JP6999531B2 - Inspection method and inspection equipment - Google Patents

Description

The present embodiment relates to an inspection method and an inspection device.

A mask inspection device having a pattern to be transferred to a wafer is equipped with a TDI (Time Delay Integration) sensor. The TDI sensor receives the laser beam that illuminates the mask through the illumination optical system, and outputs the gradation value corresponding to the mask.

Specifically, the TDI sensor has a plurality of image pickup elements arranged in two directions orthogonal to each other. The TDI sensor captures a mask in units of lines composed of image pickup elements arranged in one direction, and outputs gradation values for each image pickup element. More specifically, the TDI sensor receives the laser light illuminating the mask in order one line at a time as the mask, that is, the stage moves in the direction orthogonal to the line, and the received laser light is used as a charge. By changing and outputting, the gradation value is output. In the output of the gradation value in each line, the TDI sensor stores the gradation value output in one line as an electric charge in the next line. By accumulating the electric charge, the TDI sensor outputs the gradation value obtained by adding the gradation value of the immediately preceding line in the next line. By repeating such accumulation and addition of gradation values, the TDI sensor can output the integrated high gradation values in the final line.

After outputting the gradation value, the inspection device measures the line width of the pattern, that is, the CD (Critical Dimension) based on the captured image (hereinafter, also referred to as an optical image) of the mask having the output gradation value. Then, the inspection device inspects the defect of the mask by comparing the measured line width with the design line width.

By the way, when the amount of light of the laser beam, that is, the brightness fluctuates with the passage of time, the gradation value of the optical image based on the laser beam illuminating the mask also fluctuates. As the gradation value fluctuates, it becomes difficult to accurately measure the line width, and it becomes difficult to accurately inspect the defect of the mask. Therefore, in order to correct the fluctuation of the gradation value due to the fluctuation of the light amount of the laser light, the light amount of the laser light may be calibrated by using a photodiode.

For example, the light emitted to the TDI sensor is detected by a photodiode. The gradation value output by the TDI sensor can be expressed by a function of the output of the photodiode. Therefore, the amount of laser light is adjusted according to the output of the photodiode so that the gradation value output by the TDI sensor becomes the optimum value. The coefficient of the gradation value output by the TDI sensor with respect to the output of the photodiode is measured in advance. Thereby, the inspection device can adjust the amount of laser light so as to optimize the gradation value output by the TDI sensor by using the output of the photodiode and this coefficient.

Japanese Unexamined Patent Publication No. 2017-072393 Japanese Unexamined Patent Publication No. 2009-300426 Japanese Patent Application Laid-Open No. 2015-0221992

Normally, the coefficient between the output of the photodiode and the gradation value output by the TDI sensor is measured and determined when the TDI sensor is replaced. However, the sensitivity of the TDI sensor requires some time to stabilize. Therefore, even if the coefficient deviates from the optimum value and the amount of laser light is calibrated, the TDI sensor may not output a desired gradation value.

Further, in order to calculate the coefficient, it is necessary to actually capture an optical image with the TDI sensor. Since it takes a long time to capture an optical image with a TDI sensor, it is not realistic to frequently acquire an optical image for calculating a coefficient.

Therefore, an object of the present invention is to provide an inspection method and an inspection apparatus capable of correcting the gradation value output from the TDI sensor with high accuracy and in a short time.

In the inspection method according to the present embodiment, the illumination optical system that illuminates the first inspection target with the light from the light source, the image pickup sensor that receives the light that illuminates the first inspection target, and the amount of light received by the image pickup sensor are used. An inspection device equipped with a light amount sensor for detection, an adjustment unit for adjusting the amount of light from a light source, and a control calculation unit for controlling the adjustment unit is used.
Using the first coefficient indicating the relationship between the gradation value output from the image pickup sensor and the light amount signal output from the light amount sensor, the first range of the light amount signal corresponding to the predetermined range of the gradation value is calculated, and the light amount is calculated. The first calibration is performed on the light amount of the light from the light source so that the light amount signal from the sensor falls within the first range, and after the first calibration, the first inspection target is imaged by the image pickup sensor and the first optical image is acquired. Then, using the gradation value output from the image pickup sensor and the light amount signal output from the light amount sensor at the time of acquiring the first optical image, a second coefficient indicating the relationship between the gradation value and the light amount signal is obtained. The second range of the light amount signal corresponding to the predetermined range of the gradation value is calculated by using the second coefficient, and the light amount signal from the light amount sensor falls into the second range with respect to the light amount of the light from the light source. After the second calibration, the image sensor captures the first inspection target or the second inspection target following the first inspection target to acquire a second optical image, and the second optical image is used. It comprises inspecting the first or second inspection target.

The inspection device according to the present embodiment has an illumination optical system that illuminates the first inspection target with light from a light source, an image pickup sensor that receives the light that illuminates the first inspection target, and the amount of light received by the image pickup sensor. A first coefficient indicating the relationship between the light amount sensor to be detected, the adjusting unit for adjusting the amount of light from the light source, the gradation value output from the image pickup sensor, and the amount of light detected by the light amount sensor is calculated, and the first coefficient is calculated. It is equipped with a control calculation unit that performs the first calibration for the adjustment unit using one coefficient.
The control calculation unit uses the gradation value of the first optical image of the first inspection target captured after the first calibration and the light amount signal output from the light amount sensor to determine the relationship between the gradation value and the light amount signal. The indicated second coefficient is calculated, the adjustment unit is subjected to the second calibration using the second coefficient, and the first inspection target or the second inspection target following the first inspection target imaged after the second calibration is performed. 2 The inspection of the first or second inspection target is performed using the optical image.

The schematic diagram of the inspection apparatus according to 1st Embodiment. The schematic diagram which shows the 1st TDI sensor of the inspection apparatus of 1st Embodiment. The block diagram which shows the structural example of the optical system and the calibration circuit of an inspection apparatus. The graph which shows the relationship between the integrated gradation value output from a TDI sensor and the light amount of the laser beam detected by a photodiode. The graph which shows the relationship between the sensitivity of a TDI sensor and time. The flow chart which shows an example of the inspection method by 1st Embodiment. The flow chart which shows an example of the inspection method by the modification 1 of 1st Embodiment. The flow chart which shows an example of the inspection method by 2nd Embodiment. The flow chart which shows an example of the inspection method by the modification 2 of 2nd Embodiment. The flow chart which shows an example of the inspection method by 3rd Embodiment.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings. The present embodiment is not limited to the present invention. The drawings are schematic or conceptual, and the ratio of each part is not always the same as the actual one. In the specification and the drawings, the same elements as those described above with respect to the existing drawings are designated by the same reference numerals, and detailed description thereof will be omitted as appropriate.

(First Embodiment)
FIG. 1 is a schematic view of the inspection device 1 according to the first embodiment. The inspection device 1 includes a light source 2 and a first TDI sensor 5A and a second TDI sensor 5B, which are examples of image pickup sensors. The light source 2 emits a laser beam toward the mask 3 as the first inspection target.
The first TDI sensor 5A and the second TDI sensor 5B take an image of the mask 3 illuminated by the light of the light source 2 and output the gradation value corresponding to the mask 3.

Further, the inspection device 1 is on the first optical path 8 connecting the light source 2 and the first TDI sensor 5A, in order from the traveling direction of the light, the dimming filter 10, the 1/2 wave plate 12, and the first polarization. It includes a beam splitter 14, a first mirror 16, a first objective lens 17, a stage 18, a second polarization beam splitter 19, a second objective lens 20, and a photodiode 27A. The mask 3 is held on the stage 18. The optical members 10, 12, 14, 16, and 17 on the first optical path 8 between the light source 2 and the mask 3 constitute a transmission illumination optical system that irradiates the mask 3 with light transmitted through the mask 3. The configuration of the transmitted illumination optical system is not limited to the configuration shown in FIG.

Further, the inspection device 1 is on the second optical path 21 connecting the light source 2 and the second TDI sensor 5B, in order from the traveling direction of the light, the dimming filter 10, the 1/2 wave plate 12, and the first polarization. It includes a beam splitter 14, a second mirror 23, a second polarization beam splitter 19, a stage 18, a third objective lens 24, and a photodiode 27B. The dimming filter 10, the 1/2 wave plate 12, the first polarization beam splitter 14, and the second polarization beam splitter 19 are optical members common to the first optical path 8 and the second optical path 21. .. The optical members 10, 12, 14, 23, 19 on the second optical path 21 between the light source 2 and the mask 3 constitute a reflected illumination optical system that irradiates the mask 3 with the light reflected by the mask 3. The configuration of the reflective illumination optical system is not limited to the configuration shown in FIG.

The inspection device 1 includes a first photodiode 27A and a second photodiode 27B as a light amount sensor, a first amplifier 28A and a second amplifier 28B, and a bus 29. The first photodiode 27A is arranged between the second objective lens 20 and the first TDI sensor 5A, and receives a part of the light incident on the first TDI sensor 5A. The first amplifier 28A is arranged between the output end of the first photodiode 27A and the bus 29. The second photodiode 27B is arranged between the third objective lens 24 and the second TDI sensor 5B, and receives a part of the light incident on the second TDI sensor 5B. The second amplifier 28B is arranged between the output end of the second photodiode 27B and the bus 29.

The first photodiode 27A detects the amount of light received by the first TDI sensor 5A as an electric signal, and outputs the detected electric signal (hereinafter, also referred to as a light amount signal or a light amount) to the first amplifier 28A. The first amplifier 28A amplifies the light amount signal from the first photodiode 27A according to the set gain, and outputs the amplified light amount signal to the bus 29. The second photodiode 27B detects the amount of light received by the second TDI sensor 5B as a light amount signal, and outputs the detected light amount signal to the second amplifier 28B. The second amplifier 28B amplifies the light amount signal from the second photodiode 27B according to the set gain, and outputs the amplified light amount signal to the bus 29. As will be described later, the light intensity signal is used for calibration of the gradation value.

Further, the inspection device 1 includes an autoloader 30, an X-direction motor 31A, a Y-direction motor 31B, a θ-direction motor 31C, and a laser length measuring unit 32. The autoloader 30 automatically conveys the mask 3 onto the stage 18. The X-direction motor 31A, the Y-direction motor 31B, and the θ-direction motor 31C drive the stage 18 in the X-direction, the Y-direction, and the θ-direction, respectively, to scan the light of the light source 2 with respect to the mask 3 on the stage 18. Let me. The laser length measuring unit 32 detects the positions of the stage 18 in the X direction and the Y direction.

The inspection device 1 further includes various circuits connected to the bus 29. Specifically, the inspection device 1 includes a sensor circuit 34, an autoloader control circuit 35, a stage control circuit 36, a position circuit 38, an expansion circuit 40, a reference circuit 41, a comparison circuit 42, and a map creation circuit. 43 and.

As a part of the control calculation unit, the calibration circuit 33 includes a dimming filter 10 and a 1/2 wave plate 12 as an adjustment unit so that the gradation value output from the TDI sensors 5A and 5B is within a predetermined range. Control. At this time, the calibration circuit 33 measures the amount of light of the laser light received by the TDI sensors 5A and 5B with the photodiodes 27A and 27B, and the dimming filters 10 and 1 are based on the light amount signals from the photodiodes 27A and 27B. / 2 Controls the wavelength plate 12. The dimming filter 10 and the 1/2 wave plate 12 can adjust, attenuate or increase the amount of laser light from the light source 2. As will be described with reference to FIG. 4, the gradation value output from the TDI sensors 5A and 5B is substantially proportional to the amount of light detected by the photodiodes 27A and 27B, and is a function using a certain proportionality coefficient. It is expressed as. Therefore, the calibration circuit 33 can calibrate the gradation value output from the TDI sensors 5A and 5B by controlling the dimming filter 10 based on the light amount signals from the photodiodes 27A and 27B.

The autoloader control circuit 35 controls the autoloader 30. The stage control circuit 36 drives and controls the motors 31A to 31C. The position circuit 38 detects the position of the stage 18 in cooperation with the laser length measuring unit 32. The expansion circuit 40 and the reference circuit 41 generate a reference image as a reference for comparison with the optical image. The comparison circuit 42 compares the line width of the pattern of the optical image with the line width of the pattern of the reference image to detect the defect of the pattern. The map creation circuit 43 creates a line width error map (CD map) based on the line width obtained by the comparison circuit 42.

The inspection device 1 further includes a control computer 45, a storage device 46, and a display device 47. The control computer 45, the storage device 46, and the display device 47 are all connected to the bus 29. The control computer 45 as a control calculation unit executes various controls related to the defect inspection of the pattern. The storage device 46 stores the defect data detected by the comparison circuit 42. Further, the storage device 46 stores a coefficient indicating the relationship between the gradation value output from the TDI sensors 5A and 5B and the light amount signal output from the photodiodes 27A and 27B. This coefficient is used for calibration by the calibration circuit 33. The display device 47 displays an image of defect data stored in the storage device 46. The calibration circuit 33 and the control computer 45 function as a control calculation unit, and may be configured by, for example, one or a plurality of CPUs.

(TDI sensor 5A, 5B)
The TDI sensors 5A and 5B will be described in more detail. FIG. 2 is a schematic view showing a first TDI sensor 5A of the inspection device of the present embodiment. Although not shown, the second TDI sensor 5B also has the same configuration as the first TDI sensor 5A. As shown in FIG. 2, the first TDI sensor 5A has a plurality of image pickup devices 51, that is, CCDs (Charge Coupled Devices) arranged in the X direction, which is an example of the second direction, and the Y direction, which is an example of the first direction. Have. The image pickup device 51 receives the light illuminating the mask 3 and converts it into electric charges to take an image of the mask 3.

A plurality of image pickup elements 51 arranged in the Y direction form a line L. In the image pickup in the TDI mode in which the output of the gradation value and the accumulation of the output gradation value in the line immediately after each line are repeated, the image sensor 51 belonging to the same line L outputs the gradation value at the same time. .. The total number of lines L may be the same as the total number of arrangements (number of pixels) of the image pickup elements 51 in the X direction, or may be less than the total number of arrangements of the image pickup elements 51 in the X direction. The total number of lines L may be, for example, 1024.

In the imaging in the TDI mode, the first TDI sensor 5A moves relative to the mask 3 on the stage 18 in the −X direction as the stage 18 moves in the + X direction, for example. While moving relative to the mask 3 in the −X direction, the first TDI sensor 5A outputs the gradation value of each image sensor 51 corresponding to the mask 3 as an electric charge in order of one line L at a time. In the output of the gradation value in each line L, the first TDI sensor 5A accumulates or transfers the output charge to the immediately following line L. As a result, in the output of the gradation value in each line L, the first TDI sensor 5A adds the gradation value output in the immediately preceding line L to the gradation value output in the current line L. Output the value. By repeating such accumulation and addition of gradation values, the first TDI sensor 5A integrates the gradation values of each line L for each image sensor 51, that is, integrates the integrated values for each image sensor 51 in the final line. Output. The second TDI sensor 5B also has the same configuration as the first TDI sensor 5A. The sensor circuit 34 may execute a part of the gradation value output processing. Hereinafter, the integrated value of the gradation value output from the TDI sensors 5A and 5B is referred to as an integrated gradation value. The integrated gradation value output by the TDI sensors 5A and 5B is used in the comparison circuit 42 for detecting a pattern defect.

(Photodiodes 27A, 27B)

FIG. 3 is a block diagram showing a configuration example of the optical system of the inspection device 1 and the calibration circuit 33. The photodiodes 27A and 27B will be described in more detail with reference to FIG. As described above, the TDI sensors 5A and 5B output the integrated gradation value based on the light emitted from the light source 2 and illuminating the mask 3. When the amount of light emitted from the light source 2 fluctuates with the passage of time, the integrated gradation value output by the TDI sensors 5A and 5B is affected by the fluctuation in the amount of light. If the integrated gradation value affected by such fluctuations in the amount of light is used for inspecting a pattern defect, the comparison circuit 42 may not be able to accurately inspect the defect. Therefore, in order to accurately inspect the defect, it is desirable to monitor the amount of light and calibrate the integrated gradation value according to the amount of light.

Therefore, in the inspection of the mask 3, the calibration circuit 33 adjusts the amount of laser light emitted to the TDI sensors 5A and 5B before the inspection of the mask 3 so that an optical image having an optimum integrated gradation value can be obtained. And calibrate the integrated gradation value. That is, the calibration circuit 33 adjusts the amount of laser light so that the integrated gradation value output from the TDI sensors 5A and 5B falls within a desired range. At this time, the photodiodes 27A and 27B arranged immediately before the TDI sensors 5A and 5B detect the amount of laser light. Then, the calibration circuit 33 executes calibration of the integrated gradation value (hereinafter, also referred to as calibration) based on the light amount signals from the photodiodes 27A and 27B.

If the amount of laser light is to be adjusted based on the integrated gradation values from the TDI sensors 5A and 5B, it is necessary to acquire an optical image in order to obtain the integrated gradation values. Acquiring an optical image takes a longer time than acquiring an amount of light. Therefore, if an attempt is made to adjust the amount of laser light based on the integrated gradation value, it takes a long time for calibration. On the other hand, it takes a relatively short time to detect the amount of laser light using the photodiodes 27A and 27B. Therefore, in the present embodiment, the integrated gradation value is calibrated based on the amount of laser light detected by the photodiodes 27A and 27B.

Further, the integrated gradation value output from the TDI sensors 5A and 5B is substantially proportional to the amount of light (light amount signal) detected by the photodiodes 27A and 27B. If the proportional coefficient is measured in advance and stored in the storage device 46, the calibration circuit 33 uses the light amount signals from the photodiodes 27A and 27B to output the integrated gradation value output from the TDI sensors 5A and 5B. Can be easily calibrated. Since the calibration of the integrated gradation value using the photodiodes 27A and 27B is completed in a short time, for example, each inspection of the mask 3 is performed before the inspection is executed.

FIG. 4 is a graph showing the relationship between the integrated gradation value output from the TDI sensor 5A and the amount of laser light detected by the photodiode 27A. As shown in this graph, the integrated gradation value output from the TDI sensor 5A is proportional to the amount of laser light detected by the photodiode 27A. The coefficient of this proportional relationship is the ratio (Qpd / Qtdi) of the light amount signals Qpd from the photodiodes 27A and 27B to the integrated gradation value Qtdi from the TDI sensors 5A and 5B. This coefficient may be measured, for example, at the time of replacement of the TDI sensor 5A, and may be stored in the storage device 46 in advance.

The integrated gradation value output from the TDI sensor 5B is also substantially proportional to the amount of laser light detected by the photodiode 27B. However, due to individual differences between the TDI sensor and the photodiode, the proportionality coefficient between the integrated gradation value from the TDI sensor 5A and the light amount signal from the photodiode 27A is the integrated gradation value from the TDI sensor 5A and the photodiode 27A. It may be different from the proportionality coefficient with the light amount signal. Each of these proportional coefficients is stored in the storage device 46. Therefore, the calibration of the integrated gradation value is performed for each of the TDI sensors 5A and 5B.

Prior to the inspection of the mask 3, the calibration circuit 33 uses the light amount signals from the photodiodes 27A and 27B and the corresponding proportional coefficients to keep the integrated gradation values of the TDI sensors 5A and 5B within a predetermined range. In addition, the dimming filter 10 and the 1/2 wavelength plate 12 are adjusted. As a result, the amount of laser light from the light source 2 is adjusted, and the integrated gradation value from the TDI sensors 5A and 5B is set within a predetermined range. After executing the calibration of the integrated gradation value in this way, the optical image of the mask 3 is acquired, and the defect inspection of the mask 3 is executed using the optical image.

By the way, it takes time for the sensitivity of the TDI sensors 5A and 5B to stabilize. For example, FIGS. 5 (A) and 5 (B) are graphs showing the relationship between the sensitivity and time of the TDI sensors 5A and 5B. Time 0 is the time when the TDI sensors 5A and 5B are replaced. With reference to these graphs, it can be seen that the sensitivity of the TDI sensors 5A and 5B increases over time and then stabilizes. The sensitivity of the TDI sensor 5A has increased to time ta and is stable after time ta. The sensitivity of the TDI sensor 5B has increased to time tb and is stable after time tb. The time ta and tb vary depending on individual differences of the TDI sensors 5A and 5B, the frequency of use, and the like, but are several tens to several hundreds.

As described above, it takes time for the sensitivity of the TDI sensors 5A and 5B to stabilize. Therefore, even if the proportional coefficient between the integrated gradation value from the TDI sensors 5A and 5B and the light amount signal from the photodiodes 27A and 27B is calculated when the TDI sensors 5A and 5B are replaced, the proportional coefficient is calculated with the passage of time. May not be appropriate. Even if an attempt is made to calibrate the integrated gradation value using such an inappropriate proportional coefficient, the integrated gradation value of the TDI sensors 5A and 5B may not fall within a predetermined range. That is, even if the integrated gradation value is calibrated, the desired integrated gradation value may not be obtained.

On the other hand, the proportionality coefficient is calculated by measuring the integrated gradation value output from the TDI sensors 5A and 5B and the amount of light detected by the photodiodes 27A and 27B, and using the integrated gradation value and the light amount signal. At this time, it is necessary to acquire the optical images by the TDI sensors 5A and 5B a plurality of times while changing the parameters such as the inspection mode (imaging magnification) and the gain of the TDI sensor. Therefore, it takes a relatively long time (for example, 2 hours or more) to set the proportionality coefficient. Frequent setting of such a long-time proportional coefficient is not preferable because it leads to a long preparation time (maintenance time) before inspection.

That is, in order to perform calibration with an accurate proportional coefficient, it is not preferable to frequently set the proportional coefficient as described above. On the other hand, in the inspection device 1 according to the present embodiment, the proportionality coefficient is obtained as follows, and calibration is performed with the proportionality coefficient.

FIG. 6 is a flow chart showing an example of the inspection method according to the first embodiment. For example, at the beginning of replacement of the TDI sensors 5A and 5B, the control computer 45 calculates the first coefficient from the integrated gradation values from the TDI sensors 5A and 5B and the light amount signals from the photodiodes 27A and 27B (S10). The first coefficient is a proportional coefficient showing the relationship between the integrated gradation value from the TDI sensors 5A and 5B and the light amount signal from the photodiodes 27A and 27B. The first coefficient is measured for each of the TDI sensors 5A and 5B, for example, when the TDI sensors 5A and 5B are replaced. The first coefficient of each of the TDI sensors 5A and 5B is stored in the storage device 46.

Next, a predetermined range of integrated gradation values of the TDI sensors 5A and 5B is set (S20). A predetermined range of integrated gradation values is manually or automatically set so as to obtain an optimum optical image for defect inspection of the mask 3. The predetermined range of the integrated gradation value of the TDI sensors 5A and 5B may be set individually.

Next, the control computer 45 uses the first coefficient corresponding to each of the TDI sensors 5A and 5B to set a predetermined range of the integrated gradation value of the TDI sensors 5A and 5B to the range of the light amount signals of the photodiodes 27A and 27B. Is converted to (S30). As a result, the range of the light intensity signals of the photodiodes 27A and 27B corresponding to the predetermined range of the integrated gradation values of the TDI sensors 5A and 5B (hereinafter, also referred to as the first range) is calculated. The first range is also calculated for each of the photodiodes 27A and 27B.

Next, the amount of laser light received by the TDI sensors 5A and 5B is detected by the photodiodes 27A and 27B (S40).

Next, the control computer 45 determines whether or not the amount of laser light is included in the first range calculated in step S30 (S50). When the amount of laser light is not included in the first range (NO in S50), the calibration circuit 33 controls the dimming filter 10 and the 1/2 wave plate 12 to adjust the amount of laser light (S60). ). For example, when the light intensity signal is below the lower limit of the first range, the calibration circuit 33 controls the dimming filter 10 so as to increase the light intensity of the laser beam. On the contrary, when the light amount signal exceeds the upper limit of the first range, the calibration circuit 33 feedback-controls the dimming filter 10 and the 1/2 wave plate 12 so as to reduce the light amount of the laser light. The calibration circuit 33 adjusts the dimming filter 10 and the 1/2 wave plate 12 so that the light intensity signal falls within the first range by repeating steps S40 to S60. The time for measuring the amount of laser light with the photodiodes 27A and 27B is relatively short. Therefore, the feedback control shown in steps S40 to S60 is completed in a relatively short time.

On the other hand, when the amount of laser light is included in the first range (YES in S50), the control computer 45 determines that the integrated gradation values of the TDI sensors 5A and 5B are within the predetermined range (S70). As described above, the calibration circuit 33 uses the light amount signals from the photodiodes 27A and 27B and the first coefficient to control the light amount of the laser light so that the integrated gradation values of the TDI sensors 5A and 5B are within a predetermined range. Adjust. Hereinafter, the calibration of the integrated gradation value using the first coefficient is referred to as "first calibration".

After the first calibration, the TDI sensors 5A and 5B capture an optical image (first optical image) of the mask 3 (S80). At this time, the illumination optical system irradiates the mask 3 (the first inspection target) with the laser light from the light source 2 through the dimming filter 10 and the 1/2 wave plate 12, and the TDI sensors 5A and 5B are on the mask 3. Image the pattern. As a result, the sensor circuit 34 can acquire an optical image of the pattern of the mask 3.

Here, as described with reference to FIGS. 5A and 5B, it takes time for the sensitivities of the TDI sensors 5A and 5B to stabilize. Therefore, the first calibration is effective immediately after the first coefficient is set (for example, immediately after the replacement of the TDI sensors 5A and 5B), and the integrated gradation value from the TDI sensors 5A and 5B is sufficiently within a predetermined range. It is considered to be included in. Therefore, immediately after the first coefficient is set, a good optical image having a desired integrated gradation value can be obtained. However, if a certain amount of time has passed since the first coefficient was set, the sensitivities of the TDI sensors 5A and 5B gradually change, and even if calibration is performed using the first coefficient, the TDI sensors 5A and 5B are used. The integrated gradation value from may be out of the predetermined range. That is, by changing the sensitivity of the TDI sensors 5A and 5B over time, the first coefficient can accurately represent the relationship between the integrated gradation value from the TDI sensors 5A and 5B and the amount of light from the photodiodes 27A and 27B. It may not be possible.

Therefore, the inspection device 1 according to the present embodiment updates the first coefficient as follows. In step S80, since the TDI sensors 5A and 5B have captured the first optical image, the integrated gradation value is acquired at this time. Therefore, the control computer 45 receives the light amount signal from the photodiodes 27A and 27B after the first calibration (when YES is set in step S50) and the integrated gradation value from the TDI sensors 5A and 5B acquired in step S80. The second coefficient is calculated using and (S90). Like the first coefficient, the second coefficient is a proportional coefficient showing the relationship between the integrated gradation value output from the TDI sensors 5A and 5B and the amount of light detected by the photodiodes 27A and 27B. However, since the second coefficient is calculated using the integrated gradation value of the optical image recently captured in step S80, it is a coefficient in consideration of the change in the sensitivity of the TDI sensors 5A and 5B.

Next, the control computer 45 determines whether or not to update the first coefficient with the second coefficient based on the difference between the first coefficient and the second coefficient (S100). For example, the control computer 45 may compare the difference itself between the first coefficient and the second coefficient with the first threshold value. When the first coefficient is A and the second coefficient is B, the control computer 45 may calculate | BA | / A and compare the calculation result with the first threshold value. Further, the control computer 45 may compare other calculated values based on the difference between the first coefficient and the second coefficient with the first threshold value. The first threshold value may be set in advance according to the comparison target and stored in the storage device 46.

When the calculated value using the difference between the first coefficient and the second coefficient is less than the first threshold value (YES in S100), the difference between the first coefficient and the second coefficient is sufficiently small. This means that the sensitivities of the TDI sensors 5A and 5B do not change as much as to the left. Therefore, the control computer 45 does not update the first coefficient and does not perform the second calibration. In this case, the control computer 45 determines that the defect inspection of the mask 3 can be accurately performed even for the first optical image obtained after the first calibration. Therefore, the calibration of the integrated gradation value is completed, and the inspection device 1 performs a defect inspection of the pattern of the mask 3 using the first optical image (S110). The defect inspection is performed in the position circuit 38, the expansion circuit 40, the reference circuit 41, the comparison circuit 42 and the mapping circuit 43.

On the other hand, when the calculated value using the difference between the first coefficient and the second coefficient is equal to or greater than the first threshold value (NO in S100), the difference between the first coefficient and the second coefficient is relatively large. This means that the sensitivities of the TDI sensors 5A and 5B have changed relatively significantly. Therefore, the control computer 45 updates the first coefficient with the second coefficient (S120). In this case, the control computer 45 determines that a sufficient integrated gradation value cannot be obtained from the first optical image and the defect inspection of the mask 3 cannot be performed accurately. Therefore, the control computer 45 updates the first coefficient with the second coefficient and executes the calibration again.

After updating the first coefficient with the second coefficient, steps S30 to S90 are executed again using the second coefficient (the updated first coefficient). That is, the control computer 45 uses the amount of light from the photodiodes 27A and 27B and the second coefficient (the updated first coefficient) so that the integrated gradation values of the TDI sensors 5A and 5B are within a predetermined range. The amount of laser light from the light source 2 is adjusted, and the integrated gradation value is calibrated (second calibration). At this time, in step S30, the control computer 45 uses the second coefficient to set a predetermined range of the integrated gradation values of the TDI sensors 5A and 5B to the range of the light intensity signals of the photodiodes 27A and 27B (second range). Convert. Next, the calibration circuit 33 adjusts the dimming filter 10 and the 1/2 wave plate 12 so that the amount of laser light falls within the second range. Calibration of the integrated gradation value using the second coefficient is called "second calibration".
After performing the second calibration, in step S80, the TDI sensors 5A and 5B capture an optical image (second optical image) of the mask 3.

In the second calibration, the first coefficient is updated with the second coefficient in consideration of the sensitivity of the TDI sensors 5A and 5B, so that the next coefficient obtained at the time of capturing the second optical image in step S90 is the second coefficient. It is almost equal to (first coefficient after update). Therefore, the control computer 45 should be determined to be YES in step S100. Alternatively, when the first coefficient is updated with the second coefficient, the control computer 45 may omit the execution of the second step S90. Therefore, in step S110, the inspection device 1 inspects the pattern of the mask 3 for defects using the second optical image.

After the imaging of the mask 3 is completed, the autoloader 30 changes the mask 3 on the stage 18 to another mask (YES in S130). The inspection device 1 executes steps S20 to S130 for the other masks. When the imaging and inspection of all the masks in the autoloader 30 are completed (NO in S130), the mask inspection is completed.

As described above, in the inspection device 1 according to the present embodiment, the integrated gradation values of the TDI sensors 5A and 5B and the photodiode 27A obtained when the first optical image is captured after the first calibration with the first coefficient is performed. , The second coefficient is calculated using the light amount signal from 27B. When the first coefficient deviates greatly from the second coefficient, the inspection device 1 performs calibration (second calibration) again with the second coefficient and acquires a second optical image of the mask 3. Then, the inspection device 1 inspects the defect of the mask 3 using the second optical image. As a result, the calibration circuit 33 can update the coefficient in real time even if the sensitivity of the TDI sensors 5A and 5B changes with time, and can adjust the amount of laser light using an appropriate coefficient. As a result, the inspection device 1 can correct the integrated gradation value from the TDI sensors 5A and 5B with high accuracy at the time of imaging of the mask 3, and can accurately inspect the defect of the mask 3.

Further, when the first coefficient is updated with the second coefficient in step S100, the TDI sensors 5A and 5B re-image the mask 3 in order to acquire the second optical image. However, when the first coefficient is not updated, it is not necessary to acquire the second optical image, and the first optical image can be used as it is for the defect inspection of the mask 3. Therefore, although the time extension associated with the update of the first coefficient is unavoidable in some cases, the additional time can be minimized.

(Modification 1)
FIG. 7 is a flow chart showing an example of the inspection method according to the first modification of the first embodiment.
In the first embodiment, in step S120, the control computer 45 updates the first coefficient with the second coefficient. Therefore, in the storage device 46, the first coefficient is rewritten to the second coefficient, and the first coefficient does not remain. On the other hand, in the modification 1, the control computer 45 uses the first coefficient as a reference coefficient and corrects it with a correction coefficient to calculate the second coefficient. In this case, the storage device 46 stores the first coefficient and the correction coefficient as the reference coefficient, and the correction coefficient is rewritten. The configuration of the inspection device 1 according to the present modification 1 may be the same as the configuration of the first embodiment.

For example, after executing steps S10 to S90, the control computer 45 calculates the ratio of the second coefficient to the first coefficient as the correction coefficient R. When the first coefficient is (Qpd / Qtdi) ₁ and the second coefficient is (Qpd / Qtdi) ₂ , the control computer 45 calculates the correction coefficient R = (Qpd / Qtdi) ₂ / (Qpd / Qtdi) ₁ . (S91). The storage device 46 may store the first coefficient (Qpd / Qtdi) ₁ as the reference coefficient and the correction coefficient R. Qpd is a certain width of the light amount signal from the photodiodes 27A and 27B, and Qtdi is the width of the integrated gradation value from the TDI sensors 5A and 5B corresponding to the Qpd.

Next, the control computer 45 determines whether or not to correct the first coefficient with the correction coefficient based on the correction coefficient (S101). For example, when the correction coefficient R is within the range of 1 ± α (YES in S101), the difference between the first coefficient and the second coefficient is sufficiently small, so that the control computer 45 does not correct the first coefficient. No second calibration is performed. α is an arbitrary numerical value set in advance. In this case, the calibration is completed, and the inspection device 1 performs a defect inspection of the pattern of the mask 3 using the first optical image (S110).

On the other hand, when the correction coefficient R is outside the range of 1 ± α (NO in S101), the difference between the first coefficient and the second coefficient is relatively large, so that the control computer 45 corrects the first coefficient with the correction coefficient. Then, the second coefficient is calculated (S121). More specifically, the control computer 45 calculates the second coefficient by multiplying the first coefficient by the correction coefficient. The control computer 45 re-executes the calibration using the second coefficient. That is, steps S30 to S90 are executed again using the second coefficient to perform the second calibration. After performing the second calibration, in step S80, the TDI sensors 5A and 5B capture an optical image (second optical image) of the mask 3. In step S110, the inspection device 1 inspects the pattern of the mask 3 for defects using the second optical image.

Other operations of the present modification 1 may be the same as the operations of the first embodiment. Therefore, the present modification 1 can obtain the same effect as that of the first embodiment.

Further, in the present modification 1, the first coefficient is corrected by the correction coefficient to calculate the second coefficient. The first coefficient stored in the storage device 46 is not changed, but the correction coefficient is changed every time an optical image is taken by the TDI sensors 5A and 5B. Therefore, the control computer 45 can calculate the second coefficient by multiplying the first coefficient by the correction coefficient. In this case, the first coefficient is stored in the storage device 46 as it is as a reference coefficient, and the correction coefficient is updated. Thereby, for example, the first coefficient at the beginning of replacement of the TDI sensors 5A and 5B can be easily known.

(Second Embodiment)
FIG. 8 is a flow chart showing an example of the inspection method according to the second embodiment.
In the first embodiment, when the first coefficient is updated by the second coefficient in step S100, the inspection device 1 updates the first coefficient to the second coefficient and then re-impresses the mask 3. In this case, the TDI sensors 5A and 5B image the same mask 3 twice. Therefore, it is inevitable that the time will be prolonged due to the update of the first coefficient.

On the other hand, in the second embodiment, the first optical image is used for calculating the second coefficient and also for the defect inspection of the mask 3 as the first inspection target. The second coefficient calculated at this time is used for calibration when the next mask (second inspection target) of the mask 3 is imaged. As a result, the first optical image is used as it is for the defect inspection of the mask 3, and the second optical image is used for the defect inspection of the mask next to the mask 3. That is, in the second embodiment, the inspection device 1 does not re-image the same mask 3. The second coefficient is reflected when the next mask is imaged. This eliminates the waste of maintenance time before inspection.

For example, after the execution of steps S10 to S90, the inspection device 1 executes the defect inspection of the mask 3 as it is using the first optical image without performing the determination in step S100.

Next, when the autoloader 30 has another mask to be inspected (YES in S130), the first coefficient is updated with the second coefficient (S140). At the same time, the autoloader 30 changes the mask 3 on the stage 18 to the next mask (hereinafter referred to as the second mask). Next, the inspection device 1 executes steps S20 to S120 for this second mask. At this time, the calibration circuit 33 performs calibration (second calibration) using the second coefficient. After the second calibration, the TDI sensors 5A and 5B acquire an optical image (second optical image) of the second mask. The second optical image is used for defect inspection of the second mask. On the other hand, when the imaging of all the masks is completed (NO in S130), it is not necessary to update the coefficients, so that the control computer 45 ends without updating the coefficients.

At the time of acquiring the second optical image, the control computer 45 may further calculate the third coefficient. The third coefficient shows the relationship between the integrated gradation value from the TDI sensors 5A and 5B obtained at the time of acquiring the second optical image and the amount of light from the photodiodes 27A and 27B obtained at the time of acquiring the second optical image. It is a coefficient (proportional coefficient). The third coefficient may be used for calibration when acquiring an optical image of the mask next to the second mask.

As described above, the optical image of a certain mask is used for defect inspection of the mask and for calculating the calibration coefficient at the time of imaging of the next mask. As a result, the coefficient is updated for each mask. Therefore, even if the sensitivities of the TDI sensors 5A and 5B gradually change, the coefficients can be updated for each mask inspection so as to follow the gradual changes. Further, the sensitivities of the TDI sensors 5A and 5B do not change abruptly but change slowly. Therefore, it is considered that the coefficient calculated at the time of imaging of a certain mask can be effectively calibrated even if it is used at the time of imaging of the next mask.

(Modification 2)
FIG. 9 is a flow chart showing an example of the inspection method according to the second embodiment. According to the second embodiment, the inspection device 1 updates the coefficient at the time of imaging of the next mask without performing the determination in step S100.

On the other hand, in the modification 2, the control computer 45 determines whether or not to update the first coefficient to the second coefficient. For example, after executing steps S10 to S130 of the second embodiment, the control computer 45 determines whether or not the calculated value using the difference between the first coefficient and the second coefficient is less than the first threshold value (S135). ). When the calculated value is less than the first threshold value (YES in S135), the difference between the first coefficient and the second coefficient is sufficiently small, so that the control computer 45 does not update the first coefficient and images the next mask. Even at times, the amount of laser light is calibrated using the same first coefficient.

On the other hand, when the calculated value using the difference between the first coefficient and the second coefficient is equal to or greater than the first threshold value (NO in S135), the difference between the first coefficient and the second coefficient is relatively large, so that the control computer 45 Updates the first coefficient with the second coefficient (S140). In this case, the control computer 45 updates the first coefficient with the second coefficient, and calibrates the amount of laser light using the updated first coefficient (second coefficient) at the time of imaging of the next mask.

In this way, even if it is determined whether or not to update the first coefficient, the effect of the second embodiment is not lost. In step S140, the first coefficient may be corrected by the correction coefficient to calculate the second coefficient as in step S121 of the modification example 1.

(Third Embodiment)
FIG. 10 is a flow chart showing an example of the inspection method according to the third embodiment. The third embodiment is an embodiment in which the first embodiment and the second embodiment are combined.

For example, in step S100 of FIG. 6, when the calculated value using the difference between the first coefficient and the second coefficient is less than the first threshold value (YES in S100), the control computer 45 uses the first optical image. Along with inspecting the mask 3 (S110), the second embodiment is executed. That is, the first coefficient is updated with the second coefficient (S140), and the updated first coefficient (second coefficient) is used for calibration at the time of imaging of the mask (second inspection target) next to the mask 3. Be done. Further, by executing steps S20 to S110, an optical image (second optical image) of the next mask can be obtained. The inspection device 1 uses the second optical image to inspect the next mask for defects.

On the other hand, when the calculated value using the difference between the first coefficient and the second coefficient is equal to or greater than the first threshold value (NO in S100), the control computer 45 executes the first embodiment. That is, the inspection device 1 updates the first coefficient to the second coefficient (S120), re-images the mask 3, and obtains an optical image of the mask 3 (S20 to S90). After that, the inspection device 1 performs a defect inspection of the next mask using the optical image (second optical image) of the mask 3 (S100, S110). In this case, in step S120, the first coefficient has already been updated with the second coefficient. Therefore, in step S130 of FIG. 10, if there is another mask (YES in S130), the mask is changed as shown by the broken line, and the process returns to step S20.

In this way, when the first coefficient does not deviate so much from the second coefficient, the inspection device 1 inspects the defect of the mask 3 as it is without re-imaging the optical image of the mask 3. Used for. Moreover, the second coefficient is used in the calibration at the time of imaging of the next mask, and the second optical image is captured for defect inspection of the mask next to the mask 3. This shortens the maintenance time before inspection.

On the other hand, when the first coefficient greatly deviates from the second coefficient, the inspection device 1 updates the first coefficient to the second coefficient and then re-impresses the optical image of the mask 3. That is, the second coefficient is used in the second calibration at the time of imaging of the mask 3, and the second optical image is an optical image re-imaged for defect inspection of the mask 3. As a result, an optical image having an integrated gradation value suitable for defect inspection of the mask 3 can be obtained.

As described above, the first and second embodiments may be combined. Thereby, the third embodiment can obtain the effects of the first and second embodiments. Further, the third embodiment may be combined with the modified example 1 or the modified example 2.

At least a part of the inspection method according to the present embodiment may be configured by hardware or software. When configured by software, a program that realizes at least a part of the functions of the inspection method may be stored in a recording medium such as a flexible disk or a CD-ROM, read by a computer, and executed. The recording medium is not limited to a removable one such as a magnetic disk or an optical disk, and may be a fixed recording medium such as a hard disk device or a memory. Further, a program that realizes at least a part of the functions of the inspection method may be distributed via a communication line (including wireless communication) such as the Internet. Further, the program may be encrypted, modulated, compressed, and distributed via a wired line or a wireless line such as the Internet, or stored in a recording medium.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and variations thereof are included in the scope of the invention described in the claims and the equivalent scope thereof, as are included in the scope and gist of the invention.

1 Inspection device, 2 light sources, 3 masks, 5A, 5B TDI sensors, 10 dimming filters, 12 1/2 wave plates, 14,19 polarization beam splitters, 16, 23 mirrors, 17, 24 objective lenses, 18 stages, 19 Second polarization beam splitter, 20 second objective lens, 27A, 27B photodiode, 28A, 28B amplifier, 33 calibration circuit, 45 control computer, 46 storage device, 47 display device.

Claims (5)

An illumination optical system that illuminates the first inspection target with light from a light source, an image pickup sensor that receives the light that illuminates the first inspection target, and a light amount sensor that detects the amount of light received by the image pickup sensor. It is an inspection method using an inspection device including an adjustment unit for adjusting the amount of light from the light source and a control calculation unit for controlling the adjustment unit.
Using the first coefficient indicating the relationship between the gradation value output from the image pickup sensor and the light amount signal output from the light amount sensor, the first range of the light amount signal corresponding to the predetermined range of the gradation value is set. Calculate and
The first calibration is performed on the light amount of the light from the light source so that the light amount signal from the light amount sensor falls within the first range.
After the first calibration, the image sensor captures the first inspection target to acquire a first optical image.
A second coefficient showing the relationship between the gradation value and the light amount signal by using the gradation value output from the image pickup sensor and the light amount signal output from the light amount sensor at the time of acquiring the first optical image. Is calculated,
Using the second coefficient, the second range of the light intensity signal corresponding to the predetermined range of the gradation value is calculated.
A second calibration is performed on the amount of light from the light source so that the light amount signal from the light amount sensor falls within the second range.
After the second calibration, the image sensor captures the first inspection target or the second inspection target following the first inspection target to acquire a second optical image.
An inspection method comprising inspecting the first or second inspection target using the second optical image. After the calculation of the second coefficient, it is further provided to determine whether or not to update the first coefficient to the second coefficient based on the difference between the first coefficient and the second coefficient.
When updating the first coefficient with the second coefficient, after performing the second calibration, the first or second inspection target is inspected using the second optical image.
The inspection method according to claim 1, wherein when the first coefficient is not updated, the inspection of the first or second inspection target is performed using the first optical image without performing the second calibration. The inspection method according to claim 1 or 2, wherein the first and second coefficients are the ratio of the light amount signal from the light amount sensor to the gradation value from the image pickup sensor. When the calculated value using the difference between the first coefficient and the second coefficient is equal to or greater than the first threshold value, the first coefficient is updated with the second coefficient.
The inspection method according to claim 2 or 3, wherein the calculated value using the difference between the first coefficient and the second coefficient is less than the first threshold value, the first coefficient is not updated. The inspection device further includes a storage unit for storing a correction coefficient calculated by using the first coefficient and the second coefficient and the first coefficient.
The inspection method according to claim 1, wherein the control calculation unit corrects the first coefficient by using the first coefficient and the correction coefficient to calculate the second coefficient.

Images