JPH10325713A - Method and instrument for inspecting surface defect - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable inhibiting of variations of a data of the results of inspection while achieving a higher size sorting accuracy. SOLUTION: In this apparatus, (n) rows of false defects each comprising three false detects fg or more are arranged at a specified interval of angle in the circumferential direction of a disc 1a for calibration and the individual false defects of each row of false defects are formed in an equal size while adjacent false defects fg are arranged at a specified interval larger than the width of a laser spot Sp and discs for sensitivity calibration are so arranged to gradually increase or decrease along the circumferential direction. Now, an inspection defects of the discs for sensitivity calibration is performed to adjust the detection sensitivity according to the results of inspection. A data for relating the level of a detection signal to the size of the false defect is generated as reference data for judging the size from the results of the inspection under an adjusted condition. Thus, the size of the defect is obtained from the level of the detection signal when the disc to be inspected undergoes an inspection of defects based on the criterion reference data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect inspection method and an inspection apparatus, and more particularly, to a magnetic disk defect inspection apparatus capable of improving the size classification accuracy and suppressing the variation in inspection result data. Surface inspection method and inspection apparatus.

[0002]

2. Description of the Related Art A hard magnetic disk used as a recording medium of a computer system has a material substrate which is finished by mirror polishing, the polished disk (substrate or polished disk), and plating on the disk. Then, at each stage of a magnetic disk or a disk such as a semiconductor wafer coated with a magnetic thin film (these are referred to as disks), defects existing on the surface and their sizes are individually inspected.

FIG. 8 shows the configuration of a main part of a magnetic disk defect inspection apparatus for inspecting a disk for defects. The defect inspection device 10 shown in FIG. 8A includes a rotation mechanism 2, a detection optical system 3, and a defect detection device 4. The disk 1 to be inspected is mounted on the spindle 21 of the rotating mechanism 2 and is rotated by driving the motor (M) 22. On the other hand, the detection optical system 3 focuses the laser beam LT generated from the laser light source 311 of the light projecting system 31 by the focusing lens 312, and forms a spot Sp on the surface of the disk 1,
The surface of the disk 1 is irradiated with a spot Sp. The spot Sp moves in the direction of the radius R of the disc 1 by relatively moving the spot Sp in the X-axis direction of the disc 1, and the spot Sp moves the surface of the disc 1 by this movement and the rotation of the disc 1. Scan in a spiral. In this case, in order to make the total scanning time of the disk 1 as short as possible, the spot Sp has an elliptical shape having a short diameter φ1 and a long diameter φ2 as shown in FIG. Then, the major diameter φ2 of the spot Sp is associated with the radial direction, and the scanning width in the radial direction is set wide.

[0004] The defect F existing on the surface scatters the light of the spot Sp. The scattered light SR is condensed by the condensing lens 321 of the light receiving system 32, and a photoelectric conversion element, for example,
The light is received by a light receiver 322 including an avalanche photodiode (APD) or a photomultiplier tube (PMT). The output signal of the light receiver 322 is input to the signal processing circuit 41 of the defect detection device 4 and is used as a defect detection signal, where the defect F is detected. Further, depending on the circuit, the size of the defect is detected or classified according to the amplitude of the output signal (detection signal) from the light receiver 322. The signal processing circuit 41 for detecting a defect is a circuit for detecting a defect by so-called sampling, and includes an amplifier for amplifying an output signal from the light receiver 322 and an output signal for a defect exceeding noise among the amplified output signals. A sampling circuit that receives a pulse from the rotary encoder 23, samples the detected value, and stores the peak value of the output signal as a level value of a detection signal for a defect (hereinafter referred to as a detected defect value); An A / D converter for converting the value into a value, a position data generating circuit for generating position data on the disk in response to a pulse from the rotary encoder 23, and the like.

The signal processing circuit 41 which detects the size of a defect or classifies and detects a defect is used to simultaneously set a plurality of thresholds having different levels with respect to an output signal from the photodetector 322 to determine the magnitude of the defect. Depending on whether the output signal exceeds the threshold, the size of the defect or, for example, large,
Detects sizes specially classified as small and medium.
In this case, the size of the defect corresponding to the detected defect value can be obtained if the threshold value is set finely and stepwise, and the size-classified detected value can be obtained if the threshold value is set relatively coarsely. The detection value at this time is output in the form of a digital value. Also in this case, the position data for the defect is generated and output together with the data indicating the defect size (or the classified size) by the position data generation circuit.

The data of the size of each defect (data of the detected defect value or size data) and the data of the position on the disk are input to the data processing device 44 from the signal processing circuit 41 as digital values. The data processing device 44 is an MPU
42 and a memory 43 and the like. Here, the number is counted for each size, and the size and its count value are output to the printer (PR) 45 together with the position of the defect on the disk 1. At this time, the defect may be printed out as a map on the disk together with the shape of the disk. Separately, a display (CRT)
Similarly, on the screen 46 and the like, the position of the defect is displayed as a map together with the shape of the disk with an image corresponding to the size of the defect. The count value is also separately displayed on the screen according to the classified size. When the data processing device 44 receives a detected defect value corresponding to the level of the defect detection signal from the signal processing circuit 41, a determination process for internally classifying the detected defect value into a defect size is performed. Is detected.

[0007] The shape of the defect F on the disk 1 varies. One example is shown in FIG. In FIG. 9, the defect Fh is a dish-shaped defect (shallow pit).
it) or a saucer pit)
It has a relatively large diameter Dh, and the depth dh is smaller than this diameter Dh. The defect Fp has a well shape and a relatively small diameter Dp but a large depth dp, and is usually simply called a pit. Note that both defects Fh, F
p often exists in isolation. On the other hand, the defect Fs is a linear scratch defect. The width ws and the depth ds of the cross section vary. It should be noted that the surface of the disk 1 also has a defect having a shape other than these. Apart from such concave defects, there are also various types of heights and sizes of defects called convex foreign substances generated by the attachment of fine particles.

Therefore, in order to detect defects of various shapes and sizes in a satisfactory manner, the above-described defect inspection apparatus 10 uses the projection angle θT of the laser beam LT of the light projecting system 31 and the light receiving system. 32 light reception angle θR, light receiver 322
The voltage V applied to (APD) or the signal processing circuit 41
The items related to the level of the detection signal, such as the gain of the amplifier built in, the threshold voltage E for noise removal, and the laser output of the laser light source 311 are optimally set via the control panel 47 (including the control circuit). Then, the detection sensitivity for the defect is adjusted. Here, the size of the defect is not only the size (area) when the defect is viewed in plan view, but also the size (area) when the defect is viewed in plan view but the depth or height is different (volume). Or, those having different volumes) are also treated as having different sizes.

[0009]

The defect detection sensitivity of the defect inspection apparatus 10 is optimally set for various types of defects F, but the adjustment requires skill. In particular, adjust the detection sensitivity of the device so that it can detect from shallow defects to deep defects, or conversely, adjust the sensitivity so that it can detect the range of foreign particles with a small particle size to those with a large particle size. It is difficult to adjust. Further, the adjustment of the defect detection sensitivity does not provide data that properly classifies the size unless it is calibrated with a certain standard. Moreover,
Inspection result data on a large number of detected defects or sizes classified into large and small sizes vary depending on the current setting state of the inspection apparatus. Variations in inspection result data are likely to occur between different inspection devices.
Therefore, it becomes difficult to share inspection result data.

[0010] Calibration of the detection sensitivity of the conventional apparatus is based on the fact that an actual disk having a dish-like defect, a pit defect, a scratch defect, or the like as a sample defect each having a known size in accordance with the defect to be detected and its size. Alternatively, an actual disk having a protrusion of a specific height is used as a sample disk, and a defect is detected for each sample disk. However, since the defect on the sample disk is of a specific shape of a specific size, even if the detection sensitivity is calibrated by this, it is not appropriate for the size of the defect to be inspected in terms of the size of the defect. In many cases, it is not the case. For this reason, the detected size and its size are biased. In practice, it is also unclear to what extent the size classification test can be performed accurately.

In recent years, the range of defect sizes to be detected has been reduced with the increase in recording density of disks. Or it tends to move to tighter sizes. However, it is also difficult to obtain a sample disk for detection sensitivity calibration suitable for the size classification at that time. An object of the present invention is to provide a surface defect inspection method capable of improving the size classification accuracy and suppressing the variation in inspection result data. Another object of the present invention is to provide a surface defect inspection apparatus that can improve the size classification accuracy and suppress the variation in inspection result data.

[0012]

A feature of the defect inspection method of the present invention for achieving the above object is that three or more pseudo-inspections formed along any one of a radial direction and a circumferential direction are provided. N (where n is an integer of 2 or more) pseudo-defect rows of defects are provided at predetermined angular intervals in the circumferential direction of the calibration disk, and the pseudo-defects in each of the pseudo-defect rows are substantially Is formed as one of a convex portion and a concave portion having a size equal to, and the adjacent pseudo defects are arranged at predetermined intervals larger than the width of the laser spot, and are stepwise along the circumferential direction. A sensitivity calibration disk having an increased or decreased sensitivity is provided, the sensitivity calibration disk is inspected for defects, the detection sensitivity is adjusted according to the inspection result, and the calibration data in a state where the detection sensitivity is adjusted. Data relating the level of the detection signal and the size of the pseudo defect is generated as reference data for size determination based on the disk inspection result, and the detection is performed when the disk to be inspected is inspected for defects based on the determination reference data. The size of the defect is obtained from the signal level.

Further, the features of the defect inspection apparatus of the present invention are as follows.
The sensitivity calibration disk, sensitivity adjustment means for adjusting the detection sensitivity, and the level of the detection signal from the inspection result of the defect inspection of the sensitivity calibration disk in a state where the sensitivity adjustment means has adjusted the detection sensitivity to an appropriate level. And reference data generating means for generating data relating the size of the pseudo defect as reference data for size determination, and a level of the detection signal obtained when a disk to be inspected is inspected for defects based on the reference data. Size detecting means for obtaining the size of the defect.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An operator performs a defect inspection in a defect inspection apparatus using the above-described sensitivity calibration disk as a defect inspection target, and detects a detected value or a detected size of a defect according to a level of a detection signal. When the defect size is displayed as an inspection result on a display or the like in the classification value of the size classified according to the signal level, the defect is displayed as a map along with the shape of the disc with an image corresponding to the size. At this time, at least the image corresponding to the pseudo defect of a different size can be obtained, so that the sensitivity can be adjusted so as to obtain an image with an appropriate detection sensitivity while watching the obtained image. As a result, it is possible to adjust the detection sensitivity to suit the size classification of the defect at that time without any skill while watching the image on the calibration disk. Moreover, as long as the same calibration disk is used, the setting state of the defect detection sensitivity of the inspection apparatus can be reproduced, so that the inspection result data is hardly affected by the current setting state, and the variation is reduced. Of course, the inspection result data hardly varies even between different inspection apparatuses, and the inspection result data can be shared.

Therefore, according to the method of classifying the size of the defect to be detected, it is possible to adjust the detection sensitivity to an appropriate value while viewing the image of the pseudo defect displayed on the screen. In particular, the detection sensitivity of the device can be adjusted according to the size at the substantially central position of the size classification. In addition, at this time, it is possible to grasp to what size the size can be classified according to the sensitivity adjustment based on the display state of the pseudo defect on the screen. Detected defect value (detected signal level) corresponding to the size of the pseudo defect in the state where the sensitivity is properly adjusted.
Is obtained from the inspection result, and data that associates the obtained detected defect value with the size of the pseudo defect is generated as reference data for size determination. By determining the defect size based on the reference data, the data indicating the size of the defect detected by the disk defect inspection apparatus or the size data of the large or small classification is substantially correctly calibrated.

By calibrating the detection sensitivity using the calibration disk every time the inspection apparatus starts the inspection, or by applying the calibration to a number of other defect inspection apparatuses in the same manner, the detection sensitivity and the inspection result data generated each time the inspection is performed. Variation,
Alternatively, variations in detection sensitivity and inspection result data between inspection apparatuses can be suppressed. By the way, as the stepwise selection range of the size of the convex portion or concave portion of each pseudo defect in the pseudo defect row, when the convex portion or concave portion is square or rectangular, the length of one side thereof is in the range of 0.5 μm to 20 μm. And the depth or height is 0.01 μm to 0.75
A range of μm is appropriate. Further, when the pseudo-defect is circular, its diameter is in the range of 0.5 μm to 20 μm and its depth or height is 0.01 μm to 0.75 μm.
A range of μm is appropriate.

[0017]

FIG. 1 is a block diagram showing an embodiment of a magnetic disk defect inspection apparatus to which a surface defect inspection method according to the present invention is applied. Magnetic disk defect inspection apparatus 11 of FIG.
Has a data processing device 50 in place of the data processing device 44 in FIG. The data processing device 50 further includes an input device 48 in addition to the MPU 42 and the memory 43 of the data processing device 44 shown in FIG. The memory 43 of the data processing device 50 includes a defect data collection program 43a, a defect map display program 43b, a criterion data generation program 43c, a defect size detection program 43d, and a detected defect value / defect size table 43e. Have been.

When calibrating the detection sensitivity, the data processor 5
0 indicates a defect inspection for the sensitivity calibration disk 1a (1b) having pseudo defect rows of various sizes shown in FIGS. 4 and 6, and a detection defect value / defect size table 43e is obtained from the inspection result. create. When performing a defect inspection of the disk 1 to be inspected, the level of a detection signal obtained by the defect inspection is converted into data indicating a defect size or a classification size with reference to the detected defect value / defect size table 43e. To size the defects.
The signal processing circuit 41 is a circuit that detects a defect described in the prior art and generates a detected defect value. This includes an amplifier 41a that amplifies an output signal from the light receiver 322,
A sampling circuit 41b for sampling a peak value of an output signal of a defect exceeding noise among output signals amplified by the amplifier 41a in accordance with a pulse from the rotary encoder 23 to obtain a detected defect value; A /
It comprises a D converter (A / D) 41c, and a position data generating circuit 41d that receives pulses from the rotary encoder 23 and generates position data on the disk.

The MPU 42 of the data processing device 50 executes the defect data collection program 43a to spirally scan the disk 1 with the laser spot Sp, and according to the scanning, A / D of the detected defect value from the signal processing circuit 41. The converted values and the position data are sequentially received and stored in the working area of the memory 43 or the buffer memory unit. Disc 1
Is completed, the MPU4 is used during detection sensitivity calibration.
2 calls and executes the defect map display program 43b. Further, at the time of normal disk defect inspection, the defect size detection program 43d is called and executed.

When this program is executed by the MPU 42, the defect size detection program 43d sequentially scans each track of the disk 1 to be inspected for a defect inspection.
Upon receiving the detected defect value obtained in accordance with the defect inspection, the detected defect value is classified by converting it into data of the size of the defect with reference to the detected defect value / defect size table 43e, and the result is used as defect size data. The data is sequentially stored in the memory 43 together with the defect position data. When the entire surface of the disk 1 has been scanned, the MPU 42 further counts the number of defects for each defect size, calculates the number of defects, and
3 is stored. Next, the defect map display program 43b
Is executed, and the classified inspection results are displayed on a map and output to the printer 45 or the like. In addition, the defect size detection program 43d generates data of a defect size corresponding to a defect detection value (level of a detection signal) or data of a classification size value obtained by further subdividing the defect size at the time of detection sensitivity calibration. The size conversion data is read into the detected defect value / defect size table 43e, and is rewritten to perform data setting. The rewritten conversion data is obtained by inputting data that has been experimentally and empirically obtained in advance. Therefore, in the defect inspection at the time of the detection sensitivity calibration, the detection result is displayed on the display 46 in a form close to the actual defect state by the detected defect value / defect size table 43e having such conversion data.

When this is executed by the MPU 42, the defect map display program 43b reads out the position data of the collected defect and the size data of the defect. The size data of the defect is converted into pattern data having a size corresponding to the size indicated by the defect data to generate display data, and the pattern data is displayed at predetermined coordinates on the screen indicated by the position data. The data is developed into image data and displayed on a map on the display 46. The criterion data generation program 43c is a program for rewriting the contents of the detected defect value / defect size table 43e. When the program is executed by the MPU 42, the input of the defect size is performed from the input device 48 after waiting for the input of the defect size. Then, the display coordinate position is detected from the position on the screen specified by the mouse or the like on the screen of the display on which the map is displayed. Then, from the inspection result, the size data of the defect displayed at this display coordinate position or the defect displayed at the position closest to this position is extracted. Further, based on the extracted size data, the detected defect value /
The corresponding size data is searched in the defect size table 43e, and the searched size data is updated to the input size data. Further, a flag indicating the update is written corresponding to the updated size data. In this way, the size data of the detected defect value / defect size table 43e is updated to the input size data. Then, by deleting the past data having no flag when the update is completed, the new detected defect value /
A defect size table 43e is generated. Thereby, the detected defect value / defect size table 43 having the new judgment reference data in which the relation between the level and the size of the detection signal is calibrated.
e is generated.

In this case, a detected defect value corresponding to the size data is extracted from the detected defect value / defect size table 43e, and converted data in which the detected defect value and the inputted defect size are paired is generated. , A pair of tables can be created separately. The conversion data is updated by generating a number of pairs of detected defect values and defect sizes, and writing the contents of the generated new table into the detected defect value / defect size table 43e. Such rewriting of the size data in the detected defect value / defect size table 43e is performed at the time of detection sensitivity adjustment, that is, when a defect inspection is performed on the sensitivity calibration disk 1a (1b).

As shown in FIG. 3, the structure of the detected defect value / defect size table 43e is a comparison table in which a signal level is associated with a defect size, and a flag column indicating update data is provided. This is the first table in this table
Standard data of signal level and defect size is stored and rewritable. Therefore, for example, EEPRO
The conversion data is updated by reading data from an external storage device such as a hard disk or a flexible disk, which is formed of a rewritable memory such as M or a RAM.

A sensitivity calibration disk (hereinafter referred to as a reference disk) 1a shown in FIG. 4 (a) is an appropriately selected mirror-polished aluminum disk, which is formed by using an ion beam sputtering apparatus. Is placed at a target position of the apparatus, and an ion beam is implanted and sputtered in accordance with the position and the shape of the pseudo defect to be formed. When the disk is sputtered, a depression is formed at the sputter position where the beam is injected. By forming a large number of these depressions in a specific arrangement, pseudo concave defects are formed. Further, the pseudo defect in the concave portion can also be formed by melting the surface with a laser beam. In this case, a circular hole can be formed. In addition, the pseudo defect in the concave portion can be formed by edging using a resist as a mask.

On the other hand, pseudo-defects of the projections are formed by sputtering particles sputtered from a target by an ion beam sputtering apparatus onto a disk 1a through a perforated mask. At this time, protrusions of convex portions are formed on the disk 1a corresponding to the holes formed in the mask. Note that tungsten is preferably used as the target metal. Further, the convex portion as a pseudo defect can also be formed by vapor phase growth of aluminum so as to be selectively arranged along the radial direction or circumferential direction of the disk 1a via a photoresist by a VCD method. .

For example, as shown in FIG.
Pseudo-defect fg of concave part along 12 rays at 0 ° intervals
Can be formed by an ion beam sputtering apparatus. Each of the formed pseudo defect strings is indicated as # 1, # 2,... # 12. Each pseudo defect row #
The pseudo defects fg 1 to # 12 are respectively formed as square recesses of the same size. Pseudo-defect f
The size of g is controlled by the amount of the ion beam, the weight of the particles used, the time, the sputter position, and the like. The diameter of the ion beam used at this time is 1/4 to 1/10 with respect to one side of a square of the size of the pseudo defect as viewed in plan.
It is squeezed to the extent. Many ions are implanted in a row along one side in the horizontal direction of the concave hole forming the ion beam, and this is repeated while shifting in the vertical direction. For example, in the case of forming a square recess having a side length wg of the pseudo defect of wg = 1 μm, an ion beam having a diameter of 0.2 μm is laterally applied at least five times along one side in the horizontal direction of the square. Press,
Next, the 0.2 μm ion beam is shifted in the vertical direction by 0.2 μm, and is again implanted five times along one side in the horizontal direction.
This is shifted 4 times in the vertical direction, and 5 times in the vertical direction.
Times, a total of 25 shots. As a result, the pseudo-defect fg of the square concave portion of wg = 1 μm shown in FIG.
Is formed. This is performed by the number of pseudo defects formed at intervals larger than the radial width (long diameter) φ2 of the laser spot Sp along the radius of the disk 1a. Of course, the position of the next pseudo defect to be formed in the circumferential direction which is the scanning direction of the laser spot Sp is the laser spot Sp.
Is far away from the circumferential width (shorter diameter) φ1 of.

In the figure, Mk provided on the inner periphery of # 1 is a mark indicating that this pseudo defect row is # 1, and indicates a reference pseudo defect row. This mark is formed by a large pseudo defect having a shape as a mark. In this example, each pseudo defect row is formed so that the numerical value of the defect size increases clockwise. Thus, a group FGR of pseudo defects fg having different sizes can be identified. Note that increasing the defect size in the clockwise direction means that the defect size decreases in the counterclockwise direction.
The increase and decrease are relative. In the figure, only the pseudo defect rows # 1, # 2, and # 12 are shown in the form of a square, but the width Wg and the depth dg of one side of these squares (see FIG. # 1 to # 12 differing in the defective column
The size is gradually increasing. But,
Since the change in size is very small and the relationship between the depth and the height is involved, the relationship is not sufficiently represented in the drawing.

Therefore, the pseudo defect f in FIG.
An example of the range between the length wg of one side of g and the depth dg is as follows:
It looks like this: The size wg of the pseudo defect fg: 0.5 μm, 1 μm, 3 μm, 5 μm, 10 μm, 20 μm dg: 0.025 μm, 0.05 μm, 0.2 μm, 0.75 μm The length wg and the depth dg of these sides are determined by the actual defect F. It is in the range to cover the size enough. So, for example,
When the size of the object to be inspected is to be classified around 1 μm on a side, wg = 1 μm and wg = 0.5
.mu.m and wg = 3 .mu.m are adopted so that they have the following relationship of the pseudo defect row. First, the pseudo-defect row of # 1 is defined as wg = 0.5 μm and depth dg = 0.025 μm.
m, and the # 2 pseudo-defect row is defined as wg = 0.5 μm and depth dg.
₌ 0.05 μm, and the # 3 pseudo-defect row is represented by wg = 0.5 μm
m, the depth dg = 0.2 μm, and the # 4 pseudo defect row is wg
= 0.5 .mu.m and a depth dg = 0.75 .mu.m. Next, the pseudo-defect row of # 5 is set to wg = 1 μm, the depth dg is set to 0.025 μm, and the pseudo-defect row of # 6 is set to wg = 1 μm.
1 μm, depth dg ₌ 0.05 μm, # 7 pseudo defect row is wg = 1 μm, depth dg = 0.2 μm, # 8 pseudo defect row is wg = 1 μm, depth dg = 0.75 μm. To form Then, the pseudo-defect row of # 9 is calculated as wg = 3μ.
m, the depth dg = 0.025 μm, the # 10 pseudo-defect row is wg = 3 μm, the depth dg ₌ 0.05 μm, and # 1
The pseudo-defect row # 1 is wg = 3 μm and the depth dg = 0.2 μm, and the pseudo-defect row # 12 is wg = 3 μm and the depth dg = 0.
One having a thickness of 75 μm is formed. Here, the increase in the size of the pseudo defect in each pseudo defect row in the depth direction is 0.0%.
0.025 × 2, 0.025 ×, based on 25 μm
4, 0.025 × 20 and 0.025 × 30.

FIG. 4A shows 12 pseudo defect rows having pseudo defects fg whose size sequentially increases for each pseudo defect row formed as described above. Of course, if the center of the classification target at the time of inspection is 5 μm, the length of one side for each depth is 12 μm, which is 3 μm, 5 μm, and 10 μm. Will be created. Assuming that the interval between the pseudo defect rows is 15 °, 2
Four pseudo-defect rows can be formed on one reference disk. Thus, one reference disk can include the above two examples. That is, the above wg =
0.5 μm, wg = 1 μm, wg = 3 μm, wg = 5 μ
For all combinations of m, wg = 10 .mu.m and wg = 20 .mu.m, a pseudo defect row in which the size of the pseudo defect fg sequentially increases is provided for each pseudo defect row.

As shown in FIG. 5A, two adjacent pseudo defects fg in each pseudo defect group FGR form a gap Ga slightly larger than the major diameter φ2 (width in the disk radial direction) of the laser spot Sp. Are arranged at equal intervals. Such a reference disk 1a is mounted on the magnetic disk defect inspection device 11 of FIG. 1 or is mounted in place of the disk 1 of the magnetic disk defect inspection device 10 of FIG. When the defect size is displayed as an inspection result on the display 46 in the detected value of the defect size detected in accordance with the detection value or the classification value in which the size is subdivided in accordance with the level of the detection signal, the pseudo defect of the reference disk 1a is displayed on the display 46. The columns are displayed as defect maps. The state depends on the detection sensitivity adjustment.

For example, since the size of the convex portion or the concave portion of each pseudo defect row changes stepwise, when the detection sensitivity to the defect is high, the pseudo defect having a small size is displayed in a state corresponding to the small size. As for the pseudo defect of the size, the defect of more than a certain degree is displayed as having the same size when the level of the detection signal reaches saturation. Conversely, when the detection sensitivity is low, small-sized pseudo defects are not detected, and large-sized pseudo defects are displayed according to their sizes. Thus, by repeating the defect inspection of the reference disk 1a a plurality of times while adjusting the detection sensitivity via the control panel 47, the sensitivity can be adjusted so as to obtain an optimal display.

At this time, the detection sensitivity of the device on the control panel 47 is adjusted by adjusting the light receiver 322 (A
DP), the applied voltage V, the gain of the amplifier of the signal processing circuit 41, the threshold voltage E, the laser output of the laser light source 311 and the like. Further, the projection angle θT of the laser beam LT of the light projecting system 31 and the light receiving angle θR of the light receiving system 32 are adjusted as necessary. For example, in the first example, when the reference disk 1a having twelve pseudo-defect rows for performing size classification centering on 1 μm on a side is mounted on the defect inspection apparatus 11 and defect inspection is performed, # 1 to # 12 Control panel 4 so that the detection sensitivity is such that the pseudo-defect row is clearly displayed.
In step 7, the value of each part can be set and adjusted.
When the reference disk 1a having the 24 pseudo defect rows in the following example is mounted, the apparatus is designed so that the 12 pseudo defect rows selected according to the measured size to be classified are clearly displayed. May be adjusted. further,
At the start of each inspection, 12 defect columns of a predetermined size are displayed, and the detection sensitivity is adjusted so that they are clearly displayed, so that the variation in inspection result data for each inspection is reduced. Thus, it is possible to suppress the variation of the inspection result data between the apparatuses.

As described above, if the detection sensitivity is set to be limited to a specific size range, the defect detection range corresponding to the size required in the inspection result image of the pseudo defect row of the calibration disk is clarified. In addition, the detection sensitivity can be adjusted by the control panel 47 so that the display can be distinguished from the display of other pseudo defect columns. As a result, it is possible to preferentially detect a specific size or size range. In addition, the adjustment is easy because the image on the screen can be seen, and no skill is required. Calibration of the defect detection sensitivity on the basis of the inspection result on such a display screen can be relatively performed by anyone. In addition,
A mark which can be displayed on a display or the like at the position of the pseudo-defect row used as a reference for the calibration disc in order to display the defect size of any one of the pseudo-defect rows formed on the calibration disc as a reference. Can be provided.

Here, in each pseudo defect row,
Three or more projections or depressions of substantially the same size are formed in each row, and these projections or depressions are arranged in the disk radial direction at predetermined intervals larger than the radial width of the laser spot. It is a requirement. This is to make it possible to recognize an array of pseudo defects even when a defect exists in the calibration disk itself. That is, the arrangement of three or more defects at regular intervals rarely occurs in a normal defect. By arranging the laser spot Sp at a predetermined interval larger than the radial width of the laser spot Sp, three defects can be detected independently. Furthermore, even if another defect enters between the convex or concave portion of the pseudo defect and is arranged, two or more defects arranged in the radial direction can be detected as a pseudo defect array, and the pseudo defect array can be detected. I know that

FIG. 5B shows a state in which the disk is not rotated.
This is an example of a case where the laser spot Sp is simply scanned in the radial direction to detect a defect detection signal of the reference disk 1a in the signal processing circuit 41 in the radial direction. In the defect detection signal, three waveforms corresponding to the three pseudo defects fg appear close to each other. Even if a real defect F exists in the vicinity and appears in the detection signal, the difference between the amplitudes of the two and the difference between before and after the defect are detected. This can be ignored from the relationship. FIG. 4 (c)
It is explanatory drawing about the shape at the time of providing the pseudo defect row | line | column # 1-# 12 which makes the circular protrusion fg 'a pseudo defect instead of the pseudo defect fg of (b). The size of the pseudo defect fg 'is selected in the following range, where the diameter is wg and the height is dg'. Size wg 'of pseudo defect fg': 0.5 μm, 1 μm, 3 μm, 5 μm, 10 μm, 20 μ
mdg ': 0.025 µm, 0.05 µm, 0.2 µm, 0.75 µm, 0.0
1 μm

FIG. 6 shows a disk 1b in which square concave pseudo defects fg of the same size are arranged in an arc shape every 60 degrees in the circumferential direction. In the disk 1b, each pseudo defect row is formed concentrically, and the pseudo defects arranged in each pseudo defect row of the same concentric circle have the same size when viewed from a plane. Then, the depth of each pseudo defect row increases in the clockwise direction, and the size of the pseudo defect differs for each pseudo defect row. The interval between the pseudo defects fg is
As shown in FIG. 7A, the laser spots Sp are arranged at equal intervals Gb larger than the circumferential width φ1 of the laser spots Sp. The depth of each pseudo defect is selected from the above dg or dg '.

The size of the pseudo-defect fg in each pseudo-defect row as viewed from a plane from the outer peripheral direction toward the inner peripheral direction becomes smaller. This size is also selected from the above wg or wg '. By allocating a small size to the inner peripheral side, the number of pseudo defects in the next pseudo defect row can be reduced somewhat. Of course, the position of the next pseudo defect in the radial direction of the laser spot Sp is the laser spot Sp.
It is farther away than the radial width (major axis) φ2 of p. FIG. 7B shows a detection signal of the pseudo defect fg in the scanning direction of the laser spot Sp corresponding to FIG. 5B. In the drawing, the dotted portion in the radial direction of the disk 1b is a linear mark defect provided for dividing each pseudo defect row, and is formed by scribing with a cutter or the like. Note that the reference position mark Mk can also be marked on the outer periphery.

Next, the overall flow of the defect inspection process will be described with reference to FIG. First, the operator mounts the reference disk 1a on the spindle 21 of the rotating mechanism 2 (step 101), and sets M in response to a key input for starting a predetermined inspection.
The PU 42 calls the defect data collection program 43a and executes this program. Thus, a defect inspection is started (step 102). Then, the MPU 42 collects the data of the detected defect value and the position data of the defect (Step 103). Next, the MPU 42 determines whether or not the entire surface of the disk 1 has been scanned (step 104). If NO here, the process returns to the step 103. When the determination in step 104 is YES and the entire surface scanning of the disk 1 is completed, the MPU 42 determines whether or not the size inspection is to be performed (step 105). Here, if NO, then
It is determined whether or not the detection sensitivity is calibrated (step 106).
Here, if the determination is YES, the MPU 42 calls and executes the defect inspection result map display program 43b. As a result, the data of the detected defect value and the position data are read, and the state of the reference disk 1a is displayed on the display as a map (step 107). As described above, the detected defect value / defect size table 43e at this time is used.
Stores conversion data for displaying a defect state of the reference disk 1a in a form close to the actual state.

Next, the MPU 42 determines whether or not the size has been input (step 108). If the determination is NO, the process enters a loop waiting for detection sensitivity adjustment. It is determined whether the sensitivity adjustment end key has been input (step 109). When the operator operates the control panel 47 to adjust the detection sensitivity and presses the sensitivity adjustment end key, the MPU
The process of 42 returns to step 102, and the same inspection is repeated again. Thereby, for example, the reference disk 1
When the defective rows # 1 to # 12 of a are clearly identified and the optimum sensitivity adjustment is performed, the determination in step 108 as to whether or not the size is input is YES. Here, if the determination is YES, the MPU 42
Call 3c and execute. Then, the process enters a loop waiting for input of size data from the input device 48 (step 11).
0).

Here, when the operator inputs a size from the input device 48, the MPU 42 waits for a position input on the screen by a mouse (step 111). When the position of the mouse pointer is set on one of the defect columns # 1 to # 12 of the reference disk 1a corresponding to the size on the screen of the display 46 and the position is input, the MPU
42 extracts the size of the defect at the defect detection position corresponding to the input position from the collected size data (defect inspection result) (step 112), and detects the detected defect value / The position on the defect size table is searched (step 113), the size data at the position where the defect size matches in the table is rewritten with the input size data, updated, and a flag is set at this position (step 114). Thus, by referring to the defect inspection result (collected data), a new combination pair of the detected defect value and the defect size is generated by rewriting the defect size of the calibration disk corresponding to the position designation by the mouse. Then, the MPU 42 starts to determine whether or not the calibration is completed (Step 115). Here, if the determination is NO, the process returns to step 110.

By repeating the processing from step 110 to step 115, the defect size data for the pseudo defect columns # 1 to # 12 or the pseudo defect columns selected from them is detected in correspondence with the actual output signal level. The detected defect value / defect size table 43e is sequentially written into the value / defect size table 43e, and becomes the calibrated determination reference data. YE in the determination of step 115
In S, the MPU 42 deletes the old data by referring to the flag indicating the update data (Step 116). This completes the detection sensitivity calibration process. It should be noted that when the result of the previous step 106 is NO, the processing shifts to another processing.

Next, when the operator mounts the inspection disk 1 on the spindle 21 of the rotating mechanism 2 (step 101).
a) In step 102, the MPU 42 calls and executes the defect data collection program 43a to start the defect inspection. Then, YES in step 105
, The MPU 42 executes the defect size detection program 4
3d is called and executed, a size classification process is performed with reference to the updated detected defect value / defect size table 43e, and defects are counted according to the size (step 1).
17). As a result, calibrated size classification data corresponding to the level of the detection signal is obtained corresponding to the defect detection value collected by the disk defect inspection. Next, MPU42
Calls and executes the defect inspection result map display program 43b, reads out the size data and the position data, and displays the state of the inspection disk 1 on the display in a calibrated map (step 118). Of course, at this time, the count value for each defect may be displayed as necessary, and the inspection result may be output to the printer 45.

As described above, data calibration on the detected defect value / defect size table in the embodiment is an example, and the present invention is not limited to calibration using such a size conversion table. For example, defect size data can be output by a data conversion memory that outputs a defect size using a signal level value as an address value. In such a case, since the data on the memory becomes the defect size data, rewriting for updating the data may be performed. The data used as the criterion for determining the defect size is not limited to such size data, but corresponds to the size to be detected with respect to the defect detection signal in the signal processing circuit 41 of FIG. 1 or FIG. A threshold value or a threshold value corresponding to the size classified into large and small may be set. Further, the scanning of the disk by the laser spot Sp in the embodiment is not limited to the spiral, and the laser spot Sp may be moved relatively to the disk. Further, a light receiving element such as a CCD can be used as a light receiving device that receives scattered light from a defect.

Further, in the embodiment, the description has been made on the basis of a disc made of aluminum and mirror-polished.
Of course, a glass disk may be used as the reference disk. Also, the disk may be metal-plated. Further, the disk is not limited to a magnetic disk or its substrate, but can be applied to an apparatus for detecting a defect of a work by rotating the work and inspecting the work to be inspected. . As an example of what can be applied,
Wafer defect inspection (foreign matter inspection) can be mentioned.

[0045]

As can be understood from the above description, according to the present invention, a sensitivity calibration disk is subjected to defect inspection by a defect inspection apparatus using a disk for sensitivity inspection as an object of defect inspection, and detection is performed in accordance with the level of a detection signal. When the defect size is displayed as an inspection result on a display or the like in the detected value of the size of the detected defect or the classification value of the size classified according to the level of the detection signal, the defect is displayed along with the shape of the disc with an image corresponding to the size. Is displayed on the map. At this time, at least the image corresponding to the pseudo defect of a different size can be obtained, so that the sensitivity can be adjusted so as to obtain an image with an appropriate detection sensitivity while watching the obtained image. As a result, it is possible to adjust the detection sensitivity to suit the size classification of the defect at that time without any skill while watching the image on the calibration disk. Moreover, as long as the same calibration disk is used, the setting state of the defect detection sensitivity of the inspection apparatus can be reproduced, so that the inspection result data is hardly affected by the current setting state, and the variation is reduced. Of course, the inspection result data hardly varies even between different inspection apparatuses, and the inspection result data can be shared.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a magnetic disk defect inspection apparatus to which a surface defect inspection method according to the present invention is applied.

FIG. 2 is a flowchart of the defect inspection processing.

FIG. 3 is an explanatory diagram of a detected defect value / defect size table.

FIGS. 4A and 4B are specific examples of a sensitivity calibration disk used in the present invention, in which FIG. 4A is an arrangement diagram of a pseudo defect row in the sensitivity calibration disk, and FIG. FIG. 3C is an explanatory diagram of a formed square concave portion, and FIG. 4C is an explanatory diagram of a circular convex portion formed in a pseudo defect row.

FIG. 5A is an explanatory diagram of a relationship between an interval between pseudo defects and a scanning laser spot, and FIG. 5B is an explanatory diagram of a pseudo defect detection signal.

FIG. 6 is an explanatory diagram of another specific example of the sensitivity calibration disk.

FIG. 7A is an explanatory diagram of a relationship between an interval between pseudo defects and a scanning laser spot, and FIG. 7B is an explanatory diagram of a pseudo defect detection signal.

FIG. 8 is a schematic configuration diagram of a main part of a magnetic disk defect inspection apparatus, where (a) is an overall configuration diagram thereof,
(B) is an explanatory view of the laser spot.

FIG. 9 is an explanatory diagram of various defects existing on the surface of the magnetic disk.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Mirror disk, 1a ... Sensitivity calibration disk (reference disk), 2 ... Rotating mechanism, 10 ... Defect inspection device, 21 ...
Spindle, 22 motor, 3 detection optical system, 31 light projecting system, 311 laser light source, 312 focusing lens, 32
.., Light receiving system, 321, condensing lens, 322, light receiving device (AD
P etc.), 4 ... Defect detection processing unit, 41 ... Signal processing circuit, 4
1a ... Amplifier, 41b ... Sampling circuit, 41c ... A
/ D converters (A / D) 41c, 41d: position data generation circuit, 42: MPU, 43: memory, 44: data processing device, 45: printer (PR), 46: display, 43a: defect data collection program, 43b ... Defect map display program, 43c ... Criteria data generation program, 43d ... Defect size detection program, 43e
... Detection defect value / defect size table, 50 ... Data processing device, LT ... Laser beam, Sp ... Laser spot, φ1
… Spot minor axis, φ2… spot major axis, F… defect,
fg, fg ': simulated defect, FGR: group of simulated defects, wg: side length of square or rectangular simulated defects, dg: depth of simulated defects, wg': diameter of simulated defects, dg ': circular shape Simulated defect height.

Claims (7)

[Claims] 1. A laser spot is moved relative to a disk to scan the surface of the disk, a scattered light of the laser spot is received by a light receiver, and an output signal of the light receiver is amplified and detected. A surface defect inspection apparatus that obtains a signal and outputs the size of a defect detected according to the level of the detection signal or the size classified according to the level of the detection signal together with the position on the disk, A disk for calibrating the detection sensitivity, wherein a pseudo-defect row composed of three or more pseudo-defects formed along any one of a radial direction and a circumferential direction of the disk is provided on the disk for calibration. N (where n is an integer of 2 or more) are provided at every predetermined angle in the circumferential direction, and the pseudo defects in each of the pseudo defect rows are substantially equal in size. Formed as any one of a convex portion and a concave portion of the laser, and the adjacent pseudo defects are arranged at predetermined intervals larger than the width of the laser spot, and gradually increase or increase along the circumferential direction. A sensitivity calibration disk having a decreased sensitivity, performing a defect inspection of the sensitivity calibration disk, adjusting the detection sensitivity according to the inspection result, and inspecting the calibration disk in a state where the detection sensitivity is adjusted. Generates data relating the level of the detection signal and the size of the pseudo-defect as reference data for size determination, based on the level of the detection signal when performing a defect inspection on the disk to be inspected based on the determination reference data. A surface defect inspection method for obtaining the size of a defect. 2. The inspection result is a detected value of a defect size corresponding to a level of the detection signal or a classification value of a defect size classified according to the level of the detection signal, and is displayed on a display. 2. The surface defect inspection method according to 1. 3. The surface defect inspection method according to claim 1, wherein the criterion data is a threshold value corresponding to the size of the defect detected with respect to the level of the detection signal. 4. The disk to be inspected is a wafer,
2. The surface defect inspection method according to claim 1, wherein the calibration disk is obtained by forming the pseudo defect on a sample wafer. 5. A laser spot is moved relative to a disk to scan the surface of the disk, and scattered light of the laser spot is received by a light receiver, and an output signal of the light receiver is amplified and detected. A surface defect inspection apparatus that obtains a signal and outputs the size of a defect detected according to the level of the detection signal or the size classified according to the level of the detection signal together with the position on the disk, A disk for calibrating the detection sensitivity, wherein a pseudo-defect row composed of three or more pseudo-defects formed along any one of a radial direction and a circumferential direction of the disk is provided on the disk for calibration. N (where n is an integer of 2 or more) are provided at every predetermined angle in the circumferential direction, and the pseudo defects in each of the pseudo defect rows are substantially equal in size. Formed as any one of a convex portion and a concave portion of the laser, and the adjacent pseudo defects are arranged at predetermined intervals larger than the width of the laser spot, and gradually increase or increase along the circumferential direction. A decreasing sensitivity calibration disk; a sensitivity adjusting means for adjusting the detection sensitivity; and detecting the sensitivity calibration disk from a result of a defect inspection of the sensitivity calibration disk in a state where the sensitivity adjusting means is adjusted to an appropriate detection sensitivity. Reference data generating means for generating data relating a signal level and the size of the pseudo defect as reference data for size determination; andthe detection signal obtained when a target disk is inspected for defects based on the reference data. And a size detecting means for obtaining the size of the defect from the level. 6. A display and an input means for inputting a size of the defect, wherein the inspection result is classified according to a detected value of a defect size corresponding to a level of the detected signal or a level of the detected signal. A classification value of the defect size, which is displayed on the display, wherein the pseudo defect is
A plurality of marks are provided along the radial direction, one of the n pseudo defect rows is provided with a mark indicating a reference position, and the reference data generation means determines the size of the defect input by the input means. 6. The surface defect inspection apparatus according to claim 5, wherein the information is stored in association with a level of the selected detection signal. 7. The surface defect inspection apparatus according to claim 5, wherein the criterion data is a threshold value corresponding to the size of the defect detected with respect to the level of the detection signal.