JP5622126B2 - Surface position detection apparatus, exposure apparatus, and device manufacturing method - Google Patents

Description

  The present invention relates to a surface position detection apparatus, an exposure apparatus, and a device manufacturing method. In particular, the present invention relates to photosensitivity in a projection exposure apparatus used to transfer a mask pattern onto a photosensitive substrate in a lithography process for manufacturing a device such as a semiconductor element, a liquid crystal display element, an imaging element, or a thin film magnetic head. The present invention relates to detection of the surface position of a conductive substrate.

  2. Description of the Related Art Conventionally, an oblique incident type surface position detection device disclosed in Japanese Patent Application Laid-Open No. 2001-296105 (Patent Document 1) by the present applicant is known as a surface position detection device suitable for a projection exposure apparatus. In order to increase the detection accuracy of the surface position of the test surface in such an oblique incidence type surface position detection device, it is necessary to increase the incident angle of the light beam to the test surface (close to 90 °). is necessary. In this case, in order to avoid the restriction of the surface to be inspected about the configuration and arrangement of the projection optical system and the condensing optical system of the oblique incident type surface position detection device, In addition, it has been proposed that parallel prisms (hereinafter referred to as “rhombus prisms”) each having a pair of internal reflecting surfaces parallel to each other are arranged to keep the projection optical system and the focusing optical system away from the surface to be measured. (See FIG. 7 of Patent Document 1).

JP 2001-296105 A

  However, in the conventional surface position detection device disclosed in FIG. 7 of Patent Document 1 described above, the relative positional shift due to the polarization component of the light beam totally reflected by the two parallel internal reflection surfaces of the projection-side rhombus prism. May occur, and a clear pattern image may not be formed on the test surface. Similarly, relative positional deviation due to the polarization component also occurs in the light beam reflected from the test surface and totally reflected by the two inner reflection surfaces parallel to each other of the light receiving side rhombus prism, and the pattern secondary image is further blurred. There is a possibility.

  On the other hand, when the conventional surface position detection apparatus is applied to the detection of the surface position of a wafer (photosensitive substrate) whose surface is coated with a resist in the exposure apparatus, the reflectance with respect to light of a specific polarization component is It is known to vary depending on the thickness. As a result, in the conventional surface position detection device, the relative positional shift due to the polarization component generated in the light beam totally reflected by the inner surface reflection surface of the rhombus prism, and the change in reflectance due to the thickness of the resist layer of the photosensitive substrate As a result, a detection error of the surface position of the surface to be detected is likely to occur.

  The present invention has been made in view of the above-described problems, and a relative positional shift due to a polarization component generated in a light beam totally reflected by an inner surface reflection surface of a prism member affects detection of a surface position of a surface to be measured. An object of the present invention is to provide a surface position detection device that can detect the surface position of the surface to be measured with high accuracy while suppressing the influence. Further, the present invention uses a surface position detection device capable of detecting the surface position of the surface to be measured with high accuracy, and uses the surface pattern of the mask and the exposure surface of the photosensitive substrate with high accuracy relative to the projection optical system. An object of the present invention is to provide an exposure apparatus that can be aligned with each other.

  Another object of the present invention is to provide an optical device, a measurement device, and the like that can suppress a relative positional shift due to a polarization component of light generated on a total reflection surface of an optical member. Another object of the present invention is to provide an optical device, an adjustment method, and the like that can adjust a relative positional shift due to a polarization component of light generated on a total reflection surface of an optical member.

In order to solve the above problems, the first aspect of the present invention includes a projection system that projects a light beam from an oblique direction on a test surface, and a light receiving system that receives the light beam reflected by the test surface, In the surface position detection device for detecting the surface position of the test surface based on the output of the light receiving system,
At least one of the projection system and the light receiving system includes a total reflection prism member having an internal reflection surface for totally reflecting an incident light beam,
The total reflection prism member is formed in order to suppress the influence of the relative positional shift due to the polarization component of the light beam totally reflected by the inner reflection surface of the total reflection prism member on the detection of the surface position of the test surface. And a refractive index of the optical material to be set and an incident angle of the incident light beam with respect to the inner reflection surface of the total reflection prism member are set to satisfy a predetermined relationship.

In the second embodiment of the present invention, in an exposure apparatus for projecting and exposing a predetermined pattern onto a photosensitive substrate via a projection optical system,
A surface position detection device of a first form for detecting a surface position of the predetermined pattern surface or an exposure surface of the photosensitive substrate with respect to the projection optical system as a surface position of the test surface;
Alignment means for aligning the predetermined pattern surface or the exposure surface of the photosensitive substrate with respect to the projection optical system based on the detection result of the surface position detection device, An exposure apparatus is provided.

  In the third embodiment of the present invention, using the exposure apparatus of the second embodiment, an exposure process for exposing the photosensitive substrate to the predetermined pattern, and a development process for developing the photosensitive substrate exposed by the exposure process A device manufacturing method is provided.

In the fourth embodiment of the present invention, in an optical device in which an optical member having a total reflection surface is arranged in the optical path,
The refractive index of the optical member and the incident angle of the light with respect to the total reflection surface are set so as to suppress a relative displacement due to the polarization component of the light totally reflected by the total reflection surface of the optical member. An optical device is provided.

In the fifth embodiment of the present invention, the optical path has N internal reflection surfaces,
The incident angle of light with respect to each of the N internal reflection surfaces and the refractive index of each optical member forming the N internal reflection surfaces depend on the polarization component of the light totally reflected by the N internal reflection surfaces. There is provided an optical device characterized in that the relative positional deviation is set to be substantially zero.

In the sixth embodiment of the present invention, in the measurement apparatus in which the optical member having the total reflection surface is arranged in the measurement optical path,
The refractive index of the optical member and the incident angle of the measurement light with respect to the total reflection surface are set so as to suppress a relative positional shift due to the polarization component of the measurement light totally reflected by the total reflection surface of the optical member. A measuring device is provided.

In the seventh embodiment of the present invention, in the measuring device for measuring the test surface,
A detector for detecting measurement light from the test surface;
An optical member including a total reflection surface disposed in an optical path between the test surface and the detector;
In order to suppress the relative displacement due to the polarization component of the measurement light totally reflected by the total reflection surface of the optical member, the refractive index of the optical member and the incident angle of the measurement light with respect to the total reflection surface of the optical member Is provided, and a measuring device is provided.

In the eighth embodiment of the present invention, in the measuring device for measuring the test surface,
A projection system for guiding measurement light to the surface to be examined;
A light receiving system for receiving measurement light from the test surface;
An optical member including a total reflection surface disposed in at least one of the optical path of the projection system and the optical path of the light receiving system;
In order to suppress the relative displacement due to the polarization component of the measurement light totally reflected by the total reflection surface of the optical member, the refractive index of the optical member and the incident angle of the measurement light with respect to the total reflection surface of the optical member Is provided, and a measuring device is provided.

In the ninth aspect of the present invention, the optical path has N internal reflection surfaces,
The incident angle of light with respect to each of the N internal reflection surfaces and the refractive index of each optical member forming the N internal reflection surfaces depend on the polarization component of the light totally reflected by the N internal reflection surfaces. Provided is a measuring device characterized in that the relative positional deviation is set to be substantially zero.

In a tenth aspect of the present invention, in an exposure apparatus that exposes a predetermined pattern on a photosensitive substrate,
Provided is an exposure apparatus comprising a measurement device of a sixth form, a seventh form, an eighth form or a ninth form for measuring the position of the photosensitive substrate.

In an eleventh aspect of the present invention, an exposure step of exposing the predetermined pattern to the photosensitive substrate using the exposure apparatus of the tenth aspect;
And a development step of developing the photosensitive substrate exposed by the exposure step. A device manufacturing method is provided.

  In a twelfth aspect of the present invention, at least one optical member disposed in an optical path and having a total reflection surface, and a relative position by a polarization component of light totally reflected by the total reflection surface of the at least one optical member An optical device including an adjusting device that adjusts a deviation is provided.

  In a thirteenth aspect of the present invention, a relative displacement due to a step of disposing at least one optical member having a total reflection surface in an optical path and a polarization component of light generated on the total reflection surface of the at least one optical member. And an adjusting method including the step of adjusting.

  In the fourteenth aspect of the present invention, the relative misregistration due to the step of guiding light to at least one optical member having a total reflection surface and the polarization component of light generated on the total reflection surface of the at least one optical member is performed. There is provided an adjusting method including a detecting step and a step of adjusting an incident angle of light incident on a total reflection surface of the at least one optical member.

In the surface position detection device according to the typical embodiment of the present invention, the refractive index of the optical material forming the total reflection prism member and the incident angle of the incident light beam with respect to the inner reflection surface are set so as to satisfy a predetermined relationship. As a result, the relative displacement due to the polarization component does not substantially occur in the light beam totally reflected by the inner reflection surface. As a result, a clear primary pattern image is formed on the test surface, and a clear secondary pattern image is formed on the light receiving surface (or its conjugate surface). As a result, the surface position of the test surface is detected with high accuracy. be able to.
Thus, in the surface position detection device of the present invention, the influence of the relative positional shift due to the polarization component generated in the light beam totally reflected by the inner surface reflection surface of the prism member is suppressed on the detection of the surface position of the test surface, The surface position of the test surface can be detected with high accuracy. Therefore, when the surface position detection device of the present invention is applied to the detection of the surface position of the photosensitive substrate with respect to the projection optical system in the exposure apparatus, the surface position of the photosensitive substrate can be detected with high accuracy, and consequently the pattern surface of the mask. And the exposure surface of the photosensitive substrate can be aligned with respect to the projection optical system with high accuracy, and a good device can be manufactured.

It is a figure which shows schematically the structure of the exposure apparatus provided with the surface position detection apparatus concerning embodiment of this invention. It is an optical path diagram which shows that both the projection optical system and condensing optical system of FIG. 1 are both-side telecentric. It is a figure which shows roughly the structure between a pair of pentaprisms in the surface position detection apparatus of this embodiment. It is a perspective view which shows the state in which the primary image of the lattice pattern 3a is formed on the to-be-tested surface Wa. It is a figure which shows roughly the structure of the light-receiving slit S which has the rectangular-shaped opening parts Sa1-Sa5 elongated in the X direction. 5 is a diagram showing a state in which five silicon photodiodes PD1 to PD5 are provided on the light receiving surface 14a of the light receiving unit 14 so as to optically correspond to the openings Sa1 to Sa5 of the light receiving slit S. FIG. It is a figure which shows roughly a mode that the relative position shift by a polarization component generate | occur | produces in the light beam totally reflected by the inner surface reflective surface of the prism. It is a figure which shows the relationship between relative displacement amount (DELTA) (GHS) by the polarization component which generate | occur | produces in total reflection, and the measured value of AF surface. It is the elements on larger scale of FIG. It is a figure which shows the relationship between the incident angle to a total reflection surface, and the relative position shift amount by a polarization component. It is a figure which shows the relationship between the refractive index of a prism, and the relative position shift amount by a polarization component. It is a figure which shows roughly the principal part structure of the modification of this embodiment. It is a figure explaining the range of incident angle (theta) when the absolute value of relative positional offset amount (DELTA) is 0.3 micrometer or less when the refractive index n of a total reflection prism member is 1.45. It is a figure explaining the range of incident angle (theta) from which the absolute value of relative displacement amount (DELTA) is 0.3 micrometer or less when the refractive index n of a total reflection prism member is 1.5. It is a figure explaining the range of incident angle (theta) when the absolute value of relative positional offset amount (DELTA) is 0.3 micrometer or less when the refractive index n of a total reflection prism member is 1.6. It is a figure explaining the range of incident angle (theta) when the absolute value of relative displacement amount (DELTA) is 0.3 micrometer or less when the refractive index n of a total reflection prism member is 1.7. It is a figure which shows the example which applied this invention to the apparatus which measures the surface shape of a to-be-tested surface. It is a figure which shows the example which applied this invention to the apparatus which measures the position of a to-be-tested surface. It is a figure explaining the view which suppresses the relative positional deviation amount finally generate | occur | produced in the light which used the quadrilateral prism as a total reflection prism member, and was sequentially totally reflected by two total reflection surfaces. FIG. 14 is an enlarged view of a part of FIG. 13, illustrating a range of an incident angle θ in which the absolute value of the relative positional deviation amount Δ is 0.05 μm or less. FIG. 15 is an enlarged view of a part of FIG. 14, illustrating a range of an incident angle θ in which the absolute value of the relative positional deviation amount Δ is 0.05 μm or less. FIG. 16 is an enlarged view of a part of FIG. 15 and is a diagram for explaining a range of an incident angle θ in which an absolute value of a relative positional deviation amount Δ is 0.05 μm or less. FIG. 17 is an enlarged view of a part of FIG. 16, illustrating a range of an incident angle θ in which the absolute value of the relative positional deviation amount Δ is 0.05 μm or less. It is a flowchart of the method at the time of obtaining the semiconductor device as a microdevice. It is a flowchart of the method at the time of obtaining the liquid crystal display element as a microdevice.

  Embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 1 is a drawing schematically showing a configuration of an exposure apparatus provided with a surface position detection apparatus according to an embodiment of the present invention. FIG. 2 is an optical path diagram showing that the projection optical system and the condensing optical system of FIG. 1 are both telecentric. FIG. 3 is a diagram schematically showing a configuration between a pair of pentaprisms in the surface position detection apparatus of the present embodiment.

  In FIG. 1 and FIG. 2, the illustration of the configuration between the pair of pentaprisms 6 and 9 is omitted for the sake of clarity. In FIG. 1, the Z axis is set parallel to the optical axis AX of the projection optical system PL, the Y axis is set parallel to the paper surface of FIG. 1 in the plane perpendicular to the optical axis AX, and the X axis is set perpendicular to the paper surface of FIG. doing. In the present embodiment, the surface position detection device of the present invention is applied to the detection of the surface position of the photosensitive substrate in the projection exposure apparatus.

  The illustrated exposure apparatus includes an illumination system IL for illuminating a reticle R as a mask on which a predetermined pattern is formed with illumination light (exposure light) emitted from an exposure light source (not shown). The reticle R is held in parallel with the XY plane on a reticle stage (not shown) via a reticle holder RH. The reticle stage can be moved two-dimensionally along the reticle plane (that is, the XY plane) by the action of a drive system (not shown), its position coordinates are measured by a reticle interferometer (not shown), and position control is performed. It is configured to be.

  The light from the pattern formed on the reticle R forms a reticle pattern image on the surface (exposure surface) Wa of the wafer W, which is a photosensitive substrate, via the projection optical system PL. The wafer W is placed on the wafer holder 21, and the wafer holder 21 is supported by a holder holding mechanism 22. The holder holding mechanism 22 is controlled by three support points 22a to 22c (only two support points 22a and 22b are shown in FIG. 1) that can move in the vertical direction (Z direction) based on the control of the holder drive unit 23. Wafer holder 21 is supported.

  In this way, the holder driving unit 23 controls the vertical movement of the support points 22a to 22c of the holder holding mechanism 22, respectively, thereby leveling (leveling) and Z-direction (focusing direction) movement of the wafer holder 21 and eventually the wafer W. Leveling and movement in the Z direction. The wafer holder 21 and the holder holding mechanism 22 are further supported by a wafer stage (not shown). The wafer stage can be moved two-dimensionally along the wafer surface (that is, the XY plane) and can be rotated around the Z axis by the action of a drive system (not shown), and its position coordinate is a wafer interferometer (not shown). ) And the position is controlled.

  Here, in order to satisfactorily transfer the circuit pattern provided on the pattern surface of the reticle R to each exposure region of the exposure surface Wa of the wafer W, image formation by the projection optical system PL is performed for each exposure to each exposure region. It is necessary to align the current exposure area of the exposure surface Wa within the width of the focal depth centered on the surface. For that purpose, after accurately detecting the surface position of each point in the current exposure region, that is, the surface position along the optical axis AX of the projection optical system PL, the exposure surface Wa is within the range of the width of the depth of focus of the projection optical system PL. The wafer holder 21 may be leveled and moved in the Z direction so that the wafer holder 21 is accommodated, and the wafer W may be leveled and moved in the Z direction.

  The projection exposure apparatus of the present embodiment includes a surface position detection device for detecting the surface position of each point in the current exposure region of the exposure surface Wa. Referring to FIG. 1, the surface position detection apparatus of the present embodiment includes a light source 1 for supplying detection light. In general, the surface Wa of the wafer W, which is the test surface, is covered with a thin film such as a resist. Therefore, in order to reduce the influence of interference due to the thin film, the light source 1 is a white light source having a wide wavelength width (for example, a halogen lamp that supplies illumination light having a wavelength width of 600 to 900 nm, or illumination having a wide band similar to this. A xenon light source that supplies light is desirable. As the light source 1, a light emitting diode that supplies light in a wavelength band with low sensitivity to a resist can be used.

  The divergent light beam from the light source 1 is converted into a substantially parallel light beam through the condenser lens 2 and then enters the deflecting prism 3. The deflecting prism 3 deflects the substantially parallel light beam from the condenser lens 2 along the −Z direction by refraction. Further, on the exit side of the deflecting prism 3, a transmission type grating pattern 3a is formed in which elongated transmission portions extending in the X direction and elongated light shielding portions extending in the X direction are alternately provided at a constant pitch. Instead of the transmission type grating pattern, an uneven reflection type diffraction grating may be applied, or a reflection type grating pattern in which reflection portions and non-reflection portions are alternately formed may be applied.

  The light transmitted through the transmission type grating pattern 3a is incident on the projection optical system (4, 5) arranged along the optical axis AX1 parallel to the optical axis AX of the projection optical system. The projection optical system (4, 5) is composed of a projection condenser lens 4 and a projection objective lens 5. The light beam that has passed through the projection optical system (4, 5) enters the pentaprism 6. The pentaprism 6 is a pentagonal prism-shaped deflecting prism whose longitudinal axis extends along the X direction. The pentaprism 6 has a first transmission surface 6a for transmitting incident light as it is without being refracted along the optical axis AX1. Have. That is, the first transmission surface 6a is set perpendicular to the optical axis AX1.

  The light transmitted through the first transmission surface 6a and propagating through the pentaprism 6 along the optical axis AX1 is reflected by the first reflection surface 6b and then again by the second reflection surface 6c along the optical axis AX2. Reflected. The light reflected by the second reflecting surface 6c and propagating through the pentaprism 6 along the optical axis AX2 is transmitted without being refracted by the second transmitting surface 6d. That is, the second transmission surface 6d is set perpendicular to the optical axis AX2. Here, the pentaprism 6 is formed of a low thermal expansion and low dispersion optical material such as quartz glass, and a reflective film made of aluminum or silver is formed on the first reflective surface 6b and the second reflective surface 6c. Yes.

  Thus, the light incident in the −Z direction along the optical axis AX1 is largely deflected by the pentaprism 6 and guided to the test surface Wa along the optical axis AX2. At this time, the direction of the optical axis AX2 is set so that the incident angle to the surface Wa is sufficiently large, and the deflection angle by the pentaprism 6 is set. Specifically, as shown in FIG. 3, the light beam emitted from the pentaprism 6 along the optical axis AX <b> 2 enters the projection-side rhombus prism 7.

  The rhombus prism 7 is a quadrangular prism having a parallelogram (or rhombus) cross section, and its longitudinal axis is arranged along the X direction in the same manner as the pentaprism 6. In the rhombus prism 7, the light transmitted through the first transmission surface 7a perpendicular to the optical axis AX2 is sequentially reflected by the pair of reflection surfaces 7b and 7c parallel to each other, and then the second transmission parallel to the first transmission surface 7a. The light passes through the surface 7d and exits from the rhombus prism 7 along an optical axis AX21 parallel to the optical axis AX2. The light beam emitted from the rhombus prism 7 along the optical axis AX21 is incident on the test surface Wa.

  Here, in a state where the test surface Wa coincides with the image formation surface of the projection optical system PL, the projection optical system (4, 5) is connected to the surface on which the grating pattern 3a is formed (that is, the exit surface of the deflection prism 3) and the test surface. The inspection surface Wa is arranged in a conjugate manner. Further, the surface on which the lattice pattern 3a is formed and the test surface Wa are configured so as to satisfy the Scheinproof condition for the projection optical system (4, 5). As a result, the light from the grating pattern 3a is accurately imaged over the entire pattern image forming surface on the surface Wa to be examined via the projection optical system (4, 5).

  In addition, as shown by the broken line in FIG. 2, the projection optical system (4, 5) composed of the projection condenser lens 4 and the projection objective lens 5 is a so-called bilateral telecentric optical system. Therefore, each point on the surface on which the lattice pattern 3a is formed and each conjugate point on the test surface Wa have the same magnification over the entire surface. In this way, the primary image of the lattice pattern 3a is accurately formed over the entire surface Wa as shown in FIG.

  Referring again to FIG. 1, the light beam reflected by the test surface Wa along the optical axis AX31 symmetric with respect to the optical axis AX21 with respect to the optical axis AX of the projection optical system PL is incident on the light receiving side rhombus prism 8. Like the rhombus prism 7, the rhombus prism 8 is a quadrangular prism having a longitudinal axis along the X direction and a parallelogram (or rhombus) cross section. Therefore, in the rhombus prism 8, the light transmitted through the first transmission surface 8a perpendicular to the optical axis AX31 is sequentially reflected by the pair of reflection surfaces 8b and 8c parallel to each other, and then the first prism parallel to the first transmission surface 8a. 2 is transmitted through the transmission surface 8d and is emitted from the rhombus prism 8 along the optical axis AX3 parallel to the optical axis AX31.

  The light emitted from the rhombus prism 8 along the optical axis AX3 enters the condensing optical system (10, 11) through the pentaprism 9 having the same configuration as the pentaprism 6 described above. That is, the light reflected by the test surface Wa enters the pentaprism 9 along an optical axis AX3 that is symmetric with respect to the optical axis AX2 with respect to the optical axis AX of the projection optical system PL. In the pentaprism 9, light transmitted through the first transmission surface 9a perpendicular to the optical axis AX3 is sequentially reflected by the first reflection surface 9b and the second reflection surface 9c, and then along the optical axis AX4 extending in the Z direction. It reaches the second transmission surface 9d. The light transmitted through the second transmission surface 9d perpendicular to the optical axis AX4 enters the condensing optical system (10, 11) in the + Z direction along the optical axis AX4.

  The condensing optical system (10, 11) includes a light receiving objective lens 10 and a light receiving condensing lens 11. In the optical path between the light-receiving objective lens 10 and the light-receiving condenser lens 11, a vibrating mirror 12 is provided as scanning means. Therefore, the light incident on the light receiving objective lens 10 along the optical axis AX4 is deflected via the vibrating mirror 12 and reaches the light receiving condenser lens 11 along the optical axis AX5. In the present embodiment, the oscillating mirror 12 is disposed at a substantially pupil plane position of the condensing optical system (10, 11). However, the present invention is not limited to this, and the test surface Wa and tilt correction described later are performed. It can be arranged at an arbitrary position in the optical path between the prism 13 or in the optical path between the surface Wa to be measured and the deflecting prism 3.

  The light passing through the condensing optical system (10, 11) is incident on the tilt correction prism 13 having the same configuration as that of the deflection prism 3 described above. Here, in a state where the test surface Wa coincides with the imaging surface of the projection optical system PL, the condensing optical system (10, 11) conjugates the test surface Wa and the incident surface 13a of the tilt correction prism 13. It is comprised so that it may arrange. Thus, a secondary image of the grating pattern 3 a is formed on the incident surface 13 a of the tilt correction prism 13.

  A light receiving slit S as a light shielding means is provided on the incident surface 13a of the tilt correcting prism 13. As shown in FIG. 5, the light receiving slit S includes, for example, five rectangular openings Sa <b> 1 to Sa <b> 5 that are elongated in the X direction. Reflected light from the surface Wa to be measured through the condensing optical system (10, 11) passes through the openings Sa1 to Sa5 of the light receiving slit S and enters the tilt correction prism 13.

  Here, the number of the openings Sa of the light receiving slit S corresponds to the number of detection points on the test surface Wa. That is, in FIG. 4 showing a state where the primary image of the lattice pattern 3a is formed on the test surface Wa, the detection points (detection areas) Da1 to Da5 on the test surface Wa are light receiving slits S shown in FIG. Optically correspond to the five openings Sa1 to Sa5. Therefore, when it is desired to increase the number of detection points on the test surface Wa, it is only necessary to increase the number of openings Sa, and even if the number of detection points is increased, the configuration is not complicated.

  The imaging surface formed by the projection optical system PL and the incident surface 13a of the tilt correction prism 13 are configured to satisfy the Scheinproof condition for the condensing optical system (10, 11). Therefore, in a state in which the test surface Wa and the imaging surface coincide with each other, the light from the grating pattern 3a passes through the condensing optical system (10, 11) and the pattern image forming surface on the prism incident surface 13a. The image is accurately re-imaged throughout.

  In addition, as shown in FIG. 2, the optical path is indicated by a broken line, and the condensing optical system (10, 11) is composed of a bilateral telecentric optical system. Therefore, each point on the test surface Wa and each conjugate point on the prism incident surface 13a have the same magnification over the entire surface. Thus, a secondary image of the grating pattern 3a is accurately formed on the incident surface 13a of the tilt correction prism 13 over the entire surface.

  By the way, when the light receiving surface is arranged at the position of the incident surface 13a of the tilt correction prism 13, the incident angle θ of the light beam with respect to the test surface Wa is large, so that the incident angle of the light beam on the light receiving surface is also large. In this case, for example, if a silicon photo diode is disposed on the light receiving surface, the incident angle of the light beam on the silicon photo diode is increased, so that the surface reflection at the silicon photo diode is increased and the light beam is distorted. As a result, the amount of received light may be significantly reduced.

  In the present embodiment, in order to avoid a decrease in the amount of light received due to the incident angle of the light beam on the light receiving surface, as shown in FIG. 1, a conjugate surface with the surface Wa to be measured related to the condensing optical system (10, 11) An incident surface 13a of a tilt correction prism 13 as a deflection optical system is disposed. As a result, the light beam incident on the incident surface 13a of the tilt correction prism 13 along the optical axis AX5 via the condensing optical system (10, 11) is converted into an apex angle (of the incident surface and the exit surface) of the tilt correction prism 13. Is deflected according to the same refraction angle as the angle formed, and exits from the exit surface 13b along the optical axis AX6. Here, the exit surface 13b is set perpendicular to the optical axis AX6.

  Light emitted from the exit surface 13b of the tilt correction prism 13 along the optical axis AX6 enters a relay optical system (14a, 14b) including a pair of lenses 14a and 14b. The light passing through the relay optical system (14a, 14b) is a conjugate image of the secondary image of the grating pattern 3a formed on the incident surface 13a of the tilt correction prism 13 and the openings Sa1 to Sa5 of the light receiving slit S. It is formed on the light receiving surface 15 a of the light receiving unit 15. As shown in FIG. 6, five silicon photodiodes PD1 to PD5 are provided on the light receiving surface 15a so as to optically correspond to the openings Sa1 to Sa5 of the light receiving slit S. In place of the silicon photodiode, a CCD (two-dimensional charge coupled image sensor) or a photomultiplier can be used.

  Thus, in the present embodiment, the tilt correction prism 13 as the deflection optical system is used, so that the incident angle of the light beam incident on the light receiving surface 15a is sufficiently small, which is caused by the incident angle of the light beam on the light receiving surface 15a. A decrease in the amount of received light is avoided. The relay optical system (14a, 14b) is preferably a double-sided telecentric optical system as shown in FIG. Further, it is desirable that the incident surface 13a and the light receiving surface 15a of the tilt correction prism 13 are configured so as to satisfy the Scheimpflug condition regarding the relay optical system (14a, 14b).

  As described above, the light receiving slit S having the five openings Sa1 to Sa5 is provided on the incident surface 13a of the tilt correcting prism 13. Therefore, the secondary image of the lattice pattern 3a formed on the incident surface 13a is partially shielded through the light receiving slit S. That is, only the light beam from the secondary image of the lattice pattern 3a formed in the area of the openings Sa1 to Sa5 of the light receiving slit S is received through the tilt correction prism 13 and the relay optical system (14a, 14b). 15a is reached.

  Thus, as shown in FIG. 6, on the silicon photodiodes PD1 to PD5 disposed on the light receiving surface 15a of the light receiving portion 15, images of the openings Sa1 to Sa5 of the light receiving slit S, that is, slit images SL1 to SL5. Are formed respectively. The slit images SL1 to SL5 are set so as to be formed inside the rectangular light receiving regions of the silicon photodiodes PD1 to PD5, respectively.

  Here, when the test surface Wa moves up and down in the Z direction along the optical axis AX of the projection optical system PL, a secondary image of the grating pattern 3a formed on the incident surface 13a of the tilt correction prism 13 is detected. Corresponding to the vertical movement of the surface Wa, a lateral shift occurs in the pitch direction of the pattern. In the present embodiment, for example, the lateral shift amount of the secondary image of the lattice pattern 3a is detected by the principle of the photoelectric microscope disclosed in Japanese Patent Application Laid-Open No. 6-97045 by the present applicant, and projection optics is used based on the detected lateral shift amount. The surface position of the test surface Wa along the optical axis AX of the system PL is detected.

  A mirror driving unit 16 for driving the oscillating mirror 12, a position detecting unit 17 for synchronously detecting detection signals from the silicon photodiodes PD1 to PD5 based on an AC signal from the mirror driving unit 16, a surface to be detected A correction amount calculation unit 18 that calculates an inclination correction amount and a Z direction correction amount necessary to fit Wa within the range of the focal depth of the projection optical system PL, and a holder holding mechanism 22 based on the inclination correction amount and the Z direction correction amount. Since the operation of the holder driving unit 23 for controlling the driving of the wafer holder 21 to perform leveling and movement in the Z direction is the same as that of the apparatus disclosed in Japanese Patent Laid-Open No. 2001-296105 by the present applicant, Then, explanation is omitted.

  Japanese Patent Laid-Open No. 6-97045 discloses in detail the conditions of Scheimpflug, the configuration and operation of the deflection prism 3 and the tilt correction prism 13, and the specific application of the principle of the photoelectric microscope. The configuration and operation of the pentaprisms 6 and 9 are disclosed in detail in Japanese Patent Application Laid-Open No. 2001-296105. A configuration in which one or both of these pentaprisms 6 and 9 are omitted is also possible.

  In the present embodiment, the pentaprism 6 and the optical path between the projection optical system (4, 5) and the test surface Wa and the optical path between the condensing optical system (10, 11) and the test surface Wa. 9 is provided, and the optical path of the incident light beam to the test surface Wa and the optical path of the reflected light beam from the test surface Wa are largely bent by the action of the pentaprisms 6 and 9, and the projection optical system (4, 5) and the condensing light are collected. The optical system (10, 11) is sufficiently distant from the test surface Wa. As a result, the configuration and arrangement of the projection optical system (4, 5) and the condensing optical system (10, 11) are not substantially restricted by the surface Wa.

  In the present embodiment, the rhombus prisms 7 and 8 are attached in the optical path between the pentaprism 6 and the test surface Wa and in the optical path between the pentaprism 9 and the test surface Wa, respectively. The optical path of the incident light beam to the test surface Wa and the optical path of the reflected light beam from the test surface Wa are moved in parallel by the action of the rhombus prisms 7 and 8, respectively. As a result, the pair of pentaprisms 6 and 9 can be moved away from the test surface Wa, and the configuration and arrangement of the pair of pentaprisms 6 and 9 and their holding members are substantially restricted by the test surface Wa. Absent.

  In the surface position detection apparatus of the present embodiment, a projection-side prism member, that is, a rhombus prism 7, which is disposed in the optical path of the projection system and has a pair of internal reflection surfaces (7 b, 7 c) for translating the optical path of the incident light beam, The light receiving system is disposed so as to correspond to the projection-side prism member 7 in the light path of the light receiving system, and has a pair of internal reflecting surfaces (8b, 8c) for translating the optical path of the incident light beam from the surface Wa to be detected. A side prism member, that is, a rhombus prism 8 is provided. In this case, as described above, a relative positional shift due to the polarization component occurs in the light beam totally reflected by the two parallel internal reflection surfaces (7b, 7c) of the rhombus prism 7 on the projection side, and a clear pattern is obtained. An image is not formed on the test surface Wa. In addition, since the test surface side of the projection optical system (4, 5) and the test surface side of the condensing optical system (10, 11) are configured to be telecentric, the total reflection surface of the rhombus prism (7, 8). All chief rays incident on (7b, 7c, 8b, 8c) have the same incident angle.

  FIG. 7 is a diagram schematically showing a state in which a relative positional shift due to a polarization component occurs in a light beam (principal ray passing on the optical axis) totally reflected by the inner reflection surface of the prism. As shown in FIG. 7, the light L propagating through the prism 70 and entering the inner reflection surface 71 at an incident angle larger than a predetermined value is totally reflected by the inner reflection surface 71. At this time, among the light beams totally reflected by the inner surface reflection surface 71, the light Ls in the S polarization state having the polarization direction in the direction perpendicular to the paper surface of FIG. 7 and the polarization direction in the direction parallel to the paper surface of FIG. The light Lp in the P-polarized state propagates through the prism 70 along two parallel optical paths separated by a distance Δ.

  Here, the relative positional shift amount Δ due to the polarization component generated in the light beam totally reflected by the inner reflection surface 71 of the prism 70 is called a Goos-Haenchen Shift. In this way, P-polarized light with respect to the test surface Wa out of the light flux reaching the test surface Wa due to total reflection on the two parallel internal reflection surfaces (7b, 7c) of the rhombus prism 7 on the projection side. A relative positional deviation occurs between the S-polarized light and the S-polarized light, and as a result, a pattern image formed by the P-polarized light on the surface Wa and a pattern image formed by the S-polarized light on the surface Wa. Relative displacement occurs between the two.

  Similarly, a relative positional shift due to the polarization component also occurs in the light beam reflected from the test surface Wa and totally reflected by the two parallel internal reflecting surfaces (8b, 8c) of the rhombus prism 8 on the light receiving side. The pattern secondary image formed on the incident surface 13a of the tilt correction prism 13 is further blurred. In other words, the pattern secondary image formed by the P-polarized light on the incident surface 13a and the S-polarized light are incident due to the effect of total reflection on the inner reflection surfaces (8b, 8c) of the rhombus prism 8 on the light receiving side. A relative positional shift generated between the pattern secondary image formed on the surface 13a is promoted (doubled).

  The surface position detection apparatus according to the present embodiment has a wafer W having various surface states during the semiconductor exposure process (for example, a plurality of substances constituting a structure on the wafer W, or a structure on the wafer W itself (multilayer). This is applied to the surface position detection of a plurality of types of structures). In general, a resist is applied to the wafer surface. Under such circumstances, when there are variations in various surface conditions (for example, when the thickness of the layer formed on the wafer varies or the properties such as the purity of the material forming the layer vary), the resist thickness When there is a variation, the reflectance with respect to light of a specific polarization component (for example, P-polarized light, S-polarized light, etc.) varies depending on these variations.

  As a result, in the surface position detection apparatus of the present embodiment, unless special measures are taken, the surface position detection device generates a light beam totally reflected by the inner surface reflection surfaces (7b, 7c; 8b, 8c) of the rhombus prism (7; 8). Due to the relative positional deviation due to the polarization component and the reflectance change of the specific polarization component caused by the variation in the surface state of the wafer W and the variation in the resist thickness, an error in detecting the surface position of the surface Wa is likely to occur. .

  In recent years, with the miniaturization of the projection exposure pattern, the demand for the flatness of the wafer surface has become stricter, and the demand for the surface position detection accuracy has become very high. In addition, in an exposure apparatus using an ArF excimer laser light source, the thickness of the resist on the surface tends to be thin, and the surface position detection error due to the various surface states and variations in resist thickness cannot be ignored.

  Therefore, in this embodiment, the refractive index n of the optical material forming the rhombus prism (7; 8) as the total reflection prism member, and the incident light beam (optical axis) with respect to the inner reflection surface (7b, 7c; 8b, 8c). By setting so that the incident angle θ of the chief ray traveling along the line satisfies a predetermined relationship, the internal reflection surface (7b, 7c; 8b, 8c) of the rhombus prism (7; 8) is totally reflected. The influence of the relative position shift due to the polarization component of the light beam on the detection of the surface position of the surface Wa is suppressed. Hereinafter, the refractive index n of the optical material forming the rhombus prism (7; 8) and the incident angle θ of the incident light beam (principal ray along the optical axis) with respect to the internal reflection surfaces (7b, 7c; 8b, 8c) Explain the relationship to be satisfied.

  Now, in each internal reflection surface (7b, 7c, 8b, 8c) as a total reflection surface, the phase changes φp and φs in the components of P-polarized light and S-polarized light are expressed as follows. (0 ° ≦ θ ≦ 90 °), where λ is the wavelength of light and n is the refractive index of an optical material such as glass, these are expressed by the following equations (1) and (2).

Here, in the reflected light of the P-polarized component, a relative displacement Δp (P-polarized component Goose Henschen shift (GHS)) with respect to the reference reflection position P ₀ of the P-polarized component in the direction along the reflecting surface. .) Is a partial differential ∂φ / ∂β with β of the phase change φ of the P-polarized component, where β = n / λ · cos θ is the phase change of the P-polarized component. Further, in the reflected light of the S-polarized component, the relative displacement Δs (referred to as the S-polarized component Goose Henschen shift (GHS)) with respect to the reference reflection position P _{0 in} the direction along the reflection surface. ) Is a partial differential ∂φ / ∂β with β of the phase change φ of the S polarization component, where β = n / λ · cos θ is the phase change of the S polarization component.

  The relative positional shift amounts Δp, Δs of the reflected light of each polarization component are expressed by the following equations (3) and (4).

From these, when the difference Δ (Goose Henschen shift (GHS)) of the relative phase shift amount in the vertical direction between the light of the P-polarized component and the light of the S-polarized component in the reflected light is obtained, it is along the reflective surface. Since the relative phase shift amount GHS between the reflected lights of the respective polarization components in the direction is Δp−Δs, the relationship of the following expression (5) is established.
Δ = cos θ × (Δp−Δs) (5)

  For this reason, as shown in the following formula (6), there is a difference between the optical axes of P-polarized light and S-polarized light by Δ.

  Therefore, the relative positional shift amount Δ due to the polarization component generated in one total reflection (internal reflection) inside the prism (7; 8) can be expressed by the above equation (6). Therefore, the condition that the relative displacement due to the polarization component does not occur in total reflection, that is, the condition that the relative displacement Δ (Goose Henschen shift (GHS)) due to the polarization component generated in total reflection becomes zero. The following equation (7) is satisfied.

  In the present embodiment, the refractive index n of the optical material forming the rhombus prism (7; 8) as the total reflection prism member, and the incident angle θ of the incident light beam with respect to the inner reflection surface (7b, 7c; 8b, 8c) Is set so as to substantially satisfy the relationship shown in the equation (7), the light beam totally reflected by the inner surface reflecting surfaces (7b, 7c; 8b, 8c) of the rhombus prism (7; 8) is relative to the polarization component. No substantial misalignment occurs. As a result, a clear primary pattern image is formed on the surface Wa of the wafer W, which is the test surface, and a clear secondary image is formed on the incident surface 13a of the tilt correction prism 13, which in turn is formed on the test surface Wa. The surface position can be detected with high accuracy.

  Therefore, in the exposure apparatus of the present embodiment, the surface position of the exposure surface Wa of the wafer (photosensitive substrate) W is highly accurate using the surface position detection apparatus that can detect the surface position of the test surface with high precision. As a result, the pattern surface of the reticle (mask) R and the exposure surface Wa of the wafer W can be aligned with respect to the projection optical system PL with high accuracy.

  When the surface position detection apparatus of the present embodiment is applied to the detection of the surface position of the photosensitive substrate or the detection of the surface position of the mask in the exposure apparatus, the relative position due to the polarization component generated in one total reflection. It is practically preferable to suppress the magnitude (absolute value) of the deviation amount Δ within, for example, 0.3 μm. That is, when the following conditional expression (8) is satisfied, there is no practical problem with respect to detection of the surface position of the photosensitive substrate and detection of the surface position of the mask. However, in the equation (8), the unit of the wavelength λ of light is μm.

  Here, it will be described that it is practically preferable that the relative positional deviation amount Δ (Goose Henschen shift (GHS)) due to the polarization component generated in total reflection is 0.3 μm or less. The accuracy required for focus detection AF (surface detection, focus measurement) is required to be small with respect to the depth of focus of the projection optical system PL, but a recent large NA projection lens (depth of focus: about 300 nm). For the following, in consideration of other errors, the width of the AF error due to the resist film thickness is not allowed to be 50 nm or more at present. FIG. 8 shows how the AF measurement value changes with respect to the change in resist film thickness when the relative positional deviation amount Δ due to the polarization component is 0.3 μm. It can be seen that the AF measurement value changes by about 250 nm in width.

  FIG. 9 is an enlarged view of a part of FIG. The resist film thickness varies within about ± 10 nm depending on the performance of the film forming machine. For this reason, when the case where the resist film thickness changes by 20 nm in width in FIG. 9 is read from the drawing, it can be seen that the AF measurement value changes by about 50 nm in the worst case. Therefore, it is understood that it is preferable to set the relative positional deviation amount Δ due to the polarization component to 0.3 μm or less. Furthermore, in order to realize more stable detection while achieving high accuracy, it is more preferable to set the relative positional deviation amount Δ due to the polarization component to 0.2 μm or less as shown in the following equation (9). Even more preferred.

  Here, it will be described that it is practically preferable that the amount of relative positional deviation Δ (Goose Henschen shift (GHS)) due to the polarization component generated in total reflection is 0.2 μm or less. The change in the AF measurement value is proportional to the relative positional deviation amount Δ due to the polarization component. When the relative positional deviation amount Δ due to the polarization component is 0.2 μm or less, the change amount in the AF measurement value is Is about 30 nm. If the error is 30 nm, a margin of about 20 nm can be made with respect to the minimum required amount of 50 nm, and this amount corresponds to an AF measurement error when chromatic aberration due to a manufacturing error occurs in the AF optical system. Therefore, even if an error due to chromatic aberration occurs due to a manufacturing error, if the relative positional deviation amount Δ due to the polarization component is 0.2 μm or less, stable measurement with a high accuracy can always be realized.

  Hereinafter, according to a more specific form, the refractive index n of the optical material forming the rhombus prism (7; 8) and the incident angle θ of the incident light beam with respect to the inner reflection surfaces (7b, 7c; 8b, 8c) should be satisfied. Think about the relationship. FIG. 10 is a diagram illustrating the relationship between the incident angle on the total reflection surface and the relative positional shift amount due to the polarization component. In FIG. 10, the vertical axis represents the relative positional deviation amount Δ (μm) due to the polarization component, and the horizontal axis represents the incident angle θ (degrees) to the total reflection surface.

  The relative positional deviation amount Δ is determined when the totally reflected S-polarized light Ls and the P-polarized light Lp are in the positional relationship of FIG. 7 (the S-polarized light Ls is more than the P-polarized light Lp). If the light is in the inner side), the light Lp in the P-polarized state is more negative than the light Ls in the S-polarized state. This also applies to FIGS. 11 and 13 to 16 described later.

  FIG. 10 shows the incident angle θ and the relative positional deviation when the refractive index n of the optical material (for example, quartz) forming the prism is 1.45 and the center wavelength λc of the used light (detection light) is 750 nm. The relationship with Δ is shown. Referring to FIG. 10, total reflection starts when the incident angle θ is about 43 degrees, and the relative positional deviation amount Δ, which is a positive value, monotonously decreases as the incident angle θ becomes larger than the total reflection angle. Eventually, when the incident angle θ reaches about 53 degrees, the relative positional deviation amount Δ becomes almost zero, and as the incident angle θ further increases from about 53 degrees, the relative positional deviation amount Δ becomes a negative value and its magnitude. Increases monotonically.

  In this case, when the center wavelength λc of the used light is changed, the value of the relative positional deviation amount Δ with respect to the same incident angle θ changes. However, as is apparent with reference to the above equations (6) and (7), even when the center wavelength λc of the used light is changed, the value of the incident angle θ when the relative positional deviation amount Δ becomes 0 is Does not change substantially. In other words, the value of the incident angle θ when the relative positional deviation amount Δ is 0 does not substantially depend on the center wavelength λc of the used light.

  Thus, in the present embodiment, when the rhombus prism (7; 8) is formed of an optical material having a predetermined refractive index n, the light beam totally reflected by the inner reflection surface (7b, 7c; 8b, 8c). So that the relative misregistration amount Δ due to the polarized light component becomes substantially zero, that is, the absolute value of the relative misregistration amount Δ is, for example, 0.3 μm or less, more preferably 0.2 μm or less. The incident angle θ of the incident light beam with respect to the inner reflection surface (7b, 7c; 8b, 8c) may be determined. Referring to FIG. 10, it can be seen that the degree of change in the relative positional deviation amount Δ with respect to the change in the incident angle θ is small, and the influence of the mounting error of the rhombus prism (7; 8) on the detection accuracy is small.

  FIG. 11 is a diagram showing the relationship between the refractive index of the prism and the amount of relative positional deviation due to the polarization component. In FIG. 11, the vertical axis represents the relative positional deviation amount Δ (μm) due to the polarization component, and the horizontal axis represents the refractive index n of the optical material forming the prism. FIG. 11 shows the relationship between the refractive index n and the relative positional deviation amount Δ when the incident angle θ to the total reflection surface is 45 degrees and the center wavelength λc of the used light is 750 nm. Referring to FIG. 11, as the refractive index n increases from 1.45, the relative positional deviation amount Δ, which is a positive value, monotonously decreases. When the refractive index n reaches about 1.73, the relative positional deviation amount Δ is It becomes almost zero. As the refractive index n further increases from about 1.73, the relative positional deviation amount Δ becomes a negative value, and the magnitude increases monotonously.

  Also in this case, when the center wavelength λc of the used light is changed, the value of the relative positional deviation amount Δ with respect to the same refractive index n changes, but the value of the refractive index n when the relative positional deviation amount Δ is 0 is substantially the same. Does not change. Thus, in the present embodiment, the rhombus prism (7; 8) is configured to totally reflect light incident at a predetermined incident angle θ on the internal reflection surfaces (7b, 7c; 8b, 8c). The rhomboid prism (7;) so that the relative positional deviation amount Δ due to the polarization component of the totally reflected light beam is substantially 0, that is, the absolute value of the relative positional deviation amount Δ is, for example, 0.2 μm or less. What is necessary is just to determine the refractive index n of the optical material which forms 8).

  In the above-described embodiment shown in FIGS. 1 to 3, four total reflection surfaces (7b, 7c; 8b, 8c) are arranged in the optical path, and light from the projection optical system (4, 5) is converted into a pentaprism. 6 and the diamond prism 7 are guided to the test surface Wa, and light from the test surface Wa is guided to the condensing optical system (10, 11) via the rhombus prism 8 and the pentaprism 9. However, the number and arrangement of the total reflection surfaces are not limited, and a configuration in which, for example, two total reflection surfaces (31b; 32b) are arranged in the optical path as shown in FIG. 12 is also possible.

  In the modification of FIG. 12, the light from the projection optical system (4, 5) is guided to the test surface Wa via the triangular prism 31 as an optical path bending prism, and the light from the test surface Wa is bent by the optical path. The light is guided to the condensing optical system (10, 11) through a triangular prism 32 as a prism. In FIG. 12, the projection condenser lens 4 and the light receiving condenser lens 11 are not shown corresponding to FIG.

  In the modification of FIG. 12, the light emitted from the projection optical system (4, 5) along the optical axis AX <b> 1 enters the optical path bending prism 31. In the optical path bending prism 31, the light transmitted through the first transmission surface 31a is totally reflected by the total reflection surface 31b, then passes through the second transmission surface 31c, and is emitted from the optical path bending prism 31 along the optical axis AX2. The The light emitted from the optical path bending prism 31 along the optical axis AX2 enters the surface Wa.

  The light reflected by the test surface Wa along the optical axis AX3 that is symmetrical to the optical axis AX2 with respect to the optical axis AX of the projection optical system PL is incident on the optical path bending prism 32. In the optical path bending prism 32, the light transmitted through the first transmission surface 32a is totally reflected by the total reflection surface 32b, then passes through the second transmission surface 32c, and is emitted from the optical path bending prism 32 along the optical axis AX4. The The light emitted from the optical path bending prism 32 along the optical axis AX4 enters the condensing optical system (10, 11).

  Also in the modified example of FIG. 12, the refractive index n of the optical material forming the optical path bending prism (31; 32) as the total reflection prism member and the incident angle θ of light with respect to the total reflection surface (31b; 32b). By setting so as to substantially satisfy the relationship shown in Expression (7), the relative position by the polarization component generated in the light totally reflected by the total reflection surface (31b; 32b) of the optical path bending prism (31; 32). The influence of the shift on the detection of the surface position of the test surface Wa can be suppressed, and the surface position of the test surface Wa can be detected with high accuracy.

  In other words, the influence of the relative displacement due to the polarization component generated in the light totally reflected by the total reflection surface (31b; 32b) of the optical path bending prism (31; 32) has an effect on the detection of the surface position of the surface Wa. In order to detect the surface position of the surface Wa to be detected with high accuracy, the relative positional deviation amount Δ due to the polarization component of the light totally reflected by the total reflection surface (31b; 32b) becomes almost zero. That is, that is, the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection is, for example, 0.3 μm or less, and more preferably, 0.2 μm or less. The refractive index n of 31; 32) and the incident angle θ of the incident light with respect to the total reflection surface (31b; 32b) may be determined.

  Hereinafter, with reference to FIGS. 13 to 16, when the refractive index n of the total reflection prism member is changed, the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection is 0.3 μm or less. The range of the incident angle θ will be described. FIG. 13 shows the relationship between the relative positional deviation amount Δ and the incident angle θ due to one total reflection when the refractive index n of the total reflection prism member is 1.45 and the center wavelength λc of light is 750 nm. Yes. Referring to FIG. 13, in order to satisfy the conditional expression (8) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.3 μm or less, the range of the incident angle θ is about It can be seen that it may be set at 48 to 90 degrees. Further, the incident angle θ at which the relative positional deviation amount Δ becomes almost zero is about 53 degrees.

  FIG. 14 shows the relationship between the relative positional deviation amount Δ and the incident angle θ due to one total reflection when the refractive index n of the total reflection prism member is 1.5 and the center wavelength λc of light is 750 nm. Yes. Referring to FIG. 14, in order to satisfy the conditional expression (8) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.3 μm or less, the range of the incident angle θ is about It can be seen that it may be set to 45 degrees to 90 degrees. Further, the incident angle θ at which the relative positional deviation amount Δ becomes almost zero is about 52 degrees.

  FIG. 15 shows the relationship between the relative positional deviation amount Δ and the incident angle θ due to one total reflection when the refractive index n of the total reflection prism member is 1.6 and the center wavelength λc of light is 750 nm. Yes. Referring to FIG. 15, in order to satisfy the conditional expression (8) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.3 μm or less, the range of the incident angle θ is about It can be seen that it may be set to 42 degrees to 90 degrees. Further, the incident angle θ at which the relative positional deviation amount Δ is almost zero is about 49 degrees.

  FIG. 16 shows the relationship between the relative positional deviation amount Δ due to one total reflection and the incident angle θ when the refractive index n of the total reflection prism member is 1.7 and the center wavelength λc of light is 750 nm. Yes. Referring to FIG. 16, in order to satisfy the conditional expression (8) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.3 μm or less, the range of the incident angle θ is about It can be seen that it may be set to 39 degrees to 90 degrees. Further, the incident angle θ at which the relative positional deviation amount Δ is almost zero is about 46 degrees.

  In the above-described embodiment, an example in which the exposure apparatus includes a single surface position detection device is described. However, the present invention is not limited to this, and a plurality of surface position detection devices may be used as necessary. The detection visual field can also be divided. In this case, each device can be calibrated based on the detection result in the common visual field of the detection visual field of the first surface position detection device and the detection visual field of the second surface position detection device.

  In the above-described embodiment, the present invention is applied to the detection of the surface position of the photosensitive substrate of the projection exposure apparatus. However, the present invention may be applied to the detection of the surface position of the mask of the projection exposure apparatus. it can. In the above-described embodiment, the present invention is applied to the detection of the surface position of the photosensitive substrate in the projection exposure apparatus. However, the present invention may be applied to the detection of the surface position of a general test surface. it can.

  In the above-described embodiment, the present invention is applied to detection of the surface position of the test surface (position along the normal line of the test surface). However, the present invention is not limited to this. For example, in an optical apparatus in which an optical member having a total reflection surface is disposed in the measurement optical path, and more generally, an optical apparatus in which an optical member having a total reflection surface is disposed in the optical path. The present invention can be applied to. Specifically, an apparatus for measuring the surface shape of the test surface as shown in FIG. 17 or the position of the test surface (two-dimensional position of the test surface along the in-plane direction) as shown in FIG. The present invention can be applied to an apparatus that measures the above.

  In the measurement apparatus shown in FIG. 17, measurement light from the light source 41 is transmitted through a light transmission optical system (projection optical system) 42, for example, a total reflection optical member (an optical member having a total reflection surface) 43 such as a rhombus prism. Is incident on. The measurement light totally reflected by the total reflection surface (not shown) of the total reflection optical member 43 enters the test surface 44. The measurement light reflected by the test surface 44 enters a total reflection optical member 45 having the same configuration as that of the total reflection optical member 43, for example. The measurement light totally reflected by the total reflection surface (not shown) of the total reflection optical member 45 is incident on an image detector 47 such as a CCD via the light receiving optical system 46.

  In this apparatus, when the height position of the test surface 44 on which the measurement light is incident changes, the incident position (image position) of the measurement light on the image detector 47 changes. Accordingly, the measurement surface 44 is moved two-dimensionally along the in-plane direction to change the incident position of the measurement light on the measurement surface 44, or the vibration mirror or polygon mirror in the light transmission optical system 42 is changed. By measuring the incident position of the measurement light on the image detector 47 while changing the incident position of the measurement light on the test surface 44 by the action, the surface shape of the test surface 44 (the height for each position in the surface) is measured. Distribution) is measured. When there is a thin film on the test surface 44 or when the test surface 44 is polarized, if a relative position shift occurs due to a polarization component in total reflection by the total reflection optical members 43 and 45, the test surface 44 The surface shape cannot be measured with high accuracy.

  The surface shape of the test surface 44 is increased by suppressing the influence of the relative displacement due to the polarization component generated in the light totally reflected by the total reflection optical members 43 and 45 on the measurement of the surface shape of the test surface 44. In order to measure accurately, the relative positional deviation amount Δ due to the polarization component of the light totally reflected by the total reflection optical members 43 and 45 is substantially zero, that is, the polarization component generated in one total reflection. The refractive index n of the total reflection optical members 43 and 45 and the incident light with respect to the total reflection surface so that the absolute value of the relative positional deviation amount Δ due to the distance Δ is 0.3 μm or less, more preferably 0.2 μm or less. The incident angle θ may be determined.

  In the measuring apparatus shown in FIG. 18, the measurement light from the illumination system 51 is reflected by the half mirror 53 in the objective optical system 52, and then emitted from the objective optical system 52 to be an optical path bending prism (optical having a total reflection surface). Member) 54. For example, the measurement light totally reflected by the total reflection surface 54 a of the optical path bending prism 54 such as a triangular prism illuminates the test surface 55. Measurement light reflected by a mark (not shown) provided on the test surface 55 is incident on the optical path bending prism 54 again. The measurement light totally reflected by the total reflection surface 54 a of the optical path bending prism 54 is incident on an image detector 56 such as a CCD via the objective optical system 52 and the half mirror 53 in the objective optical system 52.

  In the measurement apparatus shown in FIG. 18, based on the output of the image detector 56, the position of the mark on the test surface 55 is eventually the position of the test surface 55 (two-dimensional of the test surface 55 along the in-plane direction). Measured position). When the test surface 55 has polarization, the relative position shift due to the polarization component occurs in the total reflection at the optical path bending prism 54, and therefore the position of the test surface 55 cannot be measured with high accuracy.

  In order to measure the position of the test surface 55 with high accuracy by suppressing the influence of the relative positional shift due to the polarization component generated in the light totally reflected by the optical path bending prism 54 on the measurement of the position of the test surface 55. Is such that the relative positional deviation amount Δ due to the polarization component of the light totally reflected by the optical path bending prism 54 becomes substantially zero, that is, the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection. The refractive index n of the optical path bending prism 54 and the incident angle θ of the incident light with respect to the total reflection surface 54a may be determined so that the value is, for example, 0.3 μm or less, more preferably 0.2 μm or less.

  As described above, when the test surface has a thin film or polarization characteristic that affects the polarization, Goose Hengshen is formed on the inner reflection surface (total reflection surface) arranged in the optical path (irradiation optical path, detection optical path, etc.). Since a shift (Goos-Haenchen Shift) occurs, the refractive index of the optical member that constitutes the internal reflection surface (total reflection surface) and the incident angle with respect to the internal reflection surface (total reflection surface) are appropriately adjusted so as not to occur. It is desirable to set.

By the way, in the above-mentioned embodiment shown in FIGS. 1 to 3, a parallelogram prism (a rhombus prism) having two total reflection surfaces parallel to each other is used as the total reflection prism member. Then, the refractive index n of the prism and the incident angle θ of the incident light with respect to each total reflection surface are determined so as to suppress the relative positional deviation amount Δ generated on each total reflection surface. However, the present invention is not limited to this. For example, a quadrilateral prism 61 having two total reflection surfaces (two total reflection surfaces that are not generally parallel) facing each other as shown in FIG. The relative positional deviation amount finally generated in the light after being sequentially totally reflected by the two total reflection surfaces 61a and 61b, that is, the relative positional deviation amount Δ ₁ generated in the first total reflection surface 61a and the second total reflection surface 61b. The incident angle θ ₁ (0 ° ≦ θ ₁ ≦ 90 °) of the incident light with respect to the refractive index n of the prism and the two total reflection surfaces so as to suppress the sum Δ ₁ + Δ ₂ with the relative positional deviation amount Δ ₂ generated in FIG. ) And θ ₂ (0 ° ≦ θ ₂ ≦ 90 °) can also be determined.

In this case, the relative positional deviation amount Δ ₁₂ = Δ ₁ + Δ ₂ finally generated in the light after being totally totally reflected by the two total reflection surfaces 61a and 61b is expressed by the following equation (10). The first term on the right side of Equation (10) corresponds to the relative displacement amount Δ ₁ generated on the first total reflection surface 61a, and the second term on the right side corresponds to the relative displacement amount Δ generated on the second total reflection surface 61b. It corresponds to ₂ .

Here, in order to suppress the influence of the relative position shift due to the polarization component of the totally reflected light on the detection accuracy, the amount of relative position shift generated on one total reflection surface is suppressed to 0.3 μm or less. When applied to the relative positional deviation amount delta ₁₂ represented by the formula (10) the concept of the above-described embodiments of ultimately relative positions occurring light which has been totally reflected successively by the two total reflection surfaces 61a and 61b as condition for minimize the deviation amount delta _12, the upper limit value is set to 0.3μm × 2 = 0.6μm, the following conditional expressions to -0.3μm × 2 = -0.6μm the lower limit (11) Is obtained.

As a specific numerical example, the quadrangular prism 61 is formed of quartz having a refractive index n = 1.45, the incident angle θ ₁ to the first total reflection surface 61a is 47 degrees, and the second total reflection surface 61b is incident. When the incident angle θ ₂ is 70 degrees, the relative positional deviation amount Δ ₁ generated on the first total reflection surface 61 a is +0.48 μm, and the relative positional deviation amount Δ ₂ generated on the second total reflection surface 61 b is −0. .21 μm. That is, the relative positional deviation amount delta ₂ of relative positional deviation amount generated in the first total reflection surface 61a delta ₁ and produced by the second total reflection surface 61b cancel each other, successively total of two total reflection surfaces 61a and 61b The relative positional deviation amount Δ ₁₂ finally generated in the light after being reflected is +0.27 μm and becomes 0.3 μm or less.

As another numerical example, the quadrilateral prism 61 is formed of quartz having a refractive index n = 1.45, the incident angle θ ₁ to the first total reflection surface 61a is 50 degrees, and the second total reflection surface 61b. When the incident angle θ ₂ is 60 degrees, the relative positional deviation amount Δ ₁ generated on the first total reflection surface 61 a is +0.155 μm, and the relative positional deviation amount Δ ₂ generated on the second total reflection surface 61 b is − 0.136 μm. That is, the relative positional deviation amount delta ₂ of relative positional deviation amount generated in the first total reflection surface 61a delta ₁ and produced by the second total reflection surface 61b cancel each other, successively total of two total reflection surfaces 61a and 61b The relative positional deviation amount Δ ₁₂ finally generated in the light after being reflected is +0.019 μm and is not more than 0.05 μm.

In the description with reference to FIG. 19, the quadrilateral prism 61 having two total reflection surfaces is used as the total reflection prism member. However, in general, a polygon prism having two or more total reflection surfaces may be used. The same can be considered. That is, the relative positional deviation amount finally generated in the light after being sequentially totally reflected by a plurality of (N: N is an integer of 2 or more) total reflection surfaces of one polygonal prism (generally an optical member). The refractive index n of the prism and the incident angle θa (0 ° ≦ θa ≦ 90 °; a = 1 to N) of incident light with respect to the a-th total reflection surface (internal reflection surface) may be determined so as to be kept small. . In this case, the relative positional deviation amount Δ _1N finally generated in the light after being totally totally reflected by the N total reflection surfaces is expressed by the following equation (12).

Here, in order to suppress the influence of the relative position shift due to the polarization component of the totally reflected light on the detection accuracy, the amount of relative position shift generated on one total reflection surface is suppressed to 0.3 μm or less. When the concept of the above-described embodiment is applied to the relative positional deviation amount Δ _1N represented by the equation (12), the relative positional deviation amount finally generated in the light after being sequentially totally reflected by the N total reflection surfaces. As a conditional expression for keeping Δ _1N small, the following conditional expression (13) is obtained in which the upper limit value is 0.3 μm × N = 0.3 Nμm and the lower limit value is −0.3 μm × N = −0.3 Nμm. It is done.

Furthermore, it is not limited to one optical member, but finally occurs in light after being sequentially totally reflected by a plurality of (N: N is an integer of 2 or more) total reflection surfaces in a plurality of optical members. In order to keep the relative positional deviation small, the refractive index na of the optical member forming the a-th total reflection surface (internal reflection surface) and the incident angle θa of incident light with respect to the a-th total reflection surface (0 ° ≦≦ θa ≦ 90 °; a = 1 to N) may be determined. In this case, the relative positional deviation amount Δ _1N finally generated in the light after being sequentially totally reflected by the N total reflection surfaces in the plurality of optical members is expressed by the following equation (14).

Here, in order to suppress the influence of the relative position shift due to the polarization component of the totally reflected light on the detection accuracy, the amount of relative position shift generated on one total reflection surface is suppressed to 0.3 μm or less. When the idea of the above-described embodiment is applied to the relative positional deviation amount Δ _1N represented by the equation (14), the relative positional deviation amount finally generated in the light after being sequentially totally reflected by the N total reflection surfaces. As a conditional expression for keeping Δ _1N small, the following conditional expression (15) is obtained in which the upper limit value is 0.3 μm × N = 0.3 Nμm and the lower limit value is −0.3 μm × N = −0.3 Nμm. It is done.

  Although the lower limit value is not set in the conditional expressions (8) and (9), the lower limit value of the conditional expression (8) is set to “−0.3 μm or more”, and the conditional expression (9) Is preferably set to “−0.2 μm or more”. If the lower limit value of conditional expression (8) is not reached, it is preferable because the incident angle is large and the incident angle is large, and the optical member such as a prism becomes complicated and large, and further, the overall apparatus becomes complicated and large. Absent. On the other hand, if the lower limit value of conditional expression (9) is not reached, there is a strong tendency for the incident angle to increase in the oblique incidence configuration, and there is a strong tendency for the optical member such as a prism to become complicated and large. Is not preferable because it tends to increase the complexity and size of the entire apparatus.

  In the conditional expression (8), the upper limit value is set to “+0.3 μm or less”. However, it is possible to simplify and compact the optical member and the apparatus, and to achieve more accurate and stable detection. To make this possible, set the upper limit of conditional expression (8) to “+0.05 μm or less” and the lower limit of conditional expression (8) to “−0.05 μm or more”. Is preferred. In other words, it is preferable to satisfy the following conditional expression (16) in order to enable more accurate and stable detection while enabling simplification and downsizing of the optical member and apparatus.

  Hereinafter, with reference to FIGS. 20 to 23 corresponding to FIGS. 13 to 16, when the refractive index n of the total reflection prism member is changed, the relative positional deviation amount due to the polarization component generated in one total reflection. The range of the incident angle θ in which the absolute value of Δ is 0.05 μm or less will be described. FIG. 20 is an enlarged view of a part of FIG. 13, in which the total reflection prism member has a refractive index n of 1.45 and a central wavelength λc of light of 750 nm. The relationship between the relative positional deviation amount Δ and the incident angle θ is shown. Referring to FIG. 20, in order to satisfy the conditional expression (16) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.05 μm or less, the range of the incident angle θ is about It can be seen that it may be set at 52 to 55 degrees.

  FIG. 21 is an enlarged view of a part of FIG. 14, and shows a case of total reflection when the refractive index n of the total reflection prism member is 1.5 and the center wavelength λc of light is 750 nm. The relationship between the relative positional deviation amount Δ and the incident angle θ is shown. Referring to FIG. 21, in order to satisfy the conditional expression (16) while suppressing the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.05 μm or less, the range of the incident angle θ is approximately It can be seen that it may be set at 49.5 to 56 degrees.

  FIG. 22 is an enlarged view of a part of FIG. 15, in which the total reflection prism member has a refractive index n of 1.6 and the light central wavelength λc is 750 nm. The relationship between the relative positional deviation amount Δ and the incident angle θ is shown. Referring to FIG. 22, in order to satisfy the conditional expression (16) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.05 μm or less, the range of the incident angle θ is about It can be seen that it may be set to 46.5 degrees to 53 degrees.

  FIG. 23 is an enlarged view of a part of FIG. 16, and the total reflection prism member has a refractive index n of 1.7 and a light central wavelength λc of 750 nm. The relationship between the relative positional deviation amount Δ and the incident angle θ is shown. Referring to FIG. 23, in order to satisfy the conditional expression (16) while keeping the absolute value of the relative positional deviation amount Δ due to the polarization component generated in one total reflection to 0.05 μm or less, the range of the incident angle θ is about It can be seen that it may be set to 43.5 degrees to 50 degrees.

  Similarly, the upper limit value of conditional expression (13) in the case where one optical member is sequentially totally reflected by a plurality of total reflection surfaces is 0.05 μm × N = 0.05 Nμm, and the lower limit value is −0.05 μm × N. = −0.05 Nμm is preferable to conditional expression (13) as a conditional expression for enabling more accurate and stable detection while enabling simplification and compactness of the optical member and apparatus. Conditional expression (17) is obtained.

  Similarly, in the case where a plurality of optical members are sequentially totally reflected by a plurality of total reflection surfaces, the upper limit value of conditional expression (15) is 0.05 μm × N = 0.05 Nμm, and the lower limit value is −0.05 μm × N. = −0.05 Nμm is preferable to conditional expression (15) as a conditional expression for enabling more accurate and stable detection while enabling simplification and downsizing of optical members and devices. Conditional expression (18) is obtained.

  In the present invention, the GHS generated on a plurality of total reflection surfaces is suppressed by appropriately setting the incident angle and the refractive index forming each total reflection surface. By using one or more (one or more of the optical members having a total reflection surface) as the configuration in which the incident angle can be adjusted by the adjusting device, the incident angle on each total reflection surface is changed, and the Goose Henschen shift. It is also possible to control the Goose Henschen shift (GHS) amount by adjusting the (GHS) amount. Here, the adjusting device can be configured so that the optical member can be adjusted manually or automatically, but the adjusting device is preferably configured so that the optical member can be automatically adjusted. In this case, the light passing through each total reflection surface is detected by the detection device, and an appropriate Goose Henschen shift (GHS) amount is automatically set based on output information (detection signal, drive signal, etc.) from the detection device. It is preferable to drive an adjusting device including a motor or the like so as to be able to do so.

  For example, in the embodiment shown in FIG. 12, a light splitting member (half mirror or the like) is disposed on the detector side of the lens 10 (above the lens 10), and the goose is branched into the optical path branched by the light splitting member. A detection device (detector or the like) for detecting the Henschen shift (GHS) amount is arranged. Then, a drive device (adjustment device) for independently tilting each optical path bending prism (31, 32) based on the output from the detection device (detector, etc.) is arranged, and this drive device (adjustment device) is arranged. Accordingly, the GHS amount of the entire apparatus can be appropriately set by independently adjusting the incident angles of light incident on the total reflection surfaces (31b, 32b) of the optical path bending prisms (31, 32). it can.

  In the exposure apparatus of the above-described embodiment, the reticle (mask) is illuminated by the illumination device (illumination process), and the transfer pattern formed on the mask is exposed to the photosensitive substrate using the projection optical system (exposure process). Thus, a micro device (semiconductor element, imaging element, liquid crystal display element, thin film magnetic head, etc.) can be manufactured. Hereinafter, referring to the flowchart of FIG. 24 for an example of a method for obtaining a semiconductor device as a micro device by forming a predetermined circuit pattern on a wafer or the like as a photosensitive substrate using the exposure apparatus of the present embodiment. I will explain.

  First, in step 301 of FIG. 24, a metal film is deposited on one lot of wafers. In the next step 302, a photoresist is applied on the metal film on the one lot of wafers. Thereafter, in step 303, using the exposure apparatus of the present embodiment, the image of the pattern on the mask is sequentially exposed and transferred to each shot area on the wafer of one lot via the projection optical system. Thereafter, in step 304, the photoresist on the one lot of wafers is developed, and in step 305, the resist pattern is etched on the one lot of wafers to form a pattern on the mask. Corresponding circuit patterns are formed in each shot area on each wafer.

  Thereafter, a device pattern such as a semiconductor element is manufactured by forming a circuit pattern of an upper layer. According to the semiconductor device manufacturing method described above, a semiconductor device having an extremely fine circuit pattern can be obtained with high throughput. In steps 301 to 305, a metal is deposited on the wafer, a resist is applied on the metal film, and exposure, development, and etching processes are performed. Prior to these processes, on the wafer. It is needless to say that after forming a silicon oxide film, a resist may be applied on the silicon oxide film, and steps such as exposure, development, and etching may be performed.

  In the exposure apparatus of this embodiment, a liquid crystal display element as a micro device can be obtained by forming a predetermined pattern (circuit pattern, electrode pattern, etc.) on a plate (glass substrate). Hereinafter, an example of the technique at this time will be described with reference to the flowchart of FIG. In FIG. 25, in a pattern forming process 401, a so-called photolithography process is performed in which the exposure pattern of the present embodiment is used to transfer and expose a mask pattern onto a photosensitive substrate (such as a glass substrate coated with a resist). By this photolithography process, a predetermined pattern including a large number of electrodes and the like is formed on the photosensitive substrate. Thereafter, the exposed substrate undergoes steps such as a developing step, an etching step, and a resist stripping step, whereby a predetermined pattern is formed on the substrate, and the process proceeds to the next color filter forming step 402.

  Next, in the color filter forming step 402, a large number of sets of three dots corresponding to R (Red), G (Green), and B (Blue) are arranged in a matrix or three of R, G, and B A color filter is formed by arranging a plurality of stripe filter sets in the horizontal scanning line direction. Then, after the color filter forming step 402, a cell assembly step 403 is executed. In the cell assembly step 403, a liquid crystal panel (liquid crystal cell) is assembled using the substrate having the predetermined pattern obtained in the pattern formation step 401, the color filter obtained in the color filter formation step 402, and the like.

  In the cell assembly step 403, for example, liquid crystal is injected between the substrate having the predetermined pattern obtained in the pattern formation step 401 and the color filter obtained in the color filter formation step 402, and a liquid crystal panel (liquid crystal cell) is obtained. ). Thereafter, in a module assembling step 404, components such as an electric circuit and a backlight for performing a display operation of the assembled liquid crystal panel (liquid crystal cell) are attached to complete a liquid crystal display element. According to the above-described method for manufacturing a liquid crystal display element, a liquid crystal display element having an extremely fine circuit pattern can be obtained with high throughput.

DESCRIPTION OF SYMBOLS 1 Light source 2 Condenser lens 3 Deflection prism 4, 5 Projection optical system 6, 9 Penta prism 7, 8 Diamond prism 10, 11 Condensing optical system 12 Oscillation mirror 13 Orientation correction prism 14a, 14b Relay optical system 15 Light-receiving part 16 Mirror drive Unit 17 position detection unit 18 correction amount calculation unit 21 wafer holder 22 holder holding mechanism 23 holder driving unit IL illumination system R reticle RH reticle holder PL projection optical system W wafer

Claims (10)

It said projected from an oblique direction against the test surface based on the light receiving result of the light reflected by the test surface, detected by the surface position detecting the surface position of the test surface with respect to the direction intersecting the predetermined surface In the device
Projection optics including an optical member on which a predetermined pattern is formed, and projecting the light through the predetermined pattern from an oblique direction with respect to the test surface to form a primary image of the predetermined pattern on the test surface The system,
A condensing optical system for condensing the light reflected by the test surface to form a secondary image of the predetermined pattern on a predetermined surface;
A detector that receives the light through the predetermined surface and detects a surface position of the test surface based on a light reception result of the light;
Arranged in the optical path of the light, and a total reflection prism member having an internal reflecting surface for totally reflecting the light,
Wherein the incident angle of the light with respect to the refractive index the internal reflecting surface of the optical material forming the total reflection prism member, said secondary image formed on the predetermined plane by the first polarization component of the light , So that a relative positional deviation does not occur with the secondary image formed on the predetermined plane by the second polarization component having a polarization direction perpendicular to the first polarization component of the light. A surface position detection device characterized by being set. The total reflection prism member includes a first prism member disposed in an optical path on the projection optical system side in an optical path of the light, and a second prism member disposed on an optical path on the condensing optical system side,
The refractive index of the optical material forming the first prism member and the incident angle of the light with respect to the first inner surface reflecting surface of the inner surface reflecting surface of the first prism member are determined by the first polarization component. It is set so that a relative positional shift does not occur between the primary image formed on the inspection surface and the primary image formed on the inspection surface by the second polarization component. The surface position detection apparatus according to claim 1, wherein The surface position detecting device according to claim 2 , wherein a refractive index of an optical material forming the first prism member and a refractive index of an optical material forming the second prism member are different from each other . The total reflection prism member has a first inner surface reflecting surface and a second inner surface reflecting surface;
The incident angle of the light with respect to the first inner surface reflecting surface and the incident angle of the light with respect to the second inner surface reflecting surface are relative to each other caused by total reflection between the first polarization component and the second polarization component. The positional deviation amount generated on the first inner surface reflecting surface and the positional deviation amount generated on the second inner surface reflecting surface are set so as to cancel each other with respect to a typical optical path positional deviation. Item 4. The surface position detection device according to any one of Items 1 to 3. The incident angle of the light with respect to the first inner reflection surface is set smaller than an incident angle at which the positional deviation does not substantially occur due to total reflection between the first polarization component and the second polarization component, The incident angle of the light with respect to the second inner surface reflection surface is set to be larger than an incident angle at which the positional deviation does not substantially occur due to total reflection between the first polarization component and the second polarization component. The surface position detecting device according to claim 4 . The detection unit detects a lateral shift amount of the secondary image along the predetermined surface based on the light reception result, and detects a surface position of the test surface based on the detection result of the lateral shift amount. The surface position detection device according to claim 1 , wherein the surface position detection device is a surface position detection device. In an exposure apparatus that projects and exposes a pattern formed on a mask onto a photosensitive substrate via a projection optical system,
The surface position detection device according to any one of claims 1 to 6, for detecting a surface position of the pattern surface of the mask or the exposure surface of the photosensitive substrate with respect to the projection optical system;
Alignment means for aligning the pattern surface of the mask or the exposure surface of the photosensitive substrate with respect to the projection optical system based on the detection result of the surface position detection device. Exposure equipment to do . In an exposure method for projecting and exposing a pattern formed on a mask onto a photosensitive substrate via a projection optical system,
Detecting the surface position of the pattern surface of the mask or the exposure surface of the photosensitive substrate with respect to the projection optical system using the surface position detection device according to any one of claims 1 to 6;
Aligning the pattern surface of the mask or the exposure surface of the photosensitive substrate with respect to the projection optical system based on the detection result of the surface position detection device;
An exposure method comprising: An exposure step of exposing the photosensitive substrate with a pattern of the mask using the exposure apparatus according to claim 7;
And a development step of developing the photosensitive substrate exposed in the exposure step . An exposure step of exposing the photosensitive substrate to the pattern of the mask using the exposure method according to claim 8;
A developing step of developing the photosensitive substrate exposed by the exposing step;
A device manufacturing method comprising:

Images