JP3300830B2 - Method and apparatus for inspecting defects such as foreign matter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting a defect such as a foreign substance in manufacturing a semiconductor device, a magnetic disk, a thin film transistor and the like, and an apparatus therefor.

[0002]

2. Description of the Related Art In a manufacturing process of a semiconductor or the like, a defect such as a foreign substance present on the surface of a product is inspected. In this inspection, for example, a product is irradiated with light, and scattered light generated from a defect such as a foreign substance present on the surface of the product is detected so that the defect such as a foreign substance can be identified.

By the way, in the inspection of a defect inspection apparatus for foreign matter or the like, a target product has various structures in terms of surface geometric shape, physical properties such as reflectance, film thickness, etc., depending on the kind and process. When illumination for inspection is performed on a surface having such various structures, the generated scattered light differs depending on the sample depending on the structure. Also, the scattered light from a defect such as a foreign substance present on the surface is affected by the surface structure of the sample, and also varies depending on the sample.

Therefore, it is necessary to set the inspection conditions for each type of the sample to be inspected and for each process. Inspection conditions are set by actually performing an inspection using an inspection device using a sample to be inspected, and changing the inspection conditions such as the detection threshold based on the obtained detection results, and detecting the scattered light from the sample surface. The setting is made so that only the scattered light from a defect such as a foreign substance is detected without performing. Trial and error adjustment is required to obtain such detection conditions.

[0005] A standard sample to which standard particles are adhered is used as a sample used for setting the inspection conditions of the inspection device or calibrating the sensitivity of the inspection device. There have been made several inventions relating to this standard sample. . For example, Japanese Unexamined Patent Publication
Japanese Patent No. 214641 proposes a method of using a standard sample to which metal particles having a polydispersed particle size distribution are adhered in order to perform calibration without depending on characteristics of an inspection apparatus. In Japanese Patent Application Laid-Open No. Hei 3-1218044, there is a device in which a surface protector (pellicle) having a transparent film formed on the surface of a sample to which standard particles are adhered is provided so that the influence of the adhered foreign matter is eliminated by the protector. . In Japanese Patent Application Laid-Open No. 5-160244, there is a method in which fine particles adhered on a sample are configured to be covered with a fixed film and can be washed, thereby enabling repeated use for a long time. In Japanese Patent Application Laid-Open No. 5-340884, there is a method in which a thin film having the same optical properties as a CVD film is spin-coated on particles to form a sample in order to prepare a standard sample of foreign substances in the film.

On the other hand, in order to investigate the cause of the occurrence of a defect such as a foreign substance and take a countermeasure, the distribution of the defect such as a foreign substance on a sample, or the size and shape of the defect such as a foreign substance are classified and determined. In the inspection device, the distribution of defects such as foreign substances on the sample can be obtained from the detection results. However, the magnitude of the detection output can be predicted to some extent from the magnitude of the detection output, but the magnitude of the generated scattered light differs according to the surface structure of the sample as described above. It is difficult to quantitatively determine the size of a defect such as a foreign substance.

[0007] For this reason, the fact is that the size of a defect such as a foreign substance is checked by a visual observation system as well as the shape. Regarding the classification and determination of the detection result in the inspection device, as disclosed in Japanese Patent Application Laid-Open No. 62-75336, the position information of the detected foreign matter is stored in order to improve the workability and reliability of the visual confirmation of the detection result, There is disclosed an example of an inspection apparatus that automatically positions a foreign substance in a visual field of a visual observation system from stored position information. Japanese Patent Application Laid-Open No. 4-109108 discloses a method for inspecting a foreign substance on a thin film, from a scattered light output to a thin film based on a particle diameter-scattered light correction curve created for correcting the output of a scattered light detector according to the optical characteristics of the thin film. An example of an inspection device for determining the size of a foreign substance attached to a surface is described.

Further, in order to predict the performance of the inspection apparatus,
In the related art, a simulation of scattered light detection in defect inspection of a foreign substance or the like is performed. For example, as a first conventional technique, Robert G. Knollenbur
g "The importance of media
a reflexive index in eva
luting liquid and surfac
e microcontainment meas
elements "Proceedingsof I
nstate of Environmental
Sciences pp. 501-511, 1986
Uses the Mie theory to simulate the amount of scattered light detected in liquid or from standard particles on a bare silicon substrate.

A second prior art is Gregor.
y L. Wojcik, DavidK. Vaugha
m, Lee K .; Galbraith "Calcul
ation of light scatter fr
om structureson silicons
urfaces "SPIE Vol.774 Las
ers in Microlithography p
p. 21-31, 1987, a simulation of a scattered light distribution from a cylindrical object or a spherical particle on a bare silicon substrate is performed using a finite element time domain difference method (FD-TD method).

In the first prior art, the calculation of the scattered light is performed using Mi.
e theory is used. Wolf about Mie theory
This book is described in a book such as "Principles of optics", pp932-2971, and is an analytical solution of scattered light from spherical particles. Therefore, it is impossible to calculate scattered light generated from a circuit pattern on a wafer other than spherical particles, for example.

In the second prior art, a numerical simulation of electromagnetic wave scattering is performed, but only a simulation of a distribution of scattered light generated from a detection target, and a simulation relating to detection is not included. It was not possible to perform a simulation considering the configuration of

[0012]

In the above prior art, setting of inspection conditions and classification / judgment of inspection results are difficult.
The work requires a skilled person and requires a large amount of work, so that there is a very troublesome problem in inspecting for defects such as foreign matter.

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, an object of the present invention is to make it possible to easily set inspection conditions and to classify and judge inspection defects easily and reliably. Another object of the present invention is to provide an inspection method, and an object of the present invention is to provide a defect inspection apparatus for foreign matter or the like which can accurately execute the above method.

[0014]

To SUMMARY OF THE INVENTION To achieve the above object, the defect inspection method of the foreign matter or the like of the present invention, a method for inspecting a defect of the substrate being patterned is formed on the surface odor
Te, light is irradiated from an oblique direction to the surface of the substrate, and detection are removed by the scattered light filter from the pattern of the scattered light from the substrate by irradiation of the light, the scattered light from the pattern The signal detected by the removal is processed to detect a defect on the substrate, and information on the detected defect is simulated.
Comparison with the defect data obtained by the
Classifies more the detected defect in the size and shape, and outputs the information of the size and shape of the defects the classification. Further, in the method of the present invention, in order to achieve the above object , data of a defect obtained by the simulation is used.
The defect data obtained by simulating the distribution of scattered light from a defect such as a foreign substance is used . In order to achieve the above object, the method of the present invention uses a spatial filter as a filter for removing the scattered light, and switches the spatial filter according to the shape of a pattern formed on the surface of the substrate. In order to achieve the above object,
Inspection for defects such as foreign substances on the
And the results of inspection by the plurality of inspection devices are centrally managed via the network.

In order to achieve the above-mentioned object, the defect inspection apparatus for foreign matter and the like according to the present invention provides a device for inspecting a substrate having a pattern formed on its surface .
An apparatus for inspecting defects of foreign matter such as the surface of the substrate, a light irradiating means for irradiating light from an oblique direction of the scattered light from the substrate that is illuminated by the light irradiation means, the scattered light from the pattern Detection optical system means for removing and detecting by a filter, a defect detection means for processing a signal detected by removing scattered light from the pattern by the detection optical system means, to detect a defect on the substrate, Object shape data and objects
The light intensity from the foreign object is calculated using the property value data and the lighting condition data.
To obtain the power distribution and detection output by simulation.
A simulation unit, data of defect obtained by simulation information of a defect detected by the defect detecting means
By comparing the detected defect with the size and shape
A defect classification means for classifying between, in which an output means for outputting information of the size and shape of the defect classified by the defect classification means <br/> provided. Further , the detection optical system means in the apparatus of the present invention includes a spatial filter for removing scattered light, and a switching unit for switching the spatial filter according to a shape of a pattern formed on the surface of the substrate.

[0016]

[0017]

According to the above-described method of the present invention, the surface of the substrate on which the pattern is formed is irradiated with a light beam from an oblique direction, and the reflected scattered light is filtered by a filter to remove the scattered light from the pattern. The presence or absence of a defect can be detected by distinguishing the light from the scattered light from the defect, and if there is a defect, the size and shape of the defect can be identified. Further, according to the apparatus of the present invention, the above-described method of the present invention can be easily carried out, and its effect can be sufficiently exhibited.

[0018]

[0019]

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. 1 to 6 show one embodiment of a defect inspection apparatus 6 for carrying out the method of the present invention. In FIG.
Reference numeral 21 denotes an inspection stage unit. The inspection stage unit 21 includes a Z stage 210 capable of moving the sample to be inspected on its upper surface by fixing means 218 and moving in the Z direction.
An X stage 211 for moving the test sample 26 in the X direction via Y and a Y stage for moving the test sample in the Y direction
Stage 212, Z stage 210, X stage 21
1, a stage drive system 213 for driving each stage of the Y stage 212 and a control system 214 for detecting a focal position for detecting the Z-direction position of the sample 26 to be inspected. Each stage is controlled so that it can always be focused with necessary accuracy during the inspection of the sample 26 to be inspected. The coordinates X, Y, and Z are directions indicated by arrows in FIG.

Reference numeral 22 denotes an illumination system. Reference numeral 221 denotes a laser light source, and reference numeral 222 denotes a condenser lens, which collects a light beam emitted from the laser light source 221 and illuminates the surface of the sample 26 to be inspected.

Reference numeral 24 denotes a detection optical system which has an objective lens 241 facing the surface of the sample 26 to be inspected, and a viewing zone lens (hereinafter referred to as a field lens) 243 provided near an image forming position of the objective lens 241. I have. The light passes through a spatial filter 244 provided on the Fourier transform surface with respect to the inspection field 215 of the sample 26 to be inspected and the imaging lens 245, and the inspection field 215 on the sample 26 is detected by the detector of the signal processing system 25 described later. 251 is formed.

The field lens 243 includes the objective lens 2
The image of the upper focal position 246 on the
4. Further, the optical path is branched by a half mirror 242 inserted in the middle of the optical path, and an upper focal position 2 on the objective lens 241 by a second field lens 246.
The image of 46 is formed on the Fourier transform plane observation detector 252. The spatial filter 244 uses the difference between the distribution of the scattered light generated from the surface of the sample 26 to be inspected and the scattered light generated from defects such as foreign substances on the Fourier transform surface, and uses the difference between the surface of the sample 26 to be inspected. This is a means for shielding the generated scattered light and distinguishing the scattered light from the scattered light generated from a defect such as a foreign matter by a light-shielding portion and a transmitting portion that transmit the scattered light. A plurality of types of spatial filters 244 are attached to switching means 249 such as a revolver or a slide, which will be described later.
The switching means 249 has a structure that can be switched according to the surface structure of the sample to be inspected. Further, instead of the spatial filter, a polarization filter may be used which distinguishes between the scattered light generated from the surface of the sample 26 to be inspected and the scattered light generated from a defect such as a foreign substance by utilizing the polarization difference.

Reference numeral 25 denotes a signal processing system.
Includes a detector 251, a binarization circuit 253 for performing a binarization process on an output of the detector 251, a Fourier transform plane observation detector 252, a microcomputer 254, and a display unit 255. .

The detector 251 photoelectrically converts incident light, and detects scattered light generated from the surface of the sample 26 while scanning the X stage 210. In this case, when a defect such as a foreign substance present on the sample 26 to be inspected exists in the detection visual field, the intensity of light incident on the detector increases, so that the output from the detector 251 also increases. If a one-dimensional solid-state imaging device is used for the detector 251, there is an advantage that the detection field of view can be widened while maintaining the spatial resolution of detection.

The Fourier transform plane observation detector 252
Is for detecting the light intensity distribution on the Fourier transform plane, and uses a photoelectric conversion element capable of detecting the light intensity distribution. For example, a one-dimensional or two-dimensional solid-state imaging device can be used.

In the binarization circuit 253, a threshold value for binarization is set in advance, and an output value which is equal to or higher than the scattered light intensity output from the detector 251 and corresponds to a foreign substance having a size to be detected is input. In such a case, the logic level "1" is output when the error occurs.

The microcomputer 254 includes two
When the value conversion circuit 253 outputs a logic level “1”, it is determined that “defective” is present, and the position information of the X stage 210 and the Y stage 211 is determined. If the detector 251 is not a single element, the pixel position in the element is determined. Is stored in the memory, and the position information of the defect calculated from the above, the detection output value of the detector 251 and the image data of the Fourier transform image of the defect such as a foreign substance obtained by the Fourier transform surface observation detector 252 are stored in the memory. It is configured to output to the display means 255.

Therefore, the defect inspection apparatus 6 is capable of detecting the scattered light generated from the surface of the sample
A defect such as a foreign substance is detected based on scattered light generated from a defect such as a foreign substance adhering to the sample or a defect portion generated in the sample 26 to be inspected.

In the defect inspection apparatus 6 according to the embodiment, a defect classification / determination mechanism for determining the size of a defect such as a foreign substance based on the inspection result 15 by the apparatus 6 and classifying / determining the shape of the defect is provided. Have. As shown in FIG. 1, the defect classification determination mechanism includes a defect detection simulator 1 such as a foreign substance, a simulation result database 2, a calculation result conversion unit 3, an apparatus parameter 4, and an inspection result comparison unit 5. have.

More specifically, as shown in FIG.
Simulates the light behavior of the surface of the sample 26 to be inspected. That is, first, a standard inspection sample 26 whose surface structure is modeled is prepared, and the object shape data 01 and physical property value data 02 of the prepared inspection sample 26, the wavelength of illumination light, the illumination angle, The scattered light distribution 104 including the scattered light intensity distribution, the phase distribution, and the polarization distribution from the surface of the sample 26 to be inspected is obtained by the electromagnetic wave simulator 111 from the illumination condition data 03 such as polarization. Similarly, the distribution of scattered light generated from a defect such as a foreign substance present on the surface of the sample 26 is calculated. According to such a procedure, the electromagnetic wave simulator 111 obtains the distribution of the scattered light generated from the surface of the reference sample and the scattered light generated from the foreign matter on the sample. This makes it possible to predict to what extent a filter should be used from the intensity and polarization distribution of the scattered light.

Next, the defect detection simulator 1 performs a detection simulation. In the detection simulation, a re-imaging simulator is used based on the scattered light distribution 104 including the intensity distribution, phase distribution, and polarization distribution of the scattered light obtained from the above calculation result, the filtering condition 105, and the detection pixel size condition 106. 112, the light intensity distribution 1
07 and a detection output 308 on the detector 251 are obtained.
Then, the spatial filter 244 or the polarizing filter 24
By simulating the detection using the condition of No. 8 as a parameter, a condition that maximizes the ratio of the detection output of the sample surface to the detection output of the particles is obtained. The detection conditions thus obtained are the optimum detection conditions.

Therefore, the defect detection simulator 1 has:
The scattered light distribution 104 obtained from the surface of the inspection sample 26 based on the object shape data 01, the physical property value data 02, and the illumination condition data 03, and the scattered light distribution generated from a defect such as a foreign substance existing on the inspection sample 26. An arithmetic unit 100 including an electromagnetic wave simulator 111 for obtaining the respective 104 and a re-imaging simulator 112 for obtaining a light intensity distribution 107 and a detection output 108 based on the obtained scattered light distribution 104, the filtering condition 105, and the detection pixel size condition 106; Object shape data 01, physical property value data 02, illumination condition data 03, scattered light distribution 04, filtering condition 05, detection pixel size condition 06, light intensity distribution 07, detection output 08
And an input / output unit 10 for appropriately inputting / outputting to / from the electromagnetic wave simulator 111 and the re-imaging simulator 112.

Note that the electromagnetic wave simulator 111 operates based on the Maxwell equation, and the re-imaging simulator 112 operates based on the Fourier transform method, which will be described later.

Further, the defect detection simulator 1 comprises:
The obtained optimum conditions are output to the display device 121, and the simulation result database 2 is used as a database.
The simulation result database 2 outputs the simulation result 13 to the external device 33 including the defect inspection device 6 via the communication means 123 while integrating and managing the simulation result.

The calculation result conversion means 3 converts the simulation result 13 obtained by the defect detection simulator 1 into device parameters. Therefore, the calculation result conversion means 3 creates a relation between the value obtained by the defect detection simulator 1 and the value of the parameter on the apparatus 6 as an apparatus parameter conversion table, and converts the simulation result 13 into an apparatus parameter based on the table. It is a means for data conversion.

The apparatus parameter setting means 4 sets the inspection conditions 14 of the defect inspection apparatus 6, that is, the inspection conditions 14 such as the type of the spatial filter 244 and the detection threshold, based on the data converted by the calculation result converting means 3. Then, the setting data is instructed to the defect inspection device 6. In this case, the defect inspection device 6 selects a desired type of spatial filter 244, a desired threshold value, and the like according to a command from the device parameter setting means 4.

The inspection result comparing means 5 is a device for inspecting the actual inspection object by the defect inspection device 6 and, when the inspection result 15 is output from the microcomputer 254, the actual inspection result 15 and the defect detection simulator 1 By comparing the magnitude with the detection output 08 of the simulation according to the above, that is, by comparing the distribution of scattered light from a defect such as a foreign substance, the size of the defect is determined, and the shape of the defect is classified and determined. I am trying to do it.

In this case, a method of classifying inspection results by the inspection result comparing means 5 will be described. The size of a defect such as a foreign substance can be determined also from the intensity distribution of scattered light. FIG. 6 shows standard particles (diameter 0.5 μm, 0.8 μm,
(1.0 μm, 2.0 μm).
It is observed that the scattered light intensity distribution varies depending on the diameter of the particles, and the number of stripes increases as the diameter increases. This is Figure 2
252 for the Fourier transform surface observation detector of the inspection apparatus shown in FIG.
Can be observed. If the polarization filter 248 can be installed before the detector 252, the polarization distribution of the scattered light can also be measured. The size of a defect such as a foreign substance can also be determined by comparing the measurement result of the scattered light distribution with the simulation result 13. Further, since the scattered light distribution shown above is related to the shape of the foreign matter, the detection simulator 1
In the above, a simulation in which the shape of a defect such as a foreign substance is changed is performed, the simulation is stored in the simulation result database 2, and the simulation result 13 is compared with the scattered light distribution from the defect such as the foreign substance obtained by the inspection apparatus. Classify the shape according to the size of defects such as
It becomes possible to determine.

Finally, a simulation method will be described. As described above with reference to FIG. 3, the defect detection simulator 1 receives the intensity, phase, and intensity of scattered light from the detection target by the electromagnetic wave simulator 111 from the input of the object shape 01 and the physical property value data 02 and the illumination condition data 03.
A distribution 04 such as polarization is calculated, and then this scattered light distribution 0 is calculated.
4, an image intensity distribution and a detection output 35 of a detection target are calculated by the re-imaging simulator 112 from detection condition data such as a filtering condition 05 including a spatial filter and a polarization filter and a condition 06 of a detection pixel size. The calculation result is displayed on the display device 121 and the external storage unit 1
The information is accumulated in the database 22 and integrated and managed. Further, the simulation result 13 is output from the external storage unit 122 to another device such as an inspection device by the communication unit 123.

Here, the electromagnetic wave simulator 111 and the re-imaging simulator 112 will be described in more detail. The electromagnetic wave simulator 111 performs a numerical calculation of an electromagnetic wave problem. Numerical calculation of electromagnetic wave problems is described in "Basic Analysis Method of Electromagnetic Wave Problems" supervised by Eikichi Yamashita, and was conventionally used for antenna problems or waveguide problems. There are various methods for numerical calculation of the electromagnetic wave problem, but when this is used for the problem of light as in the present invention, the scattered light (scattered electromagnetic field) generated when an object is illuminated with light (electromagnetic wave) is calculated. In doing so, the calculation target area is divided into small parts called elements, a discretized model equivalent to each of them is created, and then the whole is assembled from these. Methods such as the so-called finite element method, boundary element method, and difference method can be applied. FIG. 5 shows an example of a model to be calculated (particles on the substrate, 41: particles, 42: substrate).

Next, the re-imaging simulator 112 will be described. As the re-imaging simulator 112, a simulator (hereinafter, a Fourier simulator) that models an optical system by Fourier transform is used. Hereinafter, modeling of the imaging optical system by the Fourier simulator will be described with reference to FIG. Coherent light 60 of wavelength λ is added to the stripe pattern of pitch d _1.
(Coherent light), the following equation 1
Light is diffracted in the direction of the angle theta _1.

[0043]

(Equation 1)

Here, the lens 63 having the focal length f is arranged so that the position of the stripe pattern surface 61 is the position of the focal point.
1-order diffracted light 62, since it is projected to the distance S ₁ from the center of the screen that is supposed to focus position of the exit side, the number 2 of the relationship between the angle theta ₁ and the distance S _1.

[0045]

(Equation 2)

[0046] Equation 1, from equation (2), when the pitch of the fringe pattern to d ₁ is less than d _2, theta _1> theta _2, and the
S ₁ > S ₂ also on the screen surface 65. As described above, when the pitch of the stripe pattern is reduced from d ₁ to d ₂ , the distance from the center on the screen surface 65 is S ₁
Increases to S ₂ from. Here, the spatial frequencies s ₁ ′, s ₂ ′ of the stripe pattern (the number of light / dark repetitions per unit length)
Is defined as in the following equation (3).

[0047]

(Equation 3)

From the above equations (1), (2) and (3), the following relationship is obtained.

[0049]

(Equation 4)

As described above, the coherent light 60 is projected on the screen according to the spatial frequency of the pattern on the stripe pattern surface. This means that the spatial axis is converted to the frequency axis on the screen surface, and can be mathematically described by a Fourier transform.

Next, when a lens 64 having a focal length f is arranged at a position f from the screen surface 65, a stripe pattern is formed on the image plane. At this time, the relationship between the screen surface 65 and the image surface 66 is opposite to the relationship between the stripe pattern surface 61 and the screen surface 65. That is, an image can be reproduced by the inverse Fourier transform of the Fourier transform plane.

In the present detection simulator, the intensity and phase distribution of the scattered light obtained by the electromagnetic wave simulator correspond to the image on the Fourier transform plane in FIG. Then, the re-imaging simulator is used to reproduce an image to be detected from the intensity of the far scattered field and the phase distribution (image on the Fourier transform plane) obtained by the electromagnetic wave simulator. With the above-described configuration, detection simulation can be performed using the surface morphology of the sample to be inspected or the configuration of the inspection apparatus as parameters.

Incidentally, as a means for executing the Fourier simulator, a fast Fourier transform (FFT) is generally used, but other means may be used. Also for the re-imaging simulator, not only the Fourier simulator but also another frequency analysis method such as a maximum entropy method (MEM) may be used.

Since the defect inspection apparatus 6 of the embodiment has the defect separation determination mechanism as described above, its operation and effect will be described below. At the time of actual inspection, the inspection sample 26 whose surface structure is modeled is used, and based on this, the defect detection simulator 1 for foreign matter or the like of the defect separation and determination mechanism is used to inspect the inspection target shape data 01, physical property value data 02, defect inspection Device 6
Based on the device parameter data 03 of the optical system, a simulation result 13 of a defect such as a foreign substance or a scattered light detection output from an inspection target is output and stored in the simulation result database 2. The stored simulation result 13 is converted into data compatible with the device parameters through the calculation result converting means 3, and based on the converted data, the device parameter setting means 4 uses the defect inspection device 5 for foreign matter or the like.
The inspection condition 14 is set.

As a result, the defect inspection apparatus 5 is selected according to the conditions such as the type of the spatial filter 244 and the threshold value set by the apparatus parameter setting means 4.
An inspection object to be detected is set on the inspection stage unit 21 and the inspection object is inspected.

In this case, the inspection result 15 of the defect inspection device 6
Is sent to the inspection result comparing means 5, where the result is compared with the simulation result 13 to obtain the classification / determination result 16. The classification / judgment result 16 is fed back to the calculation result conversion means 3 and used for the next inspection.

In the method of the present invention, as described above, the distribution of scattered light generated from the surface of the sample to be inspected whose surface structure has been modeled in advance by the defect detection simulator 1 and the distribution of foreign substances and the like existing on the sample to be inspected are determined. Based on the distribution of the scattered light generated from the defect, the light intensity distribution and the detection output of the detector in the apparatus are obtained, and based on the light intensity distribution and the detection output, the detection output of the surface of the sample to be inspected and the detection of the defect Since there is a simulation step in which the condition for maximizing the output is obtained and the inspection condition 14 is standardized, the inspection condition 14 by the defect inspection apparatus can be easily set by the simulation step. Therefore, a trial-and-error approach is eliminated, and work efficiency and stability can be reduced, and the burden on the operator can be reduced.

Further, when the actual inspection sample is inspected and output by the defect inspection apparatus 6 under standardized conditions, the actual inspection result 15 and the detection output of the simulation process are compared with the inspection result comparing means. Compared by 5,
Since the method includes a step of classifying and determining the shape of a defect such as an actual foreign substance, the processing can be speeded up. Therefore, the classification and determination of the inspection result can be performed easily and reliably.

Further, since the inspection conditions 14 are the theoretically optimum inspection conditions obtained from the simulation results 13, the apparatus can be operated in a state where the performance of the apparatus is maximized. Further, the tolerance of the detection parameter can be estimated. For example, the output of a foreign substance on the deposition film changes depending on the thickness of the deposition film, but if the inspection device predicts how much margin it can detect, the stable operation of the inspection device becomes possible. Similarly, the simulator can calculate how the detection result changes when the condition of the device changes, for example, when the position of the spatial filter shifts. This makes it possible to estimate the cause when the device changes over time.

In addition, by comprehending the detection results on a physical model by simulation, it is easy to obtain compatibility of the inspection results between devices having different detection methods. When comparing devices with different detection methods,
Considering the fact that actual detection is performed using a sample and comparison is performed, if the detection method is different, the setting of the detection conditions will also be different, and the mutual detection between the devices will be performed based on the detection results obtained from the inspection conditions determined by trial and error. It is difficult to make comparisons and make the test results compatible. With the apparatus configuration according to the present invention, even when comparing the actual detection result with the sample to be inspected, the apparatus can be compared with the detection result under the optimum condition that the apparatus can take, so that the comparison of the detection result can be performed. It is possible to quantitatively interpret the inspection results of each device.

Further, in the apparatus of the present invention, as described above, the distribution of scattered light generated from the surface of the test sample whose surface structure is modeled and the scattering generated from defects such as foreign substances existing on the test sample. A defect detection simulator 1 for obtaining a light intensity distribution and a detection output of a detector in the apparatus based on the light distribution, and a detection output of the surface of the sample to be inspected based on the light intensity distribution and the detection output by the defect detection simulator 1 Device parameter setting means 4 for standardizing the inspection condition 14 by obtaining a condition for maximizing the ratio of the detection output of the defect to the actual output under the conditions standardized by the device parameter setting means 4. When the inspection object is inspected by the apparatus 6, the actual inspection result 15 is compared with the detection output by the defect detection simulator, and the shape of the defect such as an actual foreign substance is classified and determined. Because test results compared and a defect classification decision mechanism and a means 5 that can be performed accurately above method.

FIGS. 7 to 9 show other embodiments of the defect classification judging mechanism. 7 to 9, the same reference numerals as those in FIG. 1 indicate the same or corresponding components. In the embodiment shown in FIG. 7, the defect detection simulator 1 and the calculation result conversion means 3 are directly connected. As a result, the simulation result database 2 shown in FIG. 1 becomes unnecessary, so that the system configuration can be simplified and the cost can be reduced.

In the embodiment shown in FIG. 8, an object shape data 01, physical property value data 02,
The illumination condition data 03 is input by a measurement device 101 such as a dimension measurement and a film thickness measurement and a measurement data conversion means 102. According to this, since the measured value of the inspection sample 26 can be input to the defect detection simulator 1, it is possible to cope with an error of a process different from the design data and to set a more precise simulation and the inspection condition 14.

The embodiment shown in FIG. 9 uses one defect classification judging mechanism and a plurality of defect inspecting devices 6, and a plurality of defect inspecting devices 6 are connected to the defect classification judging device via a communication line 103 via a network. Has been According to this, it is possible to centrally manage the inspection conditions 14 and the inspection results of the defect inspection device 6.

FIGS. 10 and 11 show filter switching means in the defect inspection apparatus 6, respectively. In the embodiment shown in FIG.
A slider 2491 on which a plurality of types of spatial filters 244 are mounted in a straight line and an actuator 2492 for reciprocatingly driving the slider 2491 are arranged. 244 is positioned and selected.

In the embodiment shown in FIG. 11, the filter switching means 249 includes a revolver 2493 in which a plurality of types of spatial filters 244 are mounted in a line at the same radius, and an actuator 2494 for driving the revolver to rotate. By driving the actuator 2494, the revolver 2493 is rotated around the axis, and the desired spatial filter 24
4 is positioned and selected. FIG.
0 and FIG. 11, these actuators 249
2, 2494 may be either a motor or a cylinder. Also, a polarization filter may be used instead of the spatial filter 244.

[0067]

According to the method of the present invention, whether a defect such as a foreign substance is present on the surface of a substrate having a pattern formed on the surface, and if the defect is present, the size and shape thereof can be quickly determined. And it can be easily detected. Further, according to the apparatus of the present invention, the above-described method of the present invention can be easily carried out, and its effect can be sufficiently exhibited.

[0068]

[0069]

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a defect classification determination mechanism showing one embodiment of a defect inspection apparatus for performing a method of the present invention.

FIG. 2 is a schematic configuration diagram showing a defect inspection device.

FIG. 3 is an explanatory diagram showing a defect detection simulator of the defect classification determination mechanism.

FIG. 4 is an explanatory diagram showing an example in which an inspection target of a light scattering simulator is modeled.

FIG. 5 is an explanatory diagram showing an FFT simulator.

FIG. 6 is an explanatory diagram showing an intensity distribution of scattered light generated from particles.

FIG. 7 is a schematic configuration diagram of a defect classification determination mechanism showing another embodiment of the defect inspection apparatus for performing the method of the present invention.

FIG. 8 is a schematic configuration diagram of a defect classification determination mechanism showing still another embodiment of the defect inspection apparatus.

FIG. 9 is a schematic configuration diagram of a defect classification determining mechanism showing still another embodiment of the defect inspection apparatus.

FIGS. 10A and 10B are a front view and a plan view of a filter, respectively, showing a filter switching unit in the defect inspection apparatus.

FIGS. 11A and 11B are a front view and a plan view of the filter, respectively, showing another switching means of the filter in the defect inspection apparatus.

[Explanation of symbols]

01: Shape data of the inspection object, 02: Physical property value data of the inspection object, 03: Data of device parameters of the optical system, 1, ...
Simulator for detecting defects such as foreign substances, 2 ... Simulation result database, 3 ... Calculation result conversion means, 4 ... Device parameter setting means, 5 ... Detection result comparison means, 6 ... Defect inspection apparatus for foreign substances, 13 ... Simulation results, 14 ...
Inspection conditions, 15: inspection result, 16: classification / judgment result.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-7-72093 (JP, A) JP-A-7-92094 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01N 21/88-21/956 H01L 21/66

Claims (6)

(57) [Claims] 1. A substrate having a pattern formed on its surface .
A method for inspecting defects of foreign matter such as removed, irradiated with light from a direction oblique to the surface of the substrate, by a filter the scattered light from the pattern of the scattered light from the substrate by irradiation of the light The detected signal was processed by removing the scattered light from the pattern to detect the defect on the substrate, and the information on the detected defect was obtained by simulation.
Defects foreign matter such as you characterized by classifying in with the detected defect <br/> size and shape, outputs the information of the size and shape of the defects the classification by comparison with the defect data Inspection methods. 2. A defect obtained by the simulation.
2. The defect inspection method for a foreign matter or the like according to claim 1, wherein the data of the defect is data of a defect obtained by simulating a distribution of scattered light from a defect such as a foreign substance. 3. A filter for removing the scattered light is a spatial filter, the spatial filter, the 請 Motomeko 1 you and switches according to the shape of the pattern formed on the surface of the substrate Defect inspection method for foreign substances etc. described. Wherein no line by the inspection of the defects such as a foreign particle of the substrate a plurality of inspection apparatus, according to claim 1 and 2 the result of the inspection in said plurality of inspection devices, via a network, characterized in that centrally manage Or the defect inspection method for foreign substances and the like described in 3 . 5. A substrate having a pattern formed on its surface.Differentobject
An apparatus for inspecting defects such as:lightHand irradiating light
A step, illuminated by the light irradiating meansSaidScattered light from substrate
Of whichSaidFilter out scattered light from the pattern
Optical system means for detecting the scattered light from the pattern by the detection optical system means
Process the detected signal to detect defects on the substrate
Defect detection means,Using object shape data, physical property value data, and lighting condition data
Light intensity from foreign objects Simulate distribution and detection output
Simulation means required by the Said Defects detected by defect detection meansSimulate information
By comparing with the defect data obtained by the
The detected defectSize and shapeAnd inDefect classifier to classify
And information on the size and shape of the defect classified by the defect classification means.
And an output means for outputting a defect.
Place. Wherein said detection optical system means includes a spatial filter for removing scattered light, that has a switching unit switched according to the spatial filter to the shape of the pattern formed on the surface of the substrate A defect inspection apparatus for foreign matter or the like according to claim 5.