピッチを小さくすれば検出分解能は向上するが、検出
範囲が狭くなり、反対に上記ピッチを大きくすればその
逆の関係となる。However, in these methods, when the phase shift is more than one cycle, pitch jumping / shifting in alignment occurs, which is inaccurate.
There is a limit to the detection range in which the displacement must be within a certain range in advance. In particular, if the grating pitch of the diffraction grating is reduced, the detection resolution is improved, but the detection range is narrowed, and conversely, if the pitch is increased, the opposite relationship occurs.

【０００５】このため本出願人により相反する関係にあ
る検出分解能の向上と検出範囲の拡大を達成するため、
２つの物体に夫々格子ピッチの異なる回折格子を複数並
べて検出を行なう構成の提案を行なった。Therefore, in order to achieve the improvement of the detection resolution and the expansion of the detection range, which are contradictory to each other by the present applicant,
We proposed a structure in which a plurality of diffraction gratings with different grating pitches are arranged side by side on two objects for detection.

【０００６】又仮りに光軸方向より光を入射させた場合
に得られる回折光の±１次方向、±２次方向、±３次方
向……（該光軸を中心に対称的な２方向）から反対に光
を入射させた場合（その入射方向をこの回折次数を借り
て入射次数とここでは表現するとすると）、入射次数が
０に近い程、光路差変化量が少なくなるので前記検出範
囲が広がり、且つ低次数回折光の干渉なので回折光強度
も強いものが得られる一方で前記検出分解能は低下する
ことになる。反対にその入射次数が０から離れた値にな
る程、上記の場合とは全く逆になる。Further, if the light is incident from the direction of the optical axis, the ± 1st order, ± 2nd order, ± 3rd order of the diffracted light obtained (two directions symmetrical about the optical axis) ) From the opposite direction (when the incident direction is referred to as the incident order here by borrowing this diffraction order), the closer the incident order is to 0, the smaller the change amount of the optical path difference becomes. Is wide and the interference of low-order diffracted light is strong, so that a strong diffracted light intensity can be obtained, but the detection resolution is lowered. On the contrary, the farther the incident order is from 0, the opposite of the above case.

【０００７】そこで本出願人等は入射次数が複数となる
方式を採用した構成の提案を行ない、且つその次数の絶
対値が大きくなると、回折光強度が低下する欠点を補う
ため、回折格子に特殊形状のもの（ブレーズド格子）を
用いることで特定次数の回折光に０次回折光を重ねて回
折光強度を高める別の提案も行なった。Therefore, the applicants have proposed a structure adopting a system in which the incident orders are plural, and when the absolute value of the orders becomes large, the defect that the diffracted light intensity is lowered is compensated, so that the diffraction grating is specially designed. Another proposal was made to increase the intensity of diffracted light by superposing the 0th-order diffracted light on the diffracted light of a specific order by using a shape (blaze grating).

【０００８】しかし、格子ピッチの異なる複数の回折格
子を並べる最初の方法では、そのような回折格子の形成
に手間がかかり実用的ではないこと、又入射次数が複数
となる２つめの提案では、光学系が複雑となり過ぎるこ
と、更に特殊形状の回折格子を用いる３つめの提案は光
学系の複雑化以外に、その様な形状の回折格子の形成が
非常に難しく実用的ではないこと等の問題点が指摘され
た。However, the first method of arranging a plurality of diffraction gratings having different grating pitches is not practical because it takes time to form such a diffraction grating, and the second proposal in which a plurality of incident orders are provided is as follows. The problem is that the optical system becomes too complicated, and the third proposal using a specially shaped diffraction grating is very difficult to form a diffraction grating with such a shape, and is not practical, in addition to the complicated optical system. A point was pointed out.

【０００９】本発明は従来技術の以上の様な問題に鑑み
創案されたもので、光学系の構成及びその調整が複雑と
ならず、広い検出範囲を持ちながら検出分解能も高く、
更に検出すべき回折光強度も高いものが得られる位置ず
れ検出方法を提供し、併せて光軸の傾きがある場合にそ
の傾きの補正を行なう構成についても提案せんとするも
のである。The present invention was devised in view of the above problems of the prior art, and does not complicate the configuration and adjustment of the optical system, has a wide detection range and high detection resolution,
Further, the present invention proposes a position shift detection method that can obtain a high intensity of diffracted light to be detected and, at the same time, proposes a configuration for correcting the tilt of the optical axis when there is a tilt.

【００１０】[0010]

【課題を解決するための手段】ここで本発明の構成を説
明する前に、本願における回折格子から得られる回折光
の回折次数につき予め定義しておく。Before describing the structure of the present invention, the diffraction order of the diffracted light obtained from the diffraction grating of the present application will be defined in advance.

【００１１】図１及び図２は反射回折格子の入射角・回
折角の符合の状態を示している。まず周波数ｆの単色光
が格子ピッチＰの反射回折格子36に対し、光軸から入射
角θiの傾きを持って入射した場合、正反射となる回折
次数m、n＝０の回折光を中心にそれより光軸側に回折す
るものはm、n＝−１、−２、−３……というようにマイ
ナス次数又その反対側に回折するものはm、n＝＋１、＋
２、＋３……というようにプラス次数（これらの回折次
数に対応する回折角をθm、θnとする）になる。この時
の入射角θiと回折角θm、θnとの関係は、回折格子の
基礎公式により、次式数１及び数２の様になる。FIG. 1 and FIG. 2 show the states of coincidence of the incident angle and the diffraction angle of the reflection diffraction grating. First, when monochromatic light having a frequency f is incident on the reflection diffraction grating 36 having a grating pitch P with an inclination of an incident angle θi from the optical axis, the diffracted light with a diffraction order of m and n = 0 is specularly reflected. Those diffracting to the optical axis side are m, n = -1, -2, -3, and so on. Those diffracting to the minus order or the opposite side are m, n = +1 and +.
2, +3, and so on, plus order (diffraction angles corresponding to these diffraction orders are denoted by θm and θn). At this time, the relationship between the incident angle θi and the diffraction angles θm and θn is expressed by the following equations 1 and 2 according to the basic formula of the diffraction grating.

【００１２】[0012]

【数１】 [Equation 1]

【００１３】[0013]

【数２】 [Equation 2]

【００１４】以上の定義を基に本発明の位置ずれ検出方
法につき説明すると、図３に示される様に、光軸方向か
ら入射させた場合に取り出される回折光の±１次方向と
該光軸との間で成す角度の半分の角度θi／２、θi／２
であって該光軸に対し対称的な２方向から夫々逆に第１
の物体Ｍ及び第２の物体Ｗの各回折格子36a、36b（これ
らの格子ピッチはＰとする）に対して周波数のわずかに
異なる（即ち、ｆ₁、ｆ₂の）２波長の単色光を入射さ
せ、入射方向逆向きに回折する−１次回折光と、光軸反
対側から入射して正反射し、この−１次回折光と同一方
向に回折する０次回折光とを光軸の両側で且つ両物体
Ｍ、Ｗの夫々において取り出した上で、同一方向で取り
出された周波数の異なる−１次と０次の回折光を干渉せ
しめて第１ビート信号を光軸の両側において両物体Ｍ、
Ｗの夫々につき生成せしめると共に、光軸の両側で取り
出された周波数の異なる−１次の回折光同士を干渉せし
めて第２ビート信号を両物体Ｍ、Ｗの夫々で生成せし
め、更に両物体Ｍ、Ｗの各第１ビート信号に基づくその
位相差を測定すると共に、両物体Ｍ、Ｗの各第２ビート
信号に基づくその位相差を測定することで、両物体Ｍ、
Ｗの変位量を求めることを基本的特徴としている。The position deviation detecting method of the present invention will be described based on the above definitions. As shown in FIG. 3, the ± first-order directions of the diffracted light extracted when the light is incident from the optical axis direction and the optical axis. Half angle θi / 2, θi / 2
And from the two directions symmetrical with respect to the optical axis, respectively,
Of monochromatic light of two wavelengths (that is, f ₁ and f ₂ ) having slightly different frequencies with respect to the diffraction gratings 36a and 36b of the object M and the second object W (these grating pitches are P). The -1st-order diffracted light that is made incident and diffracts in the opposite direction of incidence and the 0th-order diffracted light that is specularly reflected by being incident from the opposite side of the optical axis are diffracted in the same direction as the -1st-order diffracted light on both sides of the optical axis. After extracting the light from each of the objects M and W, the -1st order and 0th order diffracted lights of different frequencies extracted in the same direction are interfered with each other to obtain the first beat signal on both sides of the optical axis.
W is generated for each W, and the -1st order diffracted lights with different frequencies extracted on both sides of the optical axis are interfered with each other to generate the second beat signal by each of the objects M and W. , W by measuring the phase difference based on each first beat signal and by measuring the phase difference based on each second beat signal on both objects M, W
The basic feature is to obtain the displacement amount of W.

【００１５】第２発明は上記発明の入射角θi／2の単色
光の入射によって取り出される−１次と０次の回折光を
干渉せしめ、それにより生成された両物体Ｍ、Ｗの各第
１ビート信号に基づくその位相差を測定する他、図４に
示される様に光軸両側から入射角θiの単色光の入射に
よって光軸方向に回折した−１次回折光を両物体Ｍ、Ｗ
の夫々において取り出した上でこれらの回折光を夫々干
渉せしめて第２ビート信号を両物体Ｍ、Ｗの夫々で生成
せしめ、更に両物体Ｍ、Ｗの各第２ビート信号に基づく
その位相差を測定し、これらの各位相差の測定で両物体
Ｍ、Ｗの変位量を夫々求めることを基本的特徴としてい
る。According to a second aspect of the present invention, the -1st order and 0th order diffracted lights extracted by the incidence of the monochromatic light having the incident angle θi / 2 of the above-mentioned invention are interfered with each other, and the first objects of both objects M and W generated thereby interfere with each other. In addition to measuring the phase difference based on the beat signal, the -1st order diffracted light diffracted in the optical axis direction by the incidence of monochromatic light having the incident angle θi from both sides of the optical axis as shown in FIG.
Of each of the objects M and W, the diffracted light beams of the two objects M and W are caused to interfere with each other, and the second beat signal is generated by each of the objects M and W. The basic feature is that the measurement is performed and the displacement amounts of the two objects M and W are obtained by measuring the respective phase differences.

【００１６】更に第３発明は上記両発明の夫々の構成に
おいて、光軸が傾いている場合に、その光軸の傾きの補
正を行なう構成に係り、その具体的構成は、同一方向で
取り出された周波数の異なる−１次と０次の回折光の位
相変化量と偏光強度を前記光軸の両側で比較し、両側の
回折光の位相変化量と偏光強度が等しくなるように該光
軸の傾きを補正することを特徴としている。Further, the third invention relates to a configuration for correcting the inclination of the optical axis when the optical axis is inclined in the respective configurations of the both inventions, and the specific configuration is taken out in the same direction. The phase change amount and the polarization intensity of the diffracted light of the -1st order and the 0th order having different frequencies are compared on both sides of the optical axis, and the phase change amount and the polarization intensity of the diffracted light on both sides are equalized. The feature is that the inclination is corrected.

【００１７】上記第１発明法のように光軸両側から入射
角θi／２の角度で単色光を入射させると、図３に示さ
れる様にその−１次回折光は入射方向逆向きに回折する
ことになる。又０次回折光はその光軸を挟んで反対側に
正反射する反射光であり、その回折方向はちょうど反対
側で入射された単色光の入射方向逆向き、即ち該入射光
の−１次回折光と同一方向になる。従って回折光は光軸
を挟んで対称的な２方向（これは前述の様に入射方向逆
向きとなる）に回折され、その一方でｆ₁周波数の−１
次回折光とｆ₂周波数の０次回折光が、又他方でｆ₂周波
数の−１次回折光とｆ₁周波数の０次回折光が夫々一緒
に取り出されてくる（これらの偏光面は円偏光で夫々の
回折光を取出し後に重ね合せることで干渉させる構成の
ものが良い）。両物体Ｍ、Ｗ共、各光軸に対称的な２方
向に夫々取り出されてくる周波数ｆ₁とｆ₂のこれらの回
折光を干渉せしめて第１ビート信号を生成せしめ（前述
の様に回折後取り出してから干渉させるタイプのものが
良い）、該ビート信号を検知すれば、両物体Ｍ、Ｗ間の
位置ずれは、図５(b)に示される様な位相差となって表
われ、従ってその検出範囲は±Ｐ／２とかなり広くな
る。更に、光軸の両側において夫々取り出された周波数
ｆ₁の−１次回折光と周波数ｆ₂の−１次回折光とを取り
出してこれらを干渉せしめ、両物体Ｍ、Ｗの夫々で第２
ビート信号の生成を行なって、これらのビート信号を検
知し、その位相差を測定する。図５(a)はその時の位相
差を表わしており、その検出範囲は±Ｐ／４（即ちＰ／
２）となって上述の場合の半分となるが検出分解能は逆
に２倍となる。従って第１発明法では±Ｐ／２（即ち
Ｐ）の広い検出範囲と、仮りに位相計分解能の精度を１
°とすると、（Ｐ／２）・（１°／360）という高い検
出分解能が得られることになる。When monochromatic light is made to enter from both sides of the optical axis at an incident angle θi / 2 as in the first invention method, the -1st order diffracted light is diffracted in the opposite direction to the incident direction as shown in FIG. It will be. The 0th-order diffracted light is a reflected light that is specularly reflected to the opposite side across the optical axis, and the diffraction direction is opposite to the incident direction of the monochromatic light that has just entered the opposite side, that is, the -1st-order diffracted light of the incident light. The same direction as. Therefore, the diffracted light is diffracted in two symmetrical directions with respect to the optical axis (this is opposite to the incident direction as described above), while the f ₁ frequency is −1.
The _second-order diffracted light and the 0th-order diffracted light of the f ₂ frequency are extracted together, and the −1st-order diffracted light of the f ₂ frequency and the 0th-order diffracted light of the f ₁ frequency are extracted together (these polarization planes are circularly polarized light, respectively). It is good to have a structure in which diffracted light is extracted and then overlapped to cause interference. Both the objects M and W interfere with these diffracted lights of frequencies f ₁ and f _{2 which} are respectively extracted in two directions symmetrical to each optical axis to generate the first beat signal (diffraction as described above). If the beat signal is detected, the positional deviation between the two objects M and W will appear as a phase difference as shown in FIG. 5 (b). Therefore, the detection range is considerably wide as ± P / 2. Further, the -1st-order diffracted light of the frequency f _{1 and} the -1st-order diffracted light of the frequency f ₂ extracted on both sides of the optical axis are extracted and interfered with each other.
A beat signal is generated, these beat signals are detected, and the phase difference between them is measured. FIG. 5A shows the phase difference at that time, and the detection range is ± P / 4 (that is, P /
2), which is half of the above case, but the detection resolution is doubled. Therefore, in the first invention method, a wide detection range of ± P / 2 (that is, P) and a phase meter resolution accuracy of 1 are assumed.
If the angle is °, a high detection resolution of (P / 2) · (1 ° / 360) will be obtained.

【００１８】又第２発明法は前述の様に光軸に対しその
両側で入射角θi／２の単色光の入射を行なって得られ
た−１次回折光と０次回折光を干渉させて（これらの偏
光面は円偏光で回折した段階で重なって干渉状態になっ
ているものもあれば、回折後取り出し中に重ね合わせる
ことにより干渉させる構成のものもある）生成された両
物体Ｍ、Ｗの夫々の第１ビート信号の位相差の測定を行
ない、±Ｐ／２の広い検出範囲で両物体Ｍ、Ｗのずれ量
を得ると共に、前記光軸に対しその両側で入射角θiの
単色光の入射を行なって光軸方向に回折される周波数ｆ
₁とｆ₂の−１次回折光を取り出して干渉させ（これらも
回折した段階で干渉状態になっているものもあれば、回
折後取り出し中に干渉状態とするものもある）、両物体
Ｍ、Ｗの夫々で生成された第２ビート信号の位相差の測
定を行なうと、図５(a)と同様な位相差波形が検出さ
れ、高い検出分解能が得られることになる。In the second invention method, the -1st-order diffracted light and the 0th-order diffracted light obtained by the incidence of the monochromatic light having the incident angle θi / 2 on both sides of the optical axis as described above are caused to interfere with each other. The polarization planes of are overlapped with each other when they are diffracted by circularly polarized light and some of them are in an interference state, and some of them are configured to cause interference by overlapping during extraction after diffraction. The phase difference of each first beat signal is measured to obtain the deviation amount of both objects M and W in a wide detection range of ± P / 2, and the monochromatic light having an incident angle θi on both sides of the optical axis is detected. Frequency f that is incident and diffracted in the optical axis direction
The −1st order diffracted lights of ₁ and f ₂ are extracted and interfered (some of them are in an interference state at the stage of diffracting, and some of them are in an interference state during extraction after diffraction), and both objects M, When the phase difference of the second beat signal generated in each of W is measured, the phase difference waveform similar to that in FIG. 5A is detected, and high detection resolution is obtained.

【００１９】更にθi／２の入射角で周波数ｆ₁とｆ₂の
各単色光を入射させた場合、その光軸の片側で取り出さ
れてくる周波数ｆ₁の−１次回折光及び周波数ｆ₂の０次
回折光の組と、該光軸の他方の側に取り出されてくる周
波数ｆ₁の０次回折光及び周波数ｆ₂の−１次回折光の組
とで、これらの各組の回折光の位相変化量と偏光強度を
測定すると、本来回折格子に対する光軸のずれがない限
りこれらの位相変化量と偏光強度は一致するはずである
が、光軸のずれがある場合は、該光軸の両側で得られる
回折光の位相変化量と偏光強度は異なる値となる。そこ
で第３発明法は第１発明法又は第２発明法による位置ず
れ検出を行なう前処理として、以上の回折光の位相変化
量と偏光強度を測定し、光軸の両側でこれが等しくなる
まで該光軸の補正（具体的には光学系の調整）を行なう
こととした。Furthermore when is incident each monochromatic light of a frequency f ₁ and f ₂ at an incident angle of .theta.i / 2, the frequency f ₁ coming taken out on one side of the optical axis -1 order diffracted light and the frequency f ₂ The phase change of the diffracted light of each of the sets is composed of a set of the 0th-order diffracted light and a set of the 0th-order diffracted light of the frequency f ₁ and the −1st-order diffracted light of the frequency f ₂ which are extracted to the other side of the optical axis. When the amount and the polarization intensity are measured, the phase change amount and the polarization intensity should be the same unless there is a deviation of the optical axis with respect to the diffraction grating, but if there is a deviation of the optical axis, both sides of the optical axis The obtained phase change amount of the diffracted light and the polarization intensity have different values. Therefore, the third invention method measures the amount of phase change and the polarization intensity of the diffracted light as described above as a pre-process for detecting the positional deviation according to the first invention method or the second invention method. It was decided to correct the optical axis (specifically, adjust the optical system).

【００２０】[0020]

【実施例】以下本発明法の具体的実施例につき詳述す
る。EXAMPLES Hereinafter, specific examples of the method of the present invention will be described in detail.

【００２１】図６及び図８はマスクＭとウェハＷの位置
ずれ検出を行なう本願第２発明法の実施に使用される光
学系装置構成の一例を示す斜視図と、該光学系光路詳細
図である。FIGS. 6 and 8 are a perspective view showing an example of the configuration of an optical system device used for carrying out the second invention method of the present application for detecting the positional deviation between the mask M and the wafer W, and a detailed view of the optical path of the optical system. is there.

【００２２】図６において、まず２波長直交偏光レーザ
光源12より偏光面が直交し、且つ周波数がわずかに異な
る（ｆ₁、ｆ₂）単色レーザ光（即ち、周波数ｆ₁成分に
ついては→で表わすＰ偏光、又周波数ｆ₂成分について
は↑で表わすＳ偏光）を発生させる。10は該光源12のコ
ントローラであり、電気的な処理を施して第１ＲＥＦ11
aから｜ｆ₁−ｆ₂｜の周波数の参照ビート信号が出力さ
れるようになる。尚、該光源12については音響光学素子
（ＡＯＭ）２つからなる周波数シフタ等で２周波数のも
のを得る構成に置き換えてもよい。In FIG. 6, first, monochromatic laser light whose polarization planes are orthogonal to each other from the two-wavelength orthogonal polarization laser light source 12 and frequencies are slightly different (f ₁ , f ₂ ) (that is, the frequency f ₁ component is represented by →). P-polarized light, or S-polarized light represented by ↑ for the frequency f ₂ component) is generated. Reference numeral 10 is a controller for the light source 12, which is electrically processed to produce a first REF 11
The reference beat signal having the frequency of | f ₁ −f ₂ | is output from a. The light source 12 may be replaced with a structure in which a frequency shifter including two acousto-optic elements (AOM) is used to obtain a light source having two frequencies.

【００２３】この光源12から射出されたアライメント光
は通常３〜４度程度レーザ射出口で楕円偏光になってお
り、これをλ／４板13という位相板（回転補正光学部
品）によって２周波成分の直交状態をより正しい姿勢に
直す。そしてアライメント光のビームは偏光ビームスプ
リッタ（ＰＢＳ）14に至り、そこからＳ偏光（ｆ₂周波
数）成分がレンズ15に至る。このレンズ15と後述するレ
ンズ19はビームエキスパンダになっていて所望のビーム
径に絞り込むリレーレンズである。該レンズ15より出た
ｆ₂周波数のＳ偏光成分はミラー16により光路えを変
え、その後λ／２板17に対し、結晶軸と45°の方向から
入射させ、周波数ｆ₂のＳ偏光をＰ偏光にして、前記ｆ₁
周波数のＰ偏光と同じ偏光方向に変える。Ｐ偏光になっ
たｆ₂周波数のアライメントビームは図上Ｚ方向、Ｘ方
向に移動が可能であり、あおりやθ回転も可能なミラー
18により光軸方向を変え、レンズ19により無偏光ビーム
スプリッタ（ＮＢＳ）22へ所望の角度で入射させる。The alignment light emitted from the light source 12 is normally elliptically polarized at a laser emission port of about 3 to 4 degrees, and this is converted into a two-frequency component by a phase plate (rotation correction optical component) called a λ / 4 plate 13. Correct the orthogonal state of to a more correct posture. Then, the beam of alignment light reaches the polarization beam splitter (PBS) 14, and the S-polarized (f ₂ frequency) component reaches the lens 15 from there. The lens 15 and a lens 19 described later are beam expanders and are relay lenses that narrow down to a desired beam diameter. The optical path of the S ₂ polarized light component of f ₂ frequency emitted from the lens 15 is changed by a mirror 16 and thereafter incident on a λ / 2 plate 17 from the direction of 45 ° with respect to the crystal axis, and the S polarized light of frequency f ₂ is P The polarized light is f ₁
Change to the same polarization direction as the P polarization of the frequency. The P-polarized f ₂ frequency alignment beam can be moved in the Z and X directions in the figure, and can be tilted or rotated by θ.
The optical axis direction is changed by 18 and incident on a non-polarizing beam splitter (NBS) 22 by a lens 19 at a desired angle.

【００２４】一方、偏光ビームスプリッタ14を透過した
Ｐ偏光のアライメントビームは、同じくビームエキスパ
ンダになっているリレーレンズ20、21を経由して前記無
偏光ビームスプリッタ22に達する。この偏光ビームスプ
リッタ22では前述したｆ₂周波数成分のＰ偏光と、ｆ₁周
波数成分のＰ偏光とが入射することになり、両アライメ
ントビームは平行光として一緒に２方向に分岐される。
一方のアライメントビームはレンズ23を通って該レンズ
23の後面焦点位置にあるピンホール24を通り、ディテク
タ25で干渉したヘテロダイン信号を受光する。第２ＲＥ
Ｆ11bにはこのディテクタ25で検出された信号に基づき
参照用ヘテロダイン信号が取り出される。On the other hand, the P-polarized alignment beam transmitted through the polarization beam splitter 14 reaches the non-polarization beam splitter 22 via relay lenses 20 and 21 which are also beam expanders. The P-polarized light having the f ₂ frequency component and the P-polarized light having the f ₁ frequency component are incident on the polarization beam splitter 22, and the two alignment beams are split into two directions together as parallel light.
One alignment beam passes through the lens 23
The heterodyne signal interfered by the detector 25 is received through the pinhole 24 at the focal point of the rear surface of the optical axis 23. Second RE
A reference heterodyne signal is taken out to F11b based on the signal detected by the detector 25.

【００２５】更に無偏光ビームスプリッタ22から所定角
度だけ傾いた平行光束は図６及び図７に示される様にア
ホーカル拡大鏡を構成するレンズ26、27で所定の幅（具
体的には後述する様に光軸より両側に夫々Ｆb・sinθ）
に拡大される。即ち前記レンズ26はその前面焦点位置が
無偏光ビームスプリッタ22の近傍になるように設置して
あり、又もう一方のレンズ27はその前面焦点位置にレン
ズ26があるように設置してある。従ってレンズ26から出
た光束は一度該レンズ26の後面焦点位置でそのビームが
交差し、且つレンズ27を出た所で光軸対称なものとな
る。しかもこのレンズ27により後面焦点の位置でビーム
ウエストＬＢＷ面28（光束ビームが一番絞られる面）が
作られる。ここが後述するアライメント瞳となる。この
時ＬＢＷ面28で２光束の幅は夫々光軸からＦb・sinθと
なる（尚Ｆｂは後述するフーリエ変換レンズ35の後面焦
点距離、又θはマスクＭ及びウェハＷの各回折格子にア
ライメント光を入射した場合に得られる回折光の回折角
度をさす）。尚該レンズ27と後述するレンズ29によりこ
のＬＢＷ28はテレセントリックな関係になっている。Further, the parallel light beam tilted by a predetermined angle from the non-polarizing beam splitter 22 has a predetermined width (specifically, as will be described later) by lenses 26 and 27 constituting an afocal magnifying glass as shown in FIGS. On both sides from the optical axis, Fb ・ sin θ)
Be expanded to. That is, the lens 26 is installed so that its front focus position is near the non-polarizing beam splitter 22, and the other lens 27 is installed so that the lens 26 is located at its front focus position. Therefore, the light flux emitted from the lens 26 is once symmetric with respect to the optical axis at the position where the beams intersect once at the focal position of the rear surface of the lens 26 and exit from the lens 27. Moreover, the lens 27 forms the beam waist LBW surface 28 (the surface where the light beam beam is most narrowed) at the focal point of the rear surface. This is the alignment pupil described later. At this time, the widths of the two light fluxes on the LBW surface 28 are Fb.sin.theta. From the optical axis (Fb is the rear focal length of the Fourier transform lens 35 described later, and .theta. Is the alignment light for the diffraction gratings of the mask M and the wafer W. Refers to the diffraction angle of the diffracted light obtained when incident. The LBW 28 is in a telecentric relationship by the lens 27 and a lens 29 described later.

【００２６】以上のＬＢＷ28の更に後方には交換可能な
変倍光学系が設置されており、まず先に１：１の変倍系
の場合の構成につき説明する。即ち、該ＬＢＷ28の位置
より光軸に対し平行に２光束が進み、テレセントリック
レンズ29によりこの２光束が夫々集光せずに平行なまま
視野絞りＡＦ31で交差するように通過する。この視野絞
り31はマスクＭ及びウェハＷ上の後述する回折格子36と
共役な関係（結像）にある。An interchangeable variable power optical system is installed further rearward of the LBW 28 described above. First, the structure in the case of a 1: 1 variable power system will be described. That is, two light fluxes travel in parallel to the optical axis from the position of the LBW 28, and the two light fluxes are not converged by the telecentric lens 29 but pass in parallel while intersecting each other in the field stop AF31. The field stop 31 has a conjugate relationship (image formation) with the mask M and the diffraction grating 36 described later on the wafer W.

【００２７】更に上記視野絞り31の後方にあるレンズ32
はその前面焦点位置が該視野絞り31にあるように配置さ
れていてこの視野絞り31との間でテレセントリックな関
係にあり、又その後方にある瞳面ＥＰ70上において光軸
に対して対称的にＦb・sinθ夫々離れた位置で集光でき
るように１：１の倍率となっている。Further, a lens 32 behind the field stop 31 is provided.
Is arranged such that its front focal position is located in the field stop 31 and has a telecentric relationship with this field stop 31, and is symmetrical with respect to the optical axis on the pupil plane EP70 behind it. Fb · sin θ has a magnification of 1: 1 so that the light can be collected at positions distant from each other.

【００２８】このレンズ32から出た光束は光軸に対して
平行に進み、前記フーリエ変換レンズ35前面焦点の位置
にある瞳面ＥＰ70上で上述した間隔で集束した後、この
瞳面ＥＰ70がその中にある偏光ビームスプリッタ33を透
過して、λ／４板34でその直線偏光が円偏光となり、フ
ーリエ変換レンズ35により所定の入射角度でマスクＭ及
びウェハＷの各回折格子36に入射される。この時のアラ
イメント光の入射角度は、そのビームが光軸方向から入
射した時に得られる±１次回折光の方向と該光軸との間
の角度に丁度相当し、該方向から逆向きに入射されるこ
とになる。この入射によって得られる各−１次回折光は
マスクＭ面及びウェハＷ面に垂直な方向、即ち光軸方向
に進み、且つ円偏光の回転方向が逆になって進むことに
なる。そしてその回折光は（ｆ₁周波数成分とｆ₂周波数
成分の夫々−１次回折光であって該回折時点で干渉しあ
うことになる）前述のフーリエ変換レンズ35を通って前
記λ／４板34に進み、それによって入射時の直線偏光と
90゜直交する角度の直線偏光に再度なり、偏光ビームス
プリッタ33の反射面で反射されて入射方向から単離され
る。前記λ／４板34を通過した光束は瞳面ＥＰ70で一度
集光し、該瞳面ＥＰ70が丁度その前面焦点距離の位置に
相当する位置になるように配置されたテレセントリック
なレンズ37により、穴あきミラー38を通過して回折格子
36と共役（結像）関係にあるピンホール39に結像し、光
束は更に後方に進む。The light flux emitted from the lens 32 travels in parallel to the optical axis and is focused on the pupil plane EP70 at the focal point of the front surface of the Fourier transform lens 35 at the above-mentioned interval. After passing through the polarizing beam splitter 33 inside, the linearly polarized light becomes circularly polarized light at the λ / 4 plate 34, and is made incident on each diffraction grating 36 of the mask M and the wafer W at a predetermined incident angle by the Fourier transform lens 35. . The angle of incidence of the alignment light at this time corresponds exactly to the angle between the direction of the ± first-order diffracted light obtained when the beam is incident from the optical axis direction and the optical axis, and is incident in the opposite direction from that direction. Will be. Each −first-order diffracted light obtained by this incidence advances in the direction perpendicular to the mask M surface and the wafer W surface, that is, the optical axis direction, and advances in the direction in which the circularly polarized light rotates in the opposite direction. Then, the diffracted light (each of the f ₁ frequency component and the f ₂ frequency component is −1st order diffracted light and interferes with each other at the time of diffraction) passes through the Fourier transform lens 35 and the λ / 4 plate 34. To linearly polarized light at the time of incidence
It becomes linearly polarized light of an angle orthogonal to 90 ° again, is reflected by the reflection surface of the polarization beam splitter 33, and is isolated from the incident direction. The light flux that has passed through the λ / 4 plate 34 is once condensed on the pupil plane EP70, and a hole is formed by the telecentric lens 37 arranged so that the pupil plane EP70 is exactly at the position of the front focal length. Diffraction grating passing through the open mirror 38
An image is formed on the pinhole 39, which has a conjugate (image forming) relationship with 36, and the light flux further travels backward.

【００２９】両−１次回折光を含む光束はナイフエッジ
ミラー40でマスクＭとウェハＷの光束に分割され、夫々
集光レンズ42、43を通って光電子増倍管（ＰＷ）45、46
で光ヘテロダインビート信号を得る。ここでは前記ナイ
フエッジミラー40でマスク回折光とウェハ回折光を分離
したが、２分割ディテクタ（図示なし）等を用いて受光
してもよい。A light flux containing both -1st-order diffracted light is split by a knife edge mirror 40 into light fluxes of a mask M and a wafer W, and passes through condenser lenses 42 and 43, respectively, and photomultiplier tubes (PW) 45 and 46.
To obtain an optical heterodyne beat signal. Although the mask diffracted light and the wafer diffracted light are separated by the knife edge mirror 40 here, they may be received by using a two-divided detector (not shown) or the like.

【００３０】次に、前述の変倍光学系の交換を行なって
１：３の倍率となる変倍系の場合（光軸方向から入射さ
せた場合に取り出される回折光の±１次方向と該光軸と
の間で成す角度の半分の角度であって該光軸に対称的な
２方向から夫々逆に入射させることになる）の構成につ
き説明する。Next, in the case of a variable power system in which the above variable power optical system is exchanged to obtain a magnification of 1: 3 (a ± 1st order direction of the diffracted light extracted when incident from the optical axis direction and The angle which is half the angle formed with the optical axis and is incident on the optical axis in opposite directions from two directions symmetrical to each other) will be described.

【００３１】まず倍率が１倍であった前記レンズ29を図
７に示すように前面焦点距離Ｆ30aと後面焦点距離Ｆ30b
が１：３のレンズ倍率になるレンズ30に切換える。これ
によってビームウエストＬＢＷ28の位置にあるアライメ
ント系の瞳はレンズ30の前面焦点の位置にあり、光軸に
対して対称位置にある２光束は該レンズ30によって前記
視野絞りＡＦ31上で集光せずに平行状態のまま交差して
テレセントリックな関係にあるレンズ32に進み、そこを
透過したビームは図８に示される様に光軸に平行な光束
となり、偏光ビームスプリッタ33の瞳面ＥＰ70で集光さ
れる。この時瞳面ＥＰ70上では光軸に対称でその間隔が
（Ｆb・sinθ）／２となるような配置に２光束がある。
この偏光ビームスプリッタ33を透過した光束はλ／４板
34により円偏光となり、所定の入射角度、即ち光軸方向
から入射させた場合に取り出される回折光の±１次方向
と該光軸との間で成す角度の半分の角度でマスクＭ及び
ウェハＷの各回折格子36に入射せしめられる。First, as shown in FIG. 7, the lens 29, which has a magnification of 1 ×, has a front focal length F30a and a rear focal length F30b.
Switch to lens 30 with a lens magnification of 1: 3. As a result, the pupil of the alignment system at the position of the beam waist LBW28 is at the position of the front focal point of the lens 30, and the two light beams symmetrically positioned with respect to the optical axis are not condensed by the lens 30 on the field stop AF31. The beam passing through the lens crosses the lens 32 in a telecentric relationship while being parallel to the light beam parallel to the optical axis as shown in FIG. 8 and condensed on the pupil plane EP70 of the polarization beam splitter 33. To be done. At this time, on the pupil plane EP70, there are two light fluxes arranged symmetrically with respect to the optical axis and the distance therebetween is (Fb · sin θ) / 2.
The light flux transmitted through this polarization beam splitter 33 is a λ / 4 plate
The light is circularly polarized by 34, and the mask M and the wafer W have a predetermined incident angle, that is, an angle half the angle between the ± first-order directions of the diffracted light extracted when incident from the optical axis direction and the optical axis. Are made incident on the respective diffraction gratings 36 of.

【００３２】入射方向に逆向きに進んでくるｆ₁周波数
成分の−１次回折光とｆ₂周波数成分の０次回折光及び
ｆ₂周波数成分の−１次回折光とｆ₁周波数成分の０次回
折光の２組の回折光（これらの各組では該回折時点で干
渉し合うことになる）は光軸に対して左右対称に進み、
前記フーリエ変換レンズ35を通り、λ／４板34で入射時
の直線偏光と90゜直交する角度の直線偏光に再度なり、
偏光ビームスプリッタ33の反射面で反射することで入射
方向から分離する。そしてテレセントリックなレンズ37
により光軸に平行に進み、ミラー38で反射し、回折格子
36と共役（結像）関係にあるピンホール47で交差した
後、後方光学系に進む。[0032] -1 -1 order diffracted light and f ₁ frequency component of 0-order diffracted light and f ₂ frequency components of the diffracted light and f ₂ frequency components of f ₁ frequency component coming proceed in the opposite direction to the incident direction 0-order diffracted light The two sets of diffracted light (each of which will interfere with each other at the time of diffraction) travel symmetrically with respect to the optical axis,
After passing through the Fourier transform lens 35, the λ / 4 plate 34 converts the linearly polarized light at the time of incidence into a linearly polarized light having an angle orthogonal to 90 ° again,
It is separated from the incident direction by being reflected by the reflecting surface of the polarization beam splitter 33. And a telecentric lens 37
Travels parallel to the optical axis, is reflected by the mirror 38, and
After intersecting with a pinhole 47 having a conjugate (imaging) relationship with 36, the process proceeds to the rear optical system.

【００３３】このピンホール47通過後２組の回折光は夫
々の分離角度方向に進み、プリズムミラー48、55で光路
を変え、そしてナイフエッジミラー49、56で分割され、
更に分割されたものの一方は、夫々プリズムミラー49、
56で分割され、更に分割されたものの一方は、夫々プリ
ズムミラー50、57で反射し、４つに分岐された各光束は
夫々集光レンズ51、52、58、59で集光され、光電子倍増
管53、54、60、61に受光されてそこで光ヘテロダインビ
ート信号を得る。尚上記ナイフエッジミラー49、56でマ
スク回折光とウェハ回折光を分離したが、２分割ディテ
クタ（図示なし）を用いて受光しても良い。After passing through the pinhole 47, the two sets of diffracted light travel in their respective separation angle directions, change the optical path by the prism mirrors 48 and 55, and are split by the knife edge mirrors 49 and 56.
One of the subdivided ones is a prism mirror 49,
One of the light beams divided by 56 and further divided is reflected by prism mirrors 50 and 57, respectively, and the respective light fluxes branched into four are condensed by condenser lenses 51, 52, 58 and 59, respectively, and photoelectron multiplication is performed. The light is received by the tubes 53, 54, 60 and 61, and the optical heterodyne beat signal is obtained there. Although the mask-diffracted light and the wafer-diffracted light are separated by the knife edge mirrors 49 and 56, they may be received by using a two-divided detector (not shown).

【００３４】次に以上の様な方法が実施された場合のそ
の作用につき詳述する。Next, the operation when the above method is carried out will be described in detail.

【００３５】本実施例では数十ミクロンギャップ量を持
ったマスクＭに形成された回折格子とウェハＷに形成さ
れた回折格子36にフーリエ変換レンズ35を介してコヒー
レントなレーザ光を２方向から入射させ、回折光を発生
させたものである。尚簡略のためマスクＭ（又はウェハ
Ｗ）１種類に上記レーザ光を当てた場合について、以下
説明する。In this embodiment, coherent laser light is incident on the diffraction grating formed on the mask M having a gap of several tens of microns and the diffraction grating 36 formed on the wafer W through the Fourier transform lens 35 from two directions. And diffracted light is generated. For simplification, the case where one type of mask M (or wafer W) is irradiated with the laser light will be described below.

【００３６】図９及び図１０はＬ／Ｓ（ライン＆スペー
ス比）を示すデューティ比(Ｐ−ａ)／Ｐ＝１／２、つま
り１：１比のＬ／Ｓである回折格子に対して照射を行な
った状態を示している（Ｐ：格子ピッチ、ａ：格子間の
スペース、図９(h)参照）。ここで回折格子数Ｎは６と
する（実用上は数１０〜数１００）。又上述のデューテ
ィ比は１次回折効率が１番良い１／２を選んだが、逆に
偶数次数は回折光が出ない欠点がある。9 and 10 show a duty ratio (P-a) / P = 1/2 indicating L / S (line & space ratio), that is, for a diffraction grating having L / S of 1: 1 ratio. Illumination is shown (P: lattice pitch, a: space between lattices, see FIG. 9 (h)). Here, the number of diffraction gratings N is 6 (practically several tens to several hundreds). Further, although the above-mentioned duty ratio is selected to be 1/2, which has the best first-order diffraction efficiency, conversely, there is a drawback that even-orders do not emit diffracted light.

【００３７】まず変倍光学系で前記レンズ29を使用した
場合の光ヘテロダイン干渉モデルから説明する。First, an optical heterodyne interference model when the lens 29 is used in a variable power optical system will be described.

【００３８】前記回折格子36から後面焦点距離Ｆbの位
置にある前記フーリエ変換レンズ35（回折格子パターン
像が回折次数ｍ、ｎに応じて等間隔になるように設計さ
れたレンズ）により回折格子36の回折格子像が等間隔に
ならぶ面を瞳面ＥＰ70と呼ぶ（この面は回折格子36のフ
ーリエ変換像を得た面なのでフーリエ面とも呼ばれ
る）。そしてこの瞳面ＥＰ70はフーリエ変換レンズ35の
前面焦点距離Ｆaの位置にある。本実施例ではＦa＝Ｆb
とする。Diffraction grating 36 by Fourier transform lens 35 (lens designed so that diffraction grating pattern images are equidistant according to diffraction orders m and n) at a position of rear focal length Fb from diffraction grating 36. A surface on which the diffraction grating images of are arranged at equal intervals is called a pupil surface EP70 (this surface is also a Fourier surface because it is a surface on which the Fourier transform image of the diffraction grating 36 is obtained). The pupil plane EP70 is located at the front focal length Fa of the Fourier transform lens 35. In this embodiment, Fa = Fb
And

【００３９】前述の様に仮りに光軸方向から入射して得
られる±１次回折光の方向から逆向きに上記レーザ光を
入射させ、光軸方向に干渉し合った２つの−１次回折光
を得ようとするならば、変倍光学系のレンズを前記レン
ズ29に交換し、前記瞳面ＥＰ70で光軸からＦb・sinθ
（θは上記回折光の回折角度）だけ離れた位置から夫々
入射せしめることになる。As described above, the above-mentioned laser light is made to enter in the opposite direction from the direction of the ± first-order diffracted light obtained by incidentally from the optical axis direction, and two −1st-order diffracted lights that interfere with each other in the optical axis direction are made. In order to obtain it, the lens of the variable power optical system is replaced with the lens 29, and Fb · sin θ is changed from the optical axis on the pupil plane EP70.
(Θ is the diffraction angle of the above-mentioned diffracted light).

【００４０】ｆ₁周波数成分のレーザ光はこの図では光
軸に対して左側にＦb・sinθの位置から、しかも前記瞳
面ＥＰ70では一度集光された状態となって入射され、フ
ーリエ変換レンズ35により回折格子36に対し照射され、
その結果、光軸方向（垂直方向）に−１次回折光（この
時の各回折次数をｍとする）が出る。上記瞳面ＥＰ70上
でのｆ₁周波数成分の回折光強度分布は図９(e)の様にな
り、各回折次数ｍは−３〜＋２までが等間隔で並ぶこと
になる（尚通常使用するフーリエ変換レンズ（光露光装
置では対物レンズ）の開口数ＮＡやディテクタ面の分割
素子数の制約のため実用的な範囲にあるのはこの５点と
判断し、これを明記した）。この強度分布に表われてい
る様に、本実施例における回折格子36のデューティ比
(Ｐ−ａ)／Ｐ＝１／２であるため、回折光の強度分布の
法則により偶数次数の回折光（ｍ＝−２、＋２、……）
はミッシングオーダとなり欠如している（このデューテ
ィ比を変えれば、該ミッシングオーダ次数位置でも回折
光強度を得ることができ、この欠如次数も使用でき
る）。The laser light of the f ₁ frequency component is incident on the left side with respect to the optical axis in the figure from the position of Fb · sin θ, and is once converged on the pupil plane EP70, and the Fourier transform lens 35 Is radiated to the diffraction grating 36 by
As a result, −1st-order diffracted light (each diffraction order at this time is defined as m) appears in the optical axis direction (vertical direction). The diffracted light intensity distribution of the f ₁ frequency component on the pupil plane EP70 is as shown in FIG. 9 (e), and each diffraction order m is -3 to +2 arranged at equal intervals (it is normally used. It is determined that these five points are within a practical range due to restrictions on the numerical aperture NA of the Fourier transform lens (objective lens in the light exposure apparatus) and the number of dividing elements on the detector surface, and this is clearly stated). As shown in this intensity distribution, the duty ratio of the diffraction grating 36 in this embodiment is
Since (P−a) / P = 1/2, even-order diffracted light (m = −2, +2, ...) According to the law of the intensity distribution of diffracted light.
Is a missing order and is missing (diffraction ratio can be changed to obtain diffracted light intensity even at the missing order position, and this missing order can also be used).

【００４１】同様にｆ₂周波数成分のレーザ光は同図で
は光軸右側にＦb・sinθ離れた位置から、しかも前記瞳
面ＥＰ70では一度集光された状態となって入射され、前
記フーリエ変換レンズ35により回折格子36に照射され、
その結果光軸方向に−１次回折光（この時の各回折次数
をｎとする）が出る。又この時の瞳面ＥＰ70上でのｆ₂
周波数成分の回折光強度分布は図９(f)の様になる。Similarly, the laser light of the f ₂ frequency component is made incident on the right side of the optical axis from a position away from Fb · sin θ in the figure, and is also converged once on the pupil plane EP70, and the Fourier transform lens The diffraction grating 36 is illuminated by 35,
As a result, −1st-order diffracted light (each diffraction order at this time is n) is emitted in the optical axis direction. Also, f ₂ on the pupil plane EP70 at this time
The diffracted light intensity distribution of frequency components is as shown in FIG. 9 (f).

【００４２】両回折光強度分布とも０次、±１次、±３
次の順に小さくなっていることがわかる。Both diffracted light intensity distributions are 0th order, ± 1st order, ± 3th order.
It can be seen that the size decreases in the following order.

【００４３】一方、図９(b)(c)(d)では、ｆ₁、ｆ₂の夫
々の周波数成分の回折光の強度を視覚化するため、前述
したディテクタ45、46面手前位置（瞳面ＥＰ70と共役な
位置にあるピンホール39から少し後方の面）での回折光
強度分布を示している。On the other hand, in FIGS. 9 (b) (c) (d), in order to visualize the intensities of the diffracted light components of the frequency components f ₁ and f ₂ , the detectors 45 and 46, which are located at the front position (pupil), are visualized. The figure shows the diffracted light intensity distribution on a surface slightly rearward from the pinhole 39 at a position conjugate with the surface EP70.

【００４４】更に上記両回折光（ｆ₁周波数成分のもの
とｆ₂周波数成分のもの）の干渉で得られる｜ｆ₁−ｆ₂
｜の光ヘテロダインビート信号の干渉縞を図９(a)に示
す。ディテクタ45、46上では干渉縞空間で一様な明るさ
（ワンカラー）内に調整されているが、ここでは説明の
ため干渉縞を複数描いた。又光ヘテロダイン干渉縞は高
速（数十ＫＨz〜数１００ＫＨz）では肉眼で見ることが
できず、電気的な信号に変換しなければならないが、こ
こでは敢えて干渉縞を得る次数の組合せ（ｍ＝−１、ｎ
＝−１）（ｍ＝−３、ｎ＝＋１）（ｍ＝＋１、ｎ＝−
３）の干渉縞を明確にするために、同図(a)に示してい
る。この図で描かれた光ヘテロダイン干渉縞のピッチが
丁度回折格子36のピッチの±Ｐ／４に当たる（図５(a)
参照）。Further, | f ₁ -f ₂ obtained by the interference of both diffracted light (of the f ₁ frequency component and f ₂ frequency component)
The interference fringes of the | optical heterodyne beat signal are shown in FIG. On the detectors 45 and 46, the brightness is adjusted to a uniform brightness (one color) in the interference fringe space, but a plurality of interference fringes are drawn here for the sake of explanation. Optical heterodyne interference fringes cannot be seen with the naked eye at high speeds (several tens of kHz to several hundreds of kHz) and must be converted into electrical signals, but here, the combination of orders (m = − 1, n
= -1) (m = -3, n = + 1) (m = + 1, n =-
In order to clarify the interference fringe of 3), it is shown in FIG. The pitch of the optical heterodyne interference fringes drawn in this figure corresponds to ± P / 4 of the pitch of the diffraction grating 36 (FIG. 5 (a)).
reference).

【００４５】次に変倍光学系のレンズを前記レンズ30に
交換した場合の光ヘテロダイン干渉モデルにつき図１０
を使用して説明する。FIG. 10 shows an optical heterodyne interference model when the lens of the variable power optical system is replaced with the lens 30.
To explain.

【００４６】まず仮りに光軸方向から入射して得られる
±１次回折光と該光軸との成す角度の半分の角度であっ
てこの光軸に対し対称的な２方向から夫々逆にｆ₁周波
数成分のレーザ光とｆ₂周波数成分のレーザ光を入射さ
せ、この入射方向に逆向きにｆ₁周波数成分の−１次回
折光とｆ₂周波数成分の０次回折光及びｆ₂周波数成分の
−１次回折光とｆ₁周波数成分の０次回折光の２組の光
ヘテロダインビート信号を得ようとするならば、変倍光
学系のレンズをレンズ30に変換し、前記瞳面ＥＰ70で光
軸から（Ｆb・sinθ）／２だけ離れた位置から夫々入射
せしめることになる。First, if the ± 1st-order diffracted light obtained by incidence from the optical axis direction is half the angle formed by the optical axis and is symmetrical with respect to the optical axis, f _{1 is} reversed from two directions. is incident laser light of the laser beam and f ₂ frequency components of the frequency components, -1 0-order diffracted light and f ₂ frequency components of the -1st-order diffracted light and f ₂ frequency components of f ₁ frequency component in the opposite direction to the incident direction In order to obtain two sets of optical heterodyne beat signals of the 0th-order diffracted light and the 0th-order diffracted light of the f ₁ frequency component, the lens of the variable power optical system is converted into the lens 30 and the pupil plane EP 70 is moved from the optical axis (Fb・ Sine θ) / 2 will be incident from each position.

【００４７】ｆ₁周波数成分のレーザ光はこの図では光
軸に対して左側に（Ｆb・sinθ）／２離れた位置から、
しかも前記瞳面ＥＰ70では一度集光（ビームウエスト）
された状態となって入射され、フーリエ変換レンズ35に
より回折格子36に照射される。その結果この入射方向と
同方向逆向きに−１次回折光が、又光軸反対側正反射方
向に０次回折光が取り出されることになる。その回折光
の像はフーリエ変換レンズ35の瞳面ＥＰ70上でフーリエ
変換像に変換される。このｆ₁周波数成分の該瞳面ＥＰ7
0上の回折光の強度分布を図１０(e)に示す。同図に示さ
れる様に０次と−１次の分が光軸を中心に対称的な位置
に並んでいる。In this figure, the laser light of the f ₁ frequency component is at the position (Fb · sin θ) / 2 away from the optical axis on the left side,
Moreover, the pupil plane EP70 once collects light (beam waist).
The light is incident on the diffraction grating 36 by the Fourier transform lens 35 in a changed state. As a result, the −1st order diffracted light is extracted in the same direction as this incident direction and the opposite direction, and the 0th order diffracted light is extracted in the regular reflection direction on the opposite side of the optical axis. The image of the diffracted light is converted into a Fourier transform image on the pupil plane EP70 of the Fourier transform lens 35. This pupil plane EP7 of this f ₁ frequency component
The intensity distribution of the diffracted light above 0 is shown in FIG. As shown in the figure, the 0th-order and -1st-order components are arranged in symmetrical positions with respect to the optical axis.

【００４８】同様にｆ₂周波数成分のレーザ光は同図で
は光軸右側に（Ｆb・sinθ）／２離れた位置から、しか
も瞳面ＥＰ70では一度集光された状態になって入射さ
れ、フーリエ変換レンズ35により回折格子36に照射され
る。その結果この入射方向と同方向逆向きに−１次回折
光（これは前述のｆ₁周波数成分の０次回折光の回折方
向と同じ向きになる）が、又光軸反対側正反射方向に０
次回折光（これは前述のｆ₁周波数成分の−１次回折光
の回折方向と同じ向きになる）が取り出されることにな
る。その回折光の像はフーリエ変換レンズ35の瞳面ＥＰ
70上でフーリエ変換像に変換される。このｆ₂周波数成
分の該瞳面ＥＰ70上の回折光の強度分布を図１０(f)に
示す。同図に示される様に−１次と０次の分が光軸中心
に対称的な位置に並んでいる。Similarly, the laser light of the f ₂ frequency component is incident on the right side of the optical axis from the position (Fb · sin θ) / 2 away from the right side of FIG. The conversion lens 35 illuminates the diffraction grating 36. As a result, the -1st order diffracted light (which is the same as the diffracted direction of the 0th order diffracted light of the f ₁ frequency component) in the same direction and the opposite direction to this incident direction, and 0 in the regular reflection direction on the opposite side of the optical axis.
The next-order diffracted light (which has the same direction as the diffraction direction of the -1st-order diffracted light of the f ₁ frequency component described above) is extracted. The image of the diffracted light is the pupil plane EP of the Fourier transform lens 35.
Converted to a Fourier transform image on 70. The intensity distribution of the diffracted light of the f ₂ frequency component on the pupil plane EP70 is shown in FIG. 10 (f). As shown in the figure, the -1st-order and 0th-order components are arranged in symmetrical positions with respect to the optical axis center.

【００４９】前述の場合と同様、図１０(b)(c)(d)に、
各回折光の強度分布を視覚化して示し、又同図(a)に得
られた光ヘテロダインビート信号の干渉縞を示す。Similar to the case described above, in FIGS. 10 (b) (c) (d),
The intensity distribution of each diffracted light is visualized and the interference fringes of the obtained optical heterodyne beat signal are shown in FIG.

【００５０】かくしてｆ₁周波数成分の回折次数ｍ＝−
１の回折光とｆ₂周波数成分の回折次数ｎ＝０の回折光
との光ヘテロダイン干渉縞と、ｆ₁周波数成分の回折次
数ｍ＝０の回折光とｆ₂周波数成分の回折次数ｎ＝−１
の回折光との光ヘテロダイン干渉縞が同時に、しかも光
軸に対して左右対称な位置に得られる。ここで得られた
干渉縞のピッチは回折格子36のピッチの±Ｐ／２に相当
し、その検出範囲が前述の場合と比べて２倍に拡大され
たことになる（図５(b)参照）。Thus, the diffraction order of the f ₁ frequency component m = −
Optical heterodyne interference fringes of the diffracted light of 1 and the diffracted light of the f ₂ frequency component n = 0, and the diffracted light of the d ₁ diffracted order of the f ₁ frequency component m = 0 and the d ₂ d of the f ₂ frequency component n = − 1
Optical heterodyne interference fringes with the diffracted light are simultaneously obtained and at positions symmetrical with respect to the optical axis. The pitch of the interference fringes obtained here corresponds to ± P / 2 of the pitch of the diffraction grating 36, which means that the detection range is doubled as compared with the case described above (see FIG. 5 (b)). ).

【００５１】この様に後者の入射方法によって得られる
検出範囲は前者のそれの２倍になるのは次の様な理由に
因る。The reason why the detection range obtained by the latter method of incidence is twice as large as that of the former method is as follows.

【００５２】まず振幅が略等しく周波数がわずかに異な
っており（数１０ＫＨz〜数１００ＫＨz）、且つ同一方
向に進行する２つの波ｕ₁とｕ₂の重ね合せを考える。First, let us consider a superposition of _two waves u ₁ and u _{2 which} have substantially equal amplitudes and slightly different frequencies (several tens of kHz to several hundreds of kHz) and which travel in the same direction.

【００５３】ｕ₁、ｕ₂は下式数３及び数４の様な波動方
程式に書ける。U ₁ and u ₂ can be written in wave equations such as the following equations 3 and 4.

【００５４】[0054]

【数３】 [Equation 3]

【００５５】[0055]

【数４】 [Equation 4]

【００５６】又ビート周波数（うなり）は振幅の変動の
繰り返し周波数で下式数５で表わせる。The beat frequency (beat) is the repetition frequency of amplitude fluctuation and can be expressed by the following equation (5).

【００５７】[0057]

【数５】 [Equation 5]

【００５８】ｕ₁とｕ₂の２つの波の振幅の和を２乗して
波の重ね合せ強度を求めると次式数６が得られる。If the sum of the amplitudes of the two waves u ₁ and u ₂ is squared to obtain the superposition strength of the waves, the following equation 6 is obtained.

【００５９】[0059]

【数６】 [Equation 6]

【００６０】上記数６の［Ｌ(1)／λ₁−Ｌ(2)／λ₂］項
から光路差［Ｌ(1)−Ｌ(2)］の変化によって位相項の遅
れや進みが出ることがわかる。From the [L (1) / λ ₁ -L (2) / λ ₂ ] term of the above equation 6, the phase term is delayed or advanced due to the change of the optical path difference [L (1) -L (2)]. I understand.

【００６１】光ヘテロダインアライメント方式ではこの
位相差の測定を行なうことになるが、この位相差は±18
0゜以内と角度検出範囲が固定されてしまうため、数６
の［Ｌ(1)／λ₁−Ｌ(2)／λ₂］の項で示される光路差
［Ｌ(1)−Ｌ(2)］が検出範囲を左右する項目となる（使
用される２つの周波数ｆ₁、ｆ₂のレーザ光はわずかに波
長が異なるだけなのでλ₁＝λ₂≒λとおける。従って2
π［Ｌ(1)／λ₁−Ｌ(2)／λ₂］の項は2π［Ｌ(1)−Ｌ
(2)］／λと置き換えることができる）。In the optical heterodyne alignment method, this phase difference is measured, but this phase difference is ± 18.
Since the angle detection range is fixed at 0 ° or less,
The optical path difference [L (1) -L (2)] shown in the section [L (1) / λ ₁ −L (2) / λ ₂ ] of (1) is an item that influences the detection range. Since the laser beams of the _two frequencies f ₁ and f ₂ are slightly different in wavelength, it can be said that λ ₁ = λ ₂ ≈λ.
The term of π [L (1) / λ ₁ -L (2) / λ ₂ ] is 2π [L (1) -L
(2)] / λ).

【００６２】次に図１１、図１２、図１３を基に図９の
場合の光軸方向にｆ₁周波数成分の−１次回折光とｆ₂周
波数成分の−１次回折光で得られる光路差Ｌ(1)−Ｌ(2)
と、図１０の場合のｆ₁周波数成分の−１次回折光とｆ₂
周波数成分の０次回折光及びｆ₂周波数成分の−１次回
折光とｆ₁周波数成分の０次回折光の夫々で得られる光
路差Ｌ(1)−Ｌ(2)を求めてみる。Next, based on FIG. 11, FIG. 12, and FIG. 13, the optical path difference L obtained by the −1st-order diffracted light of the f ₁ frequency component and the −1st-order diffracted light of the f ₂ frequency component in the optical axis direction in the case of FIG. (1) -L (2)
And the −1st order diffracted light of the f ₁ frequency component and f _{2 in} the case of FIG.
The optical path difference L (1) -L (2) obtained by the 0th-order diffracted light of the frequency component, the -1st-order diffracted light of the f ₂ frequency component, and the 0th-order diffracted light of the f ₁ frequency component will be obtained.

【００６３】上記図１１、図１２及び図１３はいずれも
回折格子36の設けられたマスクＭ（ウェハＷ）が右方向
にΔｘだけ移動した場合（これはマスクＭとウェハＷが
Δｘだけずれている場合に相当する）に、光路Ｌ(2)に
対する光路Ｌ(1)の光路差を示している。ここで光軸よ
り左側から入射するｆ₁周波数の光ビームの光路をＬ
(1)、その入射角をθi₁、回折角をθmとする。又光軸よ
り右側から入射するｆ₂周波数の光ビームの光路をＬ
(2)、その入射角をθi₂、回折角をθnとする。そして求
める光路差は光路Ｌ(2)に対する光路Ｌ(1)の光路差であ
る。この時の位相差δは下式数７の様になる。In all of FIGS. 11, 12 and 13, the mask M (wafer W) provided with the diffraction grating 36 is moved rightward by Δx (this is because the mask M and the wafer W are displaced by Δx). The light path difference of the light path L (1) with respect to the light path L (2) is shown in FIG. Here, the optical path of the light beam of f ₁ frequency that is incident from the left side of the optical axis is L
(1), the incident angle is θi ₁ , and the diffraction angle is θm. In addition, the optical path of the light beam of f ₂ frequency that is incident from the right side of the optical axis is L
(2), the incident angle is θi ₂ and the diffraction angle is θn. The obtained optical path difference is the optical path difference of the optical path L (1) with respect to the optical path L (2). The phase difference δ at this time is as shown in the following expression 7.

【００６４】[0064]

【数７】 [Equation 7]

【００６５】そのうち図１１は、前記図９の様な入射を
行なった際にマスクＭ（ウェハＷ）がΔｘだけ移動した
場合の夫々の光路を示しており、それによると光路Ｌ
(1)はその移動によってΔｘsinθi₁だけ長くなり、もう
一方の光路Ｌ(2)は該移動によってΔｘsinθi₂だけ短く
なる。これを上式数７に代入すると、その位相差δは次
式数８の様になる。FIG. 11 shows the respective optical paths when the mask M (wafer W) moves by Δx when the incident light as shown in FIG. 9 is carried out.
The movement (1) lengthens by Δxsin θi ₁ by the movement, and the other optical path L (2) shortens by Δxsin θi ₂ by the movement. Substituting this into the above equation 7, the phase difference δ becomes like the following equation 8.

【００６６】[0066]

【数８】 [Equation 8]

【００６７】ここでｆ₁周波数とｆ₂周波数の夫々の入射
光の入射角度θi₁、θi₂とこれらのｍ次及びｎ次回折光
の回折角度θm、θnは、前述の数１及び数２の回折の基
礎公式より、ｆ₁周波数成分については下式数９、又は
ｆ₂周波数成分については下式数１０の様に求められ
る。Here, the incident angles θi ₁ and θi ₂ of the respective incident lights of the f ₁ frequency and the f ₂ frequency and the diffraction angles θm and θn of the m-th and n-th order diffracted lights are given by the above-mentioned formulas 1 and 2. According to the basic formula of diffraction, the f ₁ frequency component is obtained by the following equation 9, or the f ₂ frequency component is obtained by the following equation 10.

【００６８】[0068]

【数９】 [Equation 9]

【００６９】[0069]

【数１０】 [Equation 10]

【００７０】ここでｆ₁周波数成分及びｆ₂周波数成分の
回折光はｍ＝−１、ｎ＝−１の時光軸方向（垂直方向）
に回折するので、その回折角度θm＝θn＝０となる。従
って上記数９及び数１０は下式数１１及び数１２の様に
なる。Here, the diffracted light of the f ₁ frequency component and the f ₂ frequency component is in the optical axis direction (vertical direction) when m = −1 and n = −1.
Since it is diffracted into, the diffraction angle θm = θn = 0. Therefore, the above equations 9 and 10 become like the following equations 11 and 12.

【００７１】[0071]

【数１１】 [Equation 11]

【００７２】[0072]

【数１２】 [Equation 12]

【００７３】そこで前記数８に上記数１１及び数１２を
代入するとその位相差δは下式数１３の様になる。Then, by substituting the equations 11 and 12 into the equation 8, the phase difference δ becomes as shown in the following equation 13.

【００７４】[0074]

【数１３】 [Equation 13]

【００７５】これを±の付く検出範囲で表現すると次式
数１４に示す様になり、これが変倍光学系でレンズ29を
使用した時の検出範囲となる（図５(a)参照）。When this is expressed by a detection range with ±, it becomes as shown in the following expression 14, and this is the detection range when the lens 29 is used in the variable power optical system (see FIG. 5 (a)).

【００７６】[0076]

【数１４】 [Equation 14]

【００７７】以上の説明を基に、変倍光学系でレンズ30
を使用した場合に光軸に対して対称的に取り出されてく
るｆ₁周波数成分の−１次回折光とｆ₂周波数成分の０次
回折光の組及びｆ₂周波数成分の−１次回折光とｆ₁周波
数成分の０次回折光の組の各光ヘテロダインビート信号
が同じ位相方向（進み、遅れ）を示すことになるのを説
明する。Based on the above description, the lens 30 is used in the variable power optical system.
-1st 0 set and f ₂ frequency components of the diffracted light of negative first order diffracted light and f ₂ frequency components of f ₁ frequency component coming is symmetrically removed with respect to the optical axis when using diffracted light and f ₁ It will be explained that each optical heterodyne beat signal of the set of 0th-order diffracted light of the frequency component will show the same phase direction (advance, lag).

【００７８】図１２は回折格子36がΔｘだけ移動した時
の光路差を示すもののうち、ｆ₁周波数成分の０次回折
光とｆ₂周波数成分の−１次回折光の状態を示す説明図
である。[0078] Figure 12 is among those showing an optical path difference when the diffraction grating 36 is moved by [Delta] x, is an explanatory diagram showing a 0 -1 state of order diffracted light of diffracted light and f ₂ frequency components of f ₁ frequency component.

【００７９】同図においてｆ₁周波数成分の光路Ｌ(1)は
回折方向が０次（正反射方向）なので、Δｘ移動した時
の光路差は、入射時にΔｘsinθi₁だけ移動前より長く
なると共に回折時にはΔｘsinθmだけ移動前より短くな
る。即ち、回折方向が０次になる光路Ｌ(1)の場合は、
片方が長くなれば他方が短くなる相補的な形になり、本
実施例の様に左右対称な光学配置であれば、変化量も同
じ（θi₁＝θm）で光路差は０になる。従って０次回折
光においては位相差は発生せず、その光路長の変化は常
に０のままである。In the figure, since the optical path L (1) of the f ₁ frequency component has a diffraction direction of the 0th order (regular reflection direction), the optical path difference when moving by Δx becomes longer by Δx sin θi ₁ at the time of incidence and diffracts. Sometimes it becomes shorter by Δxsinθm than before movement. That is, in the case of the optical path L (1) in which the diffraction direction is the 0th order,
If one side becomes longer, the other side becomes shorter, and the shape becomes complementary. If the optical arrangement is bilaterally symmetric as in the present embodiment, the amount of change is the same (θi ₁ = θm) and the optical path difference is zero. Therefore, no phase difference is generated in the 0th-order diffracted light, and the change in the optical path length thereof is always 0.

【００８０】更にｆ₂周波数成分の光路Ｌ(2)は回折方向
が−１次であり、Δｘ移動後の光路差はその移動前に比
べてΔｘsinθn（θi₂＝θn）だけ短くなる。ここでア
ライメント光はこの光路差分だけ往復するので２倍とな
り、その光路差は２Δｘsinθnとなる。従って前記数７
式によりその位相差を求めると、下式数１５の様にな
る。Further, the optical path L (2) of the f ₂ frequency component has the diffraction direction of −1st order, and the optical path difference after Δx movement becomes shorter by Δx sin θn (θi ₂ = θn) than before the movement. Since the alignment light reciprocates by this optical path difference, it is doubled, and the optical path difference becomes 2Δxsin θn. Therefore, the number 7
When the phase difference is obtained by the equation, the following equation 15 is obtained.

【００８１】[0081]

【数１５】 [Equation 15]

【００８２】ここでｆ₂周波数成分のｎ＝−１次回折の
場合も、前記数２の回折の基礎公式から下式数１６の様
になる。Here, also in the case of the n = −1st order diffraction of the f ₂ frequency component, the following formula 16 is obtained from the basic formula of the diffraction of the above formula 2.

【００８３】[0083]

【数１６】 [Equation 16]

【００８４】そして図１２の状態では入射角度と回折角
度が等しいため、θ_(n=-1)＝θi₂となり、上記数１６は
下式数１７の様になる。In the state of FIG. 12, since the incident angle and the diffraction angle are equal, θ _{(n = -1)} = θi ₂ , and the above equation 16 becomes the following equation 17.

【００８５】[0085]

【数１７】 [Equation 17]

【００８６】この数１７を前記数１５の式に代入する
と、その位相差δは下式数１８の様になる。Substituting this equation 17 into the equation 15 gives the phase difference δ as shown in the following equation 18.

【００８７】[0087]

【数１８】 [Equation 18]

【００８８】これを前記数１３の式の場合と比べると、
上記数１８の分母、即ち検出範囲の部分がＰ／２からＰ
となり、２倍に拡大されている。尚数１８の検出範囲を
±で表現すると、下式数１９の様になる。Comparing this with the case of the equation (13),
The denominator of the above equation 18, that is, the part of the detection range is from P / 2 to P
Has been doubled. If the detection range of Expression 18 is expressed by ±, Expression 19 below is obtained.

【００８９】[0089]

【数１９】 [Formula 19]

【００９０】一方図１３は同じく回折格子36がΔｘだけ
移動した時の光路差を示すもののうち、ｆ₁周波数成分
の−１次回折光とｆ₂周波数成分の０次回折光の状態を
示している。[0090] On the other hand 13 out similarly diffraction grating 36 is one showing an optical path difference when the moved by [Delta] x, shows a -1 0 state of order diffracted light of diffracted light and f ₂ frequency components of f ₁ frequency component.

【００９１】同図においてｆ₂周波数成分の光路Ｌ(2)は
回折方向が０次（正反射方向）なので、Δｘ移動した時
の光路差は、入射時にΔｘsinθnだけ移動前より短くな
ると共に回折時にはΔｘsinθi₂だけ移動前より長くな
る。即ちこの場合も（θn＝θi₂なので）光路差は０に
なる。In the figure, since the optical path L (2) of the f ₂ frequency component has a diffraction direction of the 0th order (regular reflection direction), the optical path difference when moving by Δx is shorter by Δx sin θn at the time of incidence and at the time of diffraction. It becomes longer than before the movement by Δx sin θi ₂ . That is, also in this case (since θn = θi ₂ ), the optical path difference becomes zero.

【００９２】これに対してｆ₁周波数成分の光路Ｌ(1)は
回折方向が−１次であり、Δｘ移動後の光路差はその移
動前に比べてΔｘsinθmだけ長くなり、しかも往復分あ
るので、２Δｘsinθmだけ発生する。よってその位相差
を求めると次式数２０の様になる。On the other hand, the optical path L (1) of the f ₁ frequency component has a diffraction direction of −1st order, and the optical path difference after Δx movement is longer by Δx sin θm than before the movement, and there is a round trip. 2Δxsin θm occurs. Therefore, when the phase difference is obtained, the following equation 20 is obtained.

【００９３】[0093]

【数２０】 [Equation 20]

【００９４】同様に回折基礎公式（前記数１）からｍ＝
−１の時は、次式数２１の様になる。Similarly, from the diffraction basic formula (Equation 1), m =
When it is -1, the following equation 21 is obtained.

【００９５】[0095]

【数２１】 [Equation 21]

【００９６】そして図１３の状態では入射角度θi₁と回
折角度θ_(m=-1)が等しいため、上記数２１は下式数２２
の様になる。In the state of FIG. 13, the incident angle θi ₁ and the diffraction angle θ _{(m = -1)} are equal.
It becomes like.

【００９７】[0097]

【数２２】 [Equation 22]

【００９８】この数２２を前記数２０の式に代入する
と、下式数２３の様になる。Substituting this formula 22 into the formula 20 gives the following formula 23.

【００９９】[0099]

【数２３】 [Equation 23]

【０１００】以上の図１２及び図１３における位相差δ
（数１８及び数２３参照）を比較すると両者の値は等し
いことがわかる。即ち、両者の位相方向（遅れ、進み）
は同じ方向を示していることになる。The phase difference δ in FIGS. 12 and 13 above.
Comparing (see Eqs. 18 and 23), it can be seen that both values are equal. That is, both phase directions (delay, lead)
Indicate the same direction.

【０１０１】[0101]

【発明の効果】以上詳述した本発明の位置ずれ検出方法
によれば、光学系の構成やその調整が複雑とならず、広
い検出範囲を持ちながら検出分解能も高く、更に検出す
べき回折光強度も高いものが得られるようになるという
優れた効果を有している。又第３発明法によれば位置ず
れ検出前に光学系の傾きを検出しながらその補正ができ
るため、本法をアライメントに用いればそのアライメン
ト精度を高めることが可能となる。更にいずれの発明と
も入射光の入射角度がより小さくなってその入射点の光
軸からの距離が短くなるため、大気のゆらぎ等影響を受
けにくく、入射光の波長変動が小さくなってその検出精
度を大幅に改善できる様になる。尚、本位置ずれ検出方
法を用いれば、装置の走査・振動状態の測定や試料等の
検出物体の振動の測定を行なう振動検出方法乃至装置に
も適用することが可能となる。According to the position shift detecting method of the present invention described in detail above, the structure of the optical system and its adjustment are not complicated, the detection resolution is high while having a wide detection range, and the diffracted light to be detected further. It has an excellent effect that a material having high strength can be obtained. Further, according to the third invention method, since the correction can be performed while detecting the inclination of the optical system before the positional deviation is detected, the alignment accuracy can be improved by using this method for alignment. Further, in any of the inventions, since the incident angle of the incident light becomes smaller and the distance of the incident point from the optical axis becomes shorter, it is less susceptible to atmospheric fluctuations, etc. Can be greatly improved. It should be noted that the use of this position shift detection method can also be applied to a vibration detection method or device for measuring the scanning / vibration state of the device or measuring the vibration of a detection object such as a sample.

【図面の簡単な説明】[Brief description of drawings]

【図１】光軸左側入射時の回折光回折次数の説明図であ
る。FIG. 1 is an explanatory diagram of diffraction orders of diffracted light when incident on the left side of the optical axis.

【図２】同じく光軸右側入射時の回折光回折次数の説明
図である。FIG. 2 is an explanatory diagram of a diffraction order of diffracted light when the light is incident on the right side of the optical axis.

【図３】θi／２入射角で光軸両側からアライメント光
の入射を行なった時の回折光の回折状態を示す斜視図で
ある。FIG. 3 is a perspective view showing a diffraction state of diffracted light when alignment light is incident from both sides of the optical axis at an incident angle of θi / 2.

【図４】同じくθi入射角で光軸両側からアライメント
光の入射を行なった時の回折光の回折状態を示す斜視図
である。FIG. 4 is a perspective view showing a diffracted state of diffracted light when alignment light is incident from both sides of the optical axis at the θi incident angle.

【図５】検出された第１、第２ビート信号と回折格子位
置ずれの関係を示す説明図である。FIG. 5 is an explanatory diagram showing the relationship between the detected first and second beat signals and the displacement of the diffraction grating position.

【図６】第２発明法の実施構成の一例を示す斜視図であ
る。FIG. 6 is a perspective view showing an example of an implementation configuration of a second invention method.

【図７】本構成における光学系光路の詳細説明図であ
る。FIG. 7 is a detailed explanatory diagram of an optical path of an optical system in this configuration.

【図８】上記構成における変倍光学系でレンズ交換を行
なった時の光路の詳細説明図である。FIG. 8 is a detailed explanatory diagram of an optical path when lenses are exchanged in the variable power optical system having the above configuration.

【図９】瞳面光軸両側にＦb・sinθ離れた位置からアラ
イメント光を入射しフーリエ変換レンズにより回折格子
に該光の照射を行なった時の光ヘテロダイン干渉モデル
を示す説明図である。FIG. 9 is an explanatory diagram showing an optical heterodyne interference model when alignment light is incident on both sides of the optical axis of the pupil plane from positions away from each other by Fb · sin θ and the Fourier transform lens irradiates the light to the diffraction grating.

【図１０】瞳面光軸両側に（Ｆb・sinθ）／２離れた位
置からアライメント光を入射しフーリエ変換レンズによ
り回折格子に該光の照射を行なった時の光ヘテロダイン
干渉モデルを示す説明図である。FIG. 10 is an explanatory diagram showing an optical heterodyne interference model when alignment light is incident on both sides of the optical axis of the pupil plane from positions (Fb · sin θ) / 2 apart and the Fourier transform lens irradiates the light to the diffraction grating. Is.

【図１１】前記図９のアライメント光照射の場合の光路
Ｌ(1)と光路Ｌ(2)間の光路差を示す説明図である。11 is an explanatory diagram showing an optical path difference between an optical path L (1) and an optical path L (2) in the case of the alignment light irradiation of FIG. 9;

【図１２】前記図１０のアライメント光照射の場合の光
路Ｌ(1)と光路Ｌ(2)間の光路差を示す説明図である。12 is an explanatory diagram showing an optical path difference between an optical path L (1) and an optical path L (2) in the case of the alignment light irradiation of FIG.

【図１３】同じく図１０のアライメント光照射の場合の
光路Ｌ(1)と光路Ｌ(2)間の光路差を示す説明図である。13 is an explanatory diagram showing an optical path difference between an optical path L (1) and an optical path L (2) in the case of alignment light irradiation of FIG.

【符号の説明】[Explanation of symbols]

29、30 レンズ 35 フーリエ変換レンズ 36、36a、36b 回折格子 70 瞳面ＥＰ Ｍ マスク Ｗ ウェハ 29, 30 lens 35 Fourier transform lens 36, 36a, 36b Diffraction grating 70 Pupil plane EP M mask W wafer

Claims (3)

【特許請求の範囲】[Claims] 【請求項１】 光軸方向から入射させた場合に取り出さ
れる回折光の±１次方向と該光軸との間で成す角度の半
分の角度であって該光軸に対し対称的な２方向から夫々
逆に第１の物体及び第２の物体の各回折格子に対して周
波数のわずかに異なる２波長の単色光を入射させ、入射
方向逆向きに回折する−１次回折光と、光軸反対側から
入射して正反射し、この−１次回折光と同一方向に回折
する０次回折光とを光軸の両側で且つ両物体の夫々にお
いて取り出した上で、同一方向で取り出された周波数の
異なる−１次と０次の回折光を干渉せしめて第１ビート
信号を光軸の両側において両物体の夫々につき生成せし
めると共に、光軸の両側で取り出された周波数の異なる
−１次の回折光同士を干渉せしめて第２ビート信号を両
物体の夫々で生成せしめ、更に両物体の各第１ビート信
号に基づくその位相差を測定すると共に、両物体の各第
２ビート信号に基づくその位相差を測定することで、両
物体の変位量を求めることを特徴とする位置ずれ検出方
法。1. A two-direction symmetrical with respect to the optical axis that is half the angle between the ± first-order directions of the diffracted light extracted when the light is incident from the optical axis direction and the optical axis. From the opposite, the monochromatic lights of two wavelengths having slightly different frequencies are incident on the diffraction gratings of the first object and the second object, respectively, and the -1st-order diffracted light that is diffracted in the incident direction and the optical axis opposite The -1st-order diffracted light which is incident from the side and is regularly reflected, and the 0th-order diffracted light which is diffracted in the same direction are extracted on both sides of the optical axis and at both objects, respectively, and the frequencies extracted in the same direction are different. -1st order and 0th order diffracted lights are caused to interfere with each other to generate the first beat signal for each of both objects on both sides of the optical axis, and the -1st order diffracted lights with different frequencies extracted on both sides of the optical axis. To generate a second beat signal at each of the two objects. Therefore, the displacement amount of both objects is obtained by further measuring the phase difference based on each first beat signal of both objects and measuring the phase difference based on each second beat signal of both objects. The positional deviation detection method. 【請求項２】 光軸方向から入射させた場合に取り出さ
れる回折光の±１次方向と該光軸との間で成す角度の半
分の角度であって該光軸に対し対称的な２方向から夫々
逆に第１の物体及び第２の物体の各回折格子に対して周
波数のわずかに異なる２波長の単色光を入射させ、入射
方向逆向きに回折する−１次回折光と、光軸反対側から
入射して正反射し、この−１次回折光と同一方向に回折
する０次回折光とを光軸の両側で且つ両物体の夫々にお
いて取り出した上で、同一方向で取り出された周波数の
異なる−１次と０次の回折光を干渉せしめて第１ビート
信号を光軸の両側において両物体の夫々につき生成せし
め、更に両物体の各第１ビート信号に基づくその位相差
を測定すると共に、前記光軸方向から入射させた場合に
取り出される回折光の±１次の方向であって該光軸に対
し対称的な２方向から夫々逆に第１の物体及び第２の物
体の各回折格子に対して前記の２波長の単色光を入射さ
せ、光軸方向に回折した−１次回折光を両物体の夫々に
おいて取り出した上でこれらの回折光を夫々干渉せしめ
て第２ビート信号を両物体の夫々で生成せしめ、更に両
物体の各第２ビート信号に基づくその位相差を測定し、
これらの各位相差の測定で両物体の変位量を夫々求める
ことを特徴とする位置ずれ検出方法。2. An angle half the angle formed between the ± 1st order direction of the diffracted light extracted when incident from the optical axis direction and the optical axis, and two directions symmetrical with respect to the optical axis. From the opposite, the monochromatic lights of two wavelengths having slightly different frequencies are incident on the diffraction gratings of the first object and the second object, respectively, and the -1st-order diffracted light that is diffracted in the incident direction and the optical axis opposite The -1st-order diffracted light which is incident from the side and is regularly reflected, and the 0th-order diffracted light which is diffracted in the same direction are extracted on both sides of the optical axis and at both objects, respectively, and the frequencies extracted in the same direction are different. -1st order and 0th order diffracted light are caused to interfere with each other to generate a first beat signal for each of both objects on both sides of the optical axis, and the phase difference based on each first beat signal of both objects is measured, Diffracted light extracted when incident from the optical axis direction The monochromatic light of the above-mentioned two wavelengths is made to enter the respective diffraction gratings of the first object and the second object in the opposite directions from the ± 1st-order directions and symmetrical to the optical axis. The -1st-order diffracted light diffracted in the axial direction is extracted at each of the two objects, and then the diffracted light is caused to interfere with each other to generate a second beat signal at each of the two objects. And measure its phase difference based on
A positional deviation detection method characterized in that the amount of displacement of both objects is obtained by measuring each of these phase differences. 【請求項３】 請求項第１項乃至第２項記載の位置ずれ
検出方法において、同一方向で取り出された周波数の異
なる−１次と０次の回折光の位相変化量と偏光強度を前
記光軸の両側で比較し、両側の回折光の位相変化量と偏
光強度が等しくなるように該光軸の傾きを補正すること
を特徴とする位置ずれ検出方法。3. The position shift detecting method according to claim 1, wherein the phase change amount and the polarization intensity of the −1st order and 0th order diffracted lights of different frequencies extracted in the same direction are used as the light. A method for detecting a positional deviation, characterized by comparing both sides of an axis and correcting the inclination of the optical axis so that the phase change amount of the diffracted light on both sides and the polarization intensity become equal.