TW201941325A - Bare die testing method - Google Patents

Abstract

A bare die testing method for testing the performance of the bare die. The bare die includes a substrate and a pad on the substrate. The substrate has a first outer outline, the pad has a second outer outline. The bare die testing method includes the following steps. Providing the bare die on a feeding tray. The feeding tray is located at a feeding tray transfer stage which is movable. Capturing an image of the first outer outline. Adjusting a position of the feeding tray transfer stage according to a relation between an image of the first outer outline and a base image, so that the base image and the image of the first outer outline can be matched. Transferring and locating the bare die to a testing transfer stage which is movable. Capturing an image of the second outer outline and define a coordinate of the second outer outline. Comparing the coordinate of the second outer outline to the coordinate of the probe tip to adjusting a position of the testing transfer stage, so that the coordinate of the second outer outline and the probe tip can be matched. Finally, a testing process is performed on the bare die.

Description

Bare crystal test method

The embodiment of the present invention relates to a test method for a bare crystal.

According to the continuous progress of IC manufacturing technology, the difference in IC process technology based on gate length (line width) dimensions has been changed by several micron (μm) process technology, sub-micron process technology (<1 μm), and deep sub-micron process. Technology (<0.25 μm) has even developed to nanometer (nm) process technology. From this we can see the tinyness and fineness of IC products.

In the IC manufacturing process, the front-end process includes a wafer processing (Wafer Fab) and die probing, and the back-end process includes packaging and a final test. In the bare chip probe process, after the complete wafer is cut into bare chips and before packaging is completed, each die is tested for acceptance through a probe card. The probe card is formed with the die through the probe. Conductive connection to test its electrical characteristics, thereby eliminating defective products from entering subsequent packaging processes and reducing packaging costs.

During bare chip needle testing, the probe of the probe card is directly in contact with the pad (Pad) on the die for needle testing. And because the demand for pads on bare chips is increasing, and the density is higher and higher, the size of the pads on bare chips is getting smaller and smaller. That is to say, the alignment between the probe tip of the probe card and the pad is also increased. When the offset between the probe tip and the pad is greater than expected, the probe and the pad will occur. Failure to align and pin test failure. For example, the size of the current pad edge on the bare die can be as small as 40 μm, and the size of the tip of the probe can be as small as 15-20 μm. It can be known that as long as the offset of the tip of the probe exceeds 1/2 of the length of the side of the pad, the bare crystal cannot be contacted and the needle test cannot be completed. Therefore, the present application provides a bare-die test method that can ensure the accuracy of the bare-die pin accuracy.

This application provides a bare die test method for testing the operating efficiency of the die. The bare die includes a substrate and a pad, and the pad is located on the substrate. The substrate has a first outer contour, and the pad has a second outer contour. Including: providing a bare crystal on a loading tray, the loading tray is positioned on a movable loading tray transfer platform. The image of the first outer contour is captured, and the position of the loading tray transfer stage is adjusted according to the image of the first outer contour relative to the initial positioning base image, so that the image of the first outer contour is consistent with the initial positioning base image. The die is transferred and positioned to a movable test transfer stage. Capture the image of the second outer contour and define the coordinate value of the image of the second outer contour, compare the coordinate value of the image of the second outer contour with the needle tip position coordinate value, and adjust the position of the test stage to make the second outer contour The coordinate value of the image is consistent with the coordinate value of the needle tip position. Finally, the bare crystal is tested.

In this way, through two image acquisitions and calibrations, it is ensured that the die can be accurately located at the test position for testing, and the accuracy of the needle test is improved.

Referring to FIG. 1, FIG. 1 is a method flowchart of an embodiment of a bare die test method according to the present invention, including the following steps:

Step S01: Bare crystals are provided on the loading tray. Bare chips (Bare die / Bare Chip) are semiconductor components that have undergone wafer process (Wafer) process, one by one, before entering the package stage (Package) process. . Each die includes a substrate and a solder pad, and the solder pad is located on the substrate for electrical connection with external equipment. Here, the substrate has a first outer contour, the pad has a second outer contour, and the area of the second outer contour is smaller than the area of the first outer contour. More specifically, the first outer contour of the substrate may be a square contour with a side length of 10 mm, and the second outer contour may be a square contour with a side length of 40 μm, but is not limited thereto. Here, the bare crystals are not limited to being individually arranged and arranged in the loading tray one by one, but may also be a plurality of bare crystals that are still adhered to a soft film commonly known as Blue Tape after being cut. Here, the loading tray is positioned on a movable loading tray transfer platform. When the loading tray transfer platform is displaced, the loading tray and the bareness on the loading tray can be changed simultaneously.晶 的 位置。 Crystal location.

Step S02: Capture an image of the first outer contour. In an embodiment, the first outer contour image of the bare crystal on the loading tray can be captured by a movable first imaging device. Here, the captured first outer contour image is the first Outer contour image. The first image capturing device may be a CCD (Charge Coupled Device) lens, but is not limited thereto. Further, the first image capturing device may be configured to capture images relative to the loading tray while being displaceable in the X direction. Thereby, the image capturing range of the first image capturing device is enlarged and the number of bare crystals that can be loaded on the loading tray can be increased. More specifically, the path of the first image pickup device moving in the X direction is a path that moves across the loading tray transfer stage in the Y direction perpendicular to the X direction, thereby enabling a comprehensive analysis of all bare crystals in the loading tray. Take the image.

Step S03: Adjust the position of the loading tray transfer stage according to the first outer contour image relative to the initial positioning base image, so that the first outer contour image is consistent with the initial positioning base image.

Here, please refer to FIG. 2 again. Step S03 includes:

In step S31, in an embodiment, the loading tray transfer stage performs first position compensation based on the first outer contour image relative to the initial positioning base image, so that the position of the bare crystals on the loading tray matches the initial positioning base image. The basic image for initial positioning can be stored in advance on the loading tray transfer stage, and the basic image for initial positioning can be changed according to different bare crystals. And the initial positioning base image is a preset standard image that the bare die should match before leaving the loading tray transfer stage.

Here, the method for the first position compensation of the loading tray transfer stage is as follows: the loading tray transfer stage has a basic image for initial positioning, and the loading tray transfer stage can rotate and can move along the X perpendicular to each other. Direction and Y direction linear displacement. Therefore, after the die is located on the loading tray transfer stage, the first image capturing device captures an image of the first outer contour of the die. After the first image capturing device captures the image of the first outer contour, the loading tray transfer stage compares the initial positioning basic image with the first outer contour image. When the initial positioning basic image is inconsistent with the first outer contour image, The loading tray transfer stage can be rotated or linearly displaced in the X direction and Y direction for correction, and is continuously compared and corrected until the first outer contour image is consistent with the initial positioning base image.

It is worth noting that the difference between the initial positioning base image and the first outer contour image of the loading tray transfer stage comparison is whether the comparison image has an offset in the X and Y directions, and whether the first outer contour as a whole has Angular offset. When the first outer contour image and the initial positioning base image have an offset in any of the X direction, the Y direction, or the angle, it is determined that the first outer contour image is inconsistent with the initial positioning base image and a first position compensation is required. After the loading tray transfer stage is compensated for the first position, it can be moved to the position to be transferred in the Y direction.

Furthermore, in step S31, since step S31 is to perform the first position compensation based on the first outer contour image of the bare crystal. Therefore, the unit displacement amount (ie, the minimum displacement amount) of the loading tray transfer stage of step S31 in the X direction and the Y direction is smaller than the side length of the first outer contour. When the unit displacement of the first position compensation in step S31 is smaller, the accuracy of the external position of the first outer contour image of the bare crystal can be improved. Here, the first position compensation of step S31 is the basis for improving the speed and accuracy of subsequent position correction. Therefore, it is not necessary to position the position of the bare crystal to the position of the last test in this step.

In step S32, the bare wafer is transferred to the loading position by the loading tray transfer stage. In an embodiment, the loading tray transfer stage can be displaced in the Y direction, and the loading tray transfer stage can be displaced in the Y direction to change the position of the bare crystal in the Y direction to the position to be transferred to continue subsequent actions. It is worth noting that, here, the sequence of steps S31 and S32 in FIG. 2 is only for illustration but not limited to this. The sequence of steps S31 and S32 can be replaced, that is, the loading tray is transferred. The stage may first transfer the die to the position to be transferred, and then perform the first position compensation before the die is successively transferred; it may also perform the first position compensation before transferring to the position to be transferred. As long as the first position compensation is completed before the die is transferred from the loading tray, the first outer contour image of the die conforms to the initial positioning base image. Further, in step S32, the unit displacement amount of the loading tray transfer stage in the Y direction may be greater than the unit displacement amount in step S31. Thereby, the bare crystal is quickly transferred to the position to be transferred.

Step S04: transfer the bare crystal and position it to a movable test transfer stage. In one embodiment, the first pick-and-place device transfers the die on the loading tray to a movable test transfer stage, and the die is positioned on the test transfer stage. The first pick-and-place device can be displaced in the X direction, and the path where the first pick-and-place device is displaced in the X direction spans the area where the loading tray transfer platform is distributed and the area where the test transfer platform is distributed. Thereby, the first pick-and-place device is moved along the X direction to the loading tray transfer stage to transfer the bare crystals to the test transfer stage.

In an embodiment, the test transfer stage transfers the bare crystals to the test transfer stage after the first pick-and-place device, and then sucks the bare crystals by vacuum suction to position the bare crystals on the test transfer stage. After the die is tested, the suction is stopped and the die is ejected, so that the die can be transferred to the classification area after the test. Further, the way in which the test transfer platform sucks the bare crystals is to suck the bare crystals through airflow and apply negative pressure, and when ejecting the bare crystals, the positive pressure is applied to the ejection mechanism through the airflow to push the bare crystals.

In addition, since a plurality of bare crystals can be carried on the loading carrier at one time, step S02 is to take an image of one of the bare crystals on the loading carrier. Therefore, in step S02, when the first image capturing device completes capturing one of the bare crystals and continues to perform steps S31 and S32, at the same time, the bare crystals of step S32 are completed to complete the subsequent testing process. The first imaging device can continue to repeat step S02 and subsequent procedures for other bare crystals on the loading tray. In this way, when a plurality of bare crystals are loaded on the loading tray for testing, there is no need to wait for one bare crystal to complete the entire test process and then restart the test process for the next bare crystal. The test process of most bare crystals You can perform different steps at the same time to improve testing efficiency.

In another embodiment, multiple bare crystals (for example, four bare crystals arranged in a 2x2 matrix state) can also be placed on the test transfer stage at the same time. Here, the first pick-and-place device can also pick up a plurality of bare crystals (the arrangement pitch corresponds to the bare crystal pitch on the test transfer stage) after completing the first position compensation of the loading tray, and transfer it to the test transfer stage. on. Thereby, the first pick-and-place device can simultaneously place a plurality of bare crystals on the test transfer stage, thereby improving pick-and-place efficiency.

Further, the testing area includes a testing machine, the testing machine has a probe card, and the probe card has a probe, and the testing machine in the testing area contacts the bare die pads through the probe of the probe card to form an electrical connection so as to bare The electrical properties of the crystal are tested. In an embodiment, after the die is transferred to a position corresponding to the probe card, the test transfer stage is raised for testing. Therefore, before the test, the bare chip and the probe of the probe card have a height distance perpendicular to the X direction and the Z direction of the Y direction, and the position of the pad must be aligned with the probe.

Step S05: Capture an image of the second outer contour. In an embodiment, the second image capturing device captures an image of the second outer contour of the die pad on the test stage, and the second image capturing device further defines a second image based on the image of the second outer contour. The coordinate value of the outer contour image (this is the coordinate value of the pad position).

Here, the position of the bare crystal after the first position compensation is within the range of the second imaging device, and the image of the second outer contour captured by the second imaging device is the second outer contour image. Furthermore, the second outer contour image is substantially a part of the first outer contour image. That is, since the imaging range of the second outer contour image is further limited to the acquisition range of the first outer contour image, Therefore, the first outer contour image substantially includes the second outer contour image.

The second image capturing device may be a CCD (Charge Coupled Device) lens, but is not limited thereto. Further, the second imaging device can be set at a fixed position, and the test transfer stage can be translated in the X, Y, and Z directions that are perpendicular to each other, and can be rotated on a plane formed by the X and Y directions. The test stage can be displaced relative to the second imaging device in all directions, thereby ensuring that the second imaging device has a sufficient imaging range. More specifically, each test area has an independent second imaging device that operates independently, and the relative position between the probe card and the second imaging device is fixed after installation and imaging correction, so it can be As much as possible, reduce the parameter variation caused by the abnormal movement between the devices, thereby improving the accuracy of subsequent position compensation.

Further, since the area of the first outer contour is larger than the area of the second outer contour, the magnification of capturing the first outer contour image may be smaller than the magnification of capturing the second outer contour image. Further, since the first image capturing device captures an image of a first outer contour of the bare crystal substrate, and the second image capturing device captures an image of a second outer contour of a solder pad on the bare crystal substrate. Therefore, the magnification of the second imaging device is larger than that of the first imaging device. Based on this, the second outer contour image is a part of the first outer contour image having a higher resolution than the first outer contour image.

Here, since the bare crystal of the present application acquires the second outer contour image only after the first position compensation, the bare crystal can be ensured that the bare crystal can be located in the range of the captured image after step S05. Inside. Therefore, step S05 can be performed immediately after the bare crystal is transferred to the test transfer stage, so that step S05 can perform image acquisition more quickly and correctly.

Step S06: Adjust the position of the test transfer stage according to the coordinate value of the second outer contour image relative to the position of the needle point, so that the coordinate value of the second outer contour image is consistent with the coordinate value of the position of the needle point. In an embodiment, the test transfer stage may be a coordinate value of a needle tip position of a probe of a memory probe card, and the test transfer stage may access a coordinate value of a second outer contour image defined by a second imaging device. . Here, the coordinate origin of the needle tip position is the same as the coordinate origin of the coordinate value of the second outer contour image. In this way, the test transfer stage can perform the second position compensation based on the coordinate value of the needle tip position, so that the coordinate value of the second outer contour image of the die is consistent with the coordinate value of the needle tip position. Furthermore, the test transfer stage is rotatable and can be linearly displaced in the X and Y directions which are perpendicular to each other. Therefore, after the die is transferred to the test transfer stage by the first pick-and-place device and the coordinates of the second outer contour image are acquired, the test transfer stage compares the coordinate value of the needle tip position with the coordinate value of the second outer contour image. When the coordinate value of the needle tip position is inconsistent with the coordinate value of the second outer contour image, the test transfer table can be rotated, linearly displaced in the X direction and the Y direction for correction, and continuously compared to the coordinate of the second outer contour image The value is consistent with the coordinate value of the needle tip position.

Further, referring to FIG. 3, the coordinates of the aforementioned needle tip position may be generated through step S61 depending on the specifications of the probe card. Step S61 is performed before the initial test of the probe cards of different specifications. For example, before the first test after changing the probe card. In step S61, a needle tip image of the probe card is captured. In one embodiment, a third image capturing device is used to capture the needle tip image of the probe card of the tester's probe card, and a needle tip position coordinate value is defined according to the needle tip image. Therefore, the needle tip position can be defined through step S61 before each time the probe card is replaced for testing, and subsequent position compensation can be directly adjusted based on the needle tip position. In this way, the case can be applied to different die or probe card specifications, and precise alignment is not required when installing the probe card, which can reduce the difficulty and time of installing the probe card, and completely eliminate manual installation The error caused by the needle card can improve the installation efficiency of the probe card and the accuracy of the needle measurement.

Here, the premise of comparing the coordinate value of the second outer contour image with the coordinate value of the needle tip position is that the reference origins of the two must be the same. Therefore, before the coordinate values defined by the second image capturing device and the third image capturing device are defined, if the origin of the reference coordinates of the two are different, the calibration step may be performed again to obtain the second image capturing device and the third image capturing device. The reference origin of the image is the same. In this way, after defining the coordinate value of the second outer contour image and the coordinate value of the needle tip position, the coordinate value of the second outer contour image and the coordinate value of the needle tip position can be directly compared. When the coordinate value of the second outer contour image is inconsistent with the needle-to-needle position coordinate value, it is judged that the position of the welding pad is inconsistent with the position of the needle tip, and the position must be compensated by the test transfer stage.

In addition, in step S06, as the size of the die is getting smaller and smaller, and the electronic device to which the die is applied requires more and more functions, the number of pads on the die is of course correspondingly increased, which leads to the bare die. The size of the pads on the wafer is getting smaller and smaller, and the configuration density is getting higher and higher. In this way, the alignment accuracy requirements between the die and the probe card are relatively increased. Once the positioning accuracy of the die is not as expected, the probe card cannot be contacted with the pad of the die during needle testing. Electrical connection cannot be formed, and a dilemma cannot be detected. Therefore, the unit displacement of the test transfer stage is directly related to the positioning accuracy of the die. Here, the unit displacement of the test transfer stage is in the order of μm.

Further, since the test machine performs a pin test by contacting the bare chip with the probe card by raising the test transfer table, the probe must withstand stress every time the lower pin contacts the bare chip. In addition, after the probe contacts the die, further pressure must be applied to cause the probe to generate a proper vertical displacement in the direction of the vertical die (needle stroke, Overdrive), so that the probe is relatively Bare die pads generate appropriate pin pressure, so that the probe can scratch the oxide layer on the pads to form a conductive connection.

Based on this, the stress on the probe during each probe test will cause wear and affect its life. Therefore, if you can grasp the stress that the probe is subjected to each test, you can also relatively grasp the life of the probe. For this reason, the positions of the bare crystals in the Z direction can be further considered so that each of the bare crystals is located at the same position in the Z direction when the test is performed. In order to ensure that the probe can be subjected to the same contact needle pressure when the needle is lowered and the needle test is performed, the contact needle pressure is related to the contact resistance during the needle test. In this way, the contact pin Under the same conditions, the accuracy of needle measurement can be further improved, and the wear and life of the probe can be further mastered.

Here, the test transfer stage can be further displaced in the Z direction to perform position compensation in the Z direction. Here, the basis of the position compensation of the test transfer stage in the Z direction is customized based on the lower needle distance of the probe card and the needle measurement stroke. The test stage performs position compensation in the Z direction so that each bare crystal can have the same distance from the probe tip after position compensation.

Step S07: Test the die. In one embodiment, the test transfer stage transfers the die to the test area, and the probe card in the test area tests the die. It can be known from the above description that after the bare crystal is compensated for the first position and the second position, the bare crystal can be adjusted to the same position as the standard sample, and the electrical test can be completed on the basis of accurate alignment.

Step S08: classify the bare crystals after the test. In one embodiment, after the die is tested, the test transfer stage transfers the die away from the test area. The second pick-and-place device removes the bare crystals after the test and places them into the classification area. Further, the bare crystals are classified into different collecting trays in the classification area according to the test results (good, defective, and performance differences), thereby completing the testing and classification of the bare crystals.

Based on the above, this case uses a first image based on the first image of the bare crystal to perform a large-area position calibration, so that the bare crystal can be initially adjusted close to the test position and determined to enter the range of the second image acquisition before testing. Here, the second image acquisition is based on the probe card's tip position to perform a finer position calibration directly than the die pad position, so that the die can be adjusted to an accurate test no matter where the material is fed. Complete the test at the location to improve the effectiveness of the test process.

Further, in this case, it is not necessary to directly contact the die during the first position calibration or the second position calibration, but to synchronize the position of the loading tray transfer stage and the test transfer stage by changing the positions of the loading tray transfer stage and the test transfer stage. Changing the position of the die for calibration can reduce the chance of directly contacting the die during the entire test process, which can significantly reduce the damage of the die and improve the product yield.

Although the present invention is disclosed in the foregoing embodiments as above, it is not intended to limit the present invention. Anyone skilled in the similar arts can make some changes and retouch without departing from the spirit and scope of the present invention. The scope of patent protection shall be determined by the scope of the patent application attached to this specification.

S01 ~ S08‧‧‧step

S31 ~ S32‧‧‧step

S61‧‧‧step

FIG. 1 is a method flowchart of an embodiment of a bare die testing method according to the present invention. FIG. 2 is a method flowchart of steps S31 and S32 in the bare die test method of the present invention. FIG. 3 is a schematic diagram of step S61 in a bare crystal test method of the present invention.

Claims (8)

A bare crystal test method is used to test the operation efficiency of a bare crystal. The bare crystal includes a substrate and a plurality of solder pads, the plurality of solder pads are located on the substrate, the substrate has a first outer contour, and the solder pad has a The second outer contour includes: providing the bare crystal on a feeding carrier, the feeding carrier positioned on a movable loading carrier transfer stage; capturing an image of the first outer contour; according to the The position of the first outer contour image is adjusted relative to the initial positioning base image to adjust the position of the loading tray transfer platform, so that the first outer contour image is consistent with the initial positioning base image; A moving test stage; capturing the image of the second outer contour and defining the coordinate value of the image of the second outer contour; adjusting the coordinate value of the image of the second outer contour relative to the coordinate value of a needle tip position The position of the test transfer stage so that the coordinate value of the image of the second outer contour is consistent with the coordinate value of the position of the needle tip; and testing the bare crystal. The die test method according to claim 1, wherein the test of the die is performed by contacting the pad with a probe card having a plurality of probes. Before the test, the images of the tip of the plurality of probes of the probe card are captured and the coordinate value is defined as the coordinate value of the position of the tip. The method for testing a bare crystal according to claim 1, wherein the number of the bare crystals is a plural number, and when one of the plurality of bare crystals is tested, the other die of the plurality of bare crystals is retrieved. An outline image. The bare die test method according to claim 1, wherein the magnification of the image capturing the first outer contour is smaller than the magnification of capturing the image of the second outer contour. The bare crystal test method according to claim 1, wherein the loading tray transfer table is rotatable and can be linearly adjusted to adjust the position along an X direction and a Y direction perpendicular to each other. The bare die test method according to claim 1, wherein the test transfer stage can be translated along an X direction, a Y direction, and a Z direction that are perpendicular to each other, and can be in a plane formed by the X direction and the Y direction Turn on. The bare crystal test method according to claim 1, wherein the test transfer stage locates the bare crystal by means of vacuum suction. The bare crystal test method according to claim 7, wherein after the bare crystal is tested, the test transfer stage stops vacuum attraction and ejects the bare crystal.