JP2726651B2 - Marking method for defective element - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for marking a defective element determined as defective after inspecting the electrical characteristics of the element to be measured with a probe needle. It relates to a marking method. 2. Description of the Related Art In a probe apparatus, for example, a probe apparatus for a semiconductor wafer, a probe needle is brought into contact with an electrode pad of the element in order to measure electrical characteristics of a large number of elements on the wafer. Is connected to a tester to perform measurement. Usually, the wafer is supported on a mounting table by a vacuum chuck or the like, and the mounting table is controlled to move in the θ direction, which is the circumferential direction of the X, Y-axis and Z-axis, so that the electrode of the element is controlled. The pads are aligned with the probe needles, and the probe needles are brought into contact with the electrode pads by mounting the wafer and raising the mounting table. [0004] Further, based on the measurement results of the electrical characteristics using a probe needle and a tester, marking is performed on defective elements by a scratch marker method, an inker method, or the like. Here, the above-described positioning of the electrode pad and the probe needle is important in performing a measurement with a very small measurement error, and automation has been strongly demanded in order to perform the measurement with high accuracy. . Therefore, proposals have been made to automate this kind of positioning without relying on visual observation of an operator (Japanese Patent Application Laid-Open Nos. 58-169922, 59-5641, 60-24029). Kaisho 60
-24030, JP-A-61-15340, JP-B-62-16018). [0007] The present applicant has also made proposals for automating the above-mentioned alignment (Japanese Patent Application No. 60-69554, Japanese Patent Application No.
62-106027). [0008] In the above-mentioned conventional proposals, all of the above-mentioned proposals only perform alignment of the mounting table based on the traces of the probe needles, and only perform alignment of the probe needles and the electrode pads. Was something. In addition to the above-described positioning, it is also important to adjust the contact pressure of the probe needles. If the contact pressure is too high, the life of the probe needles is shortened, or the wafer and the probe needles are damaged. There is. On the other hand, if the contact pressure is too small, an oxide film is usually formed on the wafer, so that the oxide film cannot be broken through by a probe needle, resulting in electrical contact failure and reliable measurement. Cannot be executed. Furthermore, even if accurate measurements can be made, if the above-mentioned marking on the defective element is not appropriate, it is not possible to separate non-defective products from non-defective ones after separating the devices, resulting in wasteful measurement. Become. Therefore, an object of the present invention is to solve the above-mentioned conventional problems and to provide a new method for adjusting the position of the probe needle and adjusting the contact pressure, and to provide a new method for marking at the time of marking. An object of the present invention is to provide a marking method for a defective element which can easily perform the size and the position of the marking for the defective element by applying to the adjustment of the defect. According to the present invention, a device to be measured on a substrate to be measured is measured, and marking is performed by a marking device on a defective device determined to be defective based on the measurement result. in marking method to a defective element, the following
It has a process of. First, the substrate to be measured is placed on a two-dimensional plane.
X and Y axis directions, which are the rectangular coordinates of
Movable in the intersecting Z axis and the θ direction around the Z axis
Placed on a mounting table, the marking means and the mounting table
Is set to the appropriate position set in advance
And mark the defective element by the marking means.
To run. Next, image the marking mark on the defective element
Image by means. In addition, the mark
Position of the previously stored marking marks when
Location information and appropriate size information, and the image of the markin
Compare with the information of the mark. Based on the comparison result, the following automatic
A correction process is performed. For the appropriate position information,
If the location of the markings is different, X,
The movement in the Y-axis direction and the θ direction causes the defective element and the mask to move.
Automatic correction of the relative position on the two-dimensional plane with the working means
You. Further, the markin information may be used for the appropriate size information.
If the size of the mark is different, the Z axis
Movement in a direction perpendicular to the two-dimensional plane.
The relative height position between the defective element and the marking means
Automatically correct. The method of the present invention is characterized in that the marking position and size when a defective element on the substrate to be measured is marked are corrected by detecting the marking mark. For example, when the relative positional relationship between the defective element and the marking means is set to an appropriate position in the X, Y, Z axis directions and the θ direction, the position and size of the marking mark are stored in advance. By comparing the mark mark information which is the proper value with the position and size information of the mark mark recognized by the imaging means, correction information of the alignment position and correction information of the position in the Z direction are obtained. be able to. Incidentally, the correction of the marking position can be performed by correcting the position of the defective element and the marking means on the two-dimensional plane, and the position may be corrected in the X, Y axis directions and the θ direction. On the other hand, the size of the marking mark is determined by the stopping position of the mounting table in the Z direction, that is, the height between the defective element and the marking means, and is determined by the correction information based on the size of the marking mark. By correcting the stop position in the direction, an appropriate size of the marking can be obtained. In addition, the height of the Z direction in which the switch is electrically turned ON / OFF is obtained, and by using this information together with the information, a more reliable marking size can be obtained. As described above, in the method of the present invention, a new method of adjusting the contact pressure and the alignment position from the trace of the probe needle is applied to the correction of the marking position and size. The position of the marking can be adjusted by correcting the position of the mounting table on the two-dimensional surface. The size of the marking can be adjusted by adjusting the position of the mounting table in the Z direction in both the scratch marker method and the inker method. The positioning and control operations are exactly the same.The imaging means takes an image of the marking mark, compares it with the appropriate marking mark information, and then controls the position of the mounting table etc. The position and size of the marking can be automatically corrected together with the positioning of the needle. Incidentally, such a correcting operation may be performed during the measurement of the substrate to be measured, or the position of the mounting table or the like may be corrected in advance by a dummy substrate for confirmation before the measurement. good. First, a method for probing a semiconductor wafer in a process preceding the method of the present invention will be specifically described with reference to the drawings. In this probing method, the alignment position and the overdrive amount of the mounting table are corrected and controlled based on the position and size of the needle mark with respect to the pad area of the device to be measured. First, the probe device will be described.
The wafer 1 is supported on a mounting table 2. The mounting table 2 has X and Y axes as orthogonal coordinate axes on the surface of the mounting table, and Z in the vertical direction.
Axis, if the circumferential direction of this Z axis is the θ direction, X, Y,
It is movable in the Z axis direction and the θ direction. The mounting table 2 is supported by a slide portion 12 which is movable along a bolt shaft 11 which is driven to rotate by a motor 10. The mounting position is indicated by a solid line in FIG. It is movable over the measurement position indicated by. At the measurement position, the probe needle 3 is attached to the probe card 4, and the ring insert 5 of the probe card 4 is provided in the inner hole of the support ring 6. The ring insert 5 and the support ring 6 are rotatable, and a handle 5 a provided on the outer peripheral portion of the ring insert 5 is linked with the peripheral rotation mechanism 20. Above the measurement position, a television camera 31 composed of an image pickup means, for example, an ITV is provided.
The needle mark added to the electrode pad of the chip of the wafer 1 is imaged. The television signal from the television camera 31 is input to a needle mark discriminator 32, where it is compared with a predetermined threshold level to determine the presence or absence of a needle mark.
Is stored at each address. The television camera 31, the needle mark discriminator 32, the CPU 33, and the RAM
Reference numeral 34 denotes an example of the configuration of the recognition unit 30. Further, there is provided control means 35 for correcting and controlling the relative positional relationship between the pad electrodes of the wafer 1 and the probe needles 3 and the overdrive amount of the mounting table 2 based on the recognition result of the recognition means 30. ing. The control means 35 comprises the CPU 33 and a ROM 36 storing various types of calculation information. And the R
The drive amount is calculated by the CPU 33 based on the information stored in the OM 36, and controls the drive in the X, Y, Z axis directions and the θ direction described above. Next, the operation will be described. First, the initial alignment when the probe needle 3 is replaced will be described. The mounting table 2 on which, for example, an aluminum vapor-deposited substrate on which an initial tip for confirming a needle mark is formed is raised in the Z direction until the probe needle 3 contacts, and then lowered. The initial tip to which the probe needle mark is added is moved by the motor 10 by a predetermined distance L together with the mounting table 2 and set at the measurement position. Then, at the measurement position, the television camera 3
By 1, the relationship between the position of the initial chip and the needle mark added to the initial chip is imaged. That is, recognition of the position of the needle mark with respect to the initial chip can be executed by accessing the RAM 34 which stores the output of the needle mark discriminator 32 for each address. That is, as shown in FIG.
In the area of the electrode pads 40a and 40b on the
1a and 41b are stored (usually, this kind of needle trace has an elliptical shape, but is schematically illustrated in a rectangular shape for convenience of explanation). First, the recognition of the needle trace will be described with reference to the rows and columns shown in FIG. 2. From the first row, scanning is performed as shown by arrow 42 to search for the last row. And
"1" is confirmed in the n-th line where the needle trace exists, the line number is stored, and then the line address where "1" exists is similarly searched. Similarly, by scanning the column side,
Search for the column address where the needle trace exists. As described above, the area where the needle mark exists can be confirmed from the binarized image data. Then, the outer shapes of the needle marks 41a and 41b can be recognized from the address where the aforementioned "1" exists. Then, from this outer shape, the needle trace centers R1 and R2 shown in FIG. 3 can be obtained by calculation. This needle trace center R1,
If R2 is determined, the known electrode pads 40a, 40b
Of the center Q, P (stored in the ROM 36)
Of the needle trace centers R1, R2 can be calculated for each of the X and Y axis directions.
Can be obtained in the X and Y directions. Next, the amount of rotation is calculated. Here, two pads 40a and 40b of the chip shown in FIG.
In the above, the distance OP between the centers of the pads and the shape of the chip are set values in the ROM 36. And
A straight line passing through the needle trace centers R1 and R2 intersects a straight line OP representing the distance between the centers of the pads. Therefore, as shown in FIG. 3, the intersection of the two straight lines is Q, and the distance of the X component and the distance of the Y component between R1 and R2 can be recognized by the recognizing means 30. It is possible to calculate the distance between R2. Therefore, the inclination α of the two straight lines can be calculated, and by rotating the ring insert 5 to which the probe needle 3 is fixed by driving the peripheral rotation mechanism 20, the alignment is corrected by α degrees in the θ direction. Thus, the correction of the rotation direction can be performed. This correction may be performed by rotating the mounting table 2 in the θ direction. As described above, the relative positional relationship between the electrode pad and the probe needle 3 can be automatically executed in the X, Y and θ directions. Next, the correction of the contact pressure based on the needle trace will be described. This contact pressure is reflected on the size of the needle marks 40a and 40b. Therefore, the size of the needle trace can be recognized by calculating the area of the entire area of the address where "1" where the needle trace is present. Then, by comparing with the predetermined size information stored in the ROM 36 in advance, it is possible to recognize how much larger or smaller than the reference size. The correction of the contact pressure can be executed by changing the stop position of the mounting table 2 in the Z direction. That is, the contact pressure of the probe needle 3 is determined by the overdrive amount of the mounting table 2 after the probe needle 3 contacts the chip. If the value is smaller than the standard, it is necessary to correct the overdrive amount to be large. In addition, a height in the Z direction at which electrical ON / OFF is performed is obtained, and by using this information together with the information, a more reliable contact pressure can be obtained. As described above, the positioning and the automatic correction of the contact pressure of the probe needle can be executed. It should be noted that such a correcting operation is performed when the probe mark confirmation substrate is actually mounted and the probe inspection is performed when the semiconductor wafer 1 is actually mounted on the mounting table 2. You can also In this case, it is troublesome to execute the above-mentioned correction every time the electrical characteristic inspection of each chip is completed. Therefore, this kind of probe apparatus has an inspection function for detecting a needle mark defect. For example, when a needle mark defect is determined by this inspection function (for example, five consecutive chips are determined to be defective chips). The above-described correction operation may be performed only in the above case. This makes it possible to effectively execute the alignment correction and the adjustment of the contact pressure during the execution of the probe inspection. Regarding the position correction in the X and Y axis directions θ described above, Japanese Patent Application No. 62-106 previously proposed by the present applicant.
027 may be employed. Next, an embodiment method of the present invention applied to the adjustment of the position of the marking and the adjustment of the size of the marking by the same procedure as the alignment and the correction of the contact pressure in the above-described probing method will be described. The marking of the defective element is performed by, for example, a scratch marker shown in FIG. This scratch marker is provided at one end of the holder 50 with a solenoid 51.
, And a sleeve guide 52 is supported at the other end. The solenoid 51 supports the piston 53 so that it can be inserted and removed by energization. On the other hand, a scratch needle 54 is supported and seated on the sleeve guide 52, and the movement of the piston 53 when the solenoid 51 is energized causes the scratch needle 54 to move to the sleeve guide 5.
It is configured to protrude further than 2 and perform marking by contacting the chip. Here, also in the case of the above-mentioned marking, it is necessary to align the position, as in the case of the probe needle 3 described above. For example, as shown in FIG. It is necessary to prevent that situation.
The marking is read by a television camera or the like after the probe inspection and is used for judging the quality of the chip. It is necessary to change the size of the marking according to the size of the chip. That is, as shown in FIG. 5 (b), an excessively large marking for the chip area is not required, but a marking which is too small for the chip area as shown in FIG. 5 (c) is required. There is a problem that pattern recognition is overlooked. Therefore, if the chip on which the above-mentioned marking has been performed is moved to the above-described measuring position, the marking mark is recognized by the television camera 31 or the like as the imaging means, and the position and the contact pressure are similarly corrected. The adjustment work that has been manually performed conventionally can be automated, and the burden on the operator can be greatly reduced. Also, this kind of marking is sometimes performed by a method called an inker method, which discharges ink in an ink fountain by moving a piston by energizing a solenoid. The positioning is the same as that of the scratch method, and the correction of the size of the marking can be corrected by the distance from the ink discharge end to the chip, that is, the moving position of the mounting table 2 in the Z direction. The present invention is not limited to the above embodiment, and various modifications can be made within the scope of the present invention. As described above, according to the present invention, the position and size of the marking are corrected based on the information of the marking mark, so that it is possible to appropriately select the non-defective product and the defective product later. Such marking marks can be accurately written on defective elements. [0056]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic explanatory view for explaining an embodiment in which the present invention is applied to a semiconductor wafer probe device. FIG. 2 is a schematic explanatory diagram for explaining an operation of detecting a positional shift of a needle mark in X, Y-axis directions and θ directions. FIG. 3 is a schematic explanatory diagram for explaining an operation of detecting a positional shift of a needle mark in the X, Y axis directions and the θ direction. FIG. 4 is a schematic explanatory diagram of a scratch marker which is an example of a marking unit. FIG. 5 is a schematic explanatory diagram for explaining a positional deviation and a difference in size of a marking. [Description of Signs] 1 substrate to be measured 2 mounting table 31 imaging means 50 to 54 marking means

Claims (1)

(57) [Claims] A method for marking a defective element, which measures a device to be measured on a substrate to be measured and determines a defective element determined to be defective based on the measurement result by a marking means, comprising: X, Y which are the rectangular coordinates of
An axial direction, a Z axis orthogonal to the two-dimensional plane, and a mounting table movable in the θ direction around the Z axis, and the relative position between the marking means and the mounting table is previously determined. A step of marking the defective element by the marking means in a state where the mark is set at the set appropriate position; a step of imaging a marking mark on the defective element by an image pickup means; Comparing the pre-stored proper position information and proper size information of the marking mark with information of the imaged marking mark, and when the position of the marking mark is different from the proper position information. Automatically correcting the relative position of the defective element and the marking means on a two-dimensional plane by moving the mounting table in the X, Y-axis directions, and θ directions; On the other hand, when the size of the marking mark is different, the relative height position between the defective element and the marking means in a direction orthogonal to the two-dimensional plane is determined by moving the mounting table in the Z-axis direction. A method for marking a defective element, the method comprising: automatically correcting. 2. 2. The method according to claim 1, wherein the correction is performed when a defect in the size or position of the marking mark on the defective element occurs continuously over a plurality of elements. 3. . 3. A substrate for marking trace information confirmation is placed on the mounting table, and the relative position correction on the two-dimensional surface and the correction of the position in the Z-axis direction are performed in advance before the measurement of the substrate to be measured. The mark for defective elements according to claim 1, characterized in that:
King way .