Pattern inspection device and pattern inspection method

Abstract

PURPOSE: To provide an inspection device that can suppress the accuracy deterioration of an image due to a change in image storage time.CONSTITUTION: A pattern inspection device according to an aspect of the present invention comprises an illumination optical system 170 (171) illuminating an inspection substrate to be tested on which a pattern is formed, an off-set calculation part 64 that calculate an off-set quantity that depends on the image storage time of each photosensor device in a plurality of photosensor devices that are arrayed two-dimensionally, a TDI sensor 105 having a plurality of photosensor devices, imaging the inspection substrate to be tested by receiving the transmitted light or reflected light from the inspection substrate to be tested with the plurality of photosensor devices, correcting the pixel value of the imaged optical image data using the off-set quantity, then outputting the corrected optical image data, and a comparator circuit 108 that compares the optical image constituted with the optical image data outputted from the TDI sensor with a reference image.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a pattern inspection apparatus and a pattern inspection method. For example, it relates to an apparatus and a method for inspecting pattern defects of an exposure mask used in semiconductor manufacturing.

Background Art

[0002] In recent years, with the increasing integration and large capacity of large-scale integrated circuits (LSIs), the circuit line widths required for semiconductor elements have become increasingly narrow. These semiconductor elements are manufactured by exposing and transferring a pattern onto a wafer using a reduction projection exposure apparatus called a stepper, using an original pattern (also referred to as a mask or reticle, hereinafter collectively referred to as a mask) on which a circuit pattern is formed.

[0003] And for the manufacture of LSIs that require a large manufacturing cost, improving the yield is essential. One of the major factors that reduce the yield is the pattern defects of the mask used when exposing and transferring ultra-fine patterns onto a semiconductor wafer by photolithography technology. In recent years, with the miniaturization of LSI pattern dimensions formed on a semiconductor wafer, the dimensions that must be detected as pattern defects have also become extremely small. Therefore, there is a need to improve the accuracy of pattern inspection apparatuses for inspecting defects in transfer masks used in LSI manufacturing.

[0004] As inspection methods, for example, there is "die to die inspection" in which optical image data obtained by imaging the same pattern at different locations on the same mask are compared, and "die to database inspection" in which drawing data (design data) converted into a device input format for input by a drawing device when drawing a pattern on a mask from pattern-designed CAD data is input to an inspection apparatus, and a reference image is generated based on this, and this is compared with an optical image that is measurement data obtained by imaging the pattern.

[0005] As the pattern is miniaturized, a technique of improving the transfer resolution by adjusting the transmittance or reflectance of the material of the exposure mask is adopted. On the other hand, in an inspection apparatus, if the transmittance or reflectance of the mask serving as a sample decreases, an image having sufficient information cannot be obtained. Since there is a limit to increasing the amount of light from the light source, it has been considered to cope with this by increasing the image accumulation time of each photosensor element of the imaging sensor. However, when the sensor is operated with different image accumulation times, the sensor temperature changes and the dark noise level changes. As a result, there has been a problem that the accuracy of the obtained image deteriorates. As a result, pseudo defects occur.

[0006] Although it is not a pattern inspection apparatus such as a mask, in an image recognition apparatus that recognizes a subject such as a finger or a barcode, it is disclosed that the accumulation time of the sensor is shortened when the ambient temperature is high (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] Therefore, one aspect of the present invention provides an inspection apparatus and method capable of suppressing deterioration in the accuracy of an image caused by a change in the image accumulation time.

Means for Solving the Problems

[0009] A pattern inspection apparatus according to one aspect of the present invention includes an illumination optical system that illuminates a substrate to be inspected on which a pattern is formed, an offset calculation unit that calculates an offset amount depending on the image accumulation time of each photosensor element of a plurality of photosensor elements arranged two-dimensionally, A time delay integration (TDI) sensor having a plurality of photosensor elements, imaging an inspection substrate by receiving transmitted or reflected light from the inspection substrate with the plurality of photosensor elements, correcting pixel values of optical image data imaged using an offset amount, and outputting the corrected optical image data; A comparison unit that compares an optical image composed of optical image data output from the TDI sensor with a reference image; characterized by comprising.

[0010] Further, the offset calculation unit calculates an offset amount according to the imaging timing, It is preferable that the TDI sensor corrects the optical image data using an offset amount corresponding to the imaging timing for each imaging timing.

[0011] Further, it is preferable that the TDI sensor corrects the optical image data using a constant offset amount after a predetermined period has elapsed from the switching point of the image accumulation time.

[0012] A pattern inspection apparatus according to another aspect of the present invention An illumination optical system that illuminates an inspection substrate on which a pattern is formed; A temperature change amount calculation unit that calculates a temperature change amount depending on the image accumulation time of each photosensor element of a plurality of photosensor elements arranged two-dimensionally; A time delay integration (TDI) sensor having a plurality of photosensor elements, imaging an inspection substrate by receiving transmitted or reflected light from the inspection substrate with the plurality of photosensor elements while controlling the temperature of the plurality of photosensor elements using the temperature change amount; A comparison unit that compares the captured optical image with a reference image; characterized by comprising.

[0013] Further, it is preferable that the TDI sensor has an air cooling mechanism, a water cooling mechanism, or a heat pipe mechanism.

[0014] Furthermore, it further includes an offset calculation unit that calculates an offset amount according to a temperature error when controlling the temperatures of a plurality of photosensor elements. The TDI sensor is preferably used to correct the optical image data using the offset amount.

[0015] A pattern inspection method according to one aspect of the present invention includes: a step of illuminating a substrate to be inspected on which a pattern is formed; a step of calculating an offset amount depending on the image accumulation time of each of a plurality of photosensor elements arranged two-dimensionally; a step of imaging the substrate to be inspected by receiving transmitted light or reflected light from the substrate to be inspected with a plurality of photosensor elements of a time delay integration (TDI) sensor having the plurality of photosensor elements, correcting pixel values of the imaged optical image data using the offset amount, and outputting the corrected optical image data; a step of comparing an optical image constituted by the optical image data output from the TDI sensor with a reference image; and is characterized by including the above steps.

[0016] A pattern inspection method according to another aspect of the present invention includes: a step of illuminating a substrate to be inspected on which a pattern is formed; a step of calculating a temperature change amount depending on the image accumulation time of each of a plurality of photosensor elements arranged two-dimensionally; a step of imaging the substrate to be inspected by receiving transmitted light or reflected light from the substrate to be inspected with a plurality of photosensor elements of a time delay integration (TDI) sensor having the plurality of photosensor elements while controlling the temperatures of the plurality of photosensor elements using the temperature change amount; a step of comparing the imaged optical image with a reference image; and is characterized by including the above steps.

Advantages of the Invention

[0017] According to one aspect of the present invention, it is possible to suppress deterioration in the accuracy of an image caused by a change in the image accumulation time.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Best Mode for Carrying Out the Invention

[0019] Embodiment 1. FIG. 1 is a configuration diagram showing the configuration of the pattern inspection apparatus according to Embodiment 1. In FIG. 1, an inspection apparatus 100 for inspecting defects in a pattern formed on an inspection target substrate, for example, a mask, includes an optical image acquisition mechanism 150 and a control system circuit 160.

[0020] The optical image acquisition mechanism 150 includes a light source 103, a transmission illumination optical system 170, a reflection illumination optical system 171, an XYθ table 102 arranged to be movable, a magnifying optical system 104, a beam splitter 174, an imaging optical system 176, a TDI (Time Delay Integration) sensor 105, a stripe pattern memory 123, a laser length measurement system 122, and an auto loader 130. When performing transmission inspection using transmitted light, the reflection illumination optical system 171 and the beam splitter 174 may be omitted. When performing reflection inspection using reflected light, the transmission illumination optical system 170 may be omitted.

[0021] On the XYθ table 102, a substrate 101 conveyed from the auto loader 130 is arranged. The substrate 101 includes, for example, a photomask for exposure that transfers a pattern to a semiconductor substrate such as a wafer. Further, a plurality of graphic patterns to be inspected are formed on this photomask. The substrate 101 is arranged on the XYθ table 102 with, for example, the pattern formation surface facing downward.

[0022] The TDI sensor 105 includes a photosensor array 124, a sensor circuit 125, and an offset circuit 126. The photosensor array 124 has a plurality of photosensor elements arranged two-dimensionally. When each photosensor element captures an image, a predetermined image accumulation time is set. In the TDI sensor 105, the outputs of a plurality of photosensor elements arranged in the scan direction are integrated and output. A plurality of photosensor elements arranged in the scan direction capture the same pixel while shifting in time according to the movement of the XYθ table 102.

[0023] In the control system circuit 160, a control computer 110 that controls the entire inspection apparatus 100 is connected via a bus 120 to a position circuit 107, a comparison circuit 108, a reference image creation circuit 112, an auto-loader control circuit 113, a table control circuit 114, an offset amount calculation circuit 136, a magnetic disk device 109, a memory 111, a magnetic tape device 115, a flexible disk device (FD) 116, a CRT 117, a pattern monitor 118, and a printer 119. Further, the TDI sensor 105 is connected to a stripe pattern memory 123, and the stripe pattern memory 123 is connected to a plurality of comparison circuits 108. Also, the XYθ table 102 is driven by an X-axis motor, a Y-axis motor, and a θ-axis motor. The XYθ table 102 is an example of a stage. Further, the reference image creation circuit 112 is connected to the comparison circuit 108. The offset amount calculation circuit 136 is connected to an offset circuit 126.

[0024] Note that a series of "~ circuits" such as the position circuit 107, comparison circuit 108, reference image creation circuit 112, autoloader control circuit 113, table control circuit 114, and offset amount calculation circuit 136 have a processing circuit. Such a processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device, etc. Alternatively, different processing circuits (separate processing circuits) may be used. For example, a series of "~ circuits" such as the position circuit 107, comparison circuit 108, reference image creation circuit 112, autoloader control circuit 113, table control circuit 114, and offset amount calculation circuit 136 may be configured and executed by the control computer 110. The input data or calculated results required for the position circuit 107, comparison circuit 108, reference image creation circuit 112, autoloader control circuit 113, table control circuit 114, and offset amount calculation circuit 136 are stored in a memory (not shown) within each circuit or the memory 111 each time. The input data or calculated results required for the control computer 110 are stored in a memory (not shown) within the control computer 110 or the memory 111 each time. A program for executing a processor or the like may be recorded on a recording medium such as a magnetic disk device 109, a magnetic tape device 115, an FD 116, or a ROM (read-only memory).

[0025] In the inspection apparatus 100, a high-magnification transmission inspection optical system is configured by the light source 103, XYθ table 102, transmission illumination optical system 170, magnifying optical system 104, imaging optical system 176, and TDI sensor 105. Also, a high-magnification reflection inspection optical system is configured by the light source 103, reflection illumination optical system 171, beam splitter 174, magnifying optical system 104, XYθ table 102, imaging optical system 176, and TDI sensor 105.

[0026] Also, the XYθ table 102 is driven by a table control circuit 114 under the control of a control computer 110. It is movable by a drive system such as a three-axis (X - Y - θ) motor that drives in the X direction, Y direction, and θ direction. For these X motor, Y motor, and θ motor, for example, a step motor can be used. The XYθ table 102 is movable in the horizontal direction and the rotational direction by the motors of each of the X, Y, and θ axes. Then, the moving position of the substrate 101 placed on the XYθ table 102 is measured by a laser length measuring system 122 and supplied to a position circuit 107. Also, the transfer of the substrate 101 from the autoloader 130 to the XYθ table 102 and the transfer process of the substrate 101 from the XYθ table 102 to the autoloader 130 are controlled by an autoloader control circuit 113.

[0027] Drawing data (design data) serving as a basis for pattern formation on the substrate 101 to be inspected is input from outside the inspection apparatus 100 and stored in a magnetic disk device 109. A plurality of graphic patterns are defined in the drawing data, and each graphic pattern is usually composed of a combination of a plurality of elemental graphics. Note that there may be a graphic pattern composed of one graphic. On the substrate 101 to be inspected, corresponding patterns are formed based on each graphic pattern defined in such drawing data.

[0028] Here, in FIG. 1, the components necessary for explaining Embodiment 1 are described. Needless to say, the inspection apparatus 100 may usually include other necessary configurations.

[0029] FIG. 2 is a conceptual diagram for explaining the inspection area in Embodiment 1. As shown in FIG. 2, the inspection area 10 (the entire inspection area) of the substrate 101 is virtually divided into a plurality of strip-shaped inspection stripes 20 with a scan width W of the TDI sensor 105, for example, in the Y direction. Then, in the inspection apparatus 100, an image (strip area image) is acquired for each inspection stripe 20. For each of the inspection stripes 20, an image of a graphic pattern arranged in the inspection stripe 20 in the longitudinal direction (X direction) of the stripe area is captured using a laser beam (inspection light). In order to prevent omission of images, it is preferable that the plurality of inspection stripes 20 are set so that adjacent inspection stripes 20 overlap with each other with a predetermined margin width.

[0030] As the XYθ table 102 moves, the TDI sensor 105 continuously moves relatively in the X direction while an optical image is acquired. The TDI sensor 105 continuously captures an optical image with a scan width W as shown in FIG. 2. In other words, the TDI sensor 105 captures an optical image on the surface of the substrate 101 on which a plurality of graphic patterns are formed while moving relatively in the integration direction of the TDI sensor 105. In Embodiment 1, after capturing the optical image in one inspection stripe 20, it moves to the position of the next inspection stripe 20 in the Y direction and then moves in the reverse direction while continuously capturing an optical image with a scan width W in the same manner. That is, imaging is repeated in the forward (FWD)-backward (BWD) directions going in opposite directions for the forward and return paths.

[0031] Also, in actual inspection, the stripe area image of each inspection stripe 20 is divided into images of a plurality of rectangular frame areas 30 as shown in FIG. 2. Then, inspection is performed for each image of the frame area 30. For example, it is divided into a size of 512×512 pixels. Therefore, a reference image to be compared with the frame image 31 of the frame area 30 is also created for each frame area 30.

[0032] Here, the imaging direction is not limited to the forward (FWD)-backward (BWD) repetition. Imaging may be performed from one direction. For example, the FWD-FWD repetition may be used. Alternatively, the BWD-BWD repetition may be used.

[0033] Here, as described above, a technique of improving the transfer resolution by adjusting the transmittance or reflectance of the material of the exposure mask as the pattern is miniaturized is adopted. It is desirable that the same inspection apparatus 100 can inspect a plurality of substrates 101 having different transmittances or reflectances. On the other hand, in the inspection apparatus 100, when the transmittance or reflectance of the mask serving as the substrate 101 decreases, the amount of received light by the TDI sensor 105 decreases. As a result, an image having sufficient information cannot be obtained. Since there is a limit to increasing the amount of light from the light source 103, it can be dealt with by increasing the image accumulation time of each photosensor element of the TDI sensor 105 to be imaged. For example, instead of doubling the amount of light, an image similar to the image obtained when the amount of light is doubled can be obtained by doubling the image accumulation time.

[0034] However, when each photosensor element is operated at different image accumulation times t, the sensor temperature T changes and the dark noise level Dr changes. The dark noise level Dr depends on the sensor temperature.

[0035] FIG. 3 is a diagram showing an example of a change in the dark noise level accompanying the switching of the image accumulation time of each photosensor element of the TDI sensor in the first embodiment. When the TDI sensor 105 is driven for a sufficiently long time L at the image accumulation time t1 under a certain operating condition from the state of the initial temperature T0, the sensor temperature T becomes T1 based on the operating environment and the power consumption of the photosensor element. The sensor temperature T can be defined by the following equation (1). (1) T = f1(t1, L, T0) = T1

[0036] Also, the dark noise level Dr in such a state can be defined by the following equation (2). (2) Dr = f2(T1) = Dr1

[0037] Regarding the TDI sensor 105 that was driven by the image accumulation time t1 of such a photosensor element, the image accumulation time t is switched from t1 to t2. In such a case, according to the elapsed time δ from the switching point, the sensor temperature T changes based on the operating environment and the power consumption of the photosensor element. The sensor temperature T in that case can be defined by the following formula (3). (3) T = f1(t2, δ, T1) = T2

[0038] Also, the dark noise level Dr in such a state can be defined by the following formula (4). (4) Dr = f2(T2) = Dr2

[0039] And when the elapsed time δ becomes a sufficiently long time L, the sensor temperature T2 changes as shown in the following formula (5). (5) T2 = f1(t2, L, T1)

[0040] Therefore, as shown in FIG. 3, until the elapsed time δ from the point when the image accumulation time is switched to t2 reaches L (the rising period), the dark noise level Dr2 changes according to the elapsed time δ. Specifically, the dark noise level Dr2 increases. The rising period is assumed to be about several tens of minutes. For example, about 10 to 30 minutes is assumed. And after the elapsed time δ reaches L (the stable period), the dark noise level Dr2 becomes a constant value Dr2(L).

[0041] FIG. 4 is a diagram showing an example of the correlation between the inspection signal and the sensor output with the change in the dark noise level in Embodiment 1. In FIG. 4, the vertical axis represents the inspection signal (pixel value), and the horizontal axis represents the output (voltage) of the TDI sensor 105. The amount of light incident on the TDI sensor 105 (an example of a camera) when the inspection light irradiates the inspection sample changes depending on the material of the sample. In the example of FIG. 4, the output (the output after integration of each element arranged in the scan direction) of the photosensor element showing the light-shielded portion (black portion) where the light-shielding film of Sample 1 is formed is smaller than the output of the photosensor element showing the light-shielded portion (black portion) of Sample 2. Also, the output of the photosensor element showing the transmissive portion (white portion) of the glass substrate without the light-shielding film formed on Sample 1 is smaller than the output of the photosensor element showing the transmissive portion (white portion) of Sample 2. Thus, the minimum value and the maximum value of the output of the photosensor element differ depending on the material of the sample. On the other hand, in defect inspection, the measured image of the inspection target is compared with a reference image based on design data or a die image captured at different positions of the sample. Therefore, it is necessary to normalize the output of the photosensor element according to a certain standard. Specifically, as shown in FIG. 4, the output of the photosensor element is corrected (calibrated) by linear interpolation by the sensor circuit 125 so that the inspection signal (pixel value) in the light-shielded portion (black portion) becomes the same value regardless of the material of the sample. Similarly, the output of the photosensor element is corrected (calibrated) by linear interpolation by the sensor circuit 125 so that the inspection signal (pixel value) in the transmissive portion (white portion) becomes the same value regardless of the material of the sample. Therefore, as the output of the TDI sensor 105, the corrected pixel value is output as the inspection signal.

[0042] In the above-described example, the case of transmission inspection in which the TDI sensor 105 receives the transmitted light that has passed through the substrate 101 has been described, but the present invention is not limited to this. A reflection inspection in which the inspection light irradiates the substrate 101 and the TDI sensor 105 receives the reflected light from the substrate 101 may be used. In that case, the light-shielded portion (light-shielding film) of the sample becomes the white portion, and the transmissive portion (glass substrate) becomes the black portion.

[0043] By such calibration, as shown in FIG. 4, for each sample, a correlation relationship such as a linear proportional relationship is obtained between the inspection signal and the output of the photosensor element (after integration). As shown in FIG. 4, the slope and intercept will be different for each sample.

[0044] However, if the dark noise level Dr changes, the output of the photosensor element (after integration) of each pixel at the imaging position will change by the amount of change ΔDr (= Dr2 - Dr1) of the dark noise level Dr. In the example of FIG. 4, it will increase. As a result, the inspection signal also changes. In the example of FIG. 4, it will increase. Therefore, an image with an error in the pixel value used for inspection is generated. There was a problem that the accuracy of the image obtained in this way deteriorated. As a result, pseudo defects occurred.

[0045] Therefore, in Embodiment 1, the amount of change in the inspection signal corresponding to the amount of change ΔDr (= Dr2 - Dr1) of the dark noise level Dr is obtained as the offset amount Δt, and the inspection signal is corrected.

[0046] FIG. 5 is a flowchart showing an example of the main steps of the image acquisition method in Embodiment 1. In FIG. 5, the image acquisition method in Embodiment 1 includes a correlation relationship acquisition step (S102), a Dr1 calculation step (S104), a parameter recording step (S106), a setting step (S108), a scan step (S110), a determination step (S112), a δ measurement step (S114), a T2 calculation step (S116), a Dr2 calculation step (S118), a Δt2 calculation step (S120), a determination step (S122), a Δt2 setting step (S124), a scan step (S126), a recording step (S128), and a recording step (S130), and a series of steps are performed.

[0047] As the correlation relationship acquisition step (S102), first, a pattern on the substrate is imaged by the driven TDI sensor 105 for a sufficiently long time L at the image accumulation time t1, and the correlation relationship between the calibrated inspection signal (pixel value) and the photosensor element output is acquired. When switching the image accumulation time for one substrate 101, calibration is performed using such a substrate to acquire the correlation relationship. When using a plurality of types of substrates, the correlation relationship is acquired for each substrate. Thereby, the correlation relationship in the stable period at the image accumulation time t1 can be acquired.

[0048] In calibration, the sensor circuit 125 adjusts the dynamic range of the TDI sensor 105 with a dynamic lens that sets the maximum gradation to a light amount larger than the incident light amount of the white part and sets the light amount smaller than the incident light amount of the black part to zero. For example, as the resolution of 256 gradations, the gradation level of the white part is adjusted to, for example, 200. When using a plurality of types of substrates, the sensor circuit 125 adjusts so that the gradation value level of the white part and the gradation value level of the black part become the same value for all substrates. Thereby, the correlation relationship data in the case of the dark noise level Dr1 shown in FIG. 4 can be obtained.

[0049] FIG. 6 is a block diagram showing an example of the internal configuration of the offset amount calculation circuit in Embodiment 1. In FIG. 6, in the offset amount calculation circuit 136, there are arranged a storage device 59 such as a magnetic disk, a Dr1 calculation unit 50, a recording processing unit 52, a setting unit 54, a determination unit 56, a δ measurement unit 58, a T2 calculation unit 60, a Dr2 calculation unit 62, a Δt2 calculation unit 64, a determination unit 66, and a Δt setting unit 68. A series of "~ units" such as the Dr1 calculation unit 50, the recording processing unit 52, the setting unit 54, the determination unit 56, the δ measurement unit 58, the T2 calculation unit 60, the Dr2 calculation unit 62, the Δt2 calculation unit 64, the determination unit 66, and the Δt setting unit 68 have a processing circuit. Such a processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device, etc. Also, each unit may use a common processing circuit (the same processing circuit). Or, different processing circuits (separate processing circuits) may be used. Input data or calculated results necessary for the Dr1 calculation unit 50, the recording processing unit 52, the setting unit 54, the determination unit 56, the δ measurement unit 58, the T2 calculation unit 60, the Dr2 calculation unit 62, the Δt2 calculation unit 64, the determination unit 66, and the Δt setting unit 68 are stored in a memory (not shown) or the memory 111 in the offset amount calculation circuit 136 each time.

[0050] The correlation data obtained in the correlation acquisition step (S102) is stored in the storage device 59.

[0051] As the Dr1 calculation step (S104), the Dr1 calculation unit 50 calculates the dark noise level Dr1 when the TDI sensor 105 is driven for a sufficiently long time L at the image accumulation time t1. Specifically, first, the sensor temperature T1 in such a driving state is obtained by Equation (1), and using such a sensor temperature T1, the dark noise level Dr1 may be calculated by Equation (2).

[0052] As the parameter recording step (S106), the recording processing unit 52 records the image accumulation time ta = t11, the sensor temperature Ta = T1, and the offset amount Δta = 0.

[0053] As the setting step (S108), the setting unit 54 sets the driving elapsed time δ = 0.

[0054] As a scanning process (S110), the optical image acquisition mechanism 150 acquires an optical image of the test substrate 101 on which a pattern is formed. For this purpose, first, the optical image acquisition mechanism 150 scans the inspection stripe 20 with laser light (inspection light), and for each inspection stripe 20, the TDI sensor 105 captures an image of the stripe area. Specifically, it operates as follows. The XYθ table 102 is moved to a position where the target inspection stripe 20 can be imaged. In transmission inspection, the pattern formed on the substrate 101 is irradiated with laser light (e.g., DUV light) having a wavelength in the ultraviolet range or below, which serves as inspection light, from an appropriate light source 103 through the transmission illumination optical system 170. In other words, the transmission illumination optical system 170 illuminates the test substrate on which the pattern is formed. The light transmitted through the substrate 101 is imaged as an optical image on the TDI sensor 105 (an example of a sensor) by the imaging optical system 176 through the magnifying optical system 104 and enters. Specifically, the light transmitted through the substrate 101 is imaged as an optical image on the photosensor array 124 by the imaging optical system 176 through the magnifying optical system 104 and enters.

[0055] Alternatively, in reflection inspection, the pattern formed on the substrate 101 is irradiated with laser light (e.g., DUV light) having a wavelength in the ultraviolet range or below, which serves as inspection light, from an appropriate light source 103 by the reflection illumination optical system 171 onto the beam splitter 174. The irradiated laser light is reflected by the beam splitter 174 and irradiated onto the sample 101 by the magnifying optical system 104. In other words, the illumination optical system composed of the reflection illumination optical system 171, the beam splitter 174, and the magnifying optical system 104 illuminates the test substrate 101 on which the pattern is formed. The light reflected from the sample 101 passes through the magnifying optical system 104 and the beam splitter 174 and is imaged as an optical image on the photodiode array 105 (an example of a sensor) by the imaging optical system 176 and enters. Specifically, the light reflected from the substrate 101 is imaged as an optical image on the photosensor array 124 by the imaging optical system 176 through the magnifying optical system 104 and enters.

[0056] The image of the pattern formed on the photosensor array 124 is photoelectrically converted by each photosensor element of the photosensor array 124, and further A / D (analog-digital) converted by the sensor circuit 125. At this time, the integrated output of a plurality of photosensor elements arranged in the scanning direction by the sensor circuit 125 is converted into an inspection signal (pixel value) according to the above-described correlation. Then, the inspection signal (pixel value) is offset by the offset amount Δt set by the offset circuit 126. The offset amount Δt is set to the initial offset amount Δt1 = 0. Therefore, here it is output to the stripe pattern memory 123 without being offset. Then, the pixel value data of the inspection stripe 20 of the measurement target is stored in the stripe pattern memory 123. The measurement data (pixel data) is, for example, 8-bit unsigned data and represents the brightness gradation (light amount) of each pixel.

[0057] As the determination step (S112), the determination unit 56 determines whether the image accumulation time t of the photosensor element is the same as ta in which it is recorded. If the image accumulation time t of the photosensor element is the same as ta in which it is recorded, the process returns to the setting step (S108). Then, each step from the setting step (S108) to the determination step (S112) is repeated until the image accumulation time t of the photosensor element is no longer the same as ta in which it is recorded.

[0058] When not changing the image accumulation time of the photosensor element in acquiring images of each region within the same substrate, the TDI sensor 105 acquires images of each inspection stripe 20 with the same image accumulation time t1 and an offset amount of zero. Then, the process proceeds to the comparison process described later.

[0059] The moving speed of the XYθ table 102 (stage) is changed according to the switching of the image accumulation time of the photosensor element. When increasing the image accumulation time of the photosensor element, the moving speed of the XYθ table 102 is decreased accordingly. When decreasing the image accumulation time of the photosensor element, the moving speed of the XYθ table 102 is increased accordingly.

[0060] Changing (switching) the image accumulation time of the photosensor element may occur when the inspection target (image acquisition target) is changed to another substrate 101. In other words, there may be a case where the image accumulation time of the photosensor element is switched in inspection processing units. Or there may be a case where the image accumulation time of the photosensor element is switched in units of days (hours). Or, within the same substrate, there may be a case where the image accumulation time of the photosensor element is switched between regions, for example, between inspection stripes 20. To increase the image accumulation time of the photosensor element, the moving speed of the XYθ table 102 (stage) has to be decreased accordingly. Therefore, the inspection time per substrate increases. For this reason, by inspecting a part (multiple stripes) of the substrate 101 with high sensitivity (low stage speed: long image accumulation time) and the rest with the normal stage speed (high stage speed; short image accumulation time), an increase in the inspection time can be minimized.

[0061] If the image accumulation time t of the photosensor element is not the same as ta recorded in the offset amount calculation circuit 136, in other words, if the image accumulation time of the photosensor element is switched from t1 to t2, the process proceeds to the δ measurement step (S114).

[0062] As the δ measurement step (S114), the δ measurement unit 58 measures the elapsed time δ from the time when the image accumulation time of the photosensor element is switched from t1 to t2.

[0063] As the T2 calculation step (S116), the T2 calculation unit 60 calculates the sensor temperature T2 corresponding to the elapsed time δ from the time when the image accumulation time of the photosensor element is switched from t1 to t2. The sensor temperature T1 recorded as Ta at the switching time may be used. The sensor temperature T2 can be obtained by Equation (3).

[0064] As the Dr2 calculation step (S118), the Dr2 calculation unit 62 calculates the dark noise level Dr2 in the case of the sensor temperature T2 changed by setting the image accumulation time of the photosensor element to t2. The dark noise level Dr2 can be obtained by Equation (4).

[0065] As the Δt2 calculation step (S120), the Δt2 calculation unit 64 (offset calculation unit) calculates an offset amount that depends on the image accumulation time of the photosensor element. Specifically, the Δt2 calculation unit 64 calculates an offset amount Δt2 according to the change amount of the dark noise level. In the example of FIG. 4, for example, the change amount ΔDr of the dark noise level in sample 1 can be obtained by the difference between the dark noise levels Dr1 and Dr2 in sample 1. Then, using, for example, the slope k of the linear proportionality of the correlation relationship of sample 1, the offset amount Δt2 can be defined by Equation (6). (6) Δt2 = k·ΔDr = k(Dr2 - Dr1)

[0066] As the determination step (S122), the determination unit 66 determines whether the recorded Δta is the same value as Δt2. If the recorded Δta is the same value as Δt2, the process proceeds to the recording step (S130). If the recorded Δta is not the same value as Δt2, the process proceeds to the Δt2 setting step (S124). Since initially Δta = 0 is recorded, in the first determination step (S122), the process will proceed to the Δt2 setting step (S124).

[0067] As the Δt2 setting step (S124), the Δt setting unit 68 sets the offset amount Δt2 calculated by the offset circuit 126.

[0068] As the scanning step (S126), the optical image acquisition mechanism 150 scans the inspection stripe 20 with laser light (inspection light), and for each inspection stripe 20, the TDI sensor 105 captures a stripe region image. Specifically, it operates in the same manner as the scanning step (S110). Therefore, the transmitted light that has passed through the substrate 101 forms an optical image and enters the photosensor array 124 via the magnifying optical system 104 and the imaging optical system 176. Alternatively, the reflected light reflected from the substrate 101 forms an optical image and enters the photosensor array 124 via the magnifying optical system 104 and the imaging optical system 176.

[0069] The TDI sensor 105 images the substrate under inspection by receiving transmitted or reflected light from the substrate under inspection 101 with a plurality of photosensor elements. Then, the TDI sensor 105 corrects the pixel values of the optically imaged data using the offset amount and outputs the corrected optically imaged data. Specifically, it operates as follows. The image of the pattern formed on the photosensor array 124 is photoelectrically converted by each photosensor element of the photosensor array 124 and further A / D (analog-to-digital) converted by the sensor circuit 125. At this time, the integrated output of a plurality of photosensor elements arranged in the scan direction by the sensor circuit 125 is converted into an inspection signal (pixel value) according to the above-described correlation. Then, the inspection signal (pixel value) is offset by the offset amount Δt set by the offset circuit 126. The offset amount Δt is set to Δt2. Therefore, here, correction (offset) is performed by subtracting Δt2 from the inspection signal (pixel value) of each pixel. Then, the inspection signal (pixel value) of each offset pixel is output to the stripe pattern memory 123. Then, data of the pixel values of the inspection stripe 20 of the measurement target is stored in the stripe pattern memory 123. The measurement data (pixel data) is, for example, 8-bit unsigned data and represents the brightness gradation (light amount) of each pixel.

[0070] As described with reference to FIG. 3, the dark noise level changes during the rising period. Therefore, during the rising period in FIG. 3, the offset amount calculation circuit 136 (offset calculation unit) calculates the offset amount according to the imaging timing. Then, the TDI sensor 105 corrects the optically imaged data using the offset amount corresponding to the imaging timing for each imaging timing. Therefore, it operates as follows.

[0071] As the recording step (S128), the recording processing unit 52 records (overwrites) Δta = Δt2. Then, it returns to the δ measurement step (S114). And until the Δta recorded in the determination step (S122) becomes the same value as the latest Δt2 calculated in the Δt2 calculation step (S120), each step from the δ measurement step (S114) to the recording step (S128) is repeated. Each step from the δ measurement step (S114) to the recording step (S128) corresponds to the change in the dark noise level during the rising period in FIG. 3. The offset amount is changed every time a new Δt2 is set in the Δt2 setting step (S124). For example, the offset amount is changed for each inspection stripe 20. Or, for example, the offset amount is changed during the scan within the inspection stripe 20.

[0072] And when the elapsed time δ reaches the time L when it enters the stable period, the recorded Δta becomes the same value as the latest Δt2 calculated in the Δt2 calculation step (S120).

[0073] As described with reference to FIG. 3, the dark noise level becomes constant during the stable period. Therefore, after a predetermined period (δ = L) has elapsed since the time when the image accumulation time of the photosensor element was changed from t1 to t2 by the TDI sensor 105, the optical image data is corrected using a constant offset amount. Specifically, the optical image data is corrected with Δt2 (L) during the stable period in FIG. 3. Therefore, it operates as follows.

[0074] When the Δta recorded in the determination step (S122) becomes the same value as the latest Δt2 calculated in the Δt2 calculation step (S120), proceed to the recording step (S130).

[0075] As the recording step (S130), the recording processing unit 52 records ta = t2 and Ta = T2. Then, it returns to the setting step (S108). And until the image accumulation time t of the photosensor element is not the same as the recorded ta in the determination step (S112), each step from the setting step (S108) to the determination step (S112) is repeated.

[0076] In the scanning process (S110) at ta = t2, the inspection signal (pixel value) is offset by a constant offset amount Δt2(L) during the stable period set by the offset circuit 126.

[0077] Furthermore, when changing the image accumulation time of the photosensor element from t2 to t3, t2, T2, Dr2, and Δt2 in each process after the δ measurement process (S114) are replaced with t3, T3, Dr3, and Δt3.

[0078] By operating as described above, image data of each inspection stripe 20 in which the pixel value is offset by an offset amount corresponding to the change in the dark noise level can be acquired.

[0079] FIG. 7 is a flowchart showing the main process of the inspection method in Embodiment 1. In FIG. 7, the inspection method in Embodiment 1 performs a series of processes including the above-described scanning processes (S110)(S126), a reference image creation process (S204), and a comparison process (S106). The content of the scanning processes (S110)(S126) is as described above.

[0080] As the reference image creation process (S204), the reference image creation circuit 112 creates a reference image as a reference using the graphic pattern data (design data). The creation of the reference image is performed in parallel with the scanning operation of the inspection stripe 20 for each inspection stripe 20. Specifically, it operates as follows. The reference image creation circuit 112 inputs the graphic pattern data (design data) for each frame region 30 of the target inspection stripe 20 and converts each graphic pattern defined in the graphic pattern data into binary or multi-valued image data.

[0081] The graphics defined in the graphic pattern data are, for example, those based on basic graphics such as rectangles and triangles. For example, graphic data that defines the shape, size, position, etc. of each pattern graphic with information such as the coordinates (x, y) at the reference position of the graphic, the side length, and a graphic code that is an identifier for distinguishing graphic types such as rectangles and triangles is stored.

[0082] When the design pattern data that becomes such graphic data is input to the reference image creation circuit 112, it is expanded into data for each graphic, and graphic codes, graphic dimensions, etc. indicating the graphic shape of the graphic data are interpreted. Then, it is expanded into binary or multi-valued design pattern image data as a pattern arranged in a grid with a predetermined quantization dimension as a unit and output. In other words, the design data is read, the occupancy rate occupied by the graphic in the design pattern is calculated for each grid formed by virtually dividing the frame area into grids with a predetermined dimension as a unit, and n-bit occupancy rate data (design image data) is output. For example, it is preferable to set one grid as one pixel. And if it is assumed that the resolution of 1 / 2 8 (=1 / 256) is given to one pixel, a small area of 1 / 256 is allocated for the area of the graphic arranged in the pixel, and the occupancy rate in the pixel is calculated. And it is created as 8-bit occupancy rate data. Such a grid (inspection pixel) may be matched with the pixel of the measurement data.

[0083] Next, the reference image creation circuit 112 performs a filtering process on the design image data of the design pattern, which is the image data of the graphic, using a filter function.

[0084] FIG. 8 is a diagram for explaining the filtering process in the first embodiment. Since the pixel data of the optical image captured from the substrate 101 is in a state where a filter acts due to the resolution characteristics of the optical system used for imaging, in other words, in a continuously changing analog state, for example, as shown in FIG. 8, the image intensity (shading value) is different from the developed image (design image) with digital values. On the other hand, in the graphic pattern data, as described above, since it is defined by a graphic code or the like, in the developed design image, the image intensity (shading value) may become a digital value. Therefore, the reference image creation circuit 112 performs image processing (filtering process) on the developed image to create a reference image closer to the optical image. Thereby, the design image data, which is the image data on the design side with the image intensity (shading value) being a digital value, can be matched with the image generation characteristics of the measurement data (optical image). The created reference image is output to the comparison circuit 108.

[0085] FIG. 9 is a diagram showing an example of the internal configuration of each comparison circuit in Embodiment 1. In FIG. 9, in the comparison circuit 108, storage devices 70, 72, 76 such as magnetic disk devices, a frame image creation unit 74, an alignment unit 78, and a comparison processing unit 79 are arranged. A series of "~ units" such as the frame image creation unit 74, the alignment unit 78, and the comparison processing unit 79 have a processing circuit. Such a processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device, etc. Also, each "~ unit" may use a common processing circuit (the same processing circuit). Alternatively, different processing circuits (separate processing circuits) may be used. Input data or calculated results necessary for the frame image creation unit 74, the alignment unit 78, and the comparison processing unit 79 are stored in a memory (not shown) or the memory 111 in the comparison circuit 108 each time.

[0086] The stripe data (stripe region image) input to the comparison circuit 108 is stored in the storage device 70. The reference image data input to the comparison circuit 108 is stored in the storage device 72.

[0087] In the comparison step (S106), the comparison circuit 108 (an example of a comparison unit) compares the optical image composed of the optical image data output from the TDI sensor 105 with the reference image. Specifically, it operates as follows.

[0088] In the comparison circuit 108, first, the frame image creation unit 74 generates a plurality of frame images 31 obtained by dividing the stripe region image (optical image) with a predetermined width. Specifically, as shown in FIG. 2, the stripe region image is divided into frame images of a plurality of rectangular frame regions 30. For example, it is divided into a size of 512×512 pixels. The data of each frame region 30 is stored in the storage device 76.

[0089] Next, for each frame area 30, the alignment unit 78 reads out the corresponding frame image 31 and the corresponding reference image from the storage devices 72 and 76, and aligns the frame image 31 with the corresponding reference image using a predetermined algorithm. For example, alignment is performed using the least squares method.

[0090] Then, the comparison processing unit 79 (another example of the comparison unit) compares the frame image 31 with the reference image corresponding to the frame image 31. For example, comparison is performed pixel by pixel. Here, the two are compared pixel by pixel according to a predetermined determination condition, and for example, the presence or absence of a defect such as a shape defect is determined. As the determination condition, for example, the two are compared pixel by pixel according to a predetermined algorithm to determine the presence or absence of a defect. For example, the difference value of the pixel values of the two images is calculated for each pixel, and when the difference value is greater than the threshold value Th, it is determined as a defect. Then, the comparison result may be output to, for example, the magnetic disk device 109, the magnetic tape device 115, the flexible disk device (FD) 116, the CRT 117, the pattern monitor 118, or output from the printer 119.

[0091] In the above example, the case of die-database inspection has been described, but die-die inspection may also be used. In such a case, for the frame areas to be subjected to die-die inspection among the plurality of frame areas 30, the comparison circuit 108 uses the frame image (optical image) of die 2 obtained for one of the frame areas as a reference (reference image). First, for each frame area 30 to be subjected to die-die inspection, the alignment unit 78 reads out the corresponding frame image 31 of die 1 and the frame image of die 2 from the storage device 76, and aligns the frame image 31 of die 1 with the frame image of die 2 using a predetermined algorithm. For example, alignment is performed using the least squares method. Then, for each frame area 30 to be subjected to die-die inspection, the comparison processing unit 79 (comparison unit) compares the corresponding frame image 31 of die 1 with the frame image of die 2 pixel by pixel.

[0092] As described above, according to Embodiment 1, by performing an offset corresponding to a change in the dark noise level caused by a change in the image accumulation time of the photosensor element, it is possible to suppress the deterioration of the image accuracy caused by the change in the image accumulation time of the photosensor element.

[0093] Embodiment 2. In Embodiment 1, correction was performed according to a change in the dark noise level caused by a change in the image accumulation time of the photosensor element. However, the method for suppressing image deterioration is not limited to this. In Embodiment 2, the change in the dark noise level itself when switching the image accumulation time of the photosensor element is suppressed. The content other than the points to be particularly described below is the same as that in Embodiment 1.

[0094] FIG. 10 is a configuration diagram showing the configuration of the pattern inspection apparatus in Embodiment 2. In FIG. 10, except for the difference in the configuration of the TDI sensor 105 and the arrangement of the temperature adjustment circuit 134 and the refrigerant supply device 131 instead of the offset amount calculation circuit 136, it is the same as FIG. 1.

[0095] FIG. 11 is a block diagram showing an example of the internal configuration of the temperature adjustment circuit in Embodiment 2. In FIG. 11, in the temperature adjustment circuit 134, a storage device 59 such as a magnetic disk device, a T2 calculation unit 51, a temperature change amount ΔT calculation unit 53, a refrigerant temperature adjustment unit 55, and a refrigerant flow rate adjustment unit 57 are arranged. A series of "~ units" such as the T2 calculation unit 51, the temperature change amount ΔT calculation unit 53, the refrigerant temperature adjustment unit 55, and the refrigerant flow rate adjustment unit 57 have a processing circuit. Such a processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device, etc. Also, each "~ unit" may use a common processing circuit (the same processing circuit). Alternatively, different processing circuits (separate processing circuits) may be used. The input data or calculated results required for the T2 calculation unit 51, the temperature change amount ΔT calculation unit 53, the refrigerant temperature adjustment unit 55, and the refrigerant flow rate adjustment unit 57 are stored in a memory (not shown) or the memory 111 in the temperature adjustment circuit 134 each time.

[0096] The TDI sensor 105 includes a photosensor array 124, a temperature adjustment mechanism 127, and a sensor circuit 125. In Embodiment 2, the temperature adjustment mechanism 127 suppresses changes in the dark noise level by keeping the sensor temperature of the photosensor array 124 constant.

[0097] FIG. 12 is a configuration diagram showing an example of the temperature adjustment mechanism in Embodiment 2. In FIG. 12, the temperature adjustment mechanism 127 has, for example, an air cooling mechanism. Specifically, the temperature adjustment mechanism 127 includes an intake port 40, an exhaust port 41, and a heat sink 42. A plurality of fins 44 are arranged in the heat sink 42. The photosensor array 124 has a light receiving surface that converts an inspection image into an electrical signal. The photosensor array 124 is installed in contact with a heat sink 42 having good heat conductivity. Specifically, the bottom surface of the photosensor array 124 is placed in contact with the heat sink 42. Cooling air, which is an example of a refrigerant, is supplied from a refrigerant supply device 131 to the temperature adjustment mechanism 127. The cooling air, which is an example of the refrigerant inhaled from the intake port 40, passes through the heat sink 42 and is exhausted from the exhaust port 41. The cooling air cools the photosensor array 124 through a plurality of fins 44 in the heat sink 42 and exhausts the warmed refrigerant from the exhaust port 41. By arranging the plurality of fins 44, a large surface area can be brought into contact with the cooling air. By arranging the plurality of fins 44, the heat exchange efficiency can be improved.

[0098] The T2 calculation unit 51 calculates the stable sensor temperature T2 when the elapsed time δ reaches a predetermined period L after the image accumulation time of the photosensor element is switched from t1 to t2. The sensor temperature at the switching time may use T1.

[0099] The temperature change amount calculation unit 53 calculates a temperature change amount ΔT that depends on the image accumulation time of each photosensor element of a plurality of photosensor elements arranged in a two-dimensional manner. Specifically, the temperature change amount calculation unit 53 calculates the temperature change amount ΔT (= T2 - T1) when the elapsed time δ reaches a predetermined period L after the image accumulation time of the photosensor element is switched from t1 to t2.

[0100] The refrigerant temperature adjustment unit 55 adjusts the temperature of the refrigerant using the temperature change amount ΔT. In the example of FIG. 12, it adjusts the temperature of the cooling air. Specifically, it outputs the target value of the refrigerant temperature to the refrigerant supply device 131 so that the temperature change amount ΔT generated over the elapsed time δ within the predetermined period L becomes zero.

[0101] The refrigerant flow rate adjustment unit 57 adjusts the flow rate of the refrigerant using the temperature change amount ΔT. In the example of FIG. 12, it adjusts the flow rate of the cooling air. Specifically, it outputs the set value of the refrigerant flow rate to the refrigerant supply device 131 so that the temperature change amount ΔT generated over the elapsed time δ within the predetermined period L becomes zero. The relationship among the temperature change amount ΔT, the target value of the temperature, and the set value of the flow rate generated over the elapsed time δ within the predetermined period L may be measured in advance by experiments or simulations. Specifically, to obtain the heat quantity Q that causes the temperature change amount ΔT over the predetermined period L, heat is generated at a heat quantity q (= Q / L) per unit time. Therefore, it is only necessary to obtain the relationship between the heat quantity q per unit time and the target value of the temperature and the set value of the flow rate that can perform heat exchange. The relationship among the temperature change amount ΔT, the target value of the temperature, and the set value of the flow rate measured in advance is stored in the storage device 59 in advance as refrigerant correlation data.

[0102] In the air-cooling mechanism, the relationship among the temperature change amount ΔT, the target value of the temperature, and the set value of the flow rate generated over the predetermined period L may be measured in advance by experiments or simulations using the cooling air.

[0103] Within the refrigerant supply device 131, the temperature of the refrigerant is adjusted to approach the target value by the temperature adjustment function. Then, the refrigerant supply device 131 supplies the temperature-adjusted refrigerant to the temperature adjustment mechanism 127 at the set flow rate. The refrigerant exhausted from the temperature adjustment mechanism 127 is recovered by the refrigerant supply device 131 and reused.

[0104] As described above, it is possible to suppress the temperature change of the photosensor array 124 when switching the image accumulation time of the photosensor element. By suppressing the temperature change, it is possible to suppress the change in the dark noise level itself. In other words, it is possible to keep the temperature of the photosensor array constant and prevent the change in the dark noise level.

[0105] Therefore, in the scanning process, the TDI sensor 105 receives transmitted light or reflected light from the substrate 101 to be inspected by a plurality of photosensor elements while controlling the temperature of the plurality of photosensor elements using the temperature change amount ΔT. Thereby, the substrate 101 to be inspected is imaged. Since the change in the dark noise level can be suppressed, it is possible to prevent the occurrence of gradation errors.

[0106] FIG. 13 is a configuration diagram showing another example of the temperature adjustment mechanism in the second embodiment. In FIG. 13, the temperature adjustment mechanism 127 has, for example, a water cooling mechanism. Specifically, the temperature adjustment mechanism 127 has a water inlet 80, a water outlet 81, and a heat sink 82. A cooling pipe 84 is disposed in the heat sink 82. The photosensor array 124 is installed in contact with the heat sink 82 having good heat conductivity. Specifically, the bottom surface of the photosensor array 124 is placed in contact with the heat sink 82. Cooling water, which is an example of a refrigerant, is supplied from the refrigerant supply device 131 to the temperature adjustment mechanism 127. The cooling water, which is an example of the refrigerant sucked from the water inlet 80, passes through the heat sink 82 and is drained from the water outlet 81. The cooling water cools the photosensor array 124 through the cooling pipe 84 in the heat sink 82 and drains the heated refrigerant from the water outlet 81. The cooling pipe 84 is disposed so as to meander in the heat sink 82. By meandering, heat conduction can be achieved with a large surface area. Therefore, the heat exchange efficiency can be improved.

[0107] In the water cooling mechanism, the relationship between the temperature change amount ΔT generated in a predetermined period L, the target value of the temperature, and the set value of the flow rate may be measured in advance by experiment or simulation using cooling water.

[0108] In the refrigerant supply device 131, the temperature of the cooling water is adjusted by the temperature adjustment function so as to approach the target value. Then, the refrigerant supply device 131 supplies the temperature-adjusted cooling water to the temperature adjustment mechanism 127 at a set flow rate. The cooling water drained from the temperature adjustment mechanism 127 is recovered by the refrigerant supply device 131 and reused.

[0109] FIG. 14 is a configuration diagram showing another example of the temperature adjustment mechanism in the second embodiment. In FIG. 14, the temperature adjustment mechanism 127 has, for example, a heat pipe mechanism. Specifically, the temperature adjustment mechanism 127 includes heat sinks 92, one or more heat pipes 94, and a heat sink 96. One end side of such one or more heat pipes 94 is disposed in the heat sink 92. The other end side of such one or more heat pipes 94 is disposed in the heat sink 96. The photosensor array 124 is installed so as to be in contact with the heat sink 92 having good heat conductivity. Specifically, the bottom surface of the photosensor array 124 is disposed in contact with the heat sink 92. The refrigerant supply device 131 supplies the refrigerant to the temperature adjustment mechanism 127. Then, heat exchange occurs between the photosensor array 124 and one end side of the heat pipe 94 in the heat sink 92. Then, the other end side of one or more heat pipes 94 exchanges heat with the refrigerant in the heat sink 96. The refrigerant may be cooling air or cooling water.

[0110] In the heat pipe mechanism, the relationship between the temperature change amount ΔT occurring in a predetermined period L, the target value of the temperature, and the set value of the flow rate may be measured in advance by experiment or simulation using the refrigerant to be used.

[0111] As described above, in the scanning step (S110), the change in the dark noise level can be suppressed, so that the gradation error can be prevented from occurring. The contents of each step after the reference image creation step (S204) are the same as those in the first embodiment.

[0112] As described above, according to the second embodiment, by suppressing the change in the dark noise level caused by the change in the image accumulation time of the photosensor element, it is possible to suppress the deterioration in the accuracy of the image caused by the change in the image accumulation time of the photosensor element.

[0113] Embodiment 3. In the second embodiment, the case where the change in the dark noise level is suppressed by suppressing the temperature change itself has been described. In the third embodiment, a configuration for taking measures when such a suppression error of the temperature change occurs will be described. Hereinafter, the content other than the points to be particularly described is the same as that in the first embodiment and the second embodiment.

[0114] FIG. 15 is a configuration diagram showing the configuration of the pattern inspection apparatus in the third embodiment. In FIG. 15, except for the point where the configuration of the TDI sensor 105 is different and the points where the temperature adjustment circuit 134 and the refrigerant supply device 131 are further arranged, it is the same as FIG. 1. In the third embodiment, an inspection apparatus 100 combining the first embodiment and the second embodiment is provided.

[0115] Similar to the second embodiment, when switching the image accumulation time of the photosensor element, the temperature change of the photosensor array 124 is suppressed by the temperature adjustment mechanism 127. Here, a change in the dark noise level may occur due to an error after temperature control by the temperature adjustment mechanism 127.

[0116] The TDI sensor 105 includes a photosensor array 124, a temperature adjustment mechanism 127, a sensor circuit 125, an offset circuit 126, and a temperature sensor 128. In the third embodiment, similar to the second embodiment, the change in the dark noise level is suppressed by keeping the sensor temperature of the photosensor array 124 constant by the temperature adjustment mechanism 127.

[0117] In the offset amount calculation circuit 136, the Dr2 calculation unit 62 calculates the dark noise level Dr2 in the case of the sensor temperature T2 changed by setting the image accumulation time t2 of the photosensor element. The dark noise level Dr2 can be obtained by Equation (4). Here, the sensor temperature T2 uses the measured value by the temperature sensor 128. Thereby, the dark noise level Dr2 based on the error temperature after the temperature control by the temperature adjustment mechanism 127 can be obtained.

[0118] In the above-described example, the case where the measured value by the temperature sensor 128 is used for the error temperature after the temperature control has been described, but the error temperature may be obtained by calculation. For example, the ambient temperature where the photosensor array 124 is placed, the constant temperature system capacity (heat dissipation amount per unit time), the power consumption of the photosensor array 124 (per accumulation time), the heat transfer rate of the photosensor array 124, the surface area of the photosensor array 124, and the heat capacity of the photosensor array 124 are used. These values can be obtained in advance if the embodiment of the TDI sensor 105 is determined. If the initial state (accumulation time t1, sensor temperature T1) is determined, then after the image accumulation time changes from t1 to t2, t3 ···, the temperature of the photosensor array 124 can be calculated from the above parameters.

[0119] The Δt2 calculation unit 64 (offset calculation unit) calculates an offset amount corresponding to the temperature error when controlling the temperatures of a plurality of photosensor elements. Specifically, the Δt2 calculation unit 64 calculates an offset amount Δt2 corresponding to the change amount of the dark noise level. In the example of FIG. 4, for example, the change amount ΔDr of the dark noise level in sample 1 can be obtained by the difference between the dark noise levels Dr1 and Dr2 in sample 1. Then, using, for example, the slope k of the linear proportionality of the correlation of sample 1, the offset amount Δt2 can be defined by Equation (6). (6) Δt2 = k·ΔDr = k(Dr2 - Dr1)

[0120] The TDI sensor 105 corrects the optical image data using the offset amount for such an error temperature. Specifically, it operates as follows.

[0121] In the scanning steps (S110) and (S126), the TDI sensor 105 images the substrate under inspection by receiving transmitted light or reflected light from the substrate under inspection by a plurality of photosensor elements. Then, the TDI sensor 105 corrects the pixel values of the imaged optical image data using the offset amount and outputs the corrected optical image data. Specifically, it operates as follows. The image of the pattern formed on the photosensor array 124 is photoelectrically converted by each photosensor element of the photosensor array 124 and further A / D converted by the sensor circuit 125. At this time, the integrated output of a plurality of photosensor elements arranged in the scanning direction by the sensor circuit 125 is converted into an inspection signal (pixel value) according to the above-described correlation. Then, the inspection signal (pixel value) is offset by the offset amount Δt set by the offset circuit 126. The offset amount Δt is set to Δt2. Therefore, here, correction (offset) is performed by subtracting Δt2 from the inspection signal (pixel value) of each pixel. Then, the inspection signals (pixel values) of the offset pixels are output to the stripe pattern memory 123. And the data of the pixel values of the inspection stripe 20 of the measurement target is stored in the stripe pattern memory 123. The measurement data (pixel data) is, for example, 8-bit unsigned data and represents the brightness gradation (light amount) of each pixel.

[0122] Since the change in the dark noise level can be suppressed and the change amount of the dark noise level due to the temperature control error can be offset, gradation error can be prevented from occurring. The content of each step after the reference image creation step (S204) is the same as that in the first embodiment.

[0123] As described above, according to the third embodiment, even when a control error occurs due to the temperature adjustment mechanism 127, it can be corrected by offset. Therefore, it is possible to suppress the deterioration of the image accuracy caused by the change in the image accumulation time of the photosensor element.

[0124] In addition, in each of the above-described embodiments, each circuit such as the position circuit 107, the comparison circuit 108, the reference image creation circuit 112, the auto-loader control circuit 113, the table control circuit 114, the temperature adjustment circuit 134, and the offset amount calculation circuit 136 has a processing circuit. Such a processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device, etc. Also, each "~ unit" may use a common processing circuit (the same processing circuit). For example, each process may be executed by the control computer 110. Alternatively, different processing circuits (separate processing circuits) may be used. Also, the input data or the calculated results required for each circuit are stored in a memory (not shown) or the memory 111 within the corresponding circuit each time. A program for executing a computer or a processor, etc. may be stored in the storage device 109.

[0125] As described above, the embodiments have been described with reference to specific examples. However, the present invention is not limited to these specific examples. For example, in the embodiment, when combining the transmission illumination optical system and the reflection illumination optical system, the case of imaging with the common TDI sensor 105 has been described, but it is not limited to this. It is also suitable to separately arrange a TDI sensor for imaging when performing transmission inspection and a TDI sensor for imaging when performing reflection inspection.

[0126] Also, in the above-described example, the case of imaging with a TDI sensor has been shown, but it is not limited to this. Any camera using a photosensor may be used. Since a change in the dark noise level can occur similarly with a change in the image accumulation time for a camera using a photosensor, it can be corrected in the same manner as in each of the above-described embodiments.

[0127] Also, parts that are not directly necessary for the description of the present invention, such as the device configuration and the control method, etc., have been omitted, but the required device configuration and control method can be appropriately selected and used. For example, although the description of the control unit configuration for controlling the inspection device 100 has been omitted, it goes without saying that the required control unit configuration can be appropriately selected and used.

[0128] In addition, all pattern inspection apparatuses and pattern inspection methods that include the elements of the present invention and can be appropriately designed and modified by those skilled in the art are included in the scope of the present invention.

Explanation of Reference Signs

[0129] 20 Inspection stripe 30 Frame area 31 Frame image 50 Dr1 calculation unit 51 T2 calculation unit 52 Recording processing unit 53 Temperature change amount calculation unit 54 Setting unit 55 Refrigerant temperature adjustment unit 56 Judgment unit 57 Refrigerant flow rate adjustment unit 58 δ measurement unit 59 Storage device 60 T2 calculation unit 62 Dr2 calculation unit 64 Δt2 calculation unit 66 Judgment unit 68 Δt setting unit 70,71,72,76 Storage device 74 Frame image generation unit 78 Alignment unit 79 Comparison processing unit 100 Inspection device 101 Substrate 102 XYθ table 103 Light source 104 Magnifying optical system 105 TDI sensor 109 Magnetic disk device 107 Position circuit 108 Comparison circuit 110 Control computer 111 Memory 112 Reference image creation circuit 113 Autoloader control circuit 114 Table control circuit 115 Magnetic tape device 116 FD 117 CRT 118 Pattern Monitor 119 Printer 120 Bus 122 Laser Length Measurement System 123 Stripe Pattern Memory 124 Photo Sensor Array 125 Sensor Circuit 126 Offset Circuit 127 Temperature Adjustment Mechanism 128 Temperature Sensor 130 Auto Loader 131 Refrigerant Supply Device 134 Temperature Adjustment Circuit 136 Offset Amount Calculation Circuit 150 Optical Image Acquisition Mechanism 160 Control System Circuit 170 Transmitted Illumination Optical System 171 Reflected Illumination Optical System 174 Beam Splitter 176 Imaging Optical System

Claims

1. An illumination optical system that illuminates a test substrate on which a pattern is formed, A temperature change amount calculation unit that calculates a temperature change amount depending on the image accumulation time of each of a plurality of photosensor elements arranged two-dimensionally, A time delay integration (TDI) sensor that has the plurality of photosensor elements, and receives transmitted light or reflected light from the test substrate by the plurality of photosensor elements while controlling the temperature of the plurality of photosensor elements using the temperature change amount, thereby imaging the test substrate, A comparison unit that compares the captured optical image and a reference image, A pattern inspection apparatus, characterized by comprising the above.

2. The pattern inspection apparatus according to claim 1, wherein the TDI sensor has an air cooling mechanism, a water cooling mechanism, or a heat pipe mechanism.

3. The apparatus further comprises an offset calculation unit that calculates an offset amount according to a temperature error when controlling the temperature of the plurality of photosensor elements, The pattern inspection apparatus according to claim 1 or 2, wherein the TDI sensor corrects optical image data using the offset amount.

4. A step of illuminating a test substrate on which a pattern is formed, A step of calculating a temperature change amount depending on the image accumulation time of each of a plurality of photosensor elements arranged two-dimensionally, A step of imaging the test substrate by receiving transmitted light or reflected light from the test substrate by the plurality of photosensor elements using a time delay integration (TDI) sensor having the plurality of photosensor elements while controlling the temperature of the plurality of photosensor elements using the temperature change amount, A step of comparing the captured optical image and a reference image, A pattern inspection method, characterized by comprising the above.

Images