JP2014228486A - Three-dimensional profile acquisition device, pattern inspection device, and three-dimensional profile acquisition method - Google Patents

Abstract

A three-dimensional profile acquisition apparatus, a pattern inspection apparatus, and a three-dimensional profile acquisition method capable of acquiring a three-dimensional profile representing the height distribution of a surface to be measured by relatively simple processing at relatively high speed. A three-dimensional profile acquisition device (21) includes an interferometer (42) having a camera (41); Based on the image, the scanning position (Z position) where the peak of the interference fringe appears is detected for each pixel by image analysis, and a three-dimensional profile PD is generated. The light emission control section 65 causes the light source 43 to blink in the same cycle and phase as the interference fringes via the light emission drive section 61 . Also, the exposure time when the camera 41 takes an image of the surface 12a to be measured of the wiring board 12 is set to a natural number multiple of the period of the interference fringes. [Selection drawing] Fig. 3

Description

  The present invention relates to a three-dimensional profile acquisition apparatus that optically measures the height distribution of a surface to be measured to acquire a three-dimensional profile, a pattern inspection apparatus including the same, and a three-dimensional profile acquisition method.

  As this type of three-dimensional profile acquisition apparatus, a foreign substance inspection apparatus using an interferometer (for example, Patent Document 1), a depth measuring apparatus using an interferometer (for example, Patent Document 2), and a surface shape measuring apparatus using an interferometer (For example, Patent Document 3) is known.

  For example, the foreign substance inspection apparatus described in Patent Document 1 is configured to inspect by laser interferometry using an interferometer that uses a laser light source as a light source. Therefore, if laser beams having different optical paths are combined, an interference fringe is always generated regardless of the optical path difference. Therefore, it is difficult to measure the height distribution of the inspection target surface.

  On the other hand, the measurement apparatuses described in Patent Documents 2 and 3 perform measurement by the white interference method using an interferometer provided with a light source that emits white light. That is, in the depth measuring apparatus described in Patent Document 2, a white light beam is divided into a measurement beam projected onto a sample through an objective lens by a beam splitter and a reference beam reflected by a reference mirror. The relative optical path length difference is changed by the fringe scanning means. And the displacement information based on the fringe scan signal from the light detection means (for example, line sensor) which received the interference beam generated by the signal processing circuit combining the reflected light from the sample and the reflected light from the reference mirror, The depth information of the recess is output based on the relative position information between the objective lens and the sample.

  Moreover, in the surface shape measuring apparatus described in Patent Document 3, white light from which a white light from a white light source passes through a bandpass filter and is limited to a specific frequency band is a reference surface whose relative distance is changed by a driving unit. Irradiation to the surface to be measured causes interference fringes due to the change in the optical path difference. The CCD camera images the measurement target surface together with the interference fringes. The CPU samples the intensity value of the interference light that changes at a specific location on the measurement target surface at a predetermined sampling interval. Further, a specific function having a peak position that matches the peak position of the interference light is estimated based on the bandwidth of the specific frequency band. By calculating the height of the peak position of the characteristic function, the uneven shape of the measurement target surface is measured.

Japanese Patent Laid-Open No. 10-293100 JP 2009-36563 A JP 2001-66122 A

  By the way, in the measuring apparatus using the white light interferometry described in Patent Document 2, a short sampling frequency (scanning distance of 100 nm or less) of 3/1 or less of the interference fringe wavelength by the light detection means (in the example of Patent Document 2 is 5 nm). ), It is necessary to image a plurality of times (three times or more) during one period of the interference fringe. In this case, if the scanning distance (scanning range) for scanning in the direction perpendicular to the surface to be measured is, for example, 50 μm, it is necessary to capture several hundred to 1,000 images in one scan, and the measurement time is very long. There was a problem that it took a long time.

  Moreover, in the surface shape measuring apparatus using the white light interferometry described in Patent Document 3, it is only necessary to sample at a time interval longer than the period of interference fringes, so the number of images captured by the camera in one scan is compared. Can be less. However, the calculation for estimating the specific function takes time, and the surface shape measurement process cannot be accelerated so much that the number of samplings can be reduced.

  An object of the present invention is to provide a three-dimensional profile acquisition apparatus, a pattern inspection apparatus, and a three-dimensional profile acquisition method capable of acquiring a three-dimensional profile representing the height distribution of a surface to be measured at a relatively high speed by a relatively simple process. It is in.

  A three-dimensional profile acquisition apparatus for solving the above problems is configured to be capable of scanning in a direction intersecting with a light source that emits light having a wide spectral width and a surface to be measured, and to emit light from the light source to the surface to be measured The measurement light reflected by the reference mirror and the reference light reflected by the reference mirror are separated, and the reflected light of the two separated lights are merged to generate an interference fringe based on the optical path difference between the two reflected lights. An interferometer having an imaging unit that captures an image with an exposure time equal to or longer than the period of the interference fringe, a control unit that scans the scanning unit and causes the light source to blink at the same cycle as the interference fringe, and the imaging unit An image solution for detecting a peak position of interference fringes in the scanning direction of the scanning unit for each pixel based on a plurality of captured images and generating a three-dimensional profile based on the peak position for each pixel. And it includes a part, a. The light having a wide spectral width may be any light source that can emit light having a spectral width that is wide enough to constitute a white interferometer, and the emitted light is not necessarily limited to white light.

  According to this configuration, when the light having a wide spectral width is separated during scanning of the scanning unit and the optical path difference between the two lights (measurement light and reference light) reflected by the measurement surface and the reference mirror is extremely small, Interference fringes occur due to interference between the two reflected lights that have joined. At this time, since the light source blinks in synchronization with the cycle of the interference fringe, the light of the interference fringe strikes the pixels of the imaging unit every half cycle when the light source is turned on. As a result, the imaging unit receives the interference fringe every half cycle during the exposure time, and for example, only the bright part or the dark part of the interference fringe is imaged. For this reason, the peak position of the interference fringe for each pixel can be detected using a relatively small number of images captured by the imaging unit with a relatively long exposure time equal to or longer than the cycle of the interference fringe. Therefore, a three-dimensional profile representing the height distribution of the surface to be measured can be acquired relatively easily and at high speed.

In the three-dimensional profile acquisition apparatus, it is preferable that the control unit reciprocates the scanning unit and shifts the blinking phase of the light source between forward movement and backward movement.
According to this configuration, since the phase at which the light source blinks is shifted between the forward movement and the backward movement of the scanning unit of the interferometer, the peak position of the interference fringe is determined at least during the forward movement and the backward movement. It can be detected reliably. The angle for shifting the phase is not particularly limited, but a predetermined angle in the range of 80 to 100 degrees is desirable as an example.

  The three-dimensional profile acquisition apparatus further includes an optical sensor that receives a part of the two reflected lights toward the imaging unit, and the control unit is based on a light reception signal of the optical sensor that receives the interference fringe. Preferably, the light source blinks in the same phase as the interference fringe.

  According to this configuration, the light source blinks in the same phase as the interference fringe by being controlled based on the light reception signal of the photosensor that has received the interference fringe. For this reason, a decrease in the accuracy of the three-dimensional profile due to the phase shift between the blinking of the light source and the interference fringe can be suppressed to a small level.

In the three-dimensional profile acquisition apparatus, it is preferable that the exposure time of the imaging unit is set to a value within a range of ¼ to ３ of the generation period of the interference fringe.
According to this configuration, the time interval for imaging can be made relatively longer than in the sampling method, and the number of necessary images can be reduced to a fraction of 1 to several tens of times. In addition, since the image analysis processing when the image analysis unit detects the peak position is relatively simple, it is possible to speed up the three-dimensional profile acquisition processing.

  A pattern inspection apparatus for solving the above problems is a pattern inspection apparatus provided with a pattern inspection unit for inspecting the presence or absence of a wiring defect based on an image obtained by imaging a wiring pattern of a wiring board, wherein the three-dimensional profile acquisition apparatus And a verification unit that verifies the authenticity of the defect detected by the pattern inspection unit based on the three-dimensional profile generated by the three-dimensional profile acquisition apparatus.

According to this configuration, the authenticity of the defect found by the pattern inspection can be verified from the information on the height distribution of the defect based on the three-dimensional profile.
A method for obtaining a three-dimensional profile for solving the above-described problem is to scan light that has a wide spectral width emitted from a light source by causing a scanning unit constituting an interferometer to scan in a direction intersecting the surface to be measured. An interference forming step of separating the measurement light reflected by the surface and the reference light reflected by the reference mirror and combining the reflected light of the two separated lights to generate an interference fringe based on an optical path difference; and A light emission control step of blinking at the same cycle as the interference fringe, an imaging step of imaging the images of the two reflected lights by the imaging unit of the interferometer with an exposure time longer than the cycle of the interference fringe, and the imaging unit A peak position of interference fringes in the scanning direction of the scanning unit is detected for each pixel based on a plurality of images, and a three-dimensional profile is generated based on the peak position for each pixel. It includes a profile generation step. According to this method, the same effect as that of the three-dimensional profile acquisition apparatus can be obtained.

  According to the present invention, a three-dimensional profile representing the height distribution of the surface to be measured can be acquired at a relatively high speed by a relatively simple process.

The schematic perspective view which shows the pattern inspection apparatus in one Embodiment. The schematic fragmentary sectional view of a wiring board. The schematic diagram which shows a three-dimensional profile acquisition apparatus. The schematic diagram explaining the principle of the white interferometry in a three-dimensional optical unit. The graph which shows the interference fringe by an optical path difference. The schematic perspective view which shows the wiring board which has a level | step difference. The graph which compared the peak position of the interference fringe between two pixels which image each surface where height differs. The graph which shows the relationship between an interference waveform, a LED drive waveform, a camera input light waveform, and a camera output. The graph which shows the relationship between the scanning position at the time of a measurement, an interference waveform, LED drive waveform, and camera imaging brightness. (A) is a graph showing the relationship of the interference waveform, LED drive waveform, camera input light waveform, and camera output with respect to the scanning position in one of forward scanning and backward scanning, and (b) is a similar graph on the other. The flowchart which shows a three-dimensional profile acquisition process.

Hereinafter, a pattern inspection apparatus including a three-dimensional profile acquisition apparatus (surface shape measuring apparatus for a surface to be measured) according to an embodiment will be described with reference to FIGS.
As shown in FIG. 1, the pattern inspection apparatus 11 is an apparatus that inspects the presence or absence of defects in the wiring pattern formed on the wiring board 12. The pattern inspection apparatus 11 includes a square box-shaped support base 13, a movable stage 14 provided on the upper surface of the support base 13, a gate-type support frame 15 laid across the stage 14, and a support frame And an imaging unit 17 attached in a suspended state capable of imaging the wiring board 12 with respect to the rail 16 fixed to the 15 horizontal bars 15a.

  The imaging unit 17 includes a high resolution lens 18 and a camera 19. As an example, the camera 19 uses an area camera. The pattern inspection apparatus 11 performs a pattern inspection for inspecting the presence or absence of a wiring defect based on the image of the wiring board 12 captured by the camera 19. However, the pattern inspection has a limit in defect detection accuracy, and the pattern inspection has a defect. It is necessary to verify the authenticity of the defect detected as a candidate based on the height distribution information.

  The pattern inspection apparatus 11 of this embodiment is equipped with a three-dimensional optical unit 20 for this verification. The three-dimensional optical unit 20 captures an image used for measuring the height distribution of the measurement target surface 12 a of the wiring board 12.

  The three-dimensional optical unit 20 acquires a three-dimensional profile (a group of (x, y, z) coordinates) indicating the height distribution of the measured surface 12a when the measured surface 12a is an XY plane. A part of the acquisition device 21 is configured. The three-dimensional optical unit 20 captures an image for acquiring a three-dimensional profile indicating a height distribution on the measurement target surface 12 a (two-dimensional plane) in the inspection target area of the wiring board 12.

  The pattern inspection apparatus 11 includes, for example, a controller 23 that drives and controls various components in the support base 13. For example, the controller 23 controls the pattern inspection such as the movement control of the stage 14 and the imaging control by the camera 19 and also controls the driving of the three-dimensional optical unit 20.

  The pattern inspection apparatus 11 includes a computer 25 such as a personal computer. The computer 25 includes a main body 26, an input device 27 including a keyboard and a mouse, and a monitor 28. Various programs for inspecting the wiring pattern of the wiring board 12 and verifying defect candidates are stored in a hard disk (not shown) in the main body 26. Further, the main body 26 is provided with an image processing circuit that measures the height distribution of the measurement target surface 12 a of the wiring board 12 based on the captured image input from the camera 19. The main body 26 includes an inspection unit 31, an image analysis unit 32, a verification unit 33 (verification unit), and the like shown in FIG. 1 as functional units including a CPU that executes a program and an image processing circuit.

  The inspection unit 31 illustrated in FIG. 1 performs a pattern inspection for inspecting the presence or absence of a defect in the wiring pattern of the wiring board 12 based on the image captured by the camera 19. The inspection unit 31 detects a defect by performing known processing such as pattern matching processing or delayed self-comparison processing as image processing for pattern inspection. The image analysis unit 32 analyzes a plurality of images captured by the three-dimensional optical unit 20 and measures the height distribution of the surface 12a to be measured, whereby a three-dimensional profile in which the height distribution is indicated by three-dimensional coordinates. Is generated. Further, the verification unit 33 verifies (verifies) the authenticity of suspicious defect candidates in the inspection unit 31 based on the three-dimensional profile acquired by the image analysis unit 32.

  As shown in FIG. 2, the wiring substrate 12 includes a substrate 35 and a plurality of wirings 36 formed on the surface to be measured 12 a of the substrate 35. Due to the demand for higher density of the wiring 36, the aspect ratio, which is the ratio of the height to the width of the wiring 36, is relatively high. In the pattern inspection in which a defect is detected by analyzing a two-dimensional image obtained by imaging the surface to be measured 12 a, an oxide film 36 a formed on the upper end surface of the wiring 36 and a concave portion on the upper end portion of the wiring 36, as shown in FIG. It is difficult to identify the missing portion 36c in which 36b and more than half of the wiring 36 in the height direction are missing. For example, the oxide film 36a is not a defect because the electrical resistance of the wiring 36 is ensured more than a specified value. The recess 36b is not a defect if it is very shallow, but if it is deeper than a certain depth, it increases the electrical resistance of the wiring 36 and becomes a defect. On the other hand, the missing portion 36c becomes a defect because the electrical resistance of the wiring 36 is remarkably increased. For this reason, in the present embodiment, the oxide film 36a, the concave portion 36b, and the missing portion 36c, which are difficult to identify by pattern inspection, are obtained based on the three-dimensional profile obtained by measuring the height distribution of the measured surface 12a by the white light interferometry. Whether the defect candidate is a defect is verified.

Next, details of the three-dimensional profile acquisition apparatus 21 will be described with reference to FIG.
As shown in FIG. 3, the three-dimensional profile acquisition device 21 includes a three-dimensional optical unit 20 and a control device 40. The control device 40 includes a part of a controller 23 that controls the three-dimensional optical unit 20 and a part of a computer 25 that performs processing necessary for obtaining a three-dimensional profile.

  As illustrated in FIG. 3, the three-dimensional optical unit 20 includes an interferometer 42 having a camera 41 as an example of an imaging unit. The interferometer 42 is, for example, a Michelson interferometer. The interferometer 42 includes a light source 43 (a white light source as an example) that emits light having a wide spectral width (for example, white light), a condensing lens 44, an optical filter 45, a half mirror 46, an objective lens 47, a half mirror 48, and a reference mirror 49. A piezoelectric actuator 50, an imaging lens 51, and the camera 41 described above. A half mirror 46, an imaging lens 51, and a half mirror 53 are sequentially arranged in the optical axis direction in the lens barrel 52 arranged coaxially with the optical axis below the camera 41 from the objective lens 47 side. They are arranged at intervals.

  The light source 43 shown in FIG. 3 is composed of, for example, an ultra-bright LED. The white light emitted from the light source 43 is collected by the condenser lens 44, further filtered through the optical filter 45 to light of a specific wavelength region (light having a wide spectral width), and then reflected by the lower surface of the half mirror 46. Then, the measurement surface 12 a of the wiring substrate 12 is irradiated through the objective lens 47 and the half mirror 48. The optical filter 45 has a function of adjusting an interference fringe generation period (interference possible distance) described later to an appropriate value. In particular, when the spectrum width of the light emitted from the light source 43 is very wide, the frequency is different between the vicinity of the peak of the interference fringe and the surroundings. Therefore, the color of the light from the light source 43 is limited to some extent to reduce the frequency of the interference fringe. The optical filter 45 is disposed between the light source 43 and the interferometer 42 for the purpose of bringing the frequency close to a constant frequency throughout the generation period (interference possible distance). In particular, if this type of adjustment is not necessary, the optical filter 45 may be omitted.

  The half mirror 48 disposed between the objective lens 47 and the wiring board 12 is supported so as to be movable in the Z direction together with the objective lens 47. At a position on the extension line of the reflection direction (left direction in FIG. 3) of a part of the light reflected by the upper surface of the half mirror 48, the reference mirror 49 is supported by the support member 54 fixed to the objective lens 47, It arrange | positions with the half mirror 48 in the state which can move to a Z direction. The objective lens 47, the half mirror 48, and the reference mirror 49 constitute a scanning unit 55 that can move integrally in the Z direction. The scanning unit 55 is driven by a piezo-type actuator 50 so that a small amount of a predetermined stroke (for example, a predetermined value within a range of 20 to 100 μm) in the height direction (Z direction) perpendicular to the surface 12a to be measured of the wiring board 12. Can be moved.

  A part of the light traveling downward through the objective lens 47 is transmitted through the half mirror 48 and reflected by the measurement surface 12a of the wiring board 12, and the reflected light is transmitted through the half mirror 48 again and travels down to the objective lens 47. Incident from the side. The other part of the light that passes downward through the objective lens 47 and is reflected by the upper surface of the half mirror 48 is reflected by the reference mirror 49, is reflected again by the upper surface of the half mirror 48, and is lowered to the objective lens 47. Incident from. The two reflected lights having different optical paths that merged on the upper surface of the half mirror 48 interfere with each other when the optical path difference between the two, which changes in accordance with the scanning position of the scanning unit 55, becomes smaller than or equal to a specified value, thereby causing interference fringes. Form.

  In addition, the three-dimensional optical unit 20 shown in FIG. 3 is provided with an optical sensor 56 (photosensor) that can receive a part of the reflected light that has merged on the upper surface of the half mirror 48. The merged reflected light is partially reflected by the lower surface of the half mirror 53 disposed between the imaging lens 51 and the camera 41 and enters the optical sensor 56. The optical sensor 56 outputs a light reception signal having a value proportional to the amount of light received. For this reason, when the interference fringe occurs, the light reception signal of the optical sensor 56 includes a waveform corresponding to the brightness of the interference fringe.

  The control device 40 shown in FIG. 3 includes the image analysis unit 32 described above, a control unit 60 that controls various types of control for the three-dimensional optical unit 20, a light emission drive unit 61, and a piezo drive unit 62. The control unit 60 includes an imaging control unit 63 that controls the camera 41, an interference monitor unit 64 that monitors the period and phase of the interference fringe based on the light reception signal from the optical sensor 56, and a light emission control unit 65 that controls the light source 43. I have.

  The blinking cycle of the light source 43 of the present embodiment is the same cycle as the interference fringe. For this reason, the light source 43 continues to emit light during the half period of the interference fringe. During scanning of the scanning unit 55, the interference monitoring unit 64 samples the waveform of the light reception signal of the optical sensor 56 during the light emission period of the light source 43, and acquires at least the phase of the period and phase of the interference fringe. At least the phase of the interference fringe IF is sent from the interference monitor unit 64 to the light emission control unit 65.

  The light emission control unit 65 shown in FIG. 3 is an LED drive pulse signal (light emission) having a period and phase that match the period and phase of the interference fringe based on the interference fringe period and phase data or the light reception signal from the optical sensor 56. Drive signal) is generated and output to the light emission drive unit 61. The light emission control unit 65 includes, for example, a phase locked loop (PLL (phase locked loop) circuit). The phase synchronization circuit performs feedback control based on the pulse signal having the same phase as the interference fringe generated based on the interference fringe cycle and phase data, or the light reception signal of the optical sensor 56 during the lighting period when the interference fringe occurs. An LED drive pulse signal having a pulse waveform synchronized in phase is output from an oscillator (not shown). In this case, the light emission control unit 65 generates an LED drive pulse signal for a half cycle based on the light reception signal of the half cycle of the lighting period of the light source 43, and the remaining half with the same phase as the generated LED drive pulse signal for the half cycle. An LED drive pulse signal for a period is generated.

  The light emission drive unit 61 generates an LED drive voltage pulse having a period and phase that match the period and phase of the interference fringe based on the LED drive pulse signal from the light emission control unit 65 and outputs the LED drive voltage pulse to the light source 43. As a result, the light source 43 blinks in the same cycle and phase as the interference fringe IF.

  Further, the control unit 60 generates a piezo drive signal that defines a voltage value to be applied to the actuator 50 in order to cause the scanning unit 55 to scan. The piezo drive unit 62 outputs (applies) a voltage based on the piezo drive signal from the control unit 60 to the piezo actuator 50. The actuator 50 is configured by laminating a plurality of piezo elements (for example, PZT elements) in a stretchable direction, and expands and contracts by the electrostrictive action of the piezo elements based on the input voltage. As the actuator 50 expands and contracts, the scanning unit 55 moves in the Z direction. In the present embodiment, a predetermined stroke for moving the scanning unit 55 in the Z direction. As an example, it is about 50 μm. This is because the scanning unit 55 is moved with a distance that is sufficiently longer than the assumed maximum measurement length in the Z direction, such as the depth of the recess and the height of the step (for example, the height of the wiring 36) to be measured in the wiring board 12. This is to make it happen. In the present embodiment, the scanning unit 55 is moved twice (one reciprocation) or more to measure one height distribution. In this example, the scanning unit 55 is moved twice (one reciprocation).

  Further, the imaging control unit 63 illustrated in FIG. 3 performs imaging control of the camera 41. In this imaging control, the imaging control unit 63 controls the exposure time Te when imaging one image, that is, the imaging time interval when imaging one image. Image data captured by the camera 41 is sequentially sent to the image analysis unit 32 of the control device 40. The image analysis unit 32 analyzes a plurality of images captured by the camera 41 and detects the peak position of the interference fringe for each pixel. The z coordinate of the position (x, y) corresponding to the pixel on the measured surface 12a is calculated from the peak position for each pixel.

Next, the principle of obtaining the height distribution of the measured surface 12a using the interferometer 42 will be briefly described.
As shown in FIG. 4, the light emitted from the light source 43 passes through the condenser lens 44 and the optical filter 45, is reflected by the lower surface of the half mirror 46, travels downward, and is separated into two lights by the half mirror 48. A part of light (measurement light) transmitted through the half mirror 48 is reflected by the measurement surface 12a of the wiring board 12 and returns to the half mirror 48 again. On the other hand, another part of light (reference light) reflected by the upper surface of the half mirror 48 is reflected by the reference mirror 49 and returns to the half mirror 48. The measurement light and the reference light having different optical paths merge at the upper surface of the half mirror 48. The distance between the half mirror 48 and the reference mirror 49 in the scanning unit 55 is a fixed distance Zm (hereinafter also referred to as “reference distance Zm”), and the distance Zs between the half mirror 48 and the measured surface 12a (hereinafter “ The measurement distance Zs ”also changes as the scanning unit 55 scans in the Z direction. The optical path difference between the two reflected lights during scanning by the scanning unit 55 is 2 (Zs−Zm).

  In this embodiment, the white light from the light source 43 that has passed through the optical filter 45 is light having a wide spectral width although some of the light components are removed. Therefore, the measurement light and the reference light have an optical path difference 2 (Zs− Interference occurs only in a very small range (Zs≈Zm) where Zm) is equal to or less than a specified value. Using this property, the control unit 60 scans the scanning unit 55 in the Z direction to change the optical path difference 2 (Zs−Zm), and the image analysis unit 32 generates Zs≈Zm based on a plurality of images. The scanning position (Z position) when the peak appears (that is, when Zs = Zm) within the generation period of the interference fringes is obtained for each pixel. At this time, the color of the light from the light source 43 is limited to some extent by passing through the optical filter 45, and the frequency of the interference fringe is adjusted so as to be substantially constant over the entire period (interference possible distance) of the interference fringe. . From the known information such as the X and Y positions (X, Y) of the three-dimensional optical unit 20 and the magnification of the camera 41, the imaging range of the camera 41 on the measured surface 12a of the wiring board 12 is known. A pixel in the range can be associated with a position on the measured surface 12a in the imaging range. The image analysis unit 32 calculates the distribution (x, y, z) of the height (z coordinate) of each position (x, y) corresponding to the pixel of the camera 41 on the measured surface 12a based on the Z position for each pixel. Find a 3D profile.

  FIG. 5 shows the relationship between the optical path difference and the brightness of each pixel. As shown in this graph, an interference fringe IF occurs in a range where the optical path difference (= 2 (Zs−Zm)) is equal to or less than a predetermined specified value (interference possible distance Lf), and the luminance waveform for each pixel has an optical path difference. The amplitude is varied according to the change, and the peak of the interference fringe IF is maximized when the optical path difference is zero (that is, Zs = Zm).

  In the present embodiment, the reference distance Zm is set to the focal length of the objective lens 47, for example, and when the focus is on the measurement target surface 12a of the wiring board 12 (for example, the upper surface of the wiring 36 having no defect), the measurement distance Zs. And the reference distance Zm substantially coincide with each other so that the interference fringe IF appears. With this setting, the scanning unit 55 moves a predetermined stroke (as an example, 50 μm) around the position where the objective lens 47 is focused on, for example, the measured surface 12a.

As shown in FIG. 6, the case where the position (height) of each of the low surface PA and the high surface PB in the Z direction is measured with the step 12b as an example will be described using the wiring board 12 having the step 12b as an example.
For example, in the scanning process in which the scanning unit 55 is moved in the Z direction, the optical path difference 2 (Zs−Zm) indicated by twice the difference between the measurement distance Zs between the half mirror 48 and the lower surface PA and the reference distance Zm. Within the specified range, as shown in FIG. 7, an interference fringe IF appears in the light hitting a pixel (referred to as “pixel A” in FIG. 7) that images the low surface PA of the camera 41. For this reason, in the area where the interference fringe IF is generated, the brightness of the light hitting the pixel A changes according to the scanning position. At this time, the interference fringe IF does not appear in the light hitting a pixel (referred to as “pixel B” in FIG. 7) that images the high surface PB of the camera 41.

  In the scanning process of the scanning unit 55, the optical path difference 2 (Zs−Zm) indicated by twice the difference between the measurement distance Zs between the half mirror 48 and the high surface PB and the reference distance Zm is within a specified range. Then, an interference fringe IF appears in the light that strikes the pixel B that images the high surface PB of the camera 41. For this reason, the brightness of the light striking the pixel B in the interference fringe IF generation area changes according to the scanning position. At this time, the interference fringe IF does not appear in the light hitting the pixel A that images the lower surface PA of the camera 41.

  The distance 2ΔZ between the scanning position Za of the scanning unit 55 when the interference fringe IF appears in the light hitting the pixel A and the scanning position Zb of the scanning unit 55 when the interference fringe IF appears in the light hitting the pixel B is This is twice the height ΔZ of the step 12b. If the peak positions Za and Zb in the Z direction of the interference fringes of the light impinging on the pixels A and B are respectively determined, 1/2 of the distance 2ΔZ between the peak positions Za and Zb is determined as the height ΔZ of the surface PB with respect to the surface PA. be able to.

  By the way, in the method of sampling a plurality of interference fringe waveforms per cycle as described in Patent Document 2, in order to obtain the peak of the interference fringe, at least 1/3 or 1 / of the pitch of the interference fringe IF is obtained. It is necessary to acquire a large number of images with sampling of 100 nm (nanometer) or less corresponding to 4. In this case, if the image data is acquired in the range of the movement stroke of the scanning unit 55 of 50 μm, 500 or more images are required. In this case, although accuracy on the order of nm is obtained, it takes a very long time for imaging, the required memory size becomes enormous, and the processing time for calculating the peak position for each pixel becomes very long, which is not practical. Further, it is difficult to suppress vibration within the exposure time when a large number of images are acquired in a sampling step of 100 nm or less while scanning the scanning unit 55 in the Z direction. From this point, mounting on the pattern inspection apparatus 11 is difficult. It becomes difficult. On the other hand, if the imaging speed of the camera 41 is increased to reduce the influence of vibration, the image becomes dark and the peak position detection accuracy decreases.

  On the other hand, accuracy of 100 nm order instead of nm order is sufficient for verifying defects of the wiring board 12 in the pattern inspection apparatus 11. Therefore, it is preferable to lengthen the exposure time so that the number of images to be captured can be reduced. However, since the interference fringe IF has a waveform in which the bright part and the dark part appear periodically, when the light source 43 is continuously emitted, the pixel of the camera 41 receives light with a constant exposure time Te that is equal to or longer than the period of the interference fringe IF. In this case, the bright part and the dark part of the interference fringe IF of light hitting each pixel are averaged, making it difficult to detect the peak.

  Therefore, as shown in FIG. 8, in this embodiment, the light source 43 (LED) blinks in the same cycle (in other words, the same frequency) and the same phase as the interference fringe IF waveform (hereinafter also referred to as “interference waveform”). Let Thereby, the camera 41 selectively images a bright portion (portion above the center line of the input light waveform in FIG. 8) as an example corresponding to the lighting period of the light source 43 in the interference waveform. In the example of FIG. 8, the input light waveform of the pixel of the camera 41 is equivalent to selecting the waveform of the bright part at the exposure time Te. Therefore, the pixel output of the camera 41 (camera output in FIG. 8) takes a luminance value corresponding to the integrated value (integrated value) of the amount of light received by the pixel during the exposure time Te. In the example of FIG. 8, the luminance value of the corresponding pixel in the captured image takes a value corresponding to the amount of received light obtained by selectively integrating the bright part. In this case, since the light quantity of the input light to the camera 41 when the light source 43 blinking in the same cycle as the interference fringe IF is always very low, it is an integral value of the light reception amount of the pixel in the exposure time Te. The camera output value (pixel luminance value) is substantially proportional to the height of the envelope EV (two-dot chain line in FIG. 5) of the interference fringe IF. The image analysis unit 32 performs interpolation calculation between the luminance values using the luminance values of corresponding pixels between different images based on the plurality of images, and puts them on the envelope EV with a position accuracy finer than the imaging interval. Are calculated, and the peak position of the interference fringe IF is obtained by calculating the peak position of the envelope EV using these brightness values. In FIG. 8, the camera output value (luminance value) is indicated by dots, and the temporal change in the amount of light received by the pixels of the camera 41 during the exposure time Te is indicated by a solid line.

  Further, as shown in FIG. 8, an exposure time Te of one cycle or more of the LED drive waveform of the light source 43 (LED) is set. In particular, the exposure time Te is a natural number multiple of one cycle of the LED drive waveform. In the example of FIG. 8, the exposure time Te is made equal to one cycle of the LED drive waveform. Of course, the present invention is not limited to this, and the exposure time Te may be a natural number times the blinking cycle of the light source 43. For example, the exposure time Te may be twice or three times the period of the LED drive waveform. In addition, since it is not necessary to devise a special device for extending the exposure time, the exposure time Te is set so that three or more images can be captured while the scanning unit 55 moves in the interval of the interference possible distance Lf where the interference fringe IF appears. It is desirable to set. That is, the exposure time Te is preferably set to a value within a range of ¼ to ３ of the generation period of the interference fringe IF (that is, the time for scanning the interference possible distance Lf). Of course, the exposure time Te can be set to a value exceeding 1/3 of the generation period of the interference fringe IF as long as the peak of the interference fringe IF can be detected. In this case, a special device (such as a peak detection analysis process) that can set the exposure time Te as long as possible while enabling the peak detection of interference fringes may be used in combination. For example, when the required accuracy is relatively rough, an exposure time Te that is longer than the generation period of the interference fringe IF may be set.

  By the way, in the example of FIG. 8, the bright part of the interference waveform and the lighting period of the LED drive waveform almost coincide with each other, and the phase of the interference waveform and the LED drive waveform almost coincide with each other. Can happen. For example, as shown in FIG. 10A, when the phase of the LED drive waveform is shifted by about 90 degrees with respect to the phase of the interference waveform, the interference waveform is added to the camera input light waveform during the lighting period of the LED drive waveform within the exposure time Te. Nearly half of the light and dark areas appear. In this case, even if the amplitude of the waveform is different, the difference between the bright part and the dark part with respect to the center of the amplitude is approximately the same, so that the bright part and the dark part are averaged and the correspondence in the plurality of images captured by the camera 41 is obtained. No difference occurs in the amount of received light at the exposure time Te between the pixels to be processed. As a result, it becomes difficult to calculate the peak position of the interference fringe IF from the luminance value for each pixel of the camera 41. Therefore, in the present embodiment, the following control is adopted.

  The graph shown in FIG. 9 shows the scanning position (Z position) of the scanning unit 55, the interference waveform at the pixel of the camera 41, the LED driving waveform, and the camera imaging brightness (camera output) in order from the top. The camera imaging luminance indicates a luminance value corresponding to the amount of light received by the pixel of the camera 41 at the time of measurement (at the time of imaging). As shown in FIG. 9, the scanning unit 55 performs one reciprocal scanning with respect to one measurement area, and at this time, the light emission control unit 65 controls the phase of the LED drive waveform to be shifted by 90 degrees between the forward movement process and the backward movement process. I do. In the forward scanning process in which the scanning unit 55 moves from the origin (position 0 μm) that is the scanning start position to the scanning upper limit position Zo (position 50 μm), the phase of the LED driving waveform is set in advance, and in the backward scanning process. Control is performed to shift the phase of the LED drive waveform by 90 degrees with respect to the phase of the forward scanning process.

  For example, if the phase of the interference waveform and the LED drive waveform is shifted by about 90 degrees in the forward scanning process shown in FIG. 10A, it is difficult to calculate the peak position of the interference fringe IF as described above. Even in this case, in the backward scanning process shown in FIG. 10B, the phases of the interference waveform and the LED drive waveform are the same, and only the bright part is integrated within the exposure time Te. The peak position of the interference fringe IF can be calculated. In this way, by shifting the phase at which the light source 43 blinks by 90 degrees between the forward movement and the backward movement of the scanning unit 55, the peak position of the interference fringe IF can be calculated at least one of the forward movement and the backward movement. Here, the image analysis unit 32 performs an interpolation calculation between the luminance values of the corresponding pixels in different images, using the luminance values of each pixel in the plurality of images captured by the camera 41 during the scanning of the scanning unit 55. Thus, the peak position of the envelope EV of the interference fringe IF is calculated. In addition, the peak position detection process of the interference fringe for each pixel by the image analysis unit 32 is performed at high speed by an image processing circuit in the computer 25.

Next, operations of the pattern inspection apparatus 11 and the three-dimensional profile acquisition apparatus 21 will be described.
First, a pattern inspection is performed on all the wirings 36 formed on the wiring board 12. The inspection unit 31 requests verification processing when pattern inspection is performed based on the image of the measurement target surface 12a captured by the camera 19 of the imaging unit 17 and a genuine and suspected defect candidate is detected. For example, the oxide film 36a, the concave portion 36b, and the missing portion 36c on the wiring 36 shown in FIG. 2 are detected as defect candidates in the pattern inspection and are to be verified. The control unit 60 that has received the verification request moves the stage 14 to a position in the X direction where the defect candidate can be imaged, and moves the three-dimensional optical unit 20 to a position in the Y direction where the defect candidate can be imaged.

  Before the start of scanning, the scanning unit 55 is scanned at least once with the light source 43 being lit, and calibration is performed to obtain at least the phase and the phase of the interference fringe based on the light reception signal of the optical sensor 56. The phase of the interference fringe with respect to the scanning position of the scanning unit 55 is obtained by this calibration.

  The control device 40 executes a three-dimensional profile acquisition processing routine program shown in FIG. Hereinafter, the three-dimensional profile acquisition processing routine will be described with reference to the flowchart of FIG. 11 includes control performed by the control unit 60 in the controller 23 such as light emission control of the light source 43, scanning control of the scanning unit 55, and imaging control of the camera 41, and processing by the image analysis unit 32. It is. Note that the computer 25 may perform part or all of the control performed by the control unit 60. Further, the image analysis unit 32 may be provided in the controller 23.

  In step S1, the scanning unit 55 is moved forward. That is, the control unit 60 outputs a piezo drive signal to the piezo drive unit 62 to change the voltage applied to the actuator 50, thereby changing the scanning unit 55 from the scan start position (origin = position 0 μm) to the scan end position (position 50 μm). ) At a constant speed.

  In step S2, the light source is blinked at the same cycle as the interference fringe. That is, the light emission control unit 65 generates an LED drive pulse signal having the same cycle as the cycle of the interference fringe IF, outputs this pulse signal to the light emission drive unit 61, and applies a pulse voltage to the light source 43 to interfere with the light source 43. Flashes in the same cycle as fringe IF.

  In step S3, the surface to be measured 12a is imaged with an exposure time that is a natural number times the cycle of the interference fringe. That is, the imaging control unit 63 causes the camera 41 to perform imaging for each exposure time Te. Image data captured by the camera 41 is sequentially sent to the image analysis unit 32.

  In the next step S4, a plurality of images are analyzed to detect a Z position where an interference fringe peak appears for each pixel. That is, the image analysis unit 32 stores the luminance of the pixel that images the measured surface 12a when no interference fringe appears in the memory as the reference luminance value. Then, the image analysis unit 32 calculates a difference ΔB between the luminance value and the reference luminance value for each pixel of the image data, and sequentially stores the difference ΔB for each pixel in the memory. Further, the image analysis unit 32 performs interpolation calculation using the difference ΔB for each pixel for a plurality of times, and for each position (scanning position) finer than the imaging pitch at which the scanning unit 55 moves during one imaging. The difference on the envelope EV (see FIG. 5) of the interference fringe IF for each pixel is sequentially obtained. Then, the image analysis unit 32 sequentially compares the differences on the envelope EV and detects the scanning position when the maximum value is reached as the peak position (Z position).

  It should be noted that in the appearance period of the interference fringe IF, the interference monitor section 64 can detect the period and phase of the interference fringe because the light reception signal during the lighting period of the optical sensor 56 includes the interference fringe waveform. The light emission control unit 65 generates an LED drive pulse signal in accordance with the phase of the interference fringe based on the light reception signal of the optical sensor 56. For this reason, the phase of the blinking waveform of the light source 43 matches the phase of the interference fringe in the generation period of the interference fringe IF, a relatively large difference ΔB is obtained, and the peak position detection sensitivity is increased.

  In step S5, it is determined whether or not the forward movement of the scanning unit has been completed. The control unit 60 determines whether or not the value of the piezo drive signal has reached a value when the scanning unit 55 has finished moving a predetermined stroke (for example, 50 μm) and is at the scanning upper limit position Zo. If it is before the end of the forward movement of the scanning part 55, it will return to step S3 and will repeat the imaging process of step S3 and the peak position detection process by the image analysis of S4 until it determines with the forward movement end in step S5.

  For example, when measuring the measurement target surface 12a of the wiring board 12 shown in FIG. 6, in the forward movement process of the scanning unit 55, the interference fringe IF is incident on the light hitting the pixel A that images the low surface PA at the scanning position Za shown in FIG. A peak appears, and an interference fringe IF peak appears in the light striking the pixel B that images the high surface PB at the scanning position Zb. For this reason, in step S4, peak positions Za and Zb corresponding to the heights of the surfaces PA and PB on the measured surface 12a are detected. When the forward movement of the scanning unit 55 ends, the process proceeds to step S6.

  In step S6, the scanning unit 55 is moved backward. That is, the control unit 60 outputs a piezo drive signal to the piezo drive unit 62 and changes the voltage applied to the actuator 50 in the direction opposite to that in the forward movement, thereby scanning the scan unit 55 from the scan end position (position 50 μm). The actuator is moved backward at a constant speed to the start position (origin = position 0 μm).

  In step S7, the light source is blinked at the same cycle as the interference fringe and at a phase different by 90 degrees from the time when the scanning unit 55 moves forward. That is, the period of the interference fringe previously monitored by the interference monitor unit 64 is known, and the output timing of the LED drive pulse signal by the light emission control unit 65 with respect to the scanning position of the scanning unit 55 is the output during the forward movement of the scanning unit 55. The phase of the LED driving pulse signal is shifted by 90 degrees with respect to the forward movement by shifting the timing by 1/4 period. Then, an LED drive pulse signal having the same cycle as that of the interference fringe IF with the phase shifted by 90 degrees is output to the light emission drive unit 61, and the light source 43 has the same cycle as the interference fringe IF and 90 degrees different from the forward movement of the scanning unit 55. Flashes in phase.

  Each process of next step S8 and S9 is a process similar to each process of step S2 and S3. That is, in step S8, the surface to be measured 12a is imaged by the camera with an exposure time that is a natural number times the interference fringe period. Image data captured by the camera 41 is sequentially sent to the image analysis unit 32. In step S9, a plurality of images are analyzed to detect a Z position where an interference fringe peak appears for each pixel.

  In step S10, it is determined whether or not the backward movement of the scanning unit has ended. That is, the control unit 60 determines whether or not the value of the piezo drive signal has reached the value when the scanning unit 55 is at the scanning start position. If it is before the end of the backward movement of the scanning unit 55, the process returns to step S8, and the imaging process in step S8 and the peak position detection process by the image analysis in S9 are repeated until it is determined in step S10 that the backward movement is completed. When the backward movement of the scanning unit 55 ends, the process proceeds to step S11. For example, even if the phase of the flicker waveform of the light source 43 is shifted by approximately 90 degrees with respect to the phase of the interference fringe IF on one side during the forward movement and the backward movement of the scanning unit 55, That is, the phase of the blinking waveform of the light source 43 is substantially matched or shifted by 180 degrees. For this reason, the peak position (Z position) of the envelope EV of the interference fringe IF can be detected based on the difference ΔB during the other scanning.

  In step S11, a three-dimensional profile is generated based on the Z position detected for each pixel. That is, the image analysis unit 32 compares, for each pixel, the magnitude (difference) between the peak detected when the scanning unit 55 moves forward and the peak detected when the scanning unit 55 moves backward, and the larger peak is detected. Adopt Z position. In this case, the peak size is compared for some of all the pixels, and it is determined whether the peak position to be used is for the forward movement or the backward movement of the scanning unit 55, and one of the determined peaks The structure which employ | adopts a position may be sufficient. When the Z position for each pixel to be adopted is determined, a three-dimensional profile PD is generated based on the z coordinate corresponding to the Z position.

  When the three-dimensional profile PD in which each pixel is expressed by coordinates (x, y, z) is generated in this way, the three-dimensional profile PD is sent to the verification unit 33. The verification unit 33 acquires the height distribution of the surface of the suspicious defect that is genuine from the three-dimensional profile PD, and verifies the authenticity of the defect based on the height. For example, in the case of the oxide film 36a shown in FIG. 2, since the height of the upper surface of the wiring 36 is not less than a specified height, it is verified that it is not a defect. For example, in the case of the recess 36b shown in FIG. 2, if the height of the upper surface of the wiring 36 is equal to or higher than a specified height, it is verified that it is not a defect, and if it is less than the specified height, it is verified as a defect. Further, for example, in the case of the missing portion 36c shown in FIG. 2, the height of the upper surface of the wiring 36 is less than the specified height, so that it is verified that it is a defect.

  Note that the calculation for detecting the peak position of the interference fringe for each pixel based on the plurality of images in Steps S4 and S9 ended all the imaging performed by the camera 41 during the scanning of the scanning unit 55 in Step S10 (Yes in S10). (Determination) may be performed later. In other words, in the above example, the peak position calculation process is performed during imaging of the forward path and the backward path of the scanning unit 55. However, after all the imaging by the camera 41 is completed in the forward path and the backward path of the scanning unit 55, the peak position of the interference fringe for each pixel is calculated in the forward path and the backward path based on a plurality of images stored in the memory until then, The peak position of each pixel may be detected by adopting the higher peak in the forward path and the backward path of the same pixel.

As described above in detail, according to the first embodiment, the following effects can be obtained.
(1) Since the light source 43 blinks in synchronism with the period of the interference fringe IF generated by the interferometer 42, the exposure time Te of the reflected light image that the camera 41 has passed through the interferometer 42 is longer than the period of the interference fringe IF. When the image is picked up by (1), the light of the interference fringe hits the pixel of the camera 41 every half cycle when the light source 43 is turned on. For this reason, the pixels of the camera 41 receive the interference fringe IF every half cycle. At this time, for example, only the bright part or only the dark part of the interference fringe IF is imaged. As a result, the pixel of the image data takes a luminance value reflecting the envelope EV of the interference fringe IF, and the image analysis unit 32 can detect the peak position of the interference fringe IF for each pixel based on a plurality of images. Therefore, the image analysis unit 32 can acquire the height distribution of the measured surface 12a from the peak position of the interference fringe IF detected for each pixel, and generate the three-dimensional profile PD.

  (2) Since the scanning unit 55 is reciprocally scanned and the phase at which the light source 43 blinks is shifted by approximately 90 degrees between the forward movement and the backward movement of the scanning section 55, interference occurs in at least one of the forward movement and the backward movement. The peak position of fringe IF can be detected. For example, the peak position of the interference fringe IF can be detected for each pixel and the three-dimensional profile PD can be generated even if the phase of the blinking of the light source 43 and the interference fringe cannot be strictly matched. Further, the three-dimensional profile PD can be generated with higher accuracy than the configuration in which the scanning unit 55 is scanned only once.

  (3) An optical sensor 56 that receives a part of the reflected light that has passed through the interferometer 42 is provided, and the light source 43 can be blinked at the phase of the interference fringe IF based on the light reception signal of the optical sensor 56. For this reason, it is possible to reduce the interference fringe IF peak position detection error caused by averaging the bright and dark portions of the interference fringe IF within the exposure time Te during imaging.

  (4) The exposure time Te of the camera 41 is set to a value within a range of ¼ to ３ of the generation period of the interference fringe IF determined from the scanning speed of the scanning unit 55. For this reason, the peak position of the interference fringe IF can be detected using a relatively small number of images, for example, a fraction of 1 to a few tens of times, compared to a case where a peak is detected by a sampling method. Also, the memory size required for storing image data can be relatively small. In addition, since the number of images can be reduced, the peak position detection process performed by the image analysis unit 32 using a plurality of images is a relatively simple process that does not require any special device (such as complicated peak detection analysis process). The acquisition process of the three-dimensional profile PD by the image analysis unit 32 can be speeded up.

  (5) The three-dimensional profile acquisition device 21 is applied to the pattern inspection device 11, and the verification unit 33 acquires the authenticity of the defect candidate detected by the pattern inspection of the wiring board 12 by the three-dimensional profile acquisition device 21. Verification based on the three-dimensional profile PD. Therefore, in the inspection of the wiring board 12 having a high aspect ratio due to the demand for miniaturization of the wiring 36, it is possible to accurately detect a defect in which the wiring 36 is insufficient in height.

  (6) Since the blinking cycle of the light source 43 is set to the same cycle as the interference fringe IF, the waveform of each half cycle of the interference fringe IF can be detected based on the light reception signal of the optical sensor 56 during the scanning of the scanning unit 55. The phase of the LED drive pulse signal can be matched with the phase of the interference fringe IF based on the received light signal. For this reason, the detection sensitivity of the interference fringe IF peak can be enhanced, and a three-dimensional profile PD with relatively high accuracy can be obtained. For example, the verification accuracy of defect candidates by the verification unit 33 can be increased.

Embodiment is not limited above, It can also change into the following aspects.
In the above-described embodiment, the scanning unit 55 is reciprocated in the Z direction perpendicular to the measured surface 12a. However, it is possible to perform only one scanning or only one scanning. For example, when the LED drive pulse signal having the same phase as the interference fringe can be output from the phase synchronization circuit in the light emission control unit 65 that receives the light reception signal of the optical sensor 56, the required accuracy can be obtained even with one scanning of the scanning unit 55. A three-dimensional profile PD can be acquired. Further, if the phase of the LED drive waveform can be correctly calibrated based on the information on the phase of the interference fringe IF acquired by sampling the interference fringe waveform by the optical sensor 56 and the interference monitor unit 64, the interference fringe can be detected even in one scan of the scanning unit 55. It is possible to accurately detect the IF peak position and obtain a three-dimensional profile PD having a certain level of accuracy.

  The number of scans of the scanning unit 55 when measuring the height distribution of the same area of the surface to be measured is not limited to one reciprocation (two scans), but one reciprocation half (three scans), a plurality of reciprocations (4 or more even-number scans) 5 or more odd-number scans may be used. In these cases, it is preferable to gradually shift the phase during different forward and backward movements, and it is preferable that at least one of the shifted phases includes a case where the phase is shifted by 90 degrees. Of course, a configuration in which the blinking phase of the light source 43 is not shifted may be used.

  The optical sensor 56 and the interference monitor unit 64 may be eliminated. Even in this configuration, the peak position of the interference fringe IF can be detected by changing the blinking phase of the light source 43 by 90 degrees between the forward movement and the backward movement of the scanning section 55, so that a three-dimensional profile can be generated.

  -The light source 43 should just be what can light-emit light with a wide spectrum width, and the luminescent color is not necessarily limited to white light. Any interference fringe IF can be generated in a range where the spectrum of light from the light source is wide and the optical path difference is narrow, and can be used for white light interferometry. The light source 43 is not limited to an LED, but may be a light bulb, an organic EL light emitter, or a fluorescent lamp. The light from the light source 43 may partially include infrared light or ultraviolet light that can be received by the camera 41.

The light source 43 may be blinked at a phase where the light source 43 is turned on in accordance with the dark part of the interference fringe, and the peak position of the interference fringe may be obtained based on the luminance value obtained by integrating the dark part of the interference fringe.
The angle at which the blinking phase of the light source 43 is shifted between forward scanning and backward scanning is not limited to 90 degrees, but may be other angles. For example, a value in the range of 80 to 100 degrees is preferable, but the peak position detection accuracy can be improved by either a forward movement or a backward movement only by shifting a little angle. Further, when the phase is shifted, the phase at the backward movement may be advanced or delayed with respect to the forward movement of the scanning unit 55. Even in this configuration, it is possible to reliably detect the peak position of the interference fringe at least during the forward movement and the backward movement of the scanning unit 55.

-The structure which does not control the phase which adjusts the phase at the time of blinking of the light source 43 to interference fringe may be sufficient. The peak position of the interference fringe can be detected without controlling the phase.
・ Instead of the Michelson interferometer, a small reference mirror (reflecting mirror) is attached to the center surface of the objective lens, a half mirror is placed in the middle of the focal plane, and the light from the surface to be measured and the reference mirror on the lens A Mirau interferometer in which reflected light interferes may be used.

  -At least a part of the control unit 60 may be configured by software, may be configured by hardware, or may be configured by cooperation of software and hardware. The image analysis unit 32 may be configured by software, or may be configured by cooperation of software and hardware.

  The wiring board 12 may be a wiring board provided with a wiring pattern such as a package, a TAB tape, or an IC chip, or a wiring board for a flat display. For example, it may be a flat panel wiring board for liquid crystal displays, organic electroluminescence (organic EL), and plasma displays. The wiring board may be a glass wiring board or a plastic wiring board. Further, the pattern on the surface to be measured to be inspected is not limited to the wiring 36 but may be a bump, and the pattern of the bump on the wiring board may be inspected by the pattern inspection apparatus 11.

  -A three-dimensional profile acquisition apparatus is not limited to being applied to a pattern inspection apparatus. The measurement target is not limited to the wiring substrate, and may be any surface to be measured that has at least one of a fine concave portion, a convex portion, and a step as an inspection target. In this case, it may be inspected that the fine concave portion or convex portion is less than the prescribed depth or height, or the presence or absence of a fine concave portion or convex portion on the flat surface should be inspected. You may do it. Further, it may be used for inspection of foreign matters as in Patent Document 1, or may be used for inspection of via holes as in Patent Document 2. Furthermore, the present invention may be applied to an in-line wafer bump inspection apparatus that inspects the height or volume of micro bumps on a semiconductor wafer. You may apply to the FC bump test | inspection apparatus which test | inspects the bump height or volume of the flip chip (FC) bump mounting piece or frame-shaped IC package. Moreover, you may apply to the buildup board | substrate plating thickness monitoring apparatus which monitors the plating thickness of the two-dimensional pattern in a buildup board | substrate. Furthermore, the present invention may be applied to a high-precision print monitoring apparatus that precisely monitors and feeds back the application thickness (printing thickness) of solder or conductive ink.

  The object to be measured by the three-dimensional profile acquisition apparatus is not limited to an electronic mounting component such as a wiring board, and may be any object that needs to precisely measure and inspect the height distribution. For example, the height distribution in MEMS (Micro Electro Mechanical Systems) may be measured. For example, a display device, a sensor, or an actuator using a micromachine may be set as a measurement target. Further, the depth of a micro trench for realizing a sensor function in a biochip (for example, a DNA chip) may be managed. Further, in nanoimprint that realizes microfabrication by transferring the original plate against the substrate, it may be applied to high-speed and precise inspection of the transferred microstructure. Further, the present invention may be applied to depth distribution inspection of optical communication devices, optical ICs, optical waveguides, and the like.

  DESCRIPTION OF SYMBOLS 11 ... Pattern inspection apparatus, 12 ... Wiring board, 12a ... Measuring surface, 21 ... Three-dimensional profile acquisition apparatus, 31 ... Pattern inspection part as an example of an inspection part, 32 ... Image analysis part, 33 ... Verification part, 36 ... Wiring 41... Camera as an example of imaging unit 42. Interferometer 43. Light source 48. Half mirror 49. Reference mirror 50. Actuator 55 55 Scanning unit 56 Optical sensor 60 Control unit 65: Light emission control unit constituting an example of control unit, IF: interference fringe, Te: exposure time, A, B: pixel, Za, Zb ... peak position, PD: three-dimensional profile.

Claims (6)

A light source that emits light having a wide spectral width;
It is configured to be able to scan in a direction crossing the surface to be measured, and separates the light from the light source into measurement light reflected by the surface to be measured and reference light reflected by a reference mirror, and the two separated lights An interferometer having a scanning unit capable of generating an interference fringe based on an optical path difference by combining the reflected light and an imaging unit that captures an image of the two reflected lights with an exposure time equal to or longer than the period of the interference fringe;
A control unit that scans the scanning unit and blinks the light source at the same cycle as the interference fringe;
An image analysis unit that detects a peak position of interference fringes in the scanning direction of the scanning unit for each pixel based on a plurality of images captured by the imaging unit, and generates a three-dimensional profile based on the peak position for each pixel When,
3D profile acquisition device.   The three-dimensional profile acquisition apparatus according to claim 1, wherein the control unit causes the scanning unit to reciprocate and shifts the blinking phase of the light source during forward movement and backward movement. An optical sensor that receives a part of the two reflected lights toward the imaging unit;
3. The three-dimensional profile acquisition apparatus according to claim 1, wherein the control unit causes the light source to blink at the same phase as the interference fringe based on a light reception signal of the optical sensor that has received the interference fringe. .   The three-dimensional profile acquisition apparatus according to any one of claims 1 to 3, wherein an exposure time of the imaging unit is set to a value within a range of ¼ to ３ of an occurrence period of the interference fringe. . A pattern inspection apparatus comprising a pattern inspection unit that inspects the presence or absence of a wiring defect based on an image obtained by imaging a wiring pattern of a wiring board,
The three-dimensional profile acquisition device according to any one of claims 1 to 4,
A verification unit that verifies the authenticity of the defect detected by the pattern inspection unit based on the three-dimensional profile generated by the three-dimensional profile acquisition device;
A pattern inspection apparatus comprising: The scanning part that constitutes the interferometer is scanned in the direction intersecting the surface to be measured, and the light having a wide spectral width emitted from the light source is reflected by the surface to be measured and the reference light that is reflected by the reference mirror And an interference formation step of generating an interference fringe based on an optical path difference by combining the separated reflected lights of the two lights, and
A light emission control step of blinking the light source at the same cycle as the interference fringe;
An imaging step of capturing an image of the two reflected lights by an imaging unit of the interferometer with an exposure time equal to or longer than the period of the interference fringe
Profile generation step for detecting a peak position of interference fringes in the scanning direction of the scanning unit for each pixel based on a plurality of images captured by the imaging unit, and generating a three-dimensional profile based on the peak position for each pixel When,
A three-dimensional profile acquisition method comprising: